Citation
Worksite health promotion-disease prevention

Material Information

Title:
Worksite health promotion-disease prevention a study in the diffusion of innovation
Creator:
Davis, Mary Foecke
Place of Publication:
Denver, Colo.
Publisher:
University of Colorado Denver
Publication Date:
Language:
English
Physical Description:
xvi, 222 leaves : illustrations ; 29 cm

Thesis/Dissertation Information

Degree:
Doctor of Public Administration
Degree Grantor:
University of Colorado Denver
Degree Divisions:
School of Public Affairs, CU Denver
Degree Disciplines:
Public Administration
Committee Chair:
Tynan, Eileen A.
Committee Members:
Iverson, Donald C.
Vernon, Thomas M., Jr.
Parr, John D.

Subjects

Subjects / Keywords:
Occupational health services ( lcsh )
Occupational health services ( fast )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Bibliography:
Includes bibliographical references (leaves 165-173).
General Note:
Submitted in partial fulfillment of the requirements for the degree, Doctor of Public Administration, Graduate School of Public Affairs.
Statement of Responsibility:
by Mary Foecke Davis.

Record Information

Source Institution:
University of Colorado Denver
Holding Location:
Auraria Library
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
16855523 ( OCLC )
ocm16855523
Classification:
LD1190.P86 1985d .D38 ( lcc )

Downloads

This item has the following downloads:


Full Text
WORKSITE HEALTH PROMOTION/DISEASE PREVENTION:
A STUDY IN THE DIFFUSION OF INNOVATION
by
Mary Foecke Davis
B.A., University of Nebraska, 1965
M.P.H., University of Minnesota, 1969
A thesis submitted to the
Faculty of the Graduate School of Public Affairs of the
University of Colorado in partial fulfillment
of the requirements for the degree of
Doctor of Public Administration
Graduate School of Public Affairs
hu
1985


Copyright by Mary Foecke Davis 1985
All Rights Reserved


This thesis for the Doctor of Public Administration
degree by
Mary Foecke Davis
has been approved for the
Graduate School
of Public Affairs
by
Eileen A. Tynan
Donald C. Iverson
Thomas M. Vernon, Jr.
Date
John D. Parr


Davis, Mary Foecke (D.P.A., Public Administration)
Worksite Health Promotion/Disease Prevention: A Study in the
Diffusion of Innovation
Thesis directed by Assistant Professor Eileen A. Tynan
The purpose of this study was to examine the process and
the variables related to the adoption of innovative health programs
by organizations. The study utilized diffusion theory as the
conceptual framework and focused on the adoption/non-adoption of
worksite health promotion/disease prevention (hp/dp) programs.
Twelve companies which had adopted hp/dp programs (adopters)
were studied along with 12 companies which did not adopt such pro-
grams (non-adopters). Personal interviews, following a guided
interview format, were conducted with a total of 58 persons. The
study explored conditions, attitudes and practices existing in
companies at the time of the decision to adopt or not adopt an
hp/dp program. Selected hypotheses suggested by diffusion theory
were tested. Information from the interviews was coded and analyzed
using discriminant analysis.
A number of variables suggested by diffusion theory predicted
adoption/non-adoption of hp/dp programs. Explanatory variables
derived from the rational/economic, behavioral, and systems perspec-
tives of diffusion.
Programs were more likely to develop in response to a
perceived opportunity than in response to a perceived threat. A


V
convergence of internal and external factors resulted in initial
interest in hp/dp.
Finances, employee benefits, and company philosophy vis-a-vis
responsibility for employees were common criteria used in the
decision process. When finances became the major criterion, it
was unlikely that a company would adopt a program.
Programs were likely to occur only in companies where
management was actively supportive, had a humanistic philosophy, and
espoused expanded responsibility for employees.
The majority of variables positively associated with hp/dp
programs are likely to change only over a long period of time.
Exposure to model programs, however, offers a mechanism for inter-
vention.
These findings suggest that proponents of hp/dp can effec-
tively identify companies likely to adopt programs and develop
effective policy and programmatic efforts. Hypotheses were identi-
fied for further study.
The form and content of this abstract are approved. I recommend its
publication.
Signed ___________________________________________
Faculty member in charge of thesis


vi
ACKNOWLEDGMENTS
I would like to acknowledge the assistance of the many
people without whom this study would not have been possible:
Eileen Tynan, who has given me steady support and sound advice
throughout the doctoral program; Don Iverson, who has guided this
project as well as my professional career, Tom Vernon, whose
inquiring mind and commitment to prevention have provided inspira-
tion; John Parr, whose practical perspective kept the study on
track; Karen Rosenberg, whose diligent efforts on the initial
survey made this follow-up study possible; Ed Fryes who made
discriminant analysis look easy; Gregory Christenson, who
critiques the methodology and instruments; and Toby Cohen, of
A&D Copy & Typing Service, who took a scribbled manuscript and
made it a beautiful finished product.
I would like to thank the persons interviewed in the 24
companies who willingly shared their time and experiences. They
made this project a pleasant learning experience for me.
Lastly, I would like to thank my family for the support
they have given me and the sacrifices they have made: Mike, who
assumed more than his share of housework and hassles over the past
months; Andrea, whose chats broke the monotony of writing; Nena,
whose hugs gave me energy to continue; and Molly, who had to give
up Saturday morning cartoons in the study, and many bedtime stories.


CONTENTS
CHAPTER I
INTRODUCTION ... ................................................. 1
Rationale for the Study......................................... 1
Worksite Health Promotion/
Disease Prevention .......................................... 4
Historical Development ...................................... 4
Current Program Parameters . ............................... 11
Current Level of Program Activity ............................ 15
Frequency of Program Components.............................. 18
Rate of Growth................................................ 21
Program Outcomes............................................. 24
Hypertension ............................................... 24
Smoking cessation......................................... 26
Employee assistance program ............................... 28
Exercise/fitness ........................................... 30
Summary......................................................... 33
CHAPTER II
REVIEW OF THE LITERATURE.......................................... 35
Historical Development.............................; . . 35
Definitions..................................................... 38
Diffusion Models and Perspectives .............................. 42
Rational/Ecnonomic Perspective ............................... 42
Economic model............................................. 42


viii
Structural model .......................................... 44
Behavioral Perspective ...................................... 46
Communications model ...................................... 46
Source................................................. 47
Attitudes............................................... 47
Channel of communication................................. 48
Characteristics of adopters ............................. 48
Champion model ............................................ 51
Systems Perspective ..................................... . 53
Marketing model ........................................... 53
Contextual model .......................................... 54
Developing a Framework for Prevention ......................... 56
Framework for Current Study ................................... 59
Unit of Analysis............................................. 59
Stage........................................................ 60
Diffusion Models........................................... 62
Summary........................................................ 62
CHAPTER III
RESEARCH METHODOLOGY........................................... 64
Previous Survey Sample ........... .............. ..... 65
Sample for Current Study ...................................... 70
Survey Instrument . ......................................... 76
Data Collection................................................ 83
Data Analysis ................................................. 85


ix
CHAPTER IV
RESULTS OF THE STUDY.............................................. 89
Current Health Promotion/Disease
Prevention Program Activities ................................ 91
Program Components ........................................... 91
Successful/Unsuccessful Activities ........................... 91
Departments Involved ........................................ 93
Program History ................................................ 93
Development of Interest...................................... 93
Decision Criteria and Processes .............................. 97
Finances.................................................... 97
Benefits................................................... 98
Corporate responsibility for health
of employees..............................................101
Corporate image/liability ................................. 104
Activities proposed/activities deferred ................... 107
Complexity of proposed programs ........................... 109
Decision process .......................................... 109
General Company Environment .............................. Ill
Management philosophy ..................................... Ill
Facilities..................................................116
Hypothesis Testing ........................................... 119
Human Elements................................................119
Champions/angels .......................................... 119
Staff expertise.............................................124
Economic Issues ............................................. 126


X
Financial status ...........
Return on investment ....
Marketing/Access Issues . .
Program availability ....
General level of exposure
Exposures to Model Program .
Discriminant Analysis Model .
Summary ........................
CHAPTER V
CONCLUSIONS AND RECOMMENDATIONS .
Limitations of the Study ....
Retrospective Bias ...........
Focus on Selected Variables
Classification Bias ..........
Decision Focus ...............
Non-Random Sample ............
Discussion of Findings .........
Targetting Diffusion Efforts .
Developing Positive Triggers .
Addressing Management Criteria
Reaching Key Actors ..........
Avenues for Change ...........
Government Role ................
Further Research Issues . .
Summary ........................
127
127
131
131
132
135
138
141
147
147
147
147
148
149
149
149
149
151
152
153
156
156
160
161


xi
BIBLIOGRAPHY...................................................165
APPENDIX
A. BUSINESS AND INDUSTRY.SURVEYS ............................. 174
B. INTERVIEW GUIDE.............................................195
C. DIALOGUE USED WITH CONTACT PERSONS..........................201
D. INTERVIEW RECORD FORM.......................................216


TABLES
Table
1. Companies Included/Excluded From
Sample: 1983 Survey . ...................... 68
2. Companies Included/Excluded From
Sample: Current Study..................................... . 75
3. Source of Information on Variables ........................ 87
4. Classification of Companies on
Variables Examined in Study ............................... 90
5. Association Between "Importance of
Finances" Variable and the Decision
to Adopt or Not Adopt a Worksite Health .
Promotion/Disease Prevention Program ....................... 99
6. "Importance of Finances" Variable as a
Predictor of Adoption/Non-Adoption of
Worksite Health Promotion/Disease
Prevention Program........................................... 100
7. Association Between "Importance of
Benefits" Variables and the Decision
to Adopt or Not Adopt a Worksite
Health Promotion/Disease Prevention
Program......................................................102
8. "Importance of Benefits" Variable as a
Predictor of Adoption/Non-Adoption of
Worksite Health Promotion/Disease
Prevention Program......................................... 103
9. Association Between "Responsibility
for Employee" Variable and the Decision
to Adopt or Not Adopt a Worksite Health
Promotion/Disease Prevention Program ....................... 105
10. "Responsibility for Employees" Variable
as a Predictor of Adoption/Non-Adoption
of Worksite Health Promotion/Disease
Prevention Program
106


:iii
110
112
114
115
117
118
121
122
123
125
Association Between "Program Proposed"
Variable and the Decision to Adopt or Not
Adopt a Worksite Health Promotion/Disease
Prevention Program .......................
Association Between "Decision Process"
Variable and the Decision to Adopt or
Not Adopt a Worksite Health Promotion/
Disease Prevention Program ...............
Association Between "Management Philosophy"
Variable and the Decision to Adopt or Not
Adopt a Worksite Health Promotion/Disease
Prevention Program .......................
"Management Philosophy" Variable as a
Predictor of Adoption/Non-Adoption of
Worksite Health Promotion/Disease
Prevention Program .......................
Association Between "Facilities" Variable
and the Decision to Adopt or Not Adopt a
Worksite Health Promotion/Disease
Prevention Program .......................
"Facilities" Variable as a Predictor of
Adoption/Non-Adoption of Worksite Health
Promotion/Disease Prevention Program . .
Association Between "Champion" Variable
and the Decision to Adopt or Not Adopt
a Worksite Health Promotion/Disease
Prevention Program ................... .
Association Between "Angel" Variable and
the Decision to Adopt or Not Adopt a
Worksite Health Promotion/Disease
Prevention Program .......................
"Angel" Variable as a Predictor of
Adoption/Non-Adoption of Worksite Health
Promotion/Disease Prevention Program . .
Association Between "Staff Expertise"
Variable and the Decision to Adopt or
Not Adopt a Worksite Health Promotion/
Disease Prevention Program ...............


' xiv
Table
21. Association Between "Financial Status"
Variable and the Decision to Adopt or Not
Adopt a Health Promotion/Disease Prevention
Program......................................................128
22. "Financial Status" Variable as a
Predictor of Adoption/Non-Adoption
of Worksite Health Promotion/Disease
Prevention Program ........................................ 129
23. Association Between "Program Availability"
Variable and the Decision to Adopt or Not
Adopt a Health Promotion/Disease Prevention
Program......................................................133
24. Association Between "General Exposure"
Variable and the Decision to Adopt or
Not Adopt a Worksite Health Promotion/
Disease Prevention Program ............................... 134
25. Association Between "Exposure to Model"
Variable and the Decision to Adopt or
Not Adopt a Worksite Health Promotion/
Disease Prevention Program ................................ 136
26. "Exposure to Model" Variable as Predictor
of Adoption/Non-Adoption of Worksite
Health Promotion/Disease Prevention
Program..................................................... 137
27. Discriminant Analysis Model Using
Multiple Variables .................................... 139
28.. Multiple Variables as Predictors of
Adoption/Non-Adoption of Worksite
Health Promotion/Disease Prevention
Program......................................................140
29. Discriminant Analysis Using Accessible
Varible......................................................142
30. Accessible Variables and Predictor of
Adoption/Non-Adoption of WorksiteHealth
Promotion/Disease Prevention Program ....................... 143


XV
Table
31. Variables Associated with the Decision
to Adopt or Not Adopt a\ Worksite
Health Promotion/Disease Prevention
Program ...............................................144
32. Variable Not Associated With the
Decision to Adopt or Not Adopt a
Worksite Health Promotion/Disease
Prevention Program........................................145


xv i
FIGURES
Figure
1. Perspective in Organizational Behavior
Related to Models in Diffusion Theory ..................... 43
2. Adopter Categories ........................................ 49
3. Stages in the Diffusion Process............................61
4. Inclusion of Diffusion Perspectives in
Interview Guide: Primary and Secondary
Focii......................................................79


CHAPTER I
INTRODUCTION
The purpose of this study is to examine the process by
which innovative health programs are adopted by organizations. The
study focuses specifically on the diffusion of health promotion
disease prevention (hp/dp) programs at the worksite.
Rationale for the Study
Each year vast amounts of human and fiscal resources are
dedicated to the development of new products and services. Yet,
only a relatively small percentage of these products and services
will ever he utilized by society. The-vast majority survive less
than a year and reach only a fraction of the potential population
(Rogers & Shoemaker, 1971).
Failure of programs, products, and services to be widely
utilized is a phenomenon experienced by the public as well as the
private sector. Government reports are replete with examples of
new technologies which have been developed and/or promoted by public
agencies but have not been adopted by society. An example of such
a problem is worksite health promotion/disease prevention (hp/dp)
programs.
In recent years the government, as well as the private and
the non-profit.sector, have invested a substantial amount of money


2
in health promotion/disease prevention. Research has been done to
establish the link between lifestyle and major causes of death and
disability. Strategies in smoking cessation, high blood pressure
control, fitness, alcohol/drug abuse prevention have been developed
and evaluated. To reach the adult working population with these
programs, attention has been focused on the worksite. Funds have,
been allocated for the development and evaluation of exemplary
worksite hp/dp programs, conferences have been held for key business
and industry leaders; and numerous agencies and groups have issued
policy statements promoting the worksite as an appropriate site for
prevention activities. Foundations, business coalitions, as well as
state and local public agencies have undertaken a range of activ-
ities to encourage businesses to adopt worksite hp/dp programs
(Berry, 1981; Healthy People, 1979; Parkinson & Associates, 1982;
Sehnert & Tillotson,.1978).
The response from business and industry, however, has been
less than overwhelming (Davis, Rosenberg, Iverson, Vernon, & Bauer,
1984; Fielding & Breslow, 1983). A number of exemplary programs have
been developed by large employers such as IBM, Honeywell, Control
Data, Johnson & Johnson. Yet the total number of programs appears
to remain relatively small. Few inroads have been made into the
vast majority of businesses. Nationwide, it is estimated that fewer
than 5% of businesses offer hp/dp programs (Warner & Murt, 1984).
In order to understand this limited success, and to better tailor
public policies and programs, more information is needed on the


3
factors which influence the adoption/non-adoption of worksite hp/dp
programs.
In general, diffusion theory has provided a useful framework
for examining the adoption of new products and services by society.
Within the framework, numerous social, economic, structural and
marketing variables have been identified which are thought to explain
the rate and pattern of diffusion of innovation. Models have been
constructed and used to study patterns of acceptance of a range of
products and services, including, among others, educational innova-
tion, improved farming techniques, household products and new health
care services. The framework and the principles therein aid analysis
of factors related to adoption/non-adoption of innovation and provide
guidance for administrators wishing to structure their products and
services for maximum diffusion.
Worksite hp/dp is an innovationan innovation which has been
accepted on only a limited basis. It would appear that diffusion
theory is applicable and appropriate to the study of this phenom-
enon. The intent of this research project is to use diffusion
theory as a framework within which to examine the factors related to
the adoption/non-adoption of worksite hp/dp programs. It is hoped
that the findings will provide public and other administrators with
information useful in the formulation of policy and in the design of
future hp/dp program efforts.


4
Worksite Health Promotion/Disease Prevention
Historical Development
Interest in the health of employees is not a new phenomenon
for American business. As early as 1871, Metropolitan Life Insurance
developed health and safety information packages for employees and
policyholders. After the turn of the century, other insurance
companies joined in the production of a series of pamphlets
(Follman, 1978, p. 21). Organized health education efforts within
industry, however, are a relatively recent phenomenon. The first
industrial alcoholism program was started in 1942 by E. I. DuPont
de Nemours and Company. Within a short time, several other major
companies, including Eastman Kodak, Allis-Chalmers, Consolidated
Edison, Armco Steel, and Western Electric had begun similar programs
(Dunkin, 1982, p. 7). A survey done in 1950 by the National Associ-
ation of Manufacturers revealed that nearly 80% of companies with
2,500 or more employees reported a health education component in
their industrial health program. In smaller companies with fewer
than 250 employees, the rate dropped to 21%. A survey completed of
manufacturing firms with fewer than 100 employees revealed that
fewer than 3% had health education programs (USPHS, 1958).
Throughout the 1960s, a number of model programs were
developed including a fitness program at NASA. The Chicago Heart
Association developed a high blood pressure screening and referral
program which it operated at numerous industrial sites in the
Chicago area (Alderman, Green, & Flynn, 1980, p. 106). A survey


5
of 400 members of the Industrial Medical Association revealed that
two-thirds of member companies had some type of program, mostly
individual counseling (Cassuto, 1967). A more general survey,
conducted by the Conference Board in 1972 reported that, of
companies with 500 or more employees, approximately one-third
conducted some type of health education program (Lusterman, 1974,
p. 149). The majority of the programs focused on individual
counseling, information in newsletters, and noon seminars on
specific topics of interest. Size and type of company appeared
strongly correlated with the existence of a program. Companies
with 5,000 or more employees were more likely to have programs (58%)
as opposed to companies with fewer than 1,000 employees (15%).
Fifty-two percent of manufacturing firms and 42% of financial
institutions offered health education, while only 10% of wholesale
companies reported such activities (Lusterman, 1974, pp. 49-50).
On commenting on the development of these early programs,
Nickerson (1967) notes that the majority of them tended to be
"one-shot" in nature; company interest and involvement was minimal.
Ware (1983) states that most efforts seem to have been initiated
by someone outside the agency, an insurance company, or voluntary
health agency. Companies were recipients, but not active partic-
ipants in health education. During this time period, the Occupa-
tional Safety and Health Act (OSHA) of 1970 was passed by Congress.
It required "education and training of employers and employees in
the recognition, avoidance, and prevention of unsafe or unhealthful


6
working conditions" (P.L. 95-596, Sec. 21(c)). Consequently, a
number of firms strengthened their safety education efforts, but
did not alter the scope or content (Ware, 1983).
The mid-1970s marked the beginning of a period of significant
growth and change in worksite health education programs. Several
major business organizations., including the National Chamber Founda-
tion, the Health Insurance Association of America, and the Washing-
ton Business Group of Health developed and disseminated documents
urging companies to implement a wide range of educational activities
for their employees (Berry, 1982; Sehnert & Tillotson, 1978). The
Office of Health Information/Health Promotion within the federal
government held the first national conference to focus attention on
health promotion in occupational settings. The Department of Health
and Human Services issued grants to assist in the identification and
evaluation of model worksite programs. Ware (1983) reports that the
number of journal articles dealing with occupational health educa-
tion increased substantially in the early 1970s. A number of texts
specific to worksite health education/health promotion were published
(Cunningham, 1982; Everly & Feldman, 1985; O'Donnell & Ainsworth,
1984; Parkinson et al., 1982).
Perhaps more importantly, an effort was made to change the
nature of health education efforts from programs provided by
community resources to programs in which the company actively
participated, designing programs based on the needs and risks of
their employees. In 1980, Green, Kreuter, Deeds, and Partridge


7
developed a PRECEDE framework which emphasized a diagnostic approach
to program design and implementation (Green et al., 1980). The
model begins with diagnosis of social problems to determine the
extent to which health problems contribute to the failure of society
to reach specified goals. Further analysis is then done to deter-
mine behavioral factors which contribute to the identified health
problem and the extent to which these factors are changeable through
educational intervention (see Figure 1). The purpose of the model
is to assist in accurate problem diagnosis which, in turn, will lead
to more effective program design.
This diagnostic approach to program planning is reflected
in the work of Berry (1981), Parkinson et al. (1980), Behrens (1983)
and others who apply it to issues at the worksite. They suggest a
process whereby employers identify health problems of employees and
determine the extent to which these problems contribute to the
larger concerns of business such as increased insurance premiums,
increased absenteeism, and lowered productivity. Analysis is then
conducted to determine the changeability of factors which contribute
to the health problem. For example, an industry may be experiencing
excessive absenteeism. An analysis reveals that the majority of
absences are due to back injuries which result from improper
lifting techniques. A combination of educational programs and
environmental adjustments can be made which will substantially
reduce the problem. The diagnostic approach results in the identi-
fication of problems of concern to the organization and may result


8
in a wide range of strategies designed to reduce the problem. It
counters the categorical, externally imposed program model charac-
teristic -of earlier program efforts.
The term "health promotion/disaese prevention" began to
replace the term "health education." Health education was defined
by Green et al.. (1980) as "any combination of learning experiences
designed to facilitate voluntary adaptations of behavior conducive
to health" (p. 7). The term "disease prevention" was added to
encompass screenings and programs aimed at preventing certain
diseases. The term health promotion/disease prevention was
considered a better description of more comprehensive program
efforts being advocated by leaders in the field.
Two factors appear to have contributed to the increased
interest and the new direction of worksite health promotion/
disease prevention (hp/dp) programs: (a) rising cost of illness,
and (b) increasing appreciation of lifestyle as a contributor to
premature death and disability in the population.
A report done for the Health Insurance Association of
America reported in that 1960, total medical care expenditures were
roughly $27 billion (5.3% of the Gross National Product). By 1980,
his total had risen to $243 billion (9.4% of GNP). It was esti-
mated that businesses currently paid over half this amount (Berry,
1981, p. 8). Parkinson et al. (1982, p. 1) projected that, if
unchecked, this figure could reach $462 billion by 1985. Calcu^
lations on cost of illness may be expanded to include indirect costs


9
such as absenteeism, decreased productivity, turnover, as well as
the direct cost of medical care and Workmen's Compensation. Recent
estimates presented by Fielding (1984, p. 240) suggest that indirect
costs of illness are nearly three times the direct cost of illness.
If this ratio holds, the total of direct and indirect costs of
illness in the U.S. could reach a staggering trillion dollar mark by
1985. Robert Beck (1982), examining the experience at IBM, notes
that major medical costs have risen at between ". .11 and 14
percent compound growth rate, year in and year out, with no plan
improvements" (p. 3). The overall employee health costs at IBM were
approximately $1,000 per employee, per year, and was considered a
cause for concern.
There has been no national survey which has conclusively
documented rising cost of illness as the key factor in the growth
of interest in worksite hp/dp programs. Yet, preliminary evidence
would indicate that it is important. Most major publications and
presentations which focus on worksite health promotion, begin wtih
an examination of rising health care costs. Documents produced by
the National Chamber of Commerce (Tillotson & Rosala, 1978) cite
rising health care costs as one reason to begin a worksite program.
A survey of businesses in Colorado revealed that nearly 60% of
companies cited cost containment as a very important reason for
starting an hp/dp program (Davis et al., 1984). In a summary
statement, Fielding (1979) suggests that while cost-benefit evidence
is incomplete, the potential for controlling health care costs


10
through worksite hp/dp programs has been a major factor driving
industry activity in this area.
Changing disease patterns has been the second major factor
focusing attention on worksite hp/dp programs. In the early part
of the century, the bulk of premature death and disability was due
to communicable diseases. Following the improvements in sanitation
and the introduction of better medical care, this pattern shifted
dramatically. Chronic diseases such as heart disease, cancer, and
stroke began to replace tuberculosis, influenza, diphtheria, and
gastro-intestinal infections as the major health problems of the
population. In the mid-1970s, the U.S. Department of Health,
Education, and Welfare (HEW) undertook more in-depth analysis of
these disease patterns. Borrowing the "health field concept"
developed by the Canadian Ministry of Health and Welfare (LaLonde,
1974), HEW examined the extent to which health problems in the U.S.
were due to one of four factors: lack of health care, lifestyle,
environmental hazards, and human biology. Their analysis, reported
in Healthy People: The Surgeon General's Report on Health Promotion/
Disease Prevention revealed that "... perhaps as much as half of
U.S. mortality in 1976 was due to unhealthy behavior or lifestyle;
20% to environmental factors; 20% to human biological factors; and
only 10% to inadequacies in health care" (1979, p. 9). This land-
mark document examined health problems by five age groups: infants,
children, adolescents and young adults, adults, and older adults.
The section on adults focused on those ages 25 to 64 which include


11
the bulk of the workforce in America. Major problems identified in
this age group were cardiovascular disease, cancer, alcohol abuse,
mental health, and periodontal disease. Further analysis examined
"risk factors" which increased an individual's risk for a particular
condition or disease. While a number of the factors were environ-
mental, e.g., exposure to radiation, or biological, e.g., genetic
predisposition to certain conditions, the majority were due to
behavioral factors. Chief among these were: smoking, alcohol abuse,
lack of exercise, poor nutrition, uncontrolled high blood pressure,
stress, and lack of seat belt use. Healthy People and a companion
document Promoting Health/Preventing Disease: Objectives for the
Nation (1980) outlined strategies for improving the health of
America.
The worksite became a focal point for hp/dp programs for
adults. Parkinson et al. (1982, p. 2),representing the public health
perspective,cite access to high risk workers and their families
who might not otherwise participate in such programs. Berry (1982,
p; 9) and others cite factors of convenience, social support, work-
related incentives, and quality control as reasons for increasing
preference for worksite programs. It is interesting to note, that
with some exceptions noted by Bauer (1980, pp. 123-124) that there
has been little overt resistance to worksite hp/dp programs.
Current Program Parameters
There are no generally accepted criteria as to what consti-
tutes a worksite hp/dp program. The following is a presentation of


12
major definitions and program parameters which have evolved in recent,
years.
Healthy People: The Surgeon Generals Report on Health
Promotion and Disease Prevention (1979) states:
Health promotion begins with people who are basically healthy
and seeks the development of community and individual measures
which can help them to develop lifestyles that can maintain
and enhance the state of well being, (p. 119)
In comparison it states "Disease prevention begins with a threat to
healtha disease or environmental hazardand seeks to protect as
many people as possible from the harmful consequences of that threat."
In contrast, medical care is characterized as services which
". . begin with the sick and seek(s) to keep them alive, make
them well, or minimize their disability." Healthy People further
differentiates between health promotion, health protection, arid
preventive health services. Major program categories under health
promotion include: smoking cessation, alcohol/drug abuse preven-
tion, nutrition and weight control, exercise and fitness, and stress
control. Other activities such as high blood pressure screening are
considered preventive health services. Occupational safety and
health, toxic agent control, and accident-injury control are classi-
fied under the general category of health protection.
Documents from the Office of Disease Presention and Health
Promotion within the Department of Health and Human Services
references the definitions provided in Healthy People. They state
that


13
health promotion programs usually address lifestyle and
health habits of healthy individualsnutrition, physical
fitness, stress management, weight management, smoking
cessation, etc. In addition, many health programs use a
variety of screenings to make people aware of their poten-
tial to become ill or contract a disease or detect diseases
early. (Behrens, 1983, p. 1)
They broaden the definition by stating that hp/dp programs often go
beyond lifestyle to look at factors related to occupational safety
and environmental hazards, chronic disease management, and mandatory
employee assistance programs. They further state that a geniuine
commitment to worksite health promotion involves the examination of
the corporate culture to determine company policies and practices
such as smoking policy, cafeteria foods, health benefits, etc.,
which affect employee health behavior.
In 1979, a committee made up of persons from the public and
the private sector was called together to develop guidelines for
health promotion programs at the worksite. They defined health
promotion as "a combination of educational, organizational, and
environmental activities designed to support behavior conducive to
the health of employees and their families" (Parkinson et al.,
1982, pp. 8-9). Major program components recommended (a) risk
assessment using risk appraisals and other measures to determine
potential for cardiovascular disease, cancer, stroke, mental health
problems, and accidents; (b) risk reduction measures in the areas
identified as health promotion, smoking control, alcohol/drug
abuse, nutrition and weight control, fitness and exercise, and
stress management. To this, they add intervention in high blood


14
pressure control, cancer detection, accident prevention, and
protection against environmental health hazards; and (c) supportive
environmentmechanisms in the social and physical work environment
which, such as nutritious foods, spur groups which support the
adoption of positive health behavior changes.
In a report for the Health Insurance Association of America,
Berry (1981, p. 19) suggests strategies similar to those outlined
by Parkinson et al. (1982) including: risk assessment, education,
support and motivation for change, risk reduction programs, and
environmental and social support to maintain change. Topic areas
which he considers appropriate to health promotion programs include
those already mentioned above, plus a mixture of related programs
ranging in emergency medicine, the Heimlich maneuver, benefits
utilization, low back pain prevention, immunization, dental health,
parenting, retirement, screening for glaucoma and sickle cell anemia.
While the majority of these health problems can be prevented or
reduced through changes in lifestyle, the topic areas range far
afield from health promotion strategies identified in Healthy
People as having the greatest overall potential for reducing health
problems in the population.
The National Chamber Foundation report (Sehnert & Tillotson,
1978) lists the basic areas of fitness, smoking cessation, alcohol/
chemical abuse, nutrition and weight control, stress management,
accident prevention, and screening programs including blood pressure
screening and control. In addition, it promotes programs in medical


15
self-care and wiser buying of health care services. These last two
elements, not common to all other guidelines, are consistent with
the Chamber's focus on cost-containment through health care utili-
zation.
In summary, it may be said that there appears to be a central
set of program elements which are included in most program guide-
lines. These include exercise, blood pressure control, smoking
cessation, alcohol abuse prevention, nutrition and weight control,
and stress management. A second set of elements including CPR,
accident prevention, self-care, chronic disease control, parenting,
and others are included less frequently, and appear to be related to
particular interests of the sponsoring agency. There is some
consistency among guidelines, but no single standard set of defini-
tions or guidelines appears to dominate at this juncture.
Current Level of Program Activity
There is a general consensus in the literature that there
has been a dramatic increase in the number of companies offering
health promotion programs (Behrens, 1983; Berry, 1981; Cunningham,
1982; Fielding & Breslow, 1983; Ware, 1982). Yet, there is little
information on how widespread hp/dp programs are throughout industry
in general. A 1978 survey completed by the Washington Business
Group on Health (Kiefhaber, Weinberg, & Goldbeck, 1979) revealed
that over half the companies responding offered some form of health
promotion program. The group surveyed, however, was limited to


16
WBGH membership, including mostly Fortune 500 companies. The
response rate was 37%, too low to allow the results to be general-
izable without further analysis.
A survey done by Fielding and Breslow (1983) of California
businesses with 100 or more employees revealed that 78.3% offered
one or more health promotion activities. Nearly two-thirds of the
company programs were limited to two or less activities, which could
range from simple CPR seminars to more complex hypertension control
and screening.
A survey was conducted of Minnesota employers with 50 or
more employees. Nearly 80% of employers responding offered some
type of health promotion program on a systematic and on-going basis.
The survey contained a more clear, albeit minimal definition of a
health promotion program. The response rate of 54% does not allow
the results of the survey to be generaiizable to the larger popu-
lation (Minnesota Department of Health, 1982). In both California
and Minnesota, it may be assumed that those with program activities
were more likely to respond, thus artificially inflating the figures.
A recent survey completed of businesses in Colorado with 50
or more employees revealed that approximately one quarter of the
businesses contacted had a hp/dp program which offered one or more
services on a regular on-going basis. Ah additional 50% of companies
expressed interest in developing a program in the future. Over 40%
of companies with 1,000 or more employees offered a program; this
percentage decreased dramatically with decreasing size of company.


17
Overall, it was estimated that fewer than 5% of employers in Colo-
rado offered hp/dp programs on an on-going basis (Davis et al.,
1984). Figures cited by Warner and Murt (1984, p. 110) confirm
these estimates, placing the percentage of employees nationwide
covered by hp/dp programs at between 2% and 5%.
Size of company appears to be a major factor in determining
whether or not an hp/dp program is offered. A survey of California
employers revealed that over 95% of employers with 5,000 or more
employees had an hp/dp program, typically offering 3 or more
activities, e.g., high blood pressure screening, exercise programs.
In contrast, only two-thirds of businesses with fewer than 250
employees offered programs, limited in almost all cases to one or
two activities (Fielding, 1983). The Minnesota and Colorado surveys
produced similar findings. Large employers were more likely to
offer programs. In addition, these programs were likely to be more
comprehensive than those offered by their smaller counterparts
(Davis et al., 1984; Minnesota, 1982). An analysis of the hp/dp
programs selected for review by the Committee to Develop Guidelines
for Health Promotion Programs at the Worksite reveals that 16 out
of the 17 programs have more than 1,000 employees (Parkinson et al.,
1982). The list of exemplary programs compiled by Behrens (1982)
is likewise replete with such names as Campbell, Kimberly-Clark,
NASA, IBM, Xerox, and other large companies.
Recent attempts have been made to gauge the level of hp/dp
program activity in small business. Although some reports such as


18
that produced by Health Works Northwest (McMahon, Sajewski, & Graff,
1984) and by the New York Business Group on Health (Warshaw, Wein-
garten, Barr, & Lucas, 1984) provide case studies of successful
programs, sample size, sample characteristics, and low response
rates prevent the findings from being generalizable to a larger
population. More complete and more accurate information on hp/dp
programs in small businesses is desirable, considering figures
cited by Health Works Northwest "... that over 80% of all busi-
nesses in the U.S. have fewer than 20 employees, and that 99% of
all U.S. businesses employ 500 people or less ..." (McMahon
et al., 1984, p. 2).
Frequency of Program Components
Not all program components occur with equal frequency. A
1981 survey of 424 randomly selected California employers with 100
or more employees revealed that, in those companies with programs,
the most common program components were
accident prevention, cardiovascular pulmonary resuscitation,
and choke saver. Alcohol or drug abuse program and mental
health/counseling were in place in about one-quarter of
organizations offering any program. Hypertension screening,
smoking cessation, fitness and stress management programs
were made available by 10-17% of these employers. (Fielding,
1984, p. 250)
Cancer control was conducted by fewer than 10%.
Data on hp/dp programs in 315 Minnesota businesses,
industries, school districts, and hospitals revealed a similar
pattern. Prevention of injury was the most common program component


19
offered by approximately 60% of respondents. Alcohol/Chemical
Dependency Programs and Personal and Family Counseling including
Employee Assistance Programs (EAPs) were offered by almost half of
the companies. The remainder of the programs including hyper-
tension weight controli disease screening, smoking cessation,
fitness and stress management were offered by approximately 15-20%
of the companies. Diet information and programs on positive health
attitudes were least common, being offered by fewer than 15% of the
companies. Small employers, defined as those with fewer than 800
employees, offered the above program components at roughly half the
rate of the large employers in the sample. The only exception to
this were the injury prevention programs which were offered at
basically the same rate by large and small employers (Minnesota
1982, pp. 12-14).
A 1983 Colorado survey provides information on program
components in 94 randomly selected Colorado businesses of 50 or
more employees. Hp/dp activities were categorized under 3
major headings: screening, informational programs, and preventive
health services. Pre-employment medical examinations and high
blood pressure screening were the most common screening activities
carried out by approximately 3 out of 4 companies. General and
cardiovascular risk appraisal, height/weight screening, and screen-
ing for work-related health problems were offered by roughly half
of the companies. Screening for cancer was done by fewer than a
quarter of the companies. The most common topics for informational


20
programs were exercise, stress management, smoking cessation, and
nutrition offered by 3 out of 4 companies. This was followed
closely by alcohol/drug abuse, high blood pressure, low back pain,
and work-related injury prevention information offered by two-
thirds of the companies. Group and individual instruction in
exercise and stress management were the most frequent service
programs offered by 80% and 77% of companies, respectively,
followed closely by weight control and smoking cessation offered
in 2 out of 3 companies. Low back pain was offered by slightly
more than half of the companies, while self-defense for women
was provided by approximately one quarter of the companies. Two
out of 3 companies had either an Employees Assistance Program (EAP)
or an Industrial Alcoholism Program (IAP). The 206 companies
interested in developing programs in the future were asked what
type of activities they would include. Their response indicates
that they would follow basically the same pattern as established
programs (Davis et al., 1984).
A survey conducted by the Washington Business Group on
Health provides information on hp/dp activities in 59 member
companies, mostly large businesses. As in the California survey,
CPR and safety on the job are the most common program components,
offered by over 4 out of 5 companies. Hypertension and breast
self-examination were provided by 63% of companies. Programs
such as safety (off-job), smoking cessation, nutrition and obesity
control; fitness and exercise were offered by slightly over 50% of


21
companies. In contrast to other survey findings, stress management
programs were offered by fewer companies (41%) Benefits utilization
programs were part of the activities in approximately one-third of the
companies (Kiefhaber et al., 1979).
Information on types of hp/dp activities in 31 small
businesses is provided by a study done by Health Works Northwest.
Fitness/exercise programs are offered by 75% of companies. Smoking
cessation, health risk appraisals, stress management, and nutrition
are all offered by slightly more than 60% of companies. High blood
pressure control is offered by 40% of companies, substances abuse
by slightly more than one-third. While the case studies and the
summary of activities are interesting, the sample is too small and
non-scientific to make any statement about the relative frequency
of hp/dp programs in the small business community in general
(McMahon et al., 1984, p. 15).
In the presentation of their.findings, authors are quick to
note that the level of program intensity and the strategies employed
may vary widely among programs. Subjective interpretation as to
what constitutes a program may lead to over-reporting in some
cases, and under-reporting in others.
Rate of Growth
There is relatively little baseline data against which to
compared the current level of program activity. Questions on hp/dp
were not routinely included in surveys conducted on occupational


22
health. When included, questions frequently did not address the
range of activities now considered central to hp/dp programs. In
a 1967 survey conducted of the 400 members of the Industrial Medical
Association, two-thirds reported "some type of health education
program." Individual counseling was the activity most commonly
cited (Cassuto, 1967). It is not clear that individual counseling
carried out in the context.of health services is comparable to
current hp/dp program efforts. Similarly, health-related research
done by the National Conference Board over the past decade has
focused on health services, offering only incidental information on
level and type of hp/dp activities.
Several recent surveys have attempted to document program
growth. The survey of California employers showed that nearly
three-quarters of hp/dp activities currently being offered at the
worksite were begun after 1975 (Fielding & Breslow, 1983). Colorado
data provide similar information. Roughly 85% of programs in
existence were begun in a 5-year period from 1978-1983 (Davis et al.,
1984). What is not clear from either of these surveys is the number
of programs which may have been operative in the 60s or early 70s
which were recently phased out of existence.
More specific, encouraging data are available in categorical
program areas. It was estimated that alcohol programs in industry
increased from 50 to 500 in the 3-year period from 1970 to 1973.
By 1977, over 2,000 organizations offered some type of program
(Third Special Report to U.S. Congress on Alcohol and Health,


23
1978). Trice (1977) and others note that although the numbers were
small, the increase of worksite alcoholism programs had been substan-
tial. More recent estimates by Dickman and Emener (1982, p. 55)
place the number at closer to 5,500, which is indeed indicative of
a growing acceptance in at least one of the component areas of
worksite hp/dp.
In the absence of hard data, professional estimates may be
used as indicators of growth of hp/dp. Cunningham (1982) reports
that membership in the American Association of Fitness Directors in
Business and Industry grew from 25 in 1975 to over 3,000 in 1981.
Ware (1982) likewise cites the increasing number of articles in
professional journals as evidence of increased program activity. The
increasing number of business coalitions which currently offer health
education services (56%) is likewise a measure of growth (Yenney,
198A).
While these observations are all of a positive nature and
generally consistent, they might best be considered indicators of
increased interest, rather than accurate reflections of an increased
level of programs and services actually being offered at the work-
site. The rate of growth appears limited and confined to the
"cadre of the convinced" (Davis et al., 1984). Prototypes of
current hp/dp programs were developed in the 60s and were relatively
refined by the mid-70s. Yet, by most estimates, these programs have
been adopted by fewer than 10% of employers, and reach only a small
percentage of the workforce.


24
Program Outcomes
The majority of worksite health promotion programs are
designed to address one or more of the following goals: (a) to
reduce illness and improve the overall health status of employees,
(b) to reduce costs associated with absenteeism and health benefits
utilization, and (c) to improve employee morale and productivity
(Davis et al., 1984; Fielding, 1983; Kiefhaber et al., 1979).
There is little data on the extent to which these outcomes are
systematically achieved. The evidence that exists suggests that
positive outcomes are dependent, in large part, on the intensity of
the program, the appropriateness of the strategies implemented, and
the extent to which the work environment supports and reinforces
behavior change.
Program outcomes have been more thoroughly evaluated in
several categorical program areas. This section will examine the
outcomes in four areas that have been most rigorously studied: High
blood pressure control, smoking cessation, alcohol/drug abuse, and
fitness/exercise programs at the worksite. Particular attention
will be paid to the extent to which these programs improve employee
health, control cost, and/or contribute to improved morale and
productivity. Because few studies have been conducted on nutrition,
stress management programs at the worksite, they will not be included
in this section.
Hypertension. Numerous studies have shown that individuals
with high blood pressure have an increased risk for premature death


25
and disability. Rates of heart disease and strokes are six to
seven times greater in individuals with high blood pressure, than
in those with normal blood pressure (Veterans Administration, 1967).
A second set of studies have examined intervention strate-
gies and have demonstrated effective means for controlling high
blood pressure (Hypertension Detection, 1979). Several design
variables appear to differentiate successful from unsuccessful
programs: number and intensity of contact with health care providers,
active patient participation, long-term follow-up ahd the existence
of supportive family and friends (Alderman et al., 1980; Foote &
Erfurt, 1983). Programs that include this feature have a high
potential for controlling blood pressure, thereby reducing health
problems of employees. A number of successful programs are
currently underway, including a cooperative program between the
University of Michigan and Ford Motor Company, where 80% of hyper-
tensive employees are under control (Alderman et al., 1980), and
a program at Massachusetts Mutual Life Insurance Company where the
percentage of hypertensive under control rose from 36% to 82%
after one year of the program (Fielding, 1984, p. 241). "In
general," states Fielding, "worksite based hypertension detection
and control programs have results that are superior to what is
achieved in clinical practice. . ."
It is estimated that between 15% to 30% of Americans suffer
from high blood pressure. Roughly half of this is uncontrolled,
contributing to increased cardiovascular and related disabling


26
conditions. Using the cost of treating such chronic diseases and
disability, Kristein (1982) estimates that."the 'average' hyper-
tensive over a lifetime, is 'costing' the society between $170 and
$300 per year in excess medical spending, and between $270 and $460
in lost-output" (p. 32). Measuring only the cost of outside
medical care covered under company-sponsored insurance, Fielding
(1984, p. 241) places the estimate at between $175 and $250 per
hypertensive.
Costs of high blood pressure control programs vary. Ruchlin
and Alderman (1980), p. 795), reporting on their experience in 9
clinic sites in the New York City area, placed cost of treating
hypertensives at the worksite between $150 to $200 per patient per
year. In contrast, Fielding (1984) estimates that effective screen-
ing referral and follow-up programs could be provided for an
estimated $37 per hypertensive employee. Because calculations are
based on sets of changing program parameters and characteristics
of the target population, it is difficult to arrive at set figures.
Kristein (1982, p. 34) places the ratio of cost benefit between
2 to 1 and 4 to 1.
Smoking cessation. The effects of smoking on health have
been documented through hundreds of studies. Smoking increases
risk for
heart and blood vessel diseases; chronic bronchitis and
emphysema; cancers of lung, larynx, pharynx, oral cavity,
esophagus, pancreas, and urinary bladder. . Smoking
during pregnancy also increases risks of complications of
pregnancy and retardation of fetal growth. Cigarette


27
smokers have a 70 percent greater rate of death from all
causes than non-smokers, and tobacco is associated with an
estimated 320,000 premature deaths a year. Another 10
million Americans suffer from debilitating chronic diseases
caused hy smoking. (Healthy People, 1979, pp. 121-122)
Luckily, people who quit smoking can reduce their risk for
these diseases and conditions over time. A variety of smoking
cessation strategies have been tested including: physician
counseling, motivational programs, self-help programs, skill
acquisition, peer support, and smoking prohibition. In general,
programs that offer a variety of approaches are more effective
than single-faceted programs, because they have a greater change
of meshing with individual learning styles and needs of smokers.
The success rate of smoking cessation programs in general
has been mixed. Initial quit rate generally range between 70% to
90%; this drops to 30% at 6 months, with some of the best programs
maintaining a quit rate of 40%. Worksite programs generally achieve
the same quit rate, but may experience a higher dropout rate due to
time conflicts (Fielding, 1984, p. 243). Danaker (1980), however,
points to worksite smoking cessation programs as having the potential
to reach populations not normally attracted to community programs
and to provide positive incentives for non-smokers. Warner and
Murt (1984) cite the example of Speedcall Corporation, where a
carefully structured incentive program contributed to a drop in
smoking from 67% to 20% of employees over a 4-year period. Although
there appear to be some flaws in the evaluation of the Speedcall
program, it may be indicative of the potential for employers to
encourage non-smoking behavior.


28
The cost of smoking to society and to employers specif-
ically is reasonably well-documented. In a review article, Fielding
(1984) states:
Smoking is expensive to employers, translating into higher
health insurance costs (some studies show a 50% greater use
of the health care system by smokers), higher absenteeism
(an estimated 2 to 3 additional days per year), increased
accidents at work, and a higher rate of disability reimburs-
able events. (p. 243).
Kristein (1983, p. 358), analyzing data from a variety of sources,
estimated that smokers cost the employer between $336-$601 per year
in 1980 dollars. The cost centers examined by Kristein are
insurance, medical care, absenteeism, loss of productivity, and
health risk to other employees. He notes that non-health costs of
smoking may exceed the health costs to businesses. Figures presented
by Fielding (1984, p. 243) place estimates in a range of $200 to
$500 per person.
Costs of smoking cessation programs vary from $115 self-
help programs sponsored by voluntary health agencies, to commercial
programs charging $500-$l,000 per participant. Assuming that the
average smoker costs an employer $350 per year, a smoking cessation
program that costs less than $100 per employee and has a sustained
quit rate of 33% would provide cost savings for the employer. Those
in which the employee bears half the cost would provide an even
higher return on investment.
Employee assistance programs. In the 1960s, the National
Council on Alcoholism estimated that 4% to 8% of the workforce had


29
a problem with alcoholism. Later, studies placed the estimate
between 5% and 10% (Dunkin, 1982, p. 4). In a summary article,
DuPont and Basen (1980) review estimates that have been made by
various sources. They conclude that the average is closer to 10%
with a significantly higher prevalence rate among "lower blue-
collar workers" and in male-dominated industries.
Alcohol and drug abuse is a contributing factor in an
estimated 10% of all deaths in the United States. It increases
risk for cirrhosis of the liver and certain types of cancer of the
digestive system. It is a factor in an estimated 50% of all
traffic fatalities. It is considered an indirect cause of homi-
cides and suicides (Healthy People, 1979). It can be stated that
a reduction of alcohol abuse would significantly enhance the health
and well-being of employees.
Industrial alcoholism programs began in the 1940s and have
evolved to the broader concept known as Employee Assistance Programs
which address a wide range of issues including counseling for
depression, substance abuse, family relationships, personal and
financial problems. The data on the effectiveness of such programs
has generally been fragmentary, but positive. Berry (1982, p. 28)
presents data on 8 programs with a rehabilitation rate from 70% to
80%. DuPont and Basen (1980, p. 142) report several studies that
report rehabilitation rates near 50-60%. Two specific examples
given include the Bethlehem Steel Corporation and the DuPont
Corporation, with a 60% and 66% rehabilitation rate, respectively.


30
The authors note that despite the fact that the majority of program
evaluations lack control groups and are based on limited measures of
success, most of the data show positive impact. It is likely that
there are numerous less well-designed and implemented programs with
lower rehabilitation rates that do not appear in the literature.
The cost of alcohol/drug abuse to industry is estimated at
nearly $51 billion per year based on both increased medical care
utilization and decreased productivity (Berry, 1981, p. 28; Fielding,
1984, p. 240). Estimates of cost savings to business from effective
EAP programs vary. The General Motors program is generally cited as
an example of potential saving. For the first year of operation,
they experienced a 40% decrease in time off the job, a 60% decrease
in sickness and accidents benefits, and a 50% decrease in grievances.
Overall, they estimated a 3:1 return on investment (Berry, 1982,
pp. 28-29). A 1979 supplement to Medical Care examined 12 EAP
programs focused primarily on alcohol and reported similar findings,
estimating cost savings of $1,000 per client. DuPont and Basen
(1982, p. 143) report results of a federal civil service program
that expends $15 million annually, and estimated cost savings at
between $135 and $280 million per year. In summary, it may be
said that while estimates vary, a number of exemplary programs
demonstrate the strong potential for cost savings through Employee
Assistance Programs.
Exercise/fitness. Several national surveys estimate that
only between 35% to 50% of the population exercises regularly


31
(Fielding, 1982, p. 132; Haskell & Blair, 1980, p. 109; Healthy
People, 1979, p. 132). Lack of exercise can lead to osteoporosis,
increased risk of back problems, and is thought to be linked to
increased falls in the elderly population (Fielding, 1984,
pp. 243-244). A study by Paffenbarger, Wing, and Hyde (1978) of
17,000 Harvard alumni demonstrated that those not engaged in
regular vigorous exercise programs had a 64% higher risk of heart
attack than their active counterparts. Exercise can also lead to
reduced blood pressure, improve cardiovascular efficiency, and may
prompt persons to stop smoking. Although the evidence is less
extensive, it appears that exercise plays a role in preventing
mild anxiety and depression (Haskell & Blair, 1980, p. Ill; Healthy
People, 1979, p. 132), improving attitude toward the job, overall
performance, and job satisfaction (Fielding, 1984, p. 244). A
study by Durbeck et al. (1973) of NASA employees found that
participation in exercise programs increased work capacity and
satisfaction. Studies done by Rhodes and Dunwoody (1980) among
Canadian workers demonstrated the same outcomes. Another
Canadian study by Donoghue (1977) found that participation in a
fitness program reduced absenteeism. Fielding (1982, p. 911)
reports a controlled study in a Toronto insurance company which
demonstrated a 42% decrease in absenteeism among participants as
compared to a 20% decrease among those choosing not to participate.
Fielding points out that many of the studies on the impact
of fitness programs are flawed because they do not control for


32
factors which prompted individuals initially to sign up for the
programs. Still the evidence of the effectiveness of fitness
programs in participants is generally positive both from a health
perspective and a job-related benefits view.
Exercise programs in industry vary from simple motiva-
tional programs to staffed fitness facilities at the site. A
minimum program should involve employees in vigorous aerobic exer-
cise 15 to 30 minutes at least. 3 times a week. Formats may include
jogging, swimming, walking, or any number of active sports. Exer-
cise to increase muscle tone, strength and flexibility may be
desirable from the employees point of view, but have not been
shown to decrease cardiovascular risk. The main issue in worksite
fitness programs is not benefits of exercise, but the ability of
companies to construct programs which workers find attractive.
Access and peer pressure appear to be 2 key variables which employers
have only recently begun to address.
Cost savings information for worksite fitness are more
difficult to generate. Health effects of fitness are generally
long-term. There are no immediate reductions in medical care
utilization cited as in the case of EAPs and smoking cessation.
That is not to say that these links do not exist, they simply have
not been sufficiently examined. Measuring increased work capacity,
reduced absenteeism and improved productivity in economic terms, is
a problem not specific to fitness programs alone. Most frequently,
observes Wright (1982, p. 967), fitness programs are operated or


33
justified as an employee benefit. They are viewed as producing
cost savings because people want to believe they do.
In general, then, it may be said that several worksite
health promotion programs including high blood pressure control,
smoking cessation, employee assistance programs, and exercise
programs, if properly designed and implemented, can improve
employee health. The evidence for cost-savings is somewhat weaker.
Smoking cessation, high blood pressure control, and EAPs all have
a reasonable number of studies to document this claim. Cost
savings from fitness which rests on its ability to reduce
absenteeism and increase productivity is tenuous at best. Despite
the lack of solid evidence, companies believe that hp/dp programs
result in a range of positive outcomes, including improved
employee health, reduced health care cost, and improved morale and
productivity (Davis et al., 1984; Fielding & Breslow, 1983).
Finally, in considering long range effects of hp/dp
programs, Warner and Murt (1984, p. Ill) sound a word of caution.
They point to turnover rates in industry which would result in the
loss of investment in human capitalother companies indeed might
reap the benefits of the first company's hp/dp program. They
further argue for an examination of increased life span on pension
benefits, an area not currently discussed in the literature.
Summary
The scientific basis for health promotion/disease prevention
has been evolving over the past half century. More recently,


34
effective intervention strategies have been developed which hold the
promise of reducing illness and the attendant medical care costs in
the adult population. Worksite health promotion/disease preven-
tion programs have been packaged and promoted by many groups which
view the worksite as an optimum site for the delivery of preventive
programs. The response of industry has been one of interest; yet
the adoption and implementation of programs has been limited to
large companies who employ only a small portion of the total work-
force. The existing literature documents and describes existing
programs, but does little to explain the limited acceptance by
business and industry in general. The purpose of this study is
to use diffusion theory as a framework within which to understand
the process and the variables related to adoption or non-adoption
of worksite hp/dp programs.


CHAPTER II
REVIEW OF THE LITERATURE
Diffusion theory is that body of knowledge which attempts to
explain "the pattern and rate of adoption of an innovation in a
population" (Feller & Menzel, 1977, p. 51). Over the years, a
number of theoretical models have been developed and used to study,
among other things, educational innovation, introduction of new
farming techniques, and the acceptance and diffusion of health care
products and services. This chapter will review this literature
base as a framework within which to examine the adoption/non-
adoption of worksite health promotion/disease prevention (hp/dp)
programs.
Historical Development
The beginnings of diffusion theory are traced to the early
1900s (Brown, 1981; Midgley, 1977; Rogers & Shoemaker, 1971).
Anthropologists initially studied the transmission of Western
ideas to primitive societies. Although many of these early projects
tended to be highly specific to site, they provided insight into
the impact of culture on new ideas, and an empirical data base
related to diffusion (Rogers-& Shoemaker, 1971, p. 49). At almost
the same time, a French sociologist, Tarde, produced his work on
opinion leadership and imitation behavior (Tarde, 1903). His
observations provided hypotheses to be tested by later researchers.


36
Major growth in the field of diffusion research occurred in
the 1920s when diffusion theory became a tool used by rural soci-
ologists to understand the adoption of new farming practices. A
study by Ryan and Gross (1943) on the diffusion of hybrid seed
corn is a classic work that established a pattern for the study of
innovation diffusion in agriculture. The U.S. Department of Agricul-
ture Extension Service conducted numerous studies throughout the
40s and 50s to determine the efficacy of its new program promotions.
Although rural sociology continued to dominate the field,
researchers from several other disciplines became increasingly
active in the field of innovation diffusion. Pioneering work in
the field of education was done by Paul Mort. Concerned with the
lag time in the adoption of educational innovations, he attempted to
establish administrative predictors of innovativeness in local
school districts (Mort & Connell, 1938). The number of studies on
educational diffusion continued to grow rapidly throughout the 40s
and 50s focusing on a range of topics from the effects of Sputnik
to the adoption of new math. Although there were a great many
studies of the diffusion of educational innovations, many were
judged to be of lesser quality, adding little to the field in
general (Carlson, 1968; Rogers & Shoemaker, 1971).
The study of the diffusion of medical innovations grew with
the field of medical sociology in the 1950s. A study by Coleman,
Katz, and Menzel (1957) focused on the adoption and prescription
of a new drug among physicians in two Midwestern communities.


37
Following this classic study, numerous other research efforts were
undertaken. The majority of these focused on individual behavior
of physicians and patients. The diffusion of family planning prac-
tices has been studied extensively both in the United States and
abroad.
During the 1960s considerable work in diffusion innovation
was carried out by researchers in the field of communication and
marketing. Communication experts frequently focused on the effects
of various channels of communication on the rate of adoption of
particular behaviors. Katz and Lazarsfeld (1955) examined the
relative effectiveness of personal contact, printed matter and
electronic media in promoting purchase of new products in the super-
market. Because of its dependence on the continuing production and
purchase of "new products," the fashion industry in the United
States has sponsored numerous studies on consumer behavior (Baum-
garten, 1984; Midgley, 1974). Unfortunately, most of the marketing
studies related to innovation diffusion are privately funded,
therefore the results are proprietary information not widely shared
in professional journals.
More recently, books and articles have begun to appear which
address the role of government in the diffusion process. A compre-
hensive review of the literature conducted by Tornatzky, Eveland,
Boylan, Hetzner, Johnson, Roitman and Schneider (1983) for
the National Science Foundation cites over 75 books, reports
and articles of this nature written since 1975. This body of


38
literature moves beyond the passive observation of the diffusion
process to policy analysis and an active debate over the proper
role of government in the diffusion of technological innovations.
The field of diffusion research as it exists in the 1980s
incorporates data bases and perspectives from a wide variety of
disciplines including anthropology, sociology, rural sociology,
education, medical sociology, communications, marketing, psychology,
engineering, political service and economics (Rogers & Shoemaker,
1971; Tornatzky et al., 1983). The focus of this research ranges
from micro-level concern with attributes of the innovation itself
to the measurement, of the impact of macro-economic policy on adoption
practices within specified populations. The range of disciplines
involved and the variety of perspectives used add to the richness
and the complexity of the field of diffusion theory. Unfortunately,
as noted by Tornatzky et al. (1983), "No single organizing framework
is adequate for integrating all the diverse themes to be found in
this literature" (p. vii). The remainder of this chapter will focus
on identifying central themes of diffusion and exploring research
findings from those areas and studies deemed most applicable to the
study of worksite hp/dp programs.
Definitions
Given the diversity of research traditions which have con-
tributed to diffusion theory, diversity in terminology and in


39
theoretical models is to be expected. The following is a presenta-
tion of most widely used definitions and major concepts in diffusion
theory.
Webster defines innovation as "the introduction of something
new." While strikingly simple, this definition does not provide the
specificity desired by researchers. Therefore, modifiers, quali-
fiers and alternate definitions have been developed. Rogers and
Shoemaker (1971) defined innovation as ". .an idea, practice, or
object perceived as new by an individual" (p. 19). Zaltman and
Duncan (1973), recognizing that the unit of adoption may be other
than an individual, defines innovation as ". . the first or early
use of an idea by one of a set of organizaions with similar goals"
(p. 463). Using a practical, marketing perspective, Bell (1963)
considers an innovation as anything which does not yet have 10% of
the potential market. Tornatzky et al. (1983) define innovation as
". .a technology new to a given organization" (p. 1). Pelz and
Munson (1980) use the term, "knowledge based innovation" to describe
non-material innovations. These same intangible concepts such as
management by objective (MB), new psychiatric treatment, and quality
circles can also be termed innovative "social technologies."
The result of these various perspectives is that the common
use of the term "innovation" is inclusive rather than specific. In
general, an innovation may be (a) new or perceived as new, (b) tangi-
ble or intangible, and (c) adopted by an individual or a larger unit
in society.


40
A clear, operational definition of diffusion is, likewise,
elusive. Rogers (1984) defines diffusion as . the process by
which an innovation is communicated through certain channels over
time among the members of a social system" (p. 5). From a research
standpoint, however, the question remains as to where an innovation
can be considered to have diffused through a social systemat
what point does that "overt behavior change" occur which signals
adoption (Rogers & Shoemaker, 1971). An example from education
illustrates the point. Can new math be considered to have been
adopted by a school district (a) when the school board adopts a
policy supporting its adoption or (b) when the board provides a
budget for new materials or (c) when it is piloted in one school
or (d) when it is in use throughout the district or (e) only when it
becomes a regular part of the curriculum for all schools?
To address this issue researchers began to identify and
focus on the stages of the diffusion process. Perhaps the most
widely used model was that developed by Rogers and Shoemaker (1971,
pp. 24-27) outlining four major stages (a) knowledge, (b) persuasion,
(c) decision, and (d) confirmation. Zaltman and Wallendorf (1983,
p. 530) presented a slightly expanded version including: (a) knowl-
edge and awareness, (b) attitudes formation, (c) decisions,
(d) initial implementation, and (e) contrinued-sustained imple-
mentation. Depending on the purpose of the research, these stages
may be subdivided. An example is a study done by Pelz (1981) on
the use of information by local governments. The stages he


41
identified were ". . concern or diagnosis, search, design,
appraisal or selection, building commitment, implementation (plus
incorporation), and diffusion" (p. 4).
In a meta-analysis of stage models of diffusion, Tornatzky
et al. (1983) note that all models are generally variants on the
following pattern:
(1) Awareness
(2) Matching/Selection
(3) Adoption/Commitment
(4) Implementation
(5) Routinization (p. 22)
Indeed, the most recent model developed by Rogers (1984) includes
these stages and a brief description of each:
There are five main steps in the innovation-decision process:
(1) knowledge, which occurs when an individual or some other
decision-making unit is exposed to the innovation's existence
and gains some understanding of how it functions, (2) per-
suasion, which occurs when the individual forms a favorable or
unfavorable attitude' toward the innovation, (3) decision,
which occurs when the individual engages in activities that
lead to a chance to adopt or reject the innovation, (4) imple-
mentation, which occurs when the individual puts the innovation
into use, and (5) confirmation, which occurs when the individual
seeks reinforcement of an innovation decision if exposed to
conflicting messages about the innovation. (p. 8)
Faced with the lack of universally applicable and acceptable
definitions, the tendency has been for researchers to arrive at
their own operational definitions. While some authors decry this
practice as limiting the comparability of studies (Warner, 1974),
others argue that a common terminology may not ever be reasonable
or entirely appropriate (Tornatzky et al., 1983, p. 22). What does
appear important is that researchers specify the type of innovation


42
studied, the unit of analysis and the definition of adoption used in
their study.
Diffusion Models and Perspectives
Diffusion theory is a complex and diverse literature which
incorporates findings from a number of academic disciplines. The
analytical models closely resemble the theoretical perspectives in
the field or organization theory in general (see Figure 1). They
range from the rational/economic model to the more behaviorally-
oriented communications model to models which address environmental
forces affecting diffusion, thus echoing the systems approach to
management and change. Recent studies attempting to delineate
"innovativeness" in an organization appear similar to studies
addressing the complex phenomenon of "corporate culture." This
section presents an overview of major diffusion perspectives. It
must be noted that while each of the models reviewed has a major
focus different from the rest, there is considerable overlap and
duplication of variables represented in each model.
Rational/Economic Perspective
Economic model. This model focuses on economic variables
to explain the rate and pattern of diffusion of an innovation. Key
variables identified by Griliches (1957), Mansfield (1968), and
others include "... profitability, size of required investment,


A3
RATIONAL/ECONOMIC PERSPECTIVE
Economic Model
Return on investment
Size of investment
Size of company
Certainty of return
Structural Model
Size
BEHAVIORAL PERSPECTIVE
Communications Model
Source
Innovation attributes
Channels of communications
Adopter characteristics
Champion Model
Marketing Model
Access
Pricing
Promotion
Market segmentation
Contextual Model
Financial environment
Government policy actions
Consumer behavior
Figure 1. Perspective in Organizational Behavior Related to Models
in Diffusion Theory


44
uncertainty and risk aversion . (Warner, 1974, p. 435). The
major hypothesis as stated by Mansfield (1968) is that "the rate
of adoption is a linear function of the profitability of employing
the innovation, the size of the investment required to use it, and
other unspecified variables" (p. 435). A number of studies
(Romeo, 1975; Zaltman, Duncan, & Holbik, 1973) expand, support, and
clarify but do not substantially alter Mansfield's model.
Size of firm, which has been shown to be positively related
to adoption (Kimberly, 1976; Rogers, 1983) could be considered an
economic variable. Brown (1981) suggests that size of company
serves as a surrogate measure for "... ability to raise capital,
to bear the cost of the innovation and to bear the risk of failure"
(p. 156).
In his review of the contributions of the economic model to
the understanding of diffusion, Warner notes that economic variables
do indeed explain, in part, patterns and rates of adoption. The
model is judged to have strong predictive powers when the innova-
tion is stable and when market conditions are fully operative. He
notes, however, that "Diffusion is a complex social phenomenon
which clearly involves both economic and noneconomic.factors"
(Warner, 1974, p. 438).
Structural model. The structural model of diffusion theory
focuses on structural characteristics of an organization. In line
with rational organization behavior theory, it suggests that


45
actions within an organization are influenced by structural vari-
ables.
In his review of the literature, Rogers (1983) lists and
describes six structural variables thought to be related to the
adoption of innovation:
Centralization is the degree to which power and control in a
system are concentrated in the hands of a relatively few indi-
viduals. . Complexity is the degree to which an organiza-
tions members possess a relatively high level of knowledge
and expertise. . Formalization is the degree to which an
organization emphasizes following rules and procedures. . .
Interconnectedness is the degree to which units in a social
system are linked by interpersonal networks. . Organiza-
tional slack is the degree to which uncommitted resources are
available to an organization. (pp. 359-360)
Size is the sixth structural variable. Complexity, interconnected-
ness, organizational slack and size are generally positively
associated with innovation, while centralization and formalization
appear negatively related to the adoption of innovation.
Tornatzky et al. (1983) focus on several of the same
variables including centralization, formalization, and complexity.
Zaltman and Wallendorf (1983) present these same three as key
variables to be considered. They suggest, however, that the
variables operate differently at different stages of the diffusion
process. In the initiation stage, complexity is thought to be
positively associated with innovation while formalization and
centralization are negatively associated. In the implementation
stage, the reverse is true. Complexity hinders implementation,
while formalization and centralization facilitate implementation
(p. 535).


46
There are several criticisms of the structural model. The
variables are difficult to define operationally. The structure of
units in a company may vary considerably, so it is difficult to
arrive at any overall rating. Finally, structural variables
frequently are not amenable to intervention. Work by Mintzberg
(1979) and Hage (1980) continue to refine this model for easier,
more accurate and consistent application.
Behavioral Perspective
Communications model. Within the behavioral and social
sciences research has resulted in the identification of variables
and models which complement the economic perspective outlined above.
Rogers (1962; Rogers & Shoemaker, 1971; Rogers, 1983) developed
and refined a communications model, adding social factors as key
determinants of the pattern and rate of adoption. This model
dominated the field of diffusion theory for many years as evidenced
by a 1971 review of the literature in which Rogers and Shoemaker
cite over 1,500 studies built around the communications model.
Diffusion research built on the communication model focuses
on factors which assist or impede the flow of communication and
ultimately contribute to the adoption or rejection of an innovation.
The variables, as summarized by Rogers (1983), include (a) source
of innovation, (b) attributes of the innovation, (c) channels of
communication, and (d) adopter characteristics.


47
Source. Much of the research related to the source of
innovation has focused on the requisite characteristics of change
agents. Extensive research has found that the most effective agents
are those viewed by the target population as credible, similar to
themselves and as having a client orientation or empathy (Rogers &
Shoemaker, 1971). Innovations introduced into a social system by
someone possessing these characteristics would be more acceptable
and diffuse more rapidly than those innovations introduced by
agents not perceived as credible, similar and empathetic.
Attributes. Likewise, it is believed that ". . the rate
of adoption of an innovation is positively related to the character-
istics of the innovation ..." (Rogers, 1985). Specifically those
attributes include:
(1) relative advantage, the degree.to which an innovation is
perceived as superior to the idea it replaces, (2) compati-
bility, the degree to which an innovation is perceived as
being consistent with existing values, past experience and
needs of potential adopters, (3) complexity, the degree to
which an innovation is perceived as difficult to understand
and use, (4) trialability, the degree to which an innovation
may be experimented with on a limited basis, and (5) observa-
bility, the degree to which the results of an innovation are
visible to others. (Rogers, 1984, p. 6)
Work by a number of researchers (Binham, 1976; Fliegel &
Kivlin, 1966; Mohr, 1969) has focused on characteristics of the
innovation in an attempt to develop predictive models. While this
focus does provide some useful information, Downs and Mohr (1976)
contend that no two organizations view an innovation in the same
way. Downs (1978) therefore suggests that the focus should be


48
more on how organizational characteristics influence perception of
the innovation.
Channel of communication. Pioneering work in this area was
done by Whyte (1954) who explored the impact of "word of mouth"
communication in the purchase of air conditioners. This was followed
by studies by Katz and Lazerfeld (1955) to determine the relative
effectiveness of various communication channels. A major finding
to come out of these and later studies was summarized by Rogers and
Shoemaker (1971): "Mass media channels are relatively more important
at the knowledge function, and interpersonal channels are relatively
more important at the persuasion function in the innovative-decision
process" (p. 255). In more recent work, researchers have introduced
the concept of an intermediate force "local media" which reach a
large geographic area, yet are more customized than mass media to
local target groups. More powerful in influence than mass media and
more accurate than interpersonal contact, it offers the potential
new channels for diffusion of innovation.
Characteristics of adopters. The majority of research
using the communication model has been directed toward understanding
more about the potential adopters of an innovation. The classic
categorization of adopters was done by Rogers and Shoemaker (1971,
p. 1982) and is presented in Figure 2.
The labels "innovation, early adopters, early majority, late
majority, and laggards" continue to be widely used in diffusion


1971, New York:
Nos. of People
Adopting per Period


50
research. Rogers and Shoemaker (1971) summarize salient character-
istics of each group as follows: . innovators-venturesome,
early adopters-respectable; early majority-deliberate; late
majority-skeptical; and laggards-traditional" (p. 192).
Reviewing the work of Rogers and Shoemaker and more recent
marketing studies, Midgley (1977) notes two salient adopter charac-
teristics which appear across all categories of innovation: willing-
ness to take risk and innerdirectedness. Yet he notes that "...
although there is a slight tendency for individuals to innovate in
closely-related product categories, the evidence did not suggest
the existence of generalized innovators" (p. 62).
While the work of Rogers and Shoemaker has focused largely
on individual innovative behavior, Zaltman and others have focused
on characteristics of organizations as adopters of innovation.
Zaltman and Wallendorf (1983) conclude that three structural
variables affect the adoption process: complexity, formality, and
centralization. They describe their impact as follows:
The initiation of new ideas, products, or services is facil-
itated by (1) high complexity, (2) low formalization, and
(3) low centralization . The implementation of new
ideas, products, or service is facilitated by (1) low com-
plexity, (2) high formalization, and (3) high centraliza-
tion. (Zaltman & Wallendorf, 1983, p. 534)
These generalizations were largely based on the research of
Chakrabarti and Rubenstein (1976) who reviewed the adoption of
NASA technolgoy by private industry.


51
Champion model. A second subset of research from the
behavioral school more closely examines the role of various actors
in the diffusion process. It amplifies portions of the communi-
cations model and extends it by looking at organizational roles as
well as individual predisposition to innovation.
Work done by Roberts (1969) on entrepreneurial behavior
found that entrepreneurs have high achievement and only moderate
power needs. Udell, Baker, and Albaum (1976) focused on inventors
and found them to be ". .more emotionally adventurous, achievement-
oriented, independent, resourceful, creative, and personally goals-
centered and hard-working than non-inventors" (Tornatzky &
Wallendorf, 1983, p. 102). Rogers and Shoemaker (1971, pp. 185-
189) identify 32 socioeconomic, communication and personality
variables related to innovativeness. In general, innovators tend
to be better educated, of higher social status, more empathetic/
less dogmatic, more abstract thinkers, favorably inclined toward
risk, and tend to have higher levels of achievement and higher
aspirations. In addition, they tend to participate more
socially, have more exposure to media and interpersonal communi-
cation, are more cosmopolite and more socially integrated than non-
adopters. Opinion leaders shared the same basic set of character-
istics, ". .we find that they are (1) more exposed to all forms
of external communication, (2) are more cosmopolite, (3) have higher
social status, and (4) are more innovative ..." Rogers (1983)
reiterates, "most individuals evaluate an innovation, not on the


52
basis of scientific research by experts, but on the basis of the
subjective evaluations by near-peers who have already adopted the
innovation" (p. 7).
While considerable work has been done to identify character-
istics of adopters at the individual level, much less research has
been devoted to exploring the impact that these actors have on the
adoption of innovation by organizations. Keller and Holland (1978)
indicate that persons with high job performance, high job satisfac-
tion and a penchant for new ideas, tend to act as boundary-spanners,
bringing innovation into an organization. Chakrabarti (1974) and
Keller and Holland (1978) have commented on the role of "gatekeepers"
in the adoption process. They typically are technical people
exposed to complex ideas and who have the skills to translate these
ideas, thus facilitating innovation diffusion. Chakrabarti (1974)
identified and studied the phenomenon of "product champions," indi-
viduals who promote a particular innovation throughout the diffusion
process. Several authors, including Mansfield (1971), have
attempted to examine how important individual actors have been to
the organizational process, while others, such as Hage and Dewar
(1973) contend that groups or coalitions are more common and more
important to the decision process.
Tornatzky et al. (1983) provide a succinct summary of the
current debate:
little work has been done in examining the relative importance
of individual factors as opposed to factors associated with
the environment in which innovation takes place, or the extent
to which situations determine which characteristics are important,
and which are not. (p. 109)


53
Systems Perspective
Marketing model. The economic and the behavioral models provide
useful perspectives for viewing the diffusion of innovation. Indeed,
research built on these models has added enormously to the under-
standing of the processes, elements and variables involved. Recently,
however, both the economic model and behavioral models have been
criticized for presenting an incomplete picture. Feller and Menzel
(1977) note that research has focused largely on internal character-
istics and adopter categories and has not adequately addressed the
milieu in which adoption takes place. Warner (1974, p. 449) notes
that prior models assume an adequate supply of and adequate knowledge
of the innovation which may not be the case. Brown (1981) criticizes
what he terms "the adoption perspective" (pp. 5-6). Much research,
he asserts, focuses on "the process by which adoption occurs, or the.
demand of diffusion" which, he states, "... implicitly assumes
that all have equal opportunity to adopt." Instead, he suggests
a market/infrastructure perspective which focuses on the supply
aspects of diffusion. This view holds that "... individual
behavior does not represent free will so much as choices within a
constraint set and that it is government and private institutions
which establish and control the constraints ..." (Brown, 1981,
p. 30). It raises the prospect that diffusion patterns may in large
part be explained by entrepreneurial actions rather than social
interactions. The major entrepreneurial actions discussed by Brown
are (a) improving access to the innovationsfrequently done in the


54
business world by increasing the number of outlets, (b) creative
pricing policy decisions to encourage experimentation, (c) promo-
tional communication consciously designed to reach the target popu-
lation, and (d) market segmentation which identifies and.sequen-
tially targets homogeneous subgroups in the population. Unlike
previous models it: (a) introduces access as an explanatory
variable, and (b) it views diffusion as an active rather than a
passive process in which it is incumbent upon the diffusing agency
to segment the market, manipulate pricing policies, increase promo-
tion of the product, and provide outlets. It is the thesis of Brown
(1981) that
the majority of innovations are promoted by . entities
termed propagators. These entities are profit or non-profit
motivated organizations or government agencies acting to
induce the rapid and complete diffusion of the innovation.
(p. 52)
Where propagating agencies or individuals stand to profit from the
adoption of an innovation by others, diffusion will occur more
rapidly. The presence of a profit, either economic or social, for
vendors becomes a critical and explanatory variable in diffusion.
This is in line with Merkle's hypothesis (Merkle, 1980) that
scientific management was introduced into industry not so much
because this innovation benefitted the adopting firms, but rather
because it benefitted upwardly mobile middle-class professionals
who propagated the idea.
Contextual model. The market/infrestructure perspective
presented by Brown is similar to the idea of "contextual influences"


55
presented by Tornatzky et al. (1983). In a review of research based
on this perspective, Tornatzky notes, "The integrating concept is
that all such variables are external to the innovating organization
but are assumed to affect intraorganizational processes" (Tornatzky
et al., 1983, pp. 77-78). Economic conditions, interactions among
firms, consumer behavior, government policies are among the many
factors to which innovating organizations must be sensitive. Kaluzny
and Riordan (1984) present a case in point, observing the imple-
mentation of diagnosis-related group (DRG) regulations promulgated
by the government for Medicare reimbursement to hospitals. This was
in a sense a mandated innovation. The ability of hospitals to
respond depended on both structural variables, attitudes of key
actors as well as general financial status.
The market/infrastructure or contextual models closely
resemble the systems theory perspective, emphasizing the importance
of an organization's interaction with its environment. As such, it
complements the rational/economic and behavioral/communication models
developed earlier. These models and perspectives suggest a wide and
varied array of variables which affect the diffusion of innovation.
It can be assumed, however, that the picture is not complete. The
structure of organizations, the economic climate and the relationship
of employees to the organization, and the external environment will
continue to change. In response, the factors which promote adoption
and implementation of new ideas can be expected to be subtly altered
as well.


56
The above review is not an exhaustive list of models in
diffusion theory, nor does it reflect the refinements and specific
applications of the models. Rather, it is a general framework of
selected models deemed most appropriate to this research effort.
Developing a Framework for Prevention
A number of studies have been conducted by researchers in
the field of health behavior. The studies have attempted to identify
psychological, socioeconomic, and cultural determinants of health
behavior.
Early research was done by Hochbaum (1958) to identify
factors associated with the decision to obtain an x-ray to detect
possible cases of tuberculosis. Kegeler (1963) examined the
conditions under which individuals sought preventive dental
services in the absence of symptoms. Leventhal, Hochbaum and
Rosenstock (1960) studied residents of two communities during an
influenza epidemic. The purpose of the study was to ascertain
psycho-social variables associated with taking the preventive
actions of becoming immunized against influenza. These and a number
of other studies form the basis of the Health Belief Model which
states that certain conditions increase the likelihood of an indi-
vidual's adopting preventive health practices. The conditions
include: (a) the individual's perceived susceptibility to a partic-
ular condition, (b) the perceived severity of the impending illness
or severity of the impending illness or condition, (c) the perceived


57
efficacy of action weighted against the barriers to action, and
(d) the presence of a "cue to action" (Becker, 1974) .
The Health Belief Model which focused on psycho-social
factors dominated the research on preventive health behaviors.
Other studies, however, began to focus on such issues as access
to health care services and demographic variables. In a meta-
analysis of the field, Cummings, Becker and Maile (1980, p. 135)
list 99 variables thought to influence the adoption of certain
preventive health practices. The variables were clustered in six
major areas: (a) health threat, (b) knowledge, (c) social
networks, (d) demographic characteristics, (e) evaluation of health
care, and (f) access to health care. Familiarity with variables
associated with a particular health behavior in individuals allowed
for more effective targetting and design of preventive health
programs.
This literature provides a rich and useful base to under-
standing health behavior. It includes numerous studies dealing
with compliance with medical regimens and a lesser number of studies
examining adoption of preventive practice. Unfortunately, in
almost all studies, the unit of analysis is the individual. This
literature base provides useful background, but it does not provide
a sufficient theoretical framework for the study of adoption of
preventive innovations by organizations.
Diffusion literature, by contrast, contains a sufficient
theory base dealing with the adoption of innovation by organizations.


58
Research on the diffusion of preventive innovations, however,, is
limited. Preventive innovations were defined by Rogers (1984) as
. new ideas that an individual or organization adopts at one
point in time in order to avoid the possible occurrence of an
unwanted consequence at some future time" (p. 22). Preventive
innovations, notes Rogers (1984), bring a distinct set of character-
istics and problems:
Most preventive innovations are characterized by a relatively
slow rate of adoption because potential adopters have diffi-
culty in determining an innovation's relative advantage, they
are not compatible with individuals' values and attitudes,
their cause and effect relationships are complex, trial is
difficult or impossible, and the innovations results are not
very observable because they are delayed. (p. 6)
This statement suggests that organizations might respond differ-
ently to worksite health promotion/disease prevention programs than
to other more tangible innovations. Yet a recent study examining
the adoption of Employee Assistance Programs (EAP) (NIAAA, 1984)
suggests that many of the classical processes and variables
applied. Personal communication demands, organizational autonomy,
service compatibility, exposure to a model program, and general
corporate concern for employees were all shown to be related to a
decision to adopt an EAP program. Interestingly, the authors state
(NIAAA, 1984) that "the single most important variable in explain-
ing policy adoption is the fact that the organizational key contact
admits to having an alcohol or drug problem in his/her social
network" (p. 15).


59
It is clear from this brief summary that more work needs to
be done to identify the processes and the variables which influence
the diffusion of preventive innovations in organizations.
Framework for Current Study
The current study will use diffusion theory as a framework
for examining the processes and variables related to the diffusion
of worksite health promotion/disease prevention programs. The study
must necessarily focus on only certain aspects of the process. A
complete understanding of all phases, actors and variables is
certainly desirable, but well beyond the scope of this study.
Unit of Analysis
The unit of analysis will be the organization. The study
will focus on organizational characteristics and processes which
influence the diffusion process. This decision is based on the
fact that the successful delivery of hp/dp programs at the worksite
typically require policy changes and resource allocation decisions
to be made at the organizational level. In addition, numerous
authors (Everly & Feldman, 1985; O'Donnell & Ainsworth, 1985;
Parkinson et al., 1982) cite organizational commitment as a neces-
sary first step and a key element to the development of successful
hp/dp programs. Further, given limited resources, it is likely that
the government's role in the future will be to stimulate the
development of worksite hp/dp programs by employers rather than
to serve as a direct service provider to employees. Therefore,


60
information regarding organizational processes of adoption is
expected to be more relevant to future public policy and program
initiatives.
Stage
As stated earlier in this chapter, the process of diffu-
sion is generally broken down into five stages. As presented by
Rogers (1983, p. 8), these stages include knowledge, persuasion,
decision, implementation and confirmation (see Figure 3). The
current study will focus on the decision stage, examining differ-
ences between companies which decided to adopt and those which
decided to not adopt an hp/dp program. Downs and Mohr (1976)
recommend focusing on this stage because "... the focus is thus
transferred from the tool itself to the social acts by which the
tool becomes part of the system" (p. 706). Feller (1978) concurs,
stating that the examination of "the decision to adopt" provides
a perspective on the interrelationship of the innovation and
adopter needs and processes. In addition, a decision by management
to adopt or not adopt an hp/dp program is a specific indentifiable
action culminating what might be a more diffuse, less documentable
process in the earlier stages of diffusion. Of necessity, informa-
tion will also be gathered about what happened in the awareness and
persuasion stages which "set the stage" for the eventual decision.
Implementation and confirmation are obviously important stages of
the total process, but time and resource constraints relegate the
consideration of these issues to future research.


Figure 3.
Stages in the Diffusion Process


62
Diffusion Models
Only limited research has been done regarding the diffusion
of worksite hp/dp programs. It would, therefore, be premature to
select a single model or set of hypotheses to test. Rather, the
study will be exploratory in nature. Questions will be designed to
elicit general information about the factors which affected the
decision to adopt or to not adopt a program. Some questions will
be structured around each of the major diffusion perspectives
reviewed in this chapter (see Figure 3). These questions will be
open-ended in nature so as not to preclude or predetermine which
variables were most important in the process.
Summary
In summary, diffusion theory is a rich and diverse body of
knowledge which has provided the framework for examining the rate
and pattern of adoption of innovations in society. A number of
theoretical models have evolved adding new and useful analytical
perspectives. Despite the volume of work done in diffusion, rela-
tively little attention has been focused on the adoption of pre-
ventive innovations by organizations. This research effort will
use diffusion theory as a framework for examining the adoption/
non-adoption of worksite hp/dp programs. The focus will be on the
decision phase of the diffusion process and the factors influencing
this decision. The unit of analysis will be the organization.
The research effort will be exploratory in nature attempting to


generate hypotheses for future research as well as validating
general hypotheses presented in diffusion theory.


CHAPTER III
RESEARCH METHODOLOGY
The purpose of the research project was to better under-
stand the process and the variables related to the adoption or non-
adoption of worksite health promotion/disease prevention (hp/dp)
programs. The study was largely exploratory in nature; it did,
however, include testing of selected hypotheses suggested by
general diffusion theory.
Twelve companies which had adopted worksite hp/dp programs
(adopters) were studied, along with 12 companies which had actively
considered but had not adopted such programs (non-adopters). The
research examined selected conditions, attitudes and practices
which existed in these companies at the time the decision was made
to adopt or not adopt an hp/dp program.
This type of study design is referred to in Campbell and
Stanley (1963) as a correlational, data analysis design with
X
X
1
2
0
0
X^ representing the decision in 12 companies to adopt an hp/dp
program and X2 representing the decision of 12 other companies to
not adopt a program. Both sets of companies were subsequently
examined in the same manner as represented by 0 in the figure.


65
The sample is matched for size of company, an organizational
variable known to be correlated with the adoption/non-adoption of
worksite health promotion programs (Davis et al., 1984; Fielding &
Breslow, 1983).
This study design was chosen because relatively little
research has been done regarding the diffusion of worksite hp/dp
programs and, therefore, the literature does not suggest specific
hypotheses to be tested formally. It is the intent of this project
to develop a better understanding of the context and the process by
which companies make the decision to adopt or not adopt hp/dp
programs, thus generating hypotheses for further study.
Previous Survey Sample
In 1983, the Colorado Department of Health and the Institute
for Health in collaboration with several voluntary health agencies
conducted a survey of worksite hp/dp programs in Colorado. General
purposes of the survey were: (a) to develop a profile of business
and industry hp/dp programs in Colorado, and (b) to identify
obstacles and incentives to the further development of such
programs.
A listing of all Colorado employers with 50 or more employees
was obtained from the Colorado Department of Labor and Employment.
The list contained company name, location by county, type of
business and number of employees. A stratified random sample was
drawn as follows:


66
Size of Company Number in Sample
50-99 employees 136
100-249 employees 149
250-499 employees 150
500-999 employees 141
1000 or more employees 103
Because of the small number of large businesses in the state, all
companies with 500 or more employees were included in the sample.
The 150 companies with 250-499 employees represent a 60% sample of
businesses of the size in the state, the 149 companies with 100-249
employees constitute a 15% sample, and the 136 businesses with 50-99
employees constitute a 7% sample of businesses of this size in the
state. A random sample was drawn in each of the remaining size
categories. Previous studies suggest that smaller companies are
less likely to have organized multiphased hp/dp programs. Thus,
no sample was drawn for the category of 50 or fewer employees. The
final sample included both private businesses and public agencies.
Trained volunteers contacted the chief executive office of
each company selected in the sample. The hp/dp survey was briefly
explained and interviewers then asked to be referred to the person
most knowledgeable about the company's efforts in these areas. A
phone call was then made to this identified contact person to deter-
mine: (a) if the company had an ongoing hp/dp program, or (b) if
the company was interested in developing such a program in the
future. If neither of these conditions existed, the interviewer
thanked the contact person and terminated the process. If either
of these conditions existed, the company was considered "eligible,"


67
and the contact person was asked if he/she would be willing to
participate in the project. A telephone interview was then
scheduled.
A packet including a questionnaire (see Appendix A), an
explanatory cover letter and a letter signed by the Governor of
Colorado was sent to each company contact person. Companies with
ongoing programs received a long form of the questionnaire.
Companies without programs, but interested in developing them in
the future, received an abbreviated form of the same questionnaire.
Ten days were allowed for receipt of the packet and the gathering
of program information by the contact person. The contact person
was then recontacted by the interviewer and a 15-25 minute inter-
view was conducted. '
Of the 679 companies in the original sample, 321 were
excluded for the reasons summarized in Table 1. Three hundred
interviews were completed from the sample of 358 eligible companies,
for an overall response rate of 83.8%. Of the 300 private businesses
and public agencies participating in the survey, 94 currently have
an hp/dp program. A company was considered to have an hp/dp program
if it provided health screenings, classes, and/or preventive health
services on an ongoing basis. Two hundred six companies did not
currently have hp/dp programs, but expressed interest in developing
such programs in the future.


68
Table 1
Companies Included/Excluded From Sample: 1983 Survey
Total sample drawn 679
Companies excluded 321
Reason for exclusion:
phone disconnected/no listing 54
repeats within list 76
wrong size/out of state 81
No program/no interest in developing program 110
Eligible companies 358
initial refusals 9
refusal after survey sent/incompletes 49
completed surveys 300
Companies with ongoing program 94
Companies interested in developing program 206


69
Number of Companies Responding
Existing Interested
Size of Company Program in Program
50-99 employees 2 33
100-249 employees 7 57
250-499 employees 19 42
500-999 employees 22 36
1000 or more employees 44 38
94 206
The survey examined: program goals, program activities,
company policies, program management, program outcomes, utilization
of community resources, and services needed to develop or expand
programs. Among the major findings of the study were the following:
-There is a high level of interest in hp/dp programs among
private businesses and public agencies. Nearly two-
thirds of companies contacted either had established a
program or were interested in developing such a program.
-Size of the company appears to be a major factor in
determining whether or not it has an hp/dp program.
Large companies, particularly those with over 500
employees, are much more likely to have programs than
are smaller companies.
-1978 marked the beginning of an upswing in the development
of worksite hp/dp programs. Over 70% of current programs
are less than 5 years old.
-There are a variety of reasons for starting hp/dp programs
including improved employee health/reduced illness, reduced
health care costs, improved morale, improved productivity,
and reduced turnover and absenteeism.


70
-Improved employee morale was perceived as the most common
benefit of hp/dp programs. Other commonly perceived
benefits include improved employee health/reduced health
problems, reduced illness and injuries on the job, and
increased productivity.
-The majority of programs are operated using in-house staff
and a variety of free community services. Few companies
plan any major purchase of hp/dp services.
While the survey provided an interesting profile, it did not
sufficiently explain the pattern of or reasons for adoption or non-
adoption of worksite hp/dp programs.
Sample for Current Study
The sample for the current study was a subset of companies
which had participated in the 1983 study. The term "company" as
used throughout the text refers to both private businesses and
public agencies. The companies were not selected at random, but
rather were taken from a list of companies which had participated
in the 1983 study and had signed a consent form indicating that they
would be willing to be contacted in the future. The expressed
willingness of these agencies and businesses to be contacted again
was considered key to the success of the proposed study. It was
assumed that these persons had a moderate level of interest in
hp/dp program and had given some thought to issues related to
program development. Of the 300 companies participating in the
1983 survey, 54 had signed the contact consent form.


71
The sample was further reduced by focusing only on companies
which had between 250-999 employees. As a group, companies within
this size range (a) account for 15% of the workforce nationwide,
(b) have not been studied extensively, and (c) have a reasonable
adoption rate that allows for examination of factors related to
adoption and non-adoption of hp/dp programs (Davis et al., 1984;
Fielding & Breslow, 1983).
The decision to exclude companies of 1,000 or more employees
was made on the following basis: (a) the adoption rate in this size
group is considerably higher than in the other size categories,
and (b) this size group has been the subject of most of the
research and program documentation in worksite hp/dp programs.
The decision to exclude companies with 100-249 employees was
based on a small eligible sample and the imbalance between adopters
and non-adopters. Of the 64 companies this size participating in
the 1983 survey, only 7. reported having active hp/dp programs. A
similar situation exists in companies with 50-99 employees. Of the
35 companies participating in the sample only two reported active
programs.
Of the 117 companies, with between 250-999 employees that
participated in the 1983 survey, 28 signed a consent form indicating
that they would be willing to be contacted in the future. This
group became the eligible sample. Their distribution was as
follows:


72
Companies With
Programs
Companies Interested
in Program
(Non-Adopters)
Size of Company (Adopters)
250-499 employees 4
500-999 employees 11
4
9
15
13
All 28 of these companies were contacted. Subsequent additions and
deletions to this group are summarized in Table 2. One company
with a program categorized as having 500-999. employees was dropped
from the sample when it was learned that they had a total of 2,600
employees with between 500-999 in a single location. The remaining
27 companies on the Consent to Contact list were contacted and
screened to determine whether they had been appropriately categorized
as adopters or non-adopters.
Companies were considered to have adopted an hp/dp program
1. they offered three or more health promotion activities
on a planned, on-going basis, and
2. the program was approved by management as a company
program/activity.
If companies did not meet these criteria, they were not
considered adopters. Two companies (500-999) listed as having
programs in the 1983 survey, did not meet these criteria. In both
instances, some activities were undertaken on a volunteer basis but
management had never given approval. Neither company considered
themselves to have an organized, approved hp/dp program. They were,
therefore, reassigned to the non-adopter category along with the
if:


other companies that were intererested in hp/dp but had not yet
begun a program.
73
This group was further screened to determine if they,
indeed, could be considered non-adopters. The following criteria
were established for designation as non-adopters:
1. the idea of hp/dp had been under active discussion by a
group within the organization, and
2. the idea has been presented to management and either
rejected or tabled.
These criteria were.established in order to select non-
adopters who had passed through the awareness stage and the
persuasion stage and had actually reached the decision stage in
the diffusion process.
Companies which were only in the awareness or early persuasion
stages were dropped from the sample. This resulted in the exclu-
sion of two companies with 250-499 employees and three companies
with 500-999 employees. In each case, the idea of an hp/dp program
had been the interest of the contact person for the 1983 survey,
but had not been actively discussed in the organization nor
presented to management. It was decided that information from
Awareness Sta
24 Companies
Persuasion Sta
24 Companies
Decision to Adi
12 Companiei
Decision to Re.
Table 12 Comp a:


74
these companies would be extremely limited in nature. Of all the
companies contacted, one refused to be interviewed.
In order to maintain a balance between companies with
programs and companies interested in programs, three replacements
were drawn at random in the appropriate size category from the
original list of companies which had participated in the 1983
survey. These companies were contacted, met the criteria, and
agreed to become part of the final sample. Thus the final sample
was constituted as follows:
Size of Company
Companies With
Programs
(Adopters)
Companies Interested
in Program
(Non-Adopters)
250-499 4 4
500-999 8 _8
12 12
Of the 24 companies participating, 14 were private businesses and
10 were public agencies. Of the 12 adopters, 7 were private
businesses and 5 were public agencies. Of the 12 non-adopters, 9
were private businesses and 3 were public agencies.
Adopters Non-Adopters Total
Public Agency 5 38
Private Business 7 9 16
12 12 24
Of the adopters, 4 were involved in some aspect of health care
delivery, 8 were not. Of the non-adopters, 7 were health care-
related, 5 were not.
Within each company, an effort was made to identify and
interview persons representing: (a) administration, (b) human
resources/benefits, and, if possible, (c) health services. The


75
Table 2
Companies Included/Excluded from Sample: Current Study
Adopters Non-Adopters
Number of companies (250-999)
Participating in 1983 Survey
and Sumitting Consent to Contact
Forms 15
Alterations to Original Sample:
Reason Deletion of Addition
Company too large to be in sample -1
Company misclassified/did not yet
have program -2
Company not yet in decision stage
of adoption process -
Refused to participate -1
Replacement drawn from initial
survey list -
Company misclassified/had recently
begun program +1
13
+2
-5
+3
-1
12
12


76
decisionto focus on persons from administration and human resources/
benefits was based on the findings of the 1983 Colorado survey
(David et al., 1984) and a similar survey done in California
(Fielding & Breslow, 1983) which indicated that persons from top
administration and the human resource/benefits department were most
likely to be involved in decisions to offer or not offer worksite
hp/dp programs.
A total of 58 persons were interviewed, 29 each from
adopting and non-adopting companies. A minimum of two persons were
interviewed in each company with three persons being interviewed in
10 companies.
Positions of Persons Interviewed N %
Top Management: General 10 17.2
Top Management: Human Resources 7 12.1
Human Resources: Non-Management 17 29.3
Health Service/Health Education 24 41.4
Total 58 100.0
A sizeable number of interviewees (41.4%) involved persons from the
human resources departments. Many of these persons (7 out of 24)
were also considered to be top management. Twenty-four (41.4%)
persons interviewed were from employee health service and/or health
education units. Ten persons (17.2%) were from top management
including chief executive officers (CEOs), directors of marketing,
assistant superintendents and/or board members.
Survey Instrument
A review of the worksite hp/dp programs and of the diffusion
literature was conducted to develop a conceptual framework for the


77
study and to determine if validated instruments existed which could
be used. Extensive research has been done in diffusion, yet none
of the studies dealt specifically with worksite hp/dp programs; no
relevant, validated instruments appeared in the literature. Experts
in the field of diffusion and health promotion research, Dr. Everett
Rogers, Dr. Gregory Christenson, Dr. J. D. Eveland, and Dr. Donald
Iverson, were then contacted to determine if instruments in this
area had recently been developed. Correspondence and personal
conversation with them confirmed that no such instruments existed.
Further, they suggested that, in view of the limited knowledge base
in the area, that the research focus and the survey instrument be
designed to elicit information about the process of diffusion as
it occurred or did not occur in the companies studied. As a result,
instrumentation was developed suited for use in a guided interview
(see Appendix B).
The guided interview questionnaire included the following
sections:
Section A: The purpose of this section was to establish the
current level of program activity and the involvement of different
departments in planning/developing the program.
Section B: The purpose of this section is to examine how
the interest in hp/dp initially developed, the factors which
promoted and/or discouraged the development of a program, and the
administrative environment in which the program developed.


78
Section C: The purpose of this section is to present
general hypotheses about factors thought to influence the adoption
of hp/dp programs, and to verify or refute these factors as
important in the adoption decision process of companies.
Each section contains a series of open-ended questions
including probes to be used to elicit additional information if
necessary. The relationship between diffusion theory and specific
questions is presented in Figure 4. The intent and rationale for
each item on the questionnaire is presented below.
Question 1: Would you please describe the health promotion
activities which your company offers for its employees.
This question was designed to determine whether the
company should be classified as an "adopter" or "non-adopter" based
on criteria discussed on page or whether it should be excluded
from the eligible sample.
Question 2: Would you talk a little bit about the parts of the
program which have been particularly successful.
This question was included to put people more at ease in
talking about their program. It was not designed to test any
hypothesis or gather information on any variable, thus criteria
for "successful" and "less successful" were purposely not specified.
Question 3: Can you describe the involvement of different depart-
ments and people in developing the program (or) in
planning for a program?
This question was used to determine the position of the
interviewer in the organizational structure. These data would


79
RATIONAL/ECONOMIC PERSPECTIVE
Economic Model
Primary: Question 11
Secondary: Questions 5 & 6
Structural Model
Primary: Question 9
Secondary: Question 3
BEHAVIORAL PERSPECTIVE
Communications Model
Primary: Questions 7 & 8
Secondary: Question 2
Champion Model
Primary: Question 10
Secondary: Questions 5 A 6
SYSTEMS PERSPECTIVE
Marketing Model
Primary: Question 3
Secondary: Question 5 & 6
Contextual Model
Primary: Questions 5 A 6
Secondary: Question 11
Figure 4. Inclusion of Diffusion Perspectives in Interview
Guide: Primary and Secondary Focii


80
allow for comparison of responses by management level or area of
responsibility. The question also provided the interviewer with a
sense of the extent to which the interviewee had been involved with
the adoption decision process and was therefore knowledgeable about
the relevant variables.
Question 4: When did you first start talking about a program?
If appropriate ... In what year did you actually
start the program?
This question was designed to focus the inerviewee's
attention on the time span prior to program implementation. It
also provided information as to the length of time companies had
been in the awareness and persuasion stages.
Question 5: How did interest in employee health promotion first
develop in this company?
This question was included as an open-ended question to
allow respondents to identify or speculate on the variables that
influenced the diffusion process. The focus of the question is
on the awareness stage of the diffusion process.
Question 6: When the idea, of employee health promotion was pre-
sented to management or a management committee, what
were some of the issues that came up, both "pro" and
II _HO
con :
This question provides a similar opportunity for respond-
ents to identify or speculate on key variables which arose in the
persuasion stage of the diffusion process.


81
Questions 7 and 8: Can you describe some of the first activities
you decided to offer (or) the first activities you
would offer? and What were some things that you had
considered doing, but backed off from (or) What are
some things which you would back away from?
These questions provide an open-ended opportunity for
respondents to identify product variables considered important in
the decision to adopt or not adopt. The information from this
question was also used to classify the proposed programs as being
simple or complex.
Question 9; Once the proposal for a program was written, what was
the process for approval in this company (or) would be
the process for approval in this company?
This question addressed select organizational variables
hypothesized to influence the decision process, including the
formalization and centralization of the decision process. In
addition, information was generated regarding the general management
philosophy.
Question 10: Can you describe key individuals who strongly supported
your program?
This question was designed to test the champion model as
well as the related variables of staff expertise and top management
support.


82
Question 11: What kinds of discussions (if any) did you have
regarding potential costs or potential savings for
the company?
This question assessed the extent to which financial factors
had played a role in the decision process. Information was also
used to examine corporate financial status and length of return
on investment.
Question 12: What was your experience when you were shopping
around? Were you able to find programs or activities
acceptable to you?
This question assessed the importance of program avail-
ability as a factor influencing the decision process. Information
was also sought on design characteristics which made program
components more attractive to prospective adopters.
Question 13: Can you describe seminars you attended, articles you
read or visits that you received from individuals
promoting health promotion?
This question focused on the exposure of mmnagement to
information on hp/dp programs as a variable in the diffusion
process. This included general level of exposure through conferences,
written materials and sales visits as well as exposure to any specific
model program.
Questions 14 and 15: Is there any single factor that we have
talked about or not talked about that you think has
been the key element in this company? and Do you have
any comments you would like to make?


83
These questions provide the opportunity for respondents to
elaborate on any or all of the variables discussed and identify the
single variable they felt most influenced the decision process.
The survey instrument was developed and sent to committee
members and outside experts for review. They suggested: (a) that
fewer, more open-ended questions be used, utilizing probes to
gain the more specific information, (b) that the sequencing of
questions be changed, and (c) that clearer, more straightforward
language be used in the questions. These suggestions were subse-
quently incorporated.
The design of the survey instrument followed the model
suggested by Patton in Qualitative Evaluation Methods (1980) The
instrument was piloted with three colleagues, two of whom were
familiar with issues in the field of worksite hp/dp programs.
Minor changes in wording were made and the instrument was subjected
to a second pilot involving five individuals from two companies
one company which had an hp/dp program (adopter) and one company
which had not (non-adopter). Only minor revisions were necessary
as a result of the second pilot test. Finally, the dialogue pre-
ferred to be used in the initial phone contact with companies (see
Appendix C) was established and piloted on three persons in three
different companies.
Data Collection
The sample of companies to be included in the survey was
based on representatives from companies 250-499 or 500-999 who had


84
participated in the 1983 survey and agreed to be contacted again.
Each of these persons was contacted by telephone and asked if they
would be willing to participate in the study. Of the 24 companies
participating in the study, 14 had the person contacted in the 1983
survey. In the remaining 10 companies, upon learning that the
original contact person was no longer there, the caller asked to be
referred to the person in the company who was most knowledgeable
about the development of the hp/dp program in the company or had
been involved in promoting or planning the program. These persons
were identified as the contact person for this study. Contact
persons were then asked to identify two other persons representing
other company perspectives (either management, human resources, or
health services) who had been involved in promoting and planning
the hp/dp program. Personal interviews were then scheduled.
Interviews of all persons identified were conducted by a
single interviewer. The interviewer used the survey instrument and
a guided interview process. The interviews were tape-recorded and
required between 20 to 40 minutes. In all but two instances, the
interviews were conducted in person. Persons in one company were
interviewed by telephone; the information collected was complete
and the quality comparable to that gained in personal interviews.
Several items on the questionnaire, specifically probes in
questions 11 and 12 (see items marked with "D" in Appendix B),
proved to be confusing to interviewees and did not yield consistent,
usable data. To the question "Did you see the program as costing


Full Text

PAGE 1

WORKSITE HEALTH PROMOTION/DISEASE PREVENTION: A STUDY IN THE DIFFUSION OF INNOVATION by Mary Foecke Dayis B.A., University of Nebraska, 1965 M.P.H., University of Minnesota, 1969 A thesis submitted to the Faculty of the Graduate School of Public Affairs of the University of Colorado in partial of the requirements for the degree of Doctor of Public Administration Graduate School of Public Affairs 1985

PAGE 2

Copyright by Mary Foecke Davis 1985 All Rights Reserved

PAGE 3

This thesis for the Doctor of Public Administration degree by Mary Foecke Davis has been approved for the Graduate School of Public Affairs by Eileen A. Tynan Donald C. Iverson Thomas M. Vernon, Jr. John D. Parr Date ______________________ __

PAGE 4

Davis, Mary Foecke (D.P.A., Public Administration) Worksite Health Promotion/Disease Prevention: A Study in the Diffusion of Innovation Thesis directed by Assistant Professor Eileen A. Tynan The purpose of this study was to examine the process and the variables related to the adoption of innovative health programs by organizations. The study utilized diffusion theory as the conceptual framework and focused on the adoption/non-adoption of worksite health promotion/disease prevention (hp/dp) Twelve companies which had adopted hp/dp programs (adopters) were studied along with 12 companies which did not adopt such programs (non-adopters). Personal interviews, following a guided interview format, were conducted with a total of 58 persons. The study explored attitudes and practices existing in companies at the time of the decision to adopt or not adopt an hp/dp program. Selected hypotheses suggested by diffusion theory were tested. Information from the interviews was coded and analyzed using discriminant analysis. A number of variables suggested by diffusion theory predicted adoption/non-adoption of hp/dp programs. Explanatory variables derived from the rational/economic, behavioral, and systems perspectives of diffusion. Programs were more likely to develop in response to a perceived opportunity than in response to a perceived threat. A

PAGE 5

v convergence of internal and external factors resulted in initial interest in hp/dp. Finances, benefits, and company philosophy vis-a-vis responsibility for employees were common criteria used in the decision process. When finances became the major criterion, it was unlikely that a company would adopt a program. Programs were likely to occur only in companies where management was actively supportive, had a humanistic philosophy, and espoused expanded responsibility for employees. The majority of variables positively associated with hp/dp programs are likely to change only over a long period of time. Exposure to model programs, however, a mechanism for inter-vention. These findings suggest that proponents of hp/dp can effec-tively identify companies likely to adopt programs and develop effective policy and programmatic efforts. Hypotheses were identi-fied for further study. The form and content of this abstract are approved. I recommend its publication., Signed ________ Faculty member in charge of thesis

PAGE 6

ACKNOWLEDGMENTS I would like to acknowledge the assistance of the many people without whom this study would not have been possible: Eileen Tynan, who has given me steady support and sound advice throughout the doctoral program; Don Iverson, who has guided this project as well as my professional career, Tom Vernon, whose inquiring mind and commitment to prevention have provided inspiration; John Parr, whose practical perspective kept the study on track; Karen Rosenberg, whose diligent efforts on the initial survey made this follow-up study possible; Ed Fryes who made discriminant analysis look easy; Gregory Christenson, who critiques the methodology and instruments; and Toby Cohen, of A&D Copy & Typing Service, who took a scribbled manuscript and made it a beautiful finished product. I would like to thank the persons interviewed in the 24 companies who willingly shared their time and experiences. They made this project a pleasant learning experience for me. Lastly, I would like to thank my family for the support vi they have given me and the sacrifices they have made: Mike, who assumed more than his share of housework and hassles over the past months; Andrea, whose chats broke the monotony of writing; Nena, whose hugs gave me energy to continue; and Molly, who had to give up Saturday morning cartoons in the study, and many bedtime stories.

PAGE 7

CHAPTER I INTRODUCTION.. Rationale for the Study ''''orksite Health Promotion/ Disease Prevention Historical Development CONTENTS Current Program Parameters Current Level of Program Activity Frequency of Program Components Rate of Growth Program Outcomes Hypertension Smoking cessation Employee assistance program Exercise/fitness Summary CHAPTER II REVIEW OF THE LITERATURE Historical Development Definitions Diffusion Models and Perspectives Rational/Ecnonomic Perspective Economic model 1 1 4 4 11 15 18 21 24 24 26 28 30 33 35 35 38 42 42 42

PAGE 8

Structural model Behavioral Perspective Communications model Source Channel of communication Characteristics of adopters Champion model Systems Perspective Marketing model Contextual model Developing 'a Framework for Prevention Framework for Current Study Unit of Analysis Stage Diffusion Models Summary CHAPTER III RESEARCH METHODOLOGY Previous Survey Sample Sample for Current Study Survey Instrument Data Collection Data Analysis .. :viii 44 46 46 47 47 48 48 51 53 53 54 56 59 59 60 62 62 64 65 70 76 83 85

PAGE 9

CHAPTER IV RESULTS OF THE STUDY . . . Current Health Promotion/Disease Prevention Program Activities Program Components Successful/Unsuccessful Activities Departments Involved Program History Development of Interest Decision Criteria and Processes Finances Benefits Corporate responsibility for health of employees Corporate image/liability .. Activities proposed/activities deferred Complexity of proposed programs Decision process General Company Environment Management philosophy Facilities Hypothesis Testing Human Elements Champions/angels Staff expertise Economic Issues ix 89 91 91 91 93 93 93 97 97 98 101 104 107 109 109 111 111 116 119 119 119 124 126

PAGE 10

Financial status Return on investment Marketing/Access Issues Program availability General level of exposure Exposures to Model Program Discriminant Analysis Mo.del Summary CHAPTER V CONCLUSIONS AND RECOMMENDATIONS Limitations of the Study Retrospective Bias Focus on Selected Variables Classification Bias Decision Focus Non-Random Sample Discussion of Findings Targetting Diffusion Efforts Developing Positive Triggers Addressing Management Criteria Reaching Key Actors Avenues for Change Government Role Further Research Issues Summary . x 127 127 131 131 132 135 138 141 147 147 147 147 148 149 149 149 149 151 152 153 156 156 160 161

PAGE 11

BIBLIOGRAPHY .. APPENDIX A. BUSINESS AND INDUSTRY. SURVEYS B. INT,ERVIEW GUIDE C. DIALOGUE USED WITH CONTACT PERSONS D. INTERVIEW RECORD FORM ....... xi 165 174 195 201 216

PAGE 12

Table 1. TABLES Companies Included/Excluded From Sample: 1983 Survey '.' 2. Companies Included/Excluded From Sample: Current Study .... 3. Source of Information on Variables 4. Classification of Companies on Variables Examined in Study 5. Association Between "Importance of Finances" Variable and the Decision to Adopt or Not Adopt a Worksite Health Promotion/Disease Prevention Program 6. "Importance of Finances" Variable asa 'Predictor of Adoption/Non-Adoption rif Worksite Health Promotion/Disease Prevention Program . ... 7. Association Between "Importance of Benefits" Variables and the Decision to Adopt or Not Adopt a Worksite Health Promotion/Disease Prevention Program 8. "Importance of Benefits" Variable as a Predictor of Adoption/Non-Adoption of Worksite Health Promotion/Disease Prevention Program . 9. Association Between "Responsibility for Employee" Variable and the Decision to Adopt or Not Adopt a Worksite Health Promotion/Disease Prevention Program 10. "Responsibility for Employees" Variable as a Predictor of Adoption/Non-Adoption of Worksite Health Promotion/Disease 68 75 87 90 99 100 102 103 105 Prevention Program . . . . . 106

PAGE 13

Table 11. 12. Association Between "Program Proposed" Variable and the Decision to Adopt or Not Adopt a Worksite Health Promotion/Disease Prevention Program Association Between "Decision Process" Variable and the Decision to Adopt or Not Adopt a Worksite Health Promotion/ Disease Prevention Program l3. Association Between "Management Philosophy" Variable and the Decision to Adopt or Not Adopt a Worksite Health Promotion/Disease Prevention Program . 14. "Management Philosophy" Variable as a Predictor of Adoption/Non-Adoption of Worksite Health Promotion/Disease Prevention P.rogram . 15. Association Between "Facilities" Variable and the Decision to Adopt or Not Adopt a Worksite Health Promotion/Disease 16. Prevention Program "Facilities" Variable as a Predictor of Adoption/Non-Adoption of \.Jorksite Health Promotion/Disease Prevention Program 17. Association Between "Champion" Variable and the Decision to Adopt or Not Adopt a Worksite Health Promotion/Disease Prevention Program 18. Association Between "Angel" Variable and the Decision to Adopt or Not Adopt a Worksite Health Promotion/Disease . .. . xiii 110 112 114 115 117 118 121 Prevention Program . . . 122 19. "Angel" Variable as a Predictor of Adoption/Non-Adoption of Worksite Health Promotion/Disease Prevention Program 123 20. Association Between "Staff Expertise" Variable and the Decision to Adopt or Not Adopt a Worksite Health Promotion/ Disease Prevention Program 125

PAGE 14

Table 21. 22. 23. 24. 25. Association Between "Financial Status" Variable and the Decision to Adopt or Not Adopt a Health Promotion/Disease Prevention Program . . . . "Financial Status" Variable as a Predictor of Adoption/Non-Adoption of Worksite Health Promotion/Disease Prevention Program . Association Between "Program Availability" Variable and the Decision to Adopt or Not Adopt a Health Promotion/Disease Prevention Program . . . Association Between "General Exposure" Variable and the Decision to Adopt or Not Adopt a Worksite Health Promotion/ Disease Prevention Program . Association Between "Exposure to Model" Variable and the Decision to Adopt or Not Adopt a Worksite Health Promotion/ Disease Prevention Program . 26. "Exposure to Model" Variable as Predictor of Adoption/Non-Adoption of Worksite Health Promotion/Disease Prevention Program 27. Discriminant Analysis Model Using xiv 128 129 133 134 136 137 Multiple Variables . 139 28 .. Multiple Variables as Predictors of Adoption/Non-Adoption of Worksite Health Promotion/Disease Prevention 29. 30. Program Discriminant Analysis Using Accessible Varible . ..... Accessible Variables and Predictor of Adoption/Non-Adoption of Worksite'Health Promotion/Disease Prevention Program 140 142 143

PAGE 15

Table 31. Variables Associated with the Decision to Adopt or Not Adopt. a'. Worksite Health Promotion/Disease Prevention Program 32. Variable Not Associated With the Decision to Adopt or Not Adopt a Worksite Health Promotion/Disease xv 144 Prevention Program . 145

PAGE 16

FIGURES Figure 1. Perspective in Organizational Behavior Related to Models in Diffusion Theory 2. Adopter Categories 3. Stages in the Diffusion Process 4. Inclusion of Diffusion Perspectives in Interview Guide: Primary and Secondary Focii 43 49 61 79

PAGE 17

CHAPTER I INTRODUCTION The purpose of this study is to examine the process by which innovative health programs are adopted by organizations. The study focuses specifically on the diffusion of health promotion disease prevention (hp/dp) programs at the worksite. Rationale for the Study Each year vast amounts of human and fiscal resources are dedicated to the development of new products and services. Yet, only a relatively small percentage of these products and services will ever be utilized by society. The.vast majority survive less than a year and reach only a fraction of the potential population (Rogers & Shoemaker, 1971). Failure of programs, products, and services to be widely utilized is a phenomenon experienced by the public as well as the private sector. Government reports are replete with examples of new techn'ologies which have been developed and/or promoted by public agencies but have not been adopted by society. An example of such a problem is worksite health promotion/disease prevention (hp/dp) programs. In recent years the government, as well as the private and the non-profit.sector, have invested a substantial amount of money

PAGE 18

in health promotion/disease prevention. Research has been done to establish the link between lifestyle and major causes of death and disability. Strategies in smoking cessation, high blood pressure control, fitness, alcohol/drug abuse prevention have been developed and evaluated. To reach the adult working population with these programs, attention has been focused on the worksite. Funds have been allocated for the development and evaluation of exemplary worksite hp/dp programs, conferences have been held for key business and industry leaders; and numerous agencies and groups have issued policy statements promoting the worksite as an appropriate site for prevention activities. Foundations, business coalitions, as well as state and local public agencies have undertaken a range of activities to encourage businesses to worksite hp/dp programs (Berry, 1981; Healthy People, 1979; Parkinson & Associates, 1982; Sehnert & Ti11otson,1978). 2 The response from business and industry, however, has been less than overwhelming (Davis, Rosenberg, Iverson, Vernon, & Bauer, 1984; Fielding & Breslow, 1983). A number of exemplary programs have been developed by large employers such as IBM, Honeywell, Control Data, Johnson & Johnson. Yet the total number of programs appears to remain relatively small. Few inroads have been made into the vast majority of businesses. Nationwide, it is estimated that fewer than 5% of businesses offer hp/dp programs (Warner &. Murt, 1984). In order to understand this limited success, and to better tailor public policies and programs, more information is needed on the

PAGE 19

factors which influence the adoption/non-adoption of worksite hp/dp programs. 3 In general, diffusion theory. has provided a useful framework for examining the adoption of new products and services by society. Within the framework, numerous social, economic, structural and marketing variables have been identified which are thought to explain the rate and pattern of diffusion of innovation. Models have been constructed and used to study patterns of acceptance of a range of products and services, including, among others, educational innovation, improved farming techniques, household products and new health care services. The framework and the principles therein aid analysis of factors related to adoption/non-adoption of innovation and provide guidance for administrators wishing to structure their products and services for maximum diffusion. Worksite hp/dp is an innovation--an innovation which has been accepted on only a limited basis. It would appear that diffusion theory is applicable and appropriate to the study of this phenomenon. The intent of this research project is to use diffusion theory as a framework within which to examine the factors related to the adoption/non-adoption of worksite hp/dp programs. It is hoped that the findings will provide public and other administrators with information useful in the formulation of policy and in the design of future hp/dp program efforts.

PAGE 20

4 Worksite Health Promotion/Disease Prevention Historical Development Interest in the health of employees is not a new phenomenon for American business. As early as 1871, Metropolitan Life Insurance developed health and safety information packages for employees and policyholders. After the turn of the century, other insurance companies joined in the production of a series of pamphlets (Follman, 1978, p. 21). Organized health education efforts within industry, however, are a relatively recent phenomenon. The first industrial alcoholism program was started in 1942 by E. I. DuPont ./ 'de Nemours and Company. Within a short time, several other major companies, including Eastman Kodak, Allis-Chalmers, Consolidated Edison, Armco Steel, and Western Electric had begun similar programs (Dunkin, 1982, p. 7). A survey done 1950 by .the National Association of Manufacturers revealed that nearly 80% of companies with' 2,500 or more employees reported a health education component in their industrial health program. In smaller companies with fewer than 250 employees, the rate dropped to 21%. A survey completed of manufacturing firms with fewer than 100 employees revealed that fewer than 3% had health education programs (USPHS, 1958). Throughout the 1960s, a number of model programs were developed including a fitness program at NASA. The Chicago Heart Association developed a high blood pressure screening and referral program which it operated at numerous industrial sites in the Chicago area (Alderman, Green, & Flynn, 1980, p. 106). A survey

PAGE 21

of 400 members of the Industrial Medical Association revealed that two-thirds of member companies had some type of program, mostly individual counseling (Cassuto, 1967). A more general survey conducted by the Conference Board in 1972 reported that, of companies with 500 or more employees, approximately one-third conducted some type of health education program (Lusterman, 1974, p. 149). The majority of the programs focused on individual counseling, information in newsletters, and noon seminars on specific topics of interest. Size and type of company appeared strongly correlated with the existence of a program. Companies with 5,000 or more employees were more likely to have programs (58%) as opposed to companies with fewer than 1,000 employees (15%). Fifty-two percent of manufacturing firms and 42% of financial institutions offered health education, while only 10% of wholesale companies reported such activities (Lusterman, 1974, pp. 49-50). On commenting on the development of these early programs, Nickerson (1967) notes that the majority of them tended to be "one-shot" in nature; company interest and involvement was minimal. Ware (1983) states that most efforts seem to have been initiated by someone outside the agency, an insurance company, or voluntary health agency. Companies were recipients, but not active participants in health education. During this time period, the Occupational Safety and Health Act (OSHA) of 1970 was passed by Congress. It required "education and training of employers and employees in the recognition, avoidance, and prevention of unsafe or unhealthful 5

PAGE 22

working conditions" (P.L. 95-596, Sec. 21(c. Consequently, ,a number of firms strengthened their safety education efforts, but did not alter the scope or content (Ware, 1983). 6 The mid-1970s marked the beginning of a period of significant growth and change in worksite health education programs'. Several major business organizations, including the National Chamber Foundation, the Health Insurance Association of America, and the Washington Business Group of Health developed and disseminated documents urging companies to implement a wide range of educational activities for their employees (Berry, 1982; Sehnert & Tillotson, 1978). The Office of Health Information/Health Promotion within the federal government held the first national conference to focus attention on health promotion in occupational settings. The Department of Health and Human Services issued grants to assist in the identification and evaluation of model worksite programs. Ware (1983) reports that the number of jouTIla1 articles dealing with occupational health education increased substantially in the early 1970s. A number of texts specific to worksite health education/health promotion were published (Cunningham, 1982; Everly & Feldman, 1985; O'Donnell & Ainsworth, 1984; Parkinson et a1., '1982). Perhaps more importantly, an effort was made to change the nature of health education efforts from programs provided by community resources .to programs in which the company actively particip'ated, designing programs based on the needs and risks of their employees. In 1980, Green, Kreuter, Deeds, and Partridge

PAGE 23

developed a PRECEDE framework which emphasized a diagnostic approach to program design and implementation (Green et al., 1980). The model begins with diagnosis of social problems to determine the extent to which health problems contribute to the failure of society to reach specified goals. Further analysis is then done to determine behavioral factors which contribute to the identified health problem and the extent to which these factors are changeable through educational intervention (see Figure 1). The purpose of the model is to assist in accurate problem diagnosis which, in turn, will lead to more effective program design. This diagnostic approach to program planning is reflected in the work of Berry (1981), Parkinson et al. (1980), Behrens (1983) and others who apply it to issues at the worksite. They suggest a process whereby employers identify health problems of employees and determine the extent to which these problems contribute to the larger concerns of business such as increased insurance premiums, increased absenteeism, and lowered productivity. Analysis is then c6nducted to determine the changeability of factors which contribute to the health problem. For example, an industry may be experiencing absenteeism. Ari analysis reveals that the majority of absences are due to back injuries which result from improper lifting techniques. A combination of educational programs and environmental adjustments can be made which will substantially reduce the problem. The diagnostic approach results in the identification of problems of concern to the organization and may result 7

PAGE 24

in a wide range of strategies designed toreduce the problem. It counters the categorical, externally imposed program model characteristic-of earlier program efforts. The term "health promotion/disaese prevention" began to replace the tenn "health education." Health education was defined by Green et al. (1980) as "any combination of learning experiences designed to facilitate voluntary adaptations.of behavior conducive to health" (p. 7). The term "disease prevention" was added to encompass screenings and programs aimed at preventing certain diseases. The tenn health promotion/disease prevention was considered a better description of more comprehensive program efforts being advocated by leaders in the field. 8 Two factors appear to have contributed to the increased interest and the new direction of worksite health promotion/ disease prevention (hp/dp) programs: (a) rising cost of illness, and (b) increasing appreciation of lifestyle as a contributor to premature death and disability in the population. A report done for the Health Insurance Association of America reported in that 1960, total medica], care expenditures were roughly $27 billion (5.3% of the Gross National Product). By 1980, his total had risen to $243 billion (9.4% of GNP). It was estimated that businesses currently paid over half this amount (Berry, 1981, p. 8). Parkinson et al. (1982, p. 1) projected that, if unchecked, this figure could reach $462 billion by 1985. lations on cost of illness may be expanded to include indirect costs

PAGE 25

such as absenteeism, decreased productivity, turnover, as well as the direct cost of medical care and Workmen's Compensation. Recent estimates presented by Fielding p. 240) suggest that indirect costs of illness are nearly three times the direct cost of illness. If this ratio holds, the total of direct and indirect costs of illness in the u.s. could reach a staggering trillion dollar mark by 1985. Robert Beck (1982), examining the experience at IBM, notes that major medical costs have risen at between 11 and 14 percent compound growth rate, year in and year out, with no plan improvements" (p. 3). The overall emp1oyee.hea1th costs at IBM were approximately $1,000 per employee, per year, and was considered a cause for concern. There has been no national survey which has conclusively documented rising cost of illness as the key factor in the growth 9 of interest in worksite hp/dp programs. Yet, preliminary evidence would indicate that it is important. Most major publications and presentations which focus on worksite health promotion, begin wtih an examination of rising health care costs. Documents produced by the National Chamber of Commerce (Tillotson & Rosa1a, 1978) cite rising health care costs as one reason to begin a worksite program. A survey of businesses in Colorado revealed that nearly 60% of companies cited cost containment as a very important reason for starting an hp/dp program (Davis et a1., 1984). In a summary statement, Fielding (1979) suggests that while cost-benefit evidence is incomplete, the potential for controlling health care costs

PAGE 26

through worksite hp/dp programs has been a major factor driving industry activity in thisarea. Changing disease patterns has been the second major factor focusing attention on worksite hp/dp programs. In the early part 10 of the century, the bulk ofpremature death and disability was due to communicable diseases. Following the improvements in sanitation and the introduction of better medical care, this pattern shifted dramatically. Chronic diseases such as heart disease, cancer, and stroke began to replace tuberculosis, influenza, diphtheria, and gastro-intestinal infections as the major health problems of the population. In the mid-1970s, the U.S. Department of Health, Education, and Welfare (HEW) undertook more in-depth analysis of these disease patterns. Borrowing the "health field concept" developed by the Cana4ian Ministry of Health and Welfare (LaLonde, 1974), HEW examined the extent to which. health problems in the U.S. were due to one of four factors: lack of health care, lifestyle, environmental hazards, and human biology. Their analysis, reported in Healthy People: The Surgeon General's Report on Health Promotion/ Disease Prevention revealed that". perhaps as much as half of U.S. mortality in 1976 was due to unhealthy behavior or lifestyle; 20% to environmental factors; 20% to human biological factors; and only 10% to inadequacies in health care" (1979, p. 9). This landmark document examined health problems by five age groups: infants, children, adolescent,s and young adults, adults, and older adults. The section on adults focused on those ages 25 to 64 which include

PAGE 27

11 the bulk of the workforce in America. Major problems identified in this age group were cardiovascular disease, cancer, alcohol abuse, mental health, and periodontal disease. Further analysis examined lIrisk factorsllwhich increased an individual's risk for a particular condition or disease. While a number of the factors were environmental, e.g., exposure to radiation, or biological, e.g.,genetic predisposition to certain conditions, the majority were due to behavioral factors. Chief among these were: smoking, alcohol abuse, lack of exercise, poor nutrition, uncontrolled high blood pressure, stress, and lack of seat belt use. Healthy People and a companion document Promoting Health/Preventing Disease: Objectives for the Nation (1980) outlined strategies for improving the health of America. The worksite became a focal point for hp/dp programs for adults. Parkinson et a1. (1982, p. 2),representing the public health perspective,cite access to high risk workers and their families who might not otherwise participate in such programs. Berry (1982, p. 9) and others cite factors of convenience, social support, workrelated incentives, and quality control as reasons for increasing preference for worksite programs. It is interesting to note, that with some exceptions noted by Bauer (1980, pp. 123-124) that there has been little overt resistance to worksite hp/dp programs. Current Program Parameters There are no generally accepted criteria as to what constitutes a worksite hp/dp program. The following is a presentation of

PAGE 28

12 major definitions and program parameters which have evolved in recent years. Healthy People: The Surgeon General's Report on Health Promotion and Disease Prevention (1979) states: Health promotion begins with people who are basically healthy and seeks the development of cormnunity and individual measures which can help them to develop lifestyles that can maintain and enhance the state of well being. (p. 119) In comparison it states "Disease prevention begins with a threat to hea1th--a disease or environmental hazard--and seeks to protect as many people as possible from the harmful consequences of that threat." In contrast,medica1 care is characterized as services which begin with the sick and seek(s) to keep them alive, make them well, or minimize "their disabi1ity.1I Healthy People further differentiates between health promotion, health protection, and preventive health services. Major program categories under health promotion include: smoking cessation, alcohol/drug abuse preven-tion, nutrition and weight control, exercise and fitness, and stress control. Other activities such as high blood pressure screening are considered preventive health services. Occupational safety and health, toxic agent control, and accident-injury control are c1assi-fied under the general category of health protection.' Documents from the Office of Disease Presention and Health Promotion within the Department of Health and Human Services references the definitions provided in Healthy People. They state that

PAGE 29

health promotion programs usually address lifestyle and health habits of healthy individuals--nutrition, physical fitness, stress management, weight management, smoking cessation, etc. In addition, many health programs use a variety of screenings to make people aware of their potential to become ill or contract a disease or detect diseases early. (Behrens ,1983, p. 1) They broaden the definition by stating that hp/dp programs often go beyond lifestyle to look at factors related to occupational safety 13 and environmental hazards, .chronic disease management, and mandatory employee assistance programs. They further state that a geniuine commitment to worksite health promotion involves the examination of the corporate culture to determine company policies and practices such as smoking policy, cafeteria foods, health benefits, etc., which affect employee health behavior. In 1979, a committee made up of persons from the public and the private sector was called together to develop guidelines for health promotion programs at the worksite. They defined health promotion as "a ,combination of educational, organizational, and environmental activities designed to support behavior conducive to the health of employees and their families" (Parkinson et al., 1982, pp. 8-9). Major program components recommended (a) risk assessment using risk appraisals and other measures to determine potential for cardiovascular disease, cancer, stroke, mental health problems, and accidents; (b) risk reduction measures in the areas identified as health promotion, smoking control, alcohol/drug abuse, nutrition and weight control, fitness and exercise, and stress management. To this, they add intervention in high blood

PAGE 30

, 14 pressure control, cancer detection, accident prevention, and protection against environmental health hazards; and (c) supportive environment--mechanisms in the social and physical work environment which, such as nutritious foods, spur groups which support the adoption of positive health behavior changes. In a report for the Health Insurance Association of America, Berry (1981, p. 19) suggests strategies similar to those outlined by Parkinson et al. (1982) including: risk assessment, education, support and motivation for change, risk reduction programs, and environmental and social support to maintain change. Topic areas which he considers appropriate to health promotion programs include those already mentioned above, plus a mixture of related programs ranging in emergency medicine, the Heimlich maneuver, benefits utilization, low back pain prevention, immunization, dental health, parenting, retirement, screening for glaucoma and sickle cell anemia. While the majority of these health problems can be prevented or reduced through changes in lifestyle, the topic areas range far afield from health promotion strategies identified in Healthy People as having the greatest overall potential for reducing health problems in the population. The Na,tional Chamber Foundation report (Sehnert & Tillotson, 1978) lists the basic areas of fitness, smoking cessation, a1cohol/ chemical abuse, nutrition and weight control, stress management, accident prevention, and screening programs including blood pressure screening and control. In addition, 'it promotes programs in medical

PAGE 31

self-care and wiser buying of health care services. These last two elements, not common to all other guidelines, are consistent with the Chamber's focus on cost-containment through health care utilization. 15 In summary, it may be said that there appears to be a central se.t of program elements which are included in most. program guidelines. These include exercise, blood pressure control, :smoking cessation, alcohol abuse prevention, nutrition and weight control, and stress management. A second set of elements including CPR, accident prevention, self-care, chronic disease control, parenting, and others are included less frequently, and appear to be to particular interests of the sponsoring agency. There is some consistency among guidelines, but no single standard set of definitions or guidelines appears to dominate at this juncture. Current Level of Program Activity There is a general consensus in the literature that there has been a dramatic increase in the number of companies offering health promotion programs (Behrens, 1983; aerry, 1981; Cunningham, 1982; Fielding & Breslow, 1983; Ware, 1982); Yet, there is little information on how widespread hp/dp programs are throughout industry in general. A 1978 survey completed by the Washington Business Group on Health (Kiefhaber, Weinberg, & Goldbeck, 1979) revealed that over half the companies responding offered some form of health promotion program. The group surveyed, however, was limited to

PAGE 32

WBGH membership, including mostly Fortune 500 companies. The response rate was 37%, too low to allow the results to be generalizable without further analysis. 16 A survey done by Fielding and Breslow (1983) of California businesses with 100 or more employees revealed that 78.3% offered one or more health promotion activities. Nearly two-thirds of the company programs were limited to two or less activities, which could range frolIl simple CPR seminars to more complex hypertension control and screening. A survey was conducted of Minnesota employers with 50 or more employees. Nearly 80% of employers responding offered some type of health promotion program on a systematic and on-going basis. The survey contained a more clear, albeit minimal definition of a health promotion program. The response rate of 54% does not allow the results of the survey to be generaiizable to the larger population (Minnesota Department of Health, 1982). In both California and Minnesota, it may be assumed that those with program activities were more likely to respond, thus artificially inflating the figures. A recent survey completed of businesses in Colorado with 50 or more employees revealed that approximately one quarter of the businesses contacted had a hp/dp program which offered one or more services on a regular on-going basis. An additional 50% of companies expressed interest in developing 'a program in the future. Over 40% of companies with 1,000 or more employees offered a program; this percentage decreased dramatically with decreasing size of company.

PAGE 33

Overall, it was estimated that fewer than 5% of employers in Colo rado offered hp/dp programs on an on-going basis (Davis et al., 1984). Figures cited by Warner and Murt (1984, p. 110) confirm these estimates, placing the percentage of employees nationwide covered by hp/dp programs at between 2% and 5%. 17 SiZe of company appears to be a major factor in determining whether or not an hp/dp program is offered. A survey of California employers revealed that over 95% of employers with 5,000 or more employees had an hp/dp program, typically offering 3 or more activities, e.g., high blood pressure screening, exercise programs. In contrast, only two-thirds of businesses with fewer than 250 employees offered programs, limited in almost all cases to one or two activities (Fielding, 1983). The Minnesota and Colorado surveys produced similar findings. Large employers were more likely to offer programs. In addition, these programs were likely to be more comprehensive than those offered by their smaller counterparts (Davis et al., 1984; Minnesota, 1982). An analysis of the hp/dp programs selected for review by the Committee to Develop Guidelines for Health Promotion Programs at the Worksite reveals that 16 out of the 17 programs have more than 1,000 employees (Parkinson et a1., 1982). 'The list of exemplary programs compiled by Behrens (1982) is likewise replete with such names as Campbell, Kimberly-Clark, NASA, IBM, Xerox, and other large companies. Recent attempts have been made to gauge the level of hp/dp program activity in small business. Although some reports such as

PAGE 34

18 that produced by Health Works Northwest (McMahon, Sajewski, & Graff, 1984) and by the New York Business Group on Health (Warshaw, Wein-garten, Barr, & Lucas, 1984) provide case studies of successful programs, sample size, sample characteristics, and low response rates prevent the findings from being generalizable to a larger population. More complete and more accurate information on hp/dp programs in small businesses is desirable, considering figures cited by Health Works Northwest .. that over 80% of all busi-nesses in "the U.S. have fewer than 20 employees, and that 99% of all U. S. businesses employ 500 people or less II (McMahon et al., 1984, p. 2). Frequency of Program Components Not all program components occur with equal frequency. A 1981 survey of 424 randomly selected California employers with 100 or more employees revealed that, in those companies with programs, the most common program components were accident prevention, cardiovascular pulmonary resuscitation, and choke saver. Alcohol or drug abuse program and mental health/counseling were in place in about one-quarter of organizations offering any program. Hypertension screening, smoking cessation, fitness and stress.management programs were made available by 10-17% of these employers. (Fielding, 1984, p. 250) Cancer control was conducted by fewer than 10%. Data on hp/dp programs in 3l5Minnesota businesses, industries, school districts, and hospitals revealed a similar pattern. Prevention of injury was the most common program component

PAGE 35

offered by approximately 60% of respondents. Alcohol/Chemical Dependency Programs and Personal and Family Counseling including Employee Assistance Programs (EAPs) were offered by almost half of the companies. The remainder of the programs including hypertension weight control; disease screening, smoking cessation, fitness and stress management were offered by approximately 15-20% of the companies. Diet information and programs on positive health attitudes were least cornmon, being offered by fewer than 15% of the companies. Small employers, defined as those with fewer than 800 employees, offered the above program components at roughly half the rate of the large employers in the sample. The only exception to this were the injury prevention programs which were offered at basically the same rate by large and small employers (Minnesota 1982, pp. 12-14). A 1983 Colorado survey provides information on program components in 94 randomly selected Colorado businesses of 50 or more employees. Hp/dp activities were categorized under 3 19 major headings: screening, informational programs, and preventive health services. Pre-employment medical examinations and high blood pressure screening were the most cornmon screening activities carried out by approximately 3 out of 4 companies. General and cardiovascular risk appraisal, height/weight screening, and screening for work-relat.ed health problems were offered by roughly half of the companies. Screening for cancer was done by fewer than a quarter of the companies. The most cornmon topics for informational

PAGE 36

programs were exercise, stress management, smoking cessation, and nutrition offered by 3 out of 4 companies. This was followed closely by alcohol/drug abuse, high blood pressure, low back pain, and injury prevention information offered by twothirds of the companies. Group and individual instruction in exercise and stress management were the most frequent service programs offered by 80% and 77% of companies, respectively, followed closely by weight control and smoking cessation offered in 2 out of 3 companies. Low back pain was offered by slightly more than half of the companies, while self-defense for women was provided by approximately one quarter of the companies. Two out of 3 companies had either an Employees Assistance Program (EAP) or an Industrial Alcoholism (lAP). The 206 companies interested in developing programs in the future were asked what type of activities they would include. Their response indicates that they would follow basically the same pattern as established programs (Davis et al., 1984). A survey conducted by the Washington Business Group on Health provides information on hp/dp activities in 59 member companies, mostly large businesses. As in the California survey, CPR and safety on the job are the most common program components, offered by over 4 out of 5 companies. Hypertension and breast self-examination were provided by 63% of companies. Programs such as safety (off-job), smoking cessation, nutrition and obesity control; fitness and exercise were offered by slightly over 50% of 20

PAGE 37

21 companies. In contrast to other survey findings, stress management programs were offered by fewer companies (41%). Benefits utilization programs were part of the activities in approximately one-third of the companies (Kiefhaber et al., 1979). Information on types of hp/dp activities in 31 small .businesses is provided by a study done by Health Works Northwest. Fitness/exercise programs are offered by 75% of companies. Smoking cessation, health risk appraisals, stress management, and nutrition are all offered by slightly more than 60% of companies. High blood pressure control is offered by 40% of companies, substances abuse by slightly more than one-third. While the case studies and the summary of activities are interesting, the sample is too small and non-scientific to make any statement about the relative frequency of hp/dp programs in the small business community in general (McMahon et al., 1984, p. 15). In the presentation of their findings, authors are quick to note that the level of program intensity and the strategies employed may vary widely among programs. Subjective interpretation as to what constitutes a program may lead to over-reporting in some cases, and under-reporting in others. Rate of Growth There is relatively li.ttle baseline data against which to compared the current level of program activity. Questions on hp/dp were not routinely included in surveys conducted on occupational

PAGE 38

health. When included, questions frequently did not address the range of activities now considered central to hp/dp programs. In 22 a 1967 survey conducted of the 400 members of the Industrial Medical Association, two-thirds reported "some type of health education program." Individual counseling was the activity most commonly cited (Cassuto, 1967). It is not clear that individual counseling carried out in the context of health services is comparable to current hp/dp program efforts. Similarly, health-related research done by the National Conference Board over the past decade has focused on health services, offering only incidental information on level and type ofhp/dp activities. Several recent surveys have attempted to document program growth. The survey of California employers showed that nearly three-quarters of hp/dp activities currently being offered at the worksite were begun after 1975 (Fielding & Breslow, 1983). Colorado data provide similar information. Roughly 85% of programs in existence were begun in a 5-year period from 1978-1983 (Davis et a1., 1984). What is not clear from either of these surveys is the number qf programs which may have been operative in the 60s or early 70s which were recently phased out of existence. More specific, encouraging data are available in categorical program areas. It was estimated that alcohol ,programs in industry increased from 50 to 500 in the 3-year period from 1970 to 1973. By 1977, over 2,000 organizations offered some type of program (Third Special Report to U.S. Congress on Alcohol and Health,

PAGE 39

23 1978). Trice (1977) and others note that although the numbers were small, the increase of worksite alcoholism programs had beensubstantial. More recent estimates by Dickman and Emener (1982, p. 55) place the number at closer to 5,500, which is indeed indicative of a growing acceptance in at least one of the component areas of worksite hp/dp. In the absence of hard data, professional estimates may be used as indicators of growth of hp/dp. Cunningham (1982) reports that membership in the American Association of Fitness Directors in Business and Industry grew from 25 in 1975 to over in 1981. Ware (1982) likewise cites the increasing number of articles in professional journals as evidence of increased program activity. The increasing number of business coalitions which currently offer health education services (56%) is likewise a measure of growth (Yenney, 1984). While these observations are all of a positive nature and generally consistent, they might best be considered indicators of increased interest,rather than accurate reflections of an increased level of programs and services actually being offered at the worksite. The rate of growth appears limited and confined to the "cadre of the convinced" (Davis et al., 1984). Prototypes of current hp/dp programs were developed in the 60s and were relatively refined by the mid-70s. Yet, by most estimates, these programs have been adopted by fewer than" 10% of employers, and reach only a small percentage of the workforce."

PAGE 40

Program Outcomes The majority of worksite health promotion programs are designed to address one or more of the following goals: (a) to reduce illness and improve the overall health status of employees, (b) to reduce costs associated with absenteeism and health benefits utilization, and (c) to improve employee morale and productivity (Davis et al., 1984; Fielding, 1983; Kiefhaber et al., 1979). 24 There is little data on the extent to which these outcomes are systematically achieved. The evidence that exists suggests that positive outcomes are dependent, in large part, on the intensity of the program, the appropriateness of the strategies implemented, and the extent to which the work environment supports and reinforces behavior change. Program outcomes have been more thoroughly evaluated in several categorical program areas. This section will examine the outcomes in four areas that have been most rigorously studied: High blood pressure control, smoking cessation, alcohol/drug abuse, and fitness/exercise programs at the worksite. Particular attention will be paid to the extent to which these programs improve employee health, control cos.t, and/or contribute to improved morale and productivity. Because few studies have been conducted on nutrition, stress management programs at the worksite, they will not be included in this section. Hypertension. Numerous studies have shown that individuals with high blood pressure have an increased risk for premature death

PAGE 41

25 and disability. Rates of heart disease and strokes are six to seven times greater in individuals with high blood pressure, than in those" with normal blood pressure (Veterans Administration, 1967). A second set of studies have examined intervention strate-gies and have demonstrated effective means for controlling high blood pressure (Hypertension Detection, 1979). Several design variables appear to differentiate successful from unsuccessful programs: number and intensity of contact with health care providers, active patient participation, long-term follow-up ahd the existence of supportive family and friends (Alderman et a1., 1980; Foote & Erfurt, 1983). Pr"ograms that include this feature have a high potential for controlling blood pressure, thereby reducing health problems of employees. A number of successful programs are currently underway, including a cooperative program between the University of Michigan and Ford Motor Company, where 80% of hyper-tensive employees are under control (Alderman et a1., 1980), and a program at Massachusetts Mutual Life Insurance Company where the percentage of hypertensive under control rose from 36% to 82% after one year of the program (Fielding, 1984, p. 241). "In general," states Fielding, "worksite based hypertension detection and control programs have results that are superior to what is achieved in clinical practice. ." It is estimated that between 15% to 30% of Americans suffer from high blood pressure. Roughly half of this is uncontrolled, contributing to increased cardiovascular and related disabling

PAGE 42

26 conditions. Using the cost of treating such chronic diseases and disability, Kristein (1982) estimates that."the 'average' hyper-tensive over a lifetime, is 'costing' the society between $170 and $300 per year in excess medical spending, and between $270 and $460 in lostoutput" (p. 32). Measuring only the cost of outside medical care covered under company-sponsored insurance, Fielding (1984, p. 241) places the estimate at between $175 and $250 per hypertensive. Costs of high blood pressure control programs vary. Ruch1in and Alderman (1980), p. 795), reporting on their experience in 9 clinic sites in -the New York City area, placed cost of treating hypertensives at the worksite between $150 to $200 per patient per year. In contrast, Fielding (1984) estimates that effective screen-ing referral and follow-up programs could be provided for an estimated $37 per hypertensive employee. Because calculations are based on sets of changing program parameters and characteristics of the target population, it is difficult to arrive at set figures. Kristein (1982, p. 34) places the ratio of cost benefit between 2 to 1 and 4 to1. Smoking cessation. The effects of smoking on health have been documented through hundreds of studies. Smoking increases risk for heart and blood vessel diseases; chronic bronchitis and emphysema; cancers of lung, larynx, pharynx, oral cavity, esophagus, pancreas, and urinary bladder. Smoking during pregnancy also increases risks of complications of pregnancy and retardation of fetal growth. Cigarette

PAGE 43

smokers have a 70 percent greater rate of death from all causes than non-smokers, and tobacco is associated with an estimated 320,000 premature deaths a year. Another 10 million Americans suffer from debilitating chronic diseases caused by smoking. (Healthy People, 1979, pp. 121-122) Luckily, people who quit smoking can reduce their risk for these diseases and conditions over time. A variety of smoking cessation strategies have been tested including: physician counseling, motivational programs, self-help programs, skill acquisition, peer support, and smoking In general, programs that offer a variety of approaches are more effective than single-faceted programs, because they have a greater change of meshing with individual learning styles and needs of smokers. The success rate of smoking cessation programs in general has been mixed. Initial quit rate generally range between 70% to 90%; this drops to 30% at 6 months, with some of the best programs 27 maintaining a quit rate of 40%. Worksite programs generally achieve the same quit rate, but may experience a higher dropout rate due to time conflicts (Fielding, 1984, p. 243). Danaker(1980), however, points to worksite smokingcessation programs as having the potential to reach populations not normally attracted to community programs and to provide positive incentives for non-smokers. Warner and Murt (1984) cite the example of Speedcal1 Corporation, where a carefully structured incentive program contributed to a drop in from 67% to 20% of employees over a 4-year period. Although there to be some flaws in the evaluation of the Speed call program, it may be indicative of the potential for employers to encourage non-smoking behavior.

PAGE 44

28 The cost of smoking to society and to employers specif-ically is reasonably well-documented. In a review article, Fielding (1984) states: Smoking is expensive to employers,translating into higher health insurance costs (some studies. show a 50% greater use of the health care system by smokers), higher absenteeism (an estimated 2 to 3 additional days per year), increased accidents at work, and a higher rate .of disability reimbursable events. (p. 243). Kristein (1983, p. 358), analyzing data from a variety of sources, estimated that smokers cost the employer between $336-$601 per year in 1980 dollars. The cost centers examined by Kristein are insurance, medical care, absenteeism, loss of productivity, and health risk to other employees. He notes that non-health costs of smoking may exceed the health costs to businesses. Figures presented by Fielding (1984, p. 243) place estimates in a range of $200 to $500 per. per.son. Costs of smoking cessation programs vary from $115 self-help programs sponsored by voluntary health agencies, to commercial programs charging $500-$1,000 per participant. Assuming that the average smoker costs an employer $350 per year, a smoking cessation program that costs less than $100 per employee and has a sustained quit rate of 33% would provide cost savings for the employer. Those in which the employee bears half the cost would provide an even higher return on investment. Employee assistance programs. In the 1960s, the National Council on Alcoholism estimated that 4% to 8% of the workforce had

PAGE 45

a problem with alcoholism. Later, studies placed the estimate between 5% and 10% (Dunkin, 1982, p. 4). In a summary article, DuPont and Basen (1980) review estimates that have been made by various sources. They conclude that the average is closer to 10% with a significantly higher prevalence rate among "lower bluecollar workers" and in male-dominated industries. Alcohol and drug abuse is a contributing factor in an estimated 10% of all deaths in the United States. It increases risk for cirrhosis of the liver and certain types of cancer of the digestive system. It is a factor in an estimated 50% of all traffic fatalities. It is considered an indirect cause of homicides and suicides (Healthy People, 1979). It can be stated that 29 a reduction of alcohol abuse would significantly enhance the health and well-being of employees. Industrial alcoholism programsbegan in the 1940s and have evolved to the broader concept known as Employee Assistance Programs which address a wide range of issues including counseling for depression, substance abuse, family relationships, personal and financial problems. The data on the effectiveness of such programs has generally been fragmentary, but positive. Berry (1982, p. 28) presents data on 8 programs with a rehabilitation rate from 70% to 80%. DuPont and Basen (1980, p. 142) report several studies that report rehabilitation rates near 50-60%. Two specific examples given include the Bethlehem Steel Corporation and the DuPont Corporation, with a 60% and 66% rehabilitation rate, respectively.

PAGE 46

30 The authors note that despite the fact that the majority of program evaluations lack control groups and are based on limited measures of success, most of the data show positive impact. It is likely that there are numerous less well-designed and implemented programs with lower rehabilitation rates that do not appear in the literature. The cost of alcohol/drug abuse to industry is estimated at nearly $51 billion per year based on both increased medical care utilization and decreased productivity (Berry, 1981, p. 28; Fielding, 1984, p. 240). Estimates of cost savings to business from effective EAP programs vary. The General Motors program is generally cited as an example of potential saving. For the first year of operation, they experienced a 40% decrease in time off the job, a 60% decrease in sickness and accidents benefits, and a 50% decrease in grievances. Overall, they estimated a 3:1 return on investment (Berry, 1982, pp. 28-29). A 1979 supplement to Medical Care examined 12 EAP programs focused primarily on alcohol and reported similar findings, estimating cost savings of $1,000 per client. DuPont and Basen (1982, p. 143) report results of a federal civil service program that expends $15 million annually, and estimated cost savings at between $135 and $280 million per year. In summary, it may be said that while estimates vary, a number of exemplary programs demonstrate the strong potential for 'cost savings through Employee Assistance Programs. Exercise/fitness. Several national surveys estimate that only between 35% to 50% of the population exercises regularly

PAGE 47

(Fielding, 1982, p. 132; Haskell & Blair, 1980, p. 109; Healthy People, 1979, p. 132). Lack of exercise can lead to osteoporosis, increased risk of back problems, and is thought to be linked to increased falls in the elderly population (Fielding, 1984, pp. 243-244). A study by Paffenbarger, Wing, and Hyde (1978) of 17,000 Harvard alumni demonstrated that those not engaged in regular vigorous exercise programs had a 64% higher risk of heart attack than their active Exercise can also lead to reduced blood pressure, improve cardiovascular efficiency, and may prompt persons to stop smoking. Although the evidence is less extensive, it appears that exercise plays a role :in preventing 31 mild anxiety and depression (Haskell & Blair, 1980, p. Ill; Healthy People, 1979, p. 132), improving attitude toward the job, overall performance, and job satisfaction (Fielding, 1984, p. 244). A study by Durbeck et al. (1973) of NASA employees found that participation in exercise programs increased work capacity and satisfaction. Studies done by Rhodes and Dunwoody (1980) among Canadian workers demonstrated the same outcomes. Another Canadian study by Donoghue (1977) found that participation in a fitness program reduced absenteeism. Fielding (1982, p. 911) reports a controlled study in a Toronto insurance company which demonstrated a 42% decrease in absenteeism among participants as compared to a 20% decrease among those choosing not to participate. Fielding points out that many of the studies on the impact of fitness programs are flawed because they do not control for

PAGE 48

factors which prompted individuals initially to sign up for the programs. Still the evidence of the effectiveness of fitness programs in participants is generally positive both from a health perspective and a job-related benefits view. Exercise programs in industry vary from simple motivational programs to staffed fitness facilities at the site. A minimum program should involve employees in vigorous aerobic exercise 15 to 30 minutes at least. 3 times a week. Formats may include jogging, swimming, walking, or any number of active sports. Exercise to increase muscle tone, strength and flexibility may be desirable from the employees' point of view, but have not been shown to decrease cardiovascular risk. The main issue in worksite fitness programs is not benefits of exercise, but the ability of companies to construct programs which workers find attractive 32 Access and peer pressure appear to be 2 key variables which employers have only recently begun to address. Cost savings information for worksite fitness are more difficult to generate. Health effects of fitness are generally long-term. There are no immediate reductions in medical care utilization cited as in the case of EAPs and smoking cessation. That is not to say that these links do not exist, they simply have not been sufficiently examined. Measuring increased work capacity, reduced absenteeism and improved productivity in economic terms, is a problem not specific to fitness programs alone. Most frequently, observes Wright (1982, p. 967), fitness programs are operated or

PAGE 49

justified as an employee benefit. They are viewed as producing cost savings because people want to believe they do. In general, then, it may be said that several worksite health promotion programs including high blood pressure control, smoking cessation, employee assistance programs, and exercise programs, if properly designed and implemented, can improve employee health. The evidence for cost-savings is somewhat weaker. Smoking cessation, high blood pressure control, and EAPs all have a reasonable number of studies to document this claim. Cost savings from fitness which rests on its ability to reduce absenteeism and increase productivity is tenuous at best. Despite the lack of solid evidence, cornpaniesbelieve that hp/dp programs result in a range of positive outcomes, including improved employee health, reduced health care cost, and improved morale and productivity (Davis et al., 1984; Fielding & Breslow, 1983). Finally, in considering long range effects of hp/dp programs, Warner and Murt (1984, p. 111) sound a word of caution. They point to turnover rates in industry which would result in the loss of investment in human capita1--other companies indeed might reap the benefits of the first company's hp/dp program. They further argue for an examination of increased life span on pension benefits, an area not currently discussed in the literature. Summary 33 The scientific basis for health promotion/disease prevention has been evolving over the past half century. More recently,

PAGE 50

34 effective intervention strategies have been developed which hold the promise of reducing illness and the attendant medical care costs in the adult population. Worksite health promotion/disease prevention programs have been packaged and promoted by many groups which view the worksite as an optimum site for the delivery of preventive programs. The response of industry has been one of interest; yet the adoption and implementation of programs has been limited to large companies who employ only a small portion of the total work-force. The existing literature documents and describes existing programs, but does little to explain the limited acceptance by business and industry in general. The purpose of this study is to use diffusion theory as a framework within which to understand the process and the variables related to adoption or non-adoption of worksite hp/dp programs.

PAGE 51

CHAPTER II REVIEW OF THE LITERATURE Diffusion theory is that body of knowledge which attempts to explain "the pattern and rate of adoption of an innovation ina population" (Feller & Menzel, 1977, p. 51). Over the years, a number of theoretical models have been developed and used to study, among other things, educational innovation, introduction of new farming techniques, and the acceptance and diffusion of health care products and services. This chapter will review this literature base as a framework within which to examine the adoption/nonadoption of worksite health promotion/disease prevention (hp/dp) programs. Historical Development The beginnings of diffusion theory are traced to the early 1900s (Brown, 1981; Midgley, 1977; Rogers & Shoemaker, 1971). Anthropologists initially studied the transmission of Western ideas to primitive societies. Although many of these early projects tended to be highly specific to site, they provided insight into the impact of culture on new ideas, and an empirical data base related to diffusion (Rogers.& Shoemaker, 1971, p. 49). At almost the same time, a French sociologist, Tarde, produced his work on opinion leadership and imitation behavior (Tarde, 1903). His observations provided hypotheses to be tested by later researchers.

PAGE 52

Major growth in the field of diffusion research occurred in the 1920s when diffusion theory became a tool used by rural sociologists to understand the adoption of new farming practices. A study by Ryan and Gross (1943) on the diffusion of hybrid seed 36 corn is a classic work that established a pattern for the study of innovation diffusion in agriculture. The U.S. Department of Agriculture Extension Service conducted numerous studies throughout the 40s and 50s to determine the efficacy of its new program promotions. Although rural sociology continued to dominate the field, researchers from several other disciplines became increasingly active in the field of innovation diffusion. Pioneering work in the field of education was done by Paul Mort. Concerned with the lag time in the adoption of educational innovations, he attempted to establish administrative predictors of innovativeness in local school districts (Mort & Connell, 1938). The number of studies on educational diffusion continued to grow rapidly throughout the 40s and 50s focusing on a range of topics from the effects of Sputnik to the adoption of new math. Although there were a great many studies of the diffusion of educational innovations, many were judged to be of lesser quality, adding little to the field in general (Carlson, 1968; Rogers & Shoemaker, 1971). The study of the diffusion of medical innovations grew with the field of medical sociology in the 1950s. A study by Coleman, Katz, and Menzel (1957) focused on the adoption and prescription of a new drug among physicians in two Midwestern communities ..

PAGE 53

Following this classic study, numerous other research efforts were undertaken. The majority of these focused on individual behavior of physicians and patients. The diffusion of family planning practices has been studied extensively both in the United States and abroad. 37 During the 1960s considerable work in diffusion innovation was carried out by researchers in the field of communication and marketing. Communication experts frequently focused on the effects of various channels of communication on the rate of adoption of particular behaviors. Katz and Lazarsfeld (1955) examined the relative effectiveness of personal contact, printed matter and electronic media in promoting purchase of new products in the supermarket. Because of its dependence on the continuing production and purchase of "new products," the fashion industry in the United States has sponsored numerous studies on consumer behavior (Baumgarten, 1984; Midgley, 1974). Unfortunately, most of the marketing studies related to innovation diffusion are privately funded, therefore the results are proprietary information not widely shared in professional journals. More recently, books and articles have begun to appear which address the role of government in the diffusion process. A comprehensive review of the literature conducted by Tornatzky, Eveland, Boylan, Hetzner, Johnson, Roitman and Schneider (1983) for the National Science Foundation cites over 75 books, reports and articles of this nature written since 1975. This body of

PAGE 54

literature moves beyond the passive observation of the diffusion process to policy analysis and an active debate over the proper role of government in the diffusion of technological innovations. 38 The field of diffusion research as it exists in the 1980s incorporates data bases and perspectives from a wide variety of disciplines including anthropology, sociology, rural sociology, education, medical sociology, marketing, psychology, engineering, political service and economics (Rogers & Shoemaker, 1971; Tornatzky et a1., 1983). The focus of this research ranges from micro-level concern with attributes of the innovation itself to the measurement. of the impact of macro-economic policy on adoption practices within specified populations. The range of disciplines involved and the variety of perspectives used add to the richness and the complexity of the field of diffusion theory. Unfortunately, as noted by Tornatzky et a1. (1983), "No single organizing framework is adequate for integrating all the diverse themes to be found in this literature" (p. vii). The remainder of this chapter will focus on identifying central themes of diffusion and exploring research findings from those areas and studies deemed most applicable to the study of worIrsite hp/dp programs. Definitions Given the diversity of research traditions which have contributed to diffusion theory, diversity in terminology and in

PAGE 55

39 theoretical models is to be expected. The following is a presentation of most widely used definitions and major concepts in diffusion theory. Webster defines innovation as "the introduction of something new." While strikingly simple, this definition does not provide the specificity desired by researchers. Therefore, modifiers, qualifiers alternate definitions have been developed. Rogers and Shoemaker (1971) defined innovation as ". an idea, practice, or object perceived as new by an individual" (p. 19). Zaltman and Duncan (1973), recognizing that the unit of adoption may be other than an individual, defines innovation as .. the first or early use of an idea by one of a set of organizaions with similar goals'" (p. 463) .. Using a practical, marketing perspective, Bell (1963) considers an innovation as anything which does not yet have 10% of the potential market. Tornatzky et al. (1983) define innovation as a technology new to a given organization" 1). Pelz and Munson (1980) use the term, "knowledge based innovation" to describe non-material innovations. These same intangible concepts such as management by objective (MB) new psychiatric treatment, and quality circles can also be termed innovative "social technologies." The result of these various perspectives is that the common use of the term "innovation" is inclusive rather than specific. In general, an innovation may be (a) new or perceived as new, (b) tangible or intangible, and (c) adopted by an individual or a larger unit in society.

PAGE 56

40 A clear, operational definition of diffusion is, likewise, elusive. Rogers (1984) defines diffusion as II the process by which an innovation is communicated through certain channels over time among the members of a social systemll (p. 5). From a research standpoint, however, the question remains as to where an innovation can be considered to have diffused through a social system--at what point does that 1I0vert behavior changell occur signals adoption (Rogers & Shoemaker, 1971). An example from education illustrates the point. Can new math be considered to have been adopted by a school district (a) when the school board adopts a policy supporting its adoption or (b) when the board provides a budget for new materials or (c) when it is piloted in one school or (d) when it is in use throughout the district or (e) only when it becomes a regular part of the curriculum for all schools? To address this issue researchers began to identify and focus on the stages of the diffusion process. Perhaps the most widely used model was that developed by Rogers and Shoemaker (1971, pp. 24-27) outlining four major stages (a) knowledge, (b) persuasion, (c) decision, and (d) confirmation. Za1tman and Wa11endorf (1983, p. 530) presented a slightly expanded version including: (a) knowl-edge and awareness, (b) attitudes formation, (c) decisions, (d) initial implementation, and (e) contrinued-sustained imp1e-mentation. Depending on the purpose of the research, these stages may be subdivided. An example is a study done by Pelz (1981) on the use of information by local governments. The stages he

PAGE 57

identified were II concern or diagnosis, search, design, appraisal or selection, building commitment, implementation (plus incorporation), and diffusion" (p. 4). In a meta-analysis of stage models of diffusion, Tornatzky et al. (1983) note that all models are generally variants on the following pattern: (1) A,.,areness (2) Matching/Selection (3) Adoption/Commitment (4) Implementation (5) Routinization (p. 22) Indeed, the most recent model developed by Rogers (1984) includes these stages and a brief description of each: There are five main steps in the innovation-decision process: (1) knowledge, which occurs when an individual or some other decision-making unit is exposed to the innovation's existence and gains some understanding of how it functions, (2) per suasion, which occurs when the individual forms a favorable or unfavorable attitude-toward the innovation, (3) decision, which occurs when the individual engages in activities that 41 lead to a chance to adopt or reject the innovation, (4) implementation, which occurs when the individual puts the innovation into use, and (5) confirmation, which occurs when the individual seeks reinforcement of an innovation decision if exposed to conflicting messages about the innovation. (p. 8) Faced ,.,ith the lack of universally applicable and acceptable definitions, the tendency has been for researchers to arrive at their own operational definitions. While some authors decry this practice as limiting the comparability of studies (Warner, 1974), others argue that a common terminology may not ever be reasonable or entirely appropriate (Tornatzky et al., 1983, p. 22). What does appear important is that researchers specify the type of innovation

PAGE 58

42 studied, the unit of analysis and the definition of adoption used in their study. Diffusion Models and Perspectives Diffusion theory is a complex and diverse literature which incorporates findings from a number of academic disciplines. The analytical models closely resemble the theoretical perspectives in the field or organization theory in general (see Figure 1). They range from the rational/economic model to the more behaviorallyoriented communications model to models which address environmental forces affecting diffusion, thus echoing the systems approach to management and change. Recent studies attempting to delineate "innovativeness" in an organization appear similar to studies addressing the complex phenomenon of "corporate culture." This section presents an overview of major diffusion perspectives. It must be noted that while each of the models reviewed has a major focus different from the rest, there is considerable overlap and duplication of variables represented in each model. Rational/Economic Perspective Economic model. This model focuses on economic variables to explain the rate and pattern of diffusion of an innovation. Key variables identified by Griliches (1957), Mansfield (1968), and others include profitability, size of required investment,

PAGE 59

BEHAVIORAL PERSPECTIVE Economic Model Return on investment Size of investment Size of company Certainty of return Communications Model Source Innovation attributes Channels of communications Adopter characteristics Structural Model Size Centralization Formalization Marketing Model Access Pricing Promotion Market segmentation Champion Hodel Contextual Model Financial environment Government policy Consumer behavior 43 Figure 1. Perspective in Organizational Behavior Related to Models in Diffusion Theory

PAGE 60

uncertainty and risk aversion .. (Warner, 1974, p. 435). The major hypothesis as stated by Mansfield (1968) is that "the rate of adoption is a linear function of the profitability of employing the innovation, the size of the investment required to use it, and other unspecified variables" (p. 435) A number of studies (Romeo, 1975; Zaltman, Duncan, & Holbik, 1973) expand, support, and clarify but do not substantially alter Mansfield's model. Size of firm, which has been shown to be positively related to adoption (Kimberly, 1976; Rogers, 1983) could be considered an .economic variable. Brown (1981) suggests that size of company serves as a surrogate measure for ability to raise capital, to bear the cost of the innovation and to bear the risk of failure" (p. 156). 44 In his review of the contributions of the economic model to the understanding of diffusion, Warner notes that economic variables do indeed explain, in part, patterns and rates of adoption. The model is judged to have strong predictive powers'lThen the innovation is stable and when market conditions are fully operative. He notes, however, that "Diffusion is a complex social phenomenon which clearly involves both economic and noneconomic.factors" (Warner, 1974, p. 438). Structural model. The structural model of diffusion theory focuses on structural characteristics of an organization. In line with rational organization behavior theory, it suggests that

PAGE 61

actions within an organization are influenced by structQral vari-abIes. In his review of the literature, Rogers (1983) lists and describes six structural variables thought to be related to the adoption of innovation: Centralization is the degree to which power and control in a system are concentrated in the hands of a relatively few individuals Complexity is the degree to which an organization's members possess a relatively high level of knowledge and expertise .. Formalization is the degree to which an organization emphasizes following rules and procedures .. Interconnectedness is the degree to which units in a social system are linked by interpersonal networks Organizational slack is the degree to which uncommitted resources are available to an organization. (pp. 359-360) 45 Size is the sixth structural variable. Complexity, interconnected-ness, organizational slack and size are generally positively associated with innovation, while centralization and formalization appear negatively related to the of innovation. Tornatzky et a1. (1983) focus on several of the same variables including centralization, formalization, and complexity. Zaltman and Wallendorf (1983) present these same three as key variables to be considered. They suggest, however, that the variables operate differently at different stages of the diffusion process. In the initiation stage, complexity is thought to be positively associated with innovation while formalization and centralization are negatively associated. In the implementation stage, the reverse is true. Complexity hinders implementation, while fonnalization and centralization facilitate implementation (p. 535).

PAGE 62

There are several criticisms of the structural model. The variables are difficult to define operationally. The structure of units in a company may vary considerably, so it is difficult to arrive at any overall rating. Finally, structural variables frequently are notamenab1e to intervention. Work by Mintzberg (1979) .and Hage (1980) continue to refine this model for easier, more aCGurate and consistent application. Behavioral Perspective 46 Communications model. Within the behavioral and social sciences research has resulted in the identification of variables and models which the economic perspective outlined above. Rogers (1962; Rogers & Shoemaker, 1971; Rogers, 1983) developed and refined a communications model, adding social factors as key determinants of the pattern and rate of adoption. This model dominated the field of diffusion theory for many years as evidenced by a 1971 review of the literature in which Rogers and Shoemaker cite over 1,500 studies built around the communications model. Diffusion research built on the communication model focuses on factors which assist or impede the flow of communication and ultimately contribute to the adoption or rejecti6n of an innovation. The variables, as summarized by Rogers (1983), include (a) source of innovation, (b) attributes of the innovation, (c) channels of communication, and (d) adopter characteristics.

PAGE 63

47 Source. Much of the research related to the source of innovation has on the requisite characteristics of change agents. Extensive research has found that the most effective agents are those viewed by the target population as credible, similar to themselves and as having a client orientation or empathy (Rogers & Shoemaker, 1971). Innovations introduced into a social system by someone possessing these characteristics would be more acceptable and diffuse more rapidly than those innovations introduced by agents not perceived as credible, similar and empathetic. Attributes. Like\vise, it is believed that" the rate of adoption of an innovation is positively related to the character-istics of the innovation (Rogers, 1985). Specifically those attributes include: (1) relative advantage, the degree to which an innovation is perceived as superior to the idea it replaces, (2) compatibility, the degree to which an innovation is perceived as being consistent with existing values, past experience and needs of potential adopters, (3) complexity, the degree to which an innovation is perceived as difficult to understand and use, (4) the degree to which an innovat.ion may be experimented with on a limited basis, and (5) observabi1ity, the degree to which the results of an innovation are visible to others. (Rogers, 1984, p. 6) Work by a number of researchers (Binham, 1976; Fliegel & Kivlin, 1966; Mohr, 1969) has focused on characteristics of the innovation in an attempt to develop predictive models. this focus does provide some useful information, Downs and Mohr (1976) cqntend that no two organizations view an innovation in the same way. Downs (1978) suggests that the focus should be

PAGE 64

more on how organizational characteristics influence perception of the innovation. 48 Channel of communication. Pioneering work in this area was done by Whyte (1954) who explored the impact of "word of mouth" communication in the purchase of air conditioners. This was followed by studies by Katz and Lazerfeld (1955) to determine the relative effectiveness of various communication channels. A major finding to come out of these and later studies was summarized by Rogers and Shoemaker (1971): "Mass media channels are relatively more important at the knowledge function, and interpersonal channels are relatively more important at the persuasion function in the innovative-decision process" (p. 255). In more recent "."ork, researchers have introduced the concept of an intermediate force "local media" which reach a large geographic area, yet are more than mass media to local target groups. More powerful in influence than mass media and more accurate than interpersonal contact, it offers the potential new channels for diffusion of innovation. Characteristics of adopters. The majority of research using the communication model has been directed toward understanding more about the potential adopters of an innovation. The classic categorization of adopters was done by Rogers and Shoemaker (1971, p. 1982) and is presented in Figure 2. The labels "innovation, early adopters, early majority, late majority, and laggards" continue to be widely used in diffusion

PAGE 65

49 '"C C13 til 1-1 0) tlO 0 '"C p::: til + I>: ::E: ril ..c ""' N til CXl til .--I 0) 1-1 .p., til P. 0) '-'0) 0) 1-1 til 1-1 I>: 0 00 0 0) .. -1-1 -1-1 C13 C13 1-1 U 0 0;:'-' 1-1 0) -1-1 H 0) P. Z 0 4-l '"C 0 .--I 00"1 '"C .--I til -1-1 1 C13 I>: U 1-1 0) ::I C13 '"C o 0 til U.c N U) 1 I>: .. S 0 1-1 aldoaa: 10 'soN

PAGE 66

50 research. Rogers and Shoemaker (1971) summarize salient character-is tics of each group as follows: innovators-venturesome, early adopters-respectable; early majority-deliberate;' late majority-skeptical; and laggards-traditional" (p. 192). Reviewing the work of Rogers and Shoemaker and more recent marketing.studies, Midgley (1977) notes two salient adopter charac-teristics which appear across all categories of innovation: willing-ness to take risk and innerdirectedness. Yet he notes that" although there is a slight tendency for individuals to innovate in closely-related product categories, the evidence did not suggest the existence of generalized innovators" (p. 62). While the work of Rogers and Shoemaker has focused largely on individual innovative behavior, Zaltman and others have focused on characteristics of organizations as adopters of innovation. Zaltman and Wallendorf (1983) conclude that three structural variables affect the adoption process: complexity, formality, and centralization. They describe their impact as follows: The initiation of new ideas, products, or services is facilitated by (1) high complexity, (2) low formalization, and (3) low centralization The implementation of new ideas, products, or service is facilitated by (1) low complexity, (2) high formalization, 'and (3) high centralization. (Zaltman & Wallendorf, 1983, p. 534) These generalizations were largely based on the research of Chakrabarti and Rubenstein (1976) who reviewed the adoption of NASA technolgoy by private industry.

PAGE 67

Champion model. A second subset of research from the behavioral school more closely examines the role of various actors in the diffusion process. It amplifies portions of the communications model and extends it by looking at organizational roles as well as individual predisposition to innovation. Work done by Roberts (1969) on entrepreneurial behavior found tha-t entrepreneurs have high achievement and moderate 51 power needs. Udell, Baker, and A1baum (1976) focused on inventors and found them to be ". more emotionally adventurous, achievementoriented, independent, resourceful, creative, and personally goalscentered and hard-working than non-inventors" (Tornatzky & Wa1lendorf, 1983, p. 102). Rogers and Shoemaker (1971, pp. 185189)" identify 32 socioeconomic, comnmnication and personality variables related to innovativeness. In general, innovators tend to be better educated, of higher sociai status, more empathetic/ less dogmatic, more abstract thinkers, favorably inclined toward risk, and tend to have higher levels of achievement and higher aspirations. In addition, they tend to participate more socially, have more exposure to media and interpersonal communication, are more cosmopolite and more socially integrated than nonadopters. Opinion leaders shared the same basic set of characteristics, .. we find that they are (1) more exposed to all forms of external communication, (2) are more cosmopolite, (3) have higher social status, and (4) are more innovative Rogers (1983) reiterates, "most individuals evaluate an innovation, not on the

PAGE 68

52 basis of scientific research by experts, but on the basis of the subjective evaluations by near-peers who have already adopted the innovation" (p. 7). While considerable work has been done to identify character-istics of adopters at the individual level, much less research has been devoted to exploring the impact that these actors have on the adoption of innovation by organizations. Keller and Rolland (1978) indicate that persons with high job performance, high job satisfac-tion and a penchant for new ideas, tend to act as boundary-spanners, bringing innovation into an organization. Chakrabarti (1974) and Keller and Rolland (1978) have commented on the role of "gatekeepers" in the adoption process. They typically are technical people exposed to complex ideas and who have the skills to translate these ideas, thus facilitating innovation diffusion. Chakrabarti (1974) identified and studied the phenomenon of "product champions," indi-viduals who promote a particular innovation throughout the diffusion process. Several authors, including Mansfield (1971), have attempted to examine how important individual actors have been to the organizational process, while others, such as Rage and Dewar (1973) contend that groups or coalitions are more common and more important to the decision process. Tornatzky et a1. (1983) provide a succinct summary of the current debate: little work has been done in the relative importance of individual factors as opposed to factors associated with the environment in which innovation takes place, or the extent to which situations determine which characteristics are important, and which are not. (p. 109)

PAGE 69

53 Systems Perspective Marketing model. The economic and the behavioral models provide useful perspectives for viewing thediffusion of innovation. Indeed, research built on these models has added enormously to the understanding of the processes, elements and variables involved. Recently, however, both the economic model and behavioral models have been criticized for presenting an incomplete picture. Feller and Menzel (1977) note that research has focused largely on internal characteristics and adopter categories and has not adequately addressed the milieu in which adoption takes place. Warner (1974, p. 449) notes that prior models assume an adequate supply of and adequate knowledge of the innovation which may not be the case. Brown (1981) criticizes what he terms lithe adoption perspectivell (pp. 5-6). Much research, he asserts, focuses on "the process by .which adoption occurs, or the. demand of diffusion" which, he states, II implicitly assumes that all have equal opportunity to adopLII Instead, he suggests a market/infrastructure perspective which focuses on the supply aspects of diffusion. This view holds that .. individual behavior does not represent free will so much as choices within a constraint set and that it is government and private institutions which establish and control the constraints (Brown, 1981, p. 30). It raises the prospect that diffusion patterns may in large part be explained by entrepreneurial actions rather than social interactions. The major entrepreneurial actions discussed by BroWn are (a) improving access to the innovations--frequently done in the

PAGE 70

54 business world by increasing the number of outlets, (b) creative pricing policy decisions to encourage experimentation, (c) promo-tional communication consciously designed to reach the target popu-1ation, and (d) market segmentation which identifies andsequen-tia1ly targets homogeneous subgroups in the population. Unlike previous models it: (a) introduces access as an explanatory variable, and (b) it views diffusion as an active rather than a passive process in which it is incumbent upon the diffusing agency to segment the market, manipulate pricing policies, increase promo-tion of the product, and provide outlets. It is the thesis of Brown (1981) that the majority of innovations are promoted by entities termed propagators. These entities are profit or non-profit motivated organizations or government agencies acting to induce the rapid and complete diffusion of the(irinovation. (p. 52) Where propagating agencies or individuals stand to profit from the adoption of an innovation by others, diffusion will occur more rapidly. The presence of a profit, either economic or social, for vendors becomes a critical and explanatory variable in diffusion. This is in line with Merkle's hypothesis (Merkle, 1980) that scientific management was introduced into industry not so much because this innovation benefitted the adopting firms, but rather because it benefitted upwardly mobile professionals who propagated the idea. Contextual model. The market/infrastructure perspective presented by Brown is similar to the idea of IIcontextua1 influencesll

PAGE 71

55 presented by Tornatzky et al. (1983). In a review of research based on this perspective, Tornatzky notes, liThe integrating concept is that all such variables are external to the innovating organization but are assumed to affect intraorganizational processesll (Tornatzky et al., 1983, pp. 77-78). Economic conditions, interactions among firms, consumer behavior, government policies are amon"g the many factors to which innovating organizations must be sensitive. Kaluzny and Riordan (1984) present a case in point, observing the implementation of diagnosis-related group (DRG) regulations promulgated by the government for Medicare reimbursement to hosp"itals. This was in a sense a mandated innovation. The ability of hospitals to respond depended on both structural variables, attitudes of key actors as well as general financial status. The market/infrastructure or contextual models closely resemble the systems theory perspective, emphasizing the importance of an organization's interaction with its environment. As such, it complements the rational/economic and behavioral/communication models developed earlier. These models and perspectives suggest a wide and varied array of variables which affect the diffusion of innovation. It can be assumed, however, that the picture is not complete. The structure of organizations,the economic climate and the relationship of employees to" the organization, and the external environment will continue to change. In response, the factors which promote adoption and implementation of new ideas can be expected to be subtly altered as well.

PAGE 72

The above review is not an exhaustive list of models in diffusion theory, nor does it reflect the refinements and specific applications of the models. Rather, it is a general framework of selected models deemed most appropriate to this research effort. Developing a Framework for Prevention A number of studies have been conducted by researchers in 56 the field of health behavior. The studies have attempted to identify psychological, socioeconomic, and cultural determinants of health behavior. Early research was done by Hochbaum (1958) to identify factors associated with the decision to obtain an x-ray to detect possible cases of tuberculosis. Kegeler (1963) examined the conditions under which individuals sought preventive dental services in the absence of symptoms. Leventhal, Hochbaum and Rosenstock (1960) studied residents of two communities during an influenza epidemic. The purpose of the study was to ascertain psycho-social variables associated with taking the preventive actions of becoming immunized against influenza. These and a number of other studies form the basis of the Health Belief Model which states that certain conditions increase the likelihood of an individual's adopting preventive health practices. The conditions include: (a) the individual's perceived susceptibility to a particular condition, (b) the perceived severity of the impending illness or severity of the impending illness or condition, (c) the perceived

PAGE 73

efficacy of action weighted against the barriers to action, and (d) the presence of a "cue to action" (Becker, 1974). The Health Belief Model which focused on psycho-social factors dominated the research on preventive health behaviors Other studies, however, began to focus on such issues as access 57 to health care services and demographic variables. In a metaanalysis of the field, Cummings, Becker and Maile (1980, p. 135) list 99 variables thought to influence the adoption of certain preventive health practices. The variables were clustered in six major areas: (a) health threat, (b) knowledge, (c) social networks, (d) demographic characteristics, (e) evaluation of health care, and (f) access to health care. Familiarity with variables associated with a particular health behavior in individuals allowed for more effective targetting and design of preventive health programs. This literature provides a rich and useful base to understanding health behavior. It includes numerous studies dealing with compliance with medical regimens and a lesser number of studies examining adoption of preventive practice. Unfortunately, in almost all studies, the unit of analysis is the individual. This literature base provides useful background, but it does not provide a sufficient theoretical framework for the study of adoption of preventive innovations by organizations. Diffusion literature, by contrast, contains a sufficient theory base dealing with.the adoption of innovation by organizations.

PAGE 74

58 Resea rch on the dif fusion of preventive innovations, however ,. is limited. Preventive innovations were defined by Rogers (1984) as ". new ideas that an individual or organization adopts .at one point in time in order to avoid the possible occurrence 6f an unwanted consequence at some future time" (p. 22). Preventive innovations, notes Rogers (1984), bring a distinct set of character-istics and problems: Most preventive innovations are characterized by a relatively slow rate of adoption because potential adopters have difficulty in determining an innovation's relative advantage, they are not compatible with .individuals' values and attitudes, their cause and effect relationships are complex, trial is' difficult or impossible, and the innovations results are not very observable because they are delayed. (p. 6) This statement suggests that organizations might respond. differ-ently to worksite health promotion/disease prevention programs than to other more tangible innovations. 'Yet a recent study examining the adoption of Employee Assistance Programs (EAP) (NIAAA, 1984) suggests that many of the classical processes and variables applied. Personal communication demands, organizational autonomy, service compatibility,exposure to a model program, and general corporate concern for employees were all shown to be related to a decision to adopt an EAP program. Interestingly, the authors state (NIAAA, 1984) that "the single most important variable in exp1ain-ing policy adoption is the fact that the organizational key contact admits to having an alcohol or drug problem in his/her social 'network" (p. 15).

PAGE 75

It is clear from this brief summary that more work needs to be done to identify the processes and the variables which influence the diffusion of preventive innovations in organizations. Framework for Current Study 59 The current study will use diffusion theory as a framework for examining the processes and variables related to the diffusion of worksite health promotion/disease prevention programs. The study must necessarily focus on only certain aspects of the process. A complete understanding of all phases, actors and variables is certainly desirable, but well beyond the scope of this study. Unit of Analysis The unit of analysis will be the organization. The study will focus on organizational characteristics and processes which influence the diffusion process. This decision is based on the fact that the successful delivery of hp/dp programs at the worksite typically require policy changes and resource allocation decisions to be made at the organizational level. In addition, numerous authors (Everly & Feldman, 1985; O'Donnell & Ainsworth, 1985; Parkinson et a1., 1982) cite organizational commitment as a necessary first step and a key element to the development of successful hp/dp programs. Further, given limited resources, it is likely that the government's role in the future will be to stimulate the development of worksite hp/dp programs by employers rather than to serve as a direct service provider to emp19yees. Therefore,

PAGE 76

information regarding organizational processes of adoption is expected to be more relevant to future public policy and program initiatives. Stage As stated earlier in this chapter, the process of diffusion is generally broken down into five stages. As presented by Rogers (1983, p. ), these stages include knowledge, persuasion .. decision, implementation and confirmation (see Figure 3). The current study will focus on the decision stage, examining differences between companies which decided to adopt and those which decided to not adopt an hp/dp program. Downs and Mohr (1976) reconunend focus.ing on this stage because ". the focus is thus transferred from the tool itself to the social acts by which the tool becomes part of the systemll (p. 706). Feller (1978) concurs, stating that the examination of lithe decision to adoptll provides 60 a perspective on the interrelationship of the innovation and adopter needs and processes. In addition, a decision by management to adopt or not adopt an hp/dp program is a specific indentifiable action culminating what might be a more diffuse, less documentable process in the earlier stages of diffusion. Of necessity, information \"ill also be gathered about what happened in the awareness and persuasion stages which IIset the stagell for the eventual decision. Implementation and confirmation are obviously important stages of the total process, but time and resource constraints relegate the consideration of these issues to future research.

PAGE 77

;Implementation Figure 3. Stages in the Diffusion Process Confirmation (j\ I-'

PAGE 78

Diffusion Models Only limited research has been done regarding the diffusion of worksite hp/dp programs. It would, therefore, be premature to select a single model or set of hypotheses to test. Rather, the study will be exploratory in nature. Questions will be designed to elicit general information about the factors which affected the decision to adopt or to not adopt a program. Some questions will be structured around each of the major diffusion perspectives reviewed in this chapter (see Figure 3). These questions wi'll be open-ended in nature so as not to preclude or predetermine which variables were most important in the process. Summary In summary, diffusion theory is a rich and dive.rse body of knowledge which has provided the framework for examining the rate and pattern of adoption of innovations in society. A number of theoretical models have evolved adding new and useful analytical perspectives. Despite the volume of work done in diffusion, relatively little attention has been focused on the adoption of preventive innovations by organizations. This research effort will use diffusion theory a framework for examining the adoption! non-adoption of worksite hp!dp programs. The focus will be on the decision phase of the diffusion process and the factors influencing this decision. The unit of analysis will be the organization. The research effort will be exploratory in nature attempting to 62

PAGE 79

generate hypotheses for future research as well as validating general hypotheses presented in diffusion theory. 63

PAGE 80

CHAPTER III RESEARCH METHODOLOGY The purpose of the research project was to better under-stand the process and the variables related to the adoption or non-adoption of worksite health promotion/disease prevention (hp/dp) programs. The study was largely exploratory in nature; it did, however, include testing of selected hypotheses suggested by general diffusion theory. Twelve companies which had adopted worksite hp/dp programs (adopters) were studied, along with 12 companies which had actively considered but had not adopted such programs (non-adopters). The research examined selected conditions,attitudes and practices which existed in these companies at the time the decision was made to adopt or not adopt an hp / dp program. This type of study design is referred to in and Stanley (1963) as a correlational, data analysis design with ..... Xl representing the decision in 12 companies to adopt an hp/dp program and X 2 representing the decision of 12 other companies to not adopt a program. Both sets of companies were subsequently examined in the same manner as represented by 0 in the figure.

PAGE 81

The sample is matched for size of company, an organizational variable known to be correlated with the adoption/non-adoption of worksite health promotion programs (Davis et a1., 1984; Fielding & Breslow, 1983). This study design was chosen because relatively little research has been done regarding the diffusion of worksite hp/dp programs and, therefore, the literature does not suggest specific hypotheses to be tested formally. It is the intent of this project to develop a better understanding of the context and the process by which companies make the decision to adopt or not adopt hp/dp programs, thus generating hypotheses for further study. Previous Survey Sample 65 In 1983, the Colorado Department of Health and the Institute for Health in collaboration with several voluntary health agencies conducted a survey of worksite hp/dp programs in Colorado. General purposes of the survey were: (a) to develop a profile of business and industry hp/dp programs in Colorado, and (b) to identify obstacles and incentives to the further development of such programs. A listing of all Colorado employers with 50 or more employees was obtained from the Colorado Department of Labor and Employment. The list contained company name, location by county, type of business and number of employees. A stratified random sample was drawn as follows:

PAGE 82

66 Size of Number in SamE1e 50-99 employees 136 100-249 employees 149 250-499 employees 150 500-999 employees 141 1000 or more employees 103 Because of the small number of large businesses in the state, all companies with 500 or more employees were included in the sample. The 150 companies with 250-499 employees represent a 60% sample of businesses of the size in the state, the 149 companies with 100-249 employees constitute a 15% sample, and the 136 businesses with 50-99 employees constitute a 7% sample of businesses of this size in the state. A random sample was drawn in each of the remaining size categories. Previous studies suggest that smaller companies are less likely to have organized multiphased hp/dp programs. Thus, no sample was drawn for the category of 50 or fewer employees. The final sample included both private businesses and public agencies. Trained volunteers contacted the chief executive office of each company selected in the sample. The hp/dp survey was briefly explained and interviewers then asked to be referred to the person most knowledgeable about the company's efforts in these areas. A phone call was then made to this identified contact person to deter-mine: .(a) if the company had an ongoing hp/dp program, or (b) if the company was interested in developing such a program in the future. If neither of these conditions existed, the interviewer thanked the contact person and terminated the process. If either of these conditions existed, the company was considered "eligible,"

PAGE 83

and the contact person was asked if he/she would be willing to participate in the project. A telephone interview was then scheduled. A packet including a questionnaire (see Appendix A), an explanatory cover letter and a letter signed by the Governor of Colorado was sent to each company contact person. Companies with ongoing programs received a long form of the questionnaire. Companies without programs, but interested in developing them in the future, received an abbreviated form of the same questionnaire. Ten days were allowed for receipt of the packet and the gathering of program information by the contact person. The contact person was then recontacted by the interviewer and a 15-25 minute interview was conducted. 67 Of the 679 companies in the original sample, 321 were excluded for the reasons summarized in Table 1. Three hundred interviews were completed from the sample of 358 eligible companies, for an overall response rate of 83.8%. Of the 300 private businesses and public agencies participating in the survey, 94 currently have an hp/dp program. A company was considered to have an hp/dp program if it provided health screenings, classes, and/or preventive health services on an ongoing basis. Two hundred six companies did not currently have hp/dp programs, but expressed interest in developing such programs in the future.

PAGE 84

Table 1 Companies Included/Excluded From Sample: 1983 Survey Total sample drawn 679 Companies excluded 321 Reason for exclusion: phone disconnected/no listing repeats within list wrong size/out of state No program/no interest in developing program Eligible companies initial refusals refusal after survey sent/incomp1etes completed surveys Companies with ongoing program Companies interested in developing program 54 76 81 110 358 9 49 300 94 206 68

PAGE 85

Size of Company 50-99 employees 100-249 employees 250-499 employees 500-999 employees 1000 or more employees Number of Companies Responding Existing Interested Program in Program 2 7 19 22 44 33 57 42 36 38 206 The survey examined: program goals, program activities, company policies, program management, program outcomes, utilization of community resources, and services needed to develop or expand 69 programs. Among the major findings of the study were the following: -There is a high level of interest in hp/dp programs among private businesses and public agencies. Nearly two-thirds of companies contacted either had established a program or were interested in developing such a program. -Size of the company appears to be a major factor in determining whether or not it has an hp/dp program. Large companies, particularly those with over 500 employees, are much more likely to have programs than are smaller companies. -1978 marked the beginning of an upswing in the development of worksite hp/dp programs. Over 70% of current programs are less than 5 years old. -There are a variety of reasons for starting hp/dp programs including improved employee health/reduced illness, reduced health care costs, improved morale, improved productivity, and reduced turnover and absenteeism.

PAGE 86

-Improved employee morale was perceived as the most common benefit of hp/dp programs. Other commonly perceived benefits include improved employee health/reduced health problems, reduced illness and injuries on the job, and increased productivity. -The majority of programs are operated using in-house staff and a variety of free community services. Few companies plan any major purchase of hp/dp services. 70 While the survey provided an interesting profile, it did not sufficiently explain the pattern of or reasons for adoption or nonadoption of worksite hp/dp programs. Sample for Current Study The sample for the current study was a subset of companies which had participated in the 1983 study. The term "company" as used throughout the text refers to both private businesses and public agencies. The companies were not selected at random, but rather were taken from a list of companies which had participated in the 1983 study and had signed a consent form indicating that they would be willing to be contacted in the future. The expressed willingness of these agencies and businesses to be contacted again was considered key to the success of the proposed study. It was assumed that these persons had a moderate level of interest in hp/dp program and had given some thought to issues related to program development. Of the 300 companies participating in the 1983 survey, 54 had signed the contact consent form.

PAGE 87

71 The sample was further reduced by focusing only on companies which had between 250-999 employees. As a group, companies within this size range (a) account for 15% of the workforce nationwide, (b) have not been studied extensively, and (c) have a reasonable adoption rate that allows for examination of factors related to adoption and non-adoption of hp/dp programs (Davis et a1., 1984; Fielding & Breslow, 1983). The decision to exclude companies of 1,000 or more employees was made on the following basis: (a) the adoption rate in this size group is considerably higher than in the other size categories, and (b) this size group has been the subject of most of the research and program documentation in worksite hp/dp programs. The decision to exclude companies with 100-249 employees was based on a small eligible sample and the impa1ance between adopters and non-adopters. Of the 64 companies this size participating in the 1983 survey, only 7. reported having active hp/dp programs. A similar situation exists in companies with .50-99 employees. Of the 35 companies participating in the sample only two reported active programs. Of the 117 companies, with between 250-999 employees that participated in the 1983 survey, 28 signed a consent form indicating that they would be willing to be contacted in the future. This group became the eligible sample. Their distribution was as follows:

PAGE 88

Size of Company 250-499 employees 500-999 employees Companies With Programs (Adopters) 4 11 15 Companies Interested in Program (Non-Adopters) 4 9 13 All 28 of these companies were contacted. Subsequent additions and deletions to this group are summarized in Table 2. One company with a program categorized as having 500-999. employees was dropped from the sample when it was learned that they had a total of 2,600 employees with between 500-999 in a single location. The remaining 27 companies on the Consent to Contact list were contacted and 72 screened to determine whether they had been appropriately categorized as adopters or non-adopters. Companies were considered to have adopted an hp/dp program if: 1. they offered three or more health promotion activities on a planned, on-going basis, and 2. the program was approved by management as a company program/activity. If companies did not meet these criteria. they were not considered adopters. Two companies (500-999) listed as having programs in the 1983 survey, did not meet these criteria. In both instances, some activities were undertaken on a volunteer basis but management had never given approval. Neither company themselves to have an organized, approved hp/dp program. They were, therefore. reassigned to the non-adopter category along with the

PAGE 89

other companies that were intererested in hp/dp but had not yet begun a program. This group was further screened to determine if they, 'indeed, could be considered non-adopters. The following criteria were established for designation as non-adopters: 1. the idea of hp/dp had been under active discussion by a group within the organization, and 2. the idea has been presented to management and either rejected or tabled. These criteria were, established in order to select non-adopters who had passed through the 'awareness stage and the persuasion stage and had actually reached the decision stage in the diffusion process. Decision Table 12 Companies which were only in the awareness or early persuasion stages were dropped from the sample. This resulted in the exc1u-sion of two companies with 250-499 employees and three companies 73 with 500-999 employees., In each case, the idea of an hp/dp program had been the interest of the contact person for the 1983 survey, but had not been actively discussed in the organization nor presented to management. It was decided that information from

PAGE 90

these companies would be extremely limited in nature. Of all the companies contacted, one refused to be interviewed. In order to maintain a balance between companies with programs and companies interested in programs, three replacements were drawn at random in the appropriate size category from the original list of companies which had participated in the 1983 survey. These companies were contacted, met the criteria, and agreed to become part of the final sample; Thus the final sample was constituted as follows: Size of Company 250-499 500-999 Companies With Programs (Adopters) 4 8 12 Companies Interested in Program (Non-Adopters) 4 8 12 Of the 24 companies participating, 14 were private businesses and 10 were public agencies. Of the 12 adopters, 7 were private businesses and 5 were public agencies Of the 12 non-adopters, 9 were private businesses and 3 were public agencies. Public Agency Private Business Adopters 5 7 12 Non-Adopters 3 9 12 Total 8 16 Of the adopters, 4 were involved in some aspect of health care delivery, 8 were not. Of the non-adopters, 7 were health care-related, 5 were not. Withln each company, an effort was made to identify and interview persons representing: (a) administration, (b) human resources/benefits, and, if possible, (c) health services. The 74

PAGE 91

Table 2 Companies Included/Excluded from Sample: Current Study Number of companies (250-999) Participating in 1983 Survey and Sumitting Consent to Contact Fanus Alterations to Original Sample: Reason Deletion at Addition Company too large to be in sample Company misclassified/did not yet have program Company not "yet in decision stage of adoption process Refused to participate Replacement drawn from initial survey list Company misclassified/had recently begun program Adopters 15 -1 -2 -1 +1 12 75 Non-Adopters 13 +2 -5 +3 -1 12

PAGE 92

76 decision to focus on persons from administration and human resources/ benefits was based on the findings of the 1983 Colorado survey (David et a1., 1984) and a similar survey done in California (Fielding & Breslow, 1983) which indicated that persons from top administration and the human-resource/benefits department were most likely to be involved in decisions to offer or not offer worksite hp/dp programs. A total of 58" persons were interviewed, 29 each from adopting and non-adopting companies. A minimum of two persons were interviewed in each company with three persons being interviewed in 10 companies. Positions of Persons Interviewed Top Management: General Top Management: Human Resources Human Resources: Non-Management Health Service/Health Education Total N 10 7 17 24 58 % 17.2 12.1 29.3 41.4 100.0 A sizeable number of interviewees (41.4%) involved persons from the human resources departments. Many of these persons (7 out of 24) were also considered to be top management. Twenty-four (41.4%) persons interviewed were from employee health service and/or health education units. Ten persons (17.2%) were from top management including chief executive officers (CEOs), directors of marketing, assistant superintendents and/or board members. Survey Instrument A review of the worksite hp/dp programs and of the diffusion literature was conducted to develop a conceptual framework for the

PAGE 93

study and to determine if validated instruments existed which could be used. Extensive research has been done in diffusion, yet none 77 of the studies dealt specifically with worksite hp/dp programs; no relevant, validated instruments appeared in the literature. Experts in the field of diffusion and health promotion research, Dr. Everett Rogers, Dr. Gregory Christenson, Dr. J. D. Eveland, and Dr. Donald Iverson, were then contacted to determine if instruments in this area had recently been developed. Correspondence and personal conversation with them confirmed that no such instruments existed. Further, they suggested that, in view of the limited knowledge base in the area, that .the research focus and the survey instrument be designed to elicit information about the process of diffusion as it occurred or did not occur in the companies studied. As a result, instrumentation was developed suited for use in a guided interview (see Appendix B) The guided interview questionnaire included the following sections: Section A: The purpose of this section was to establish the current level of program activity and the involvement of different departments in planning/developing the program. Section B: The purpose of this section is to examine how the interest in hp/dp initially developed, the factors which promoted and/or discouraged the development of a program, and the administrative environment in which the program developed.

PAGE 94

Section C: The purpose of this section is to present general hypotheses about factors thought to influence the adoption of hp/dp programs, and to verify or refute these factors as important in the adoption decision process of companies. Each section contains a series of open-ended questions including probes to be used to elicit additional information if necessary. The relationship between diffusion theory and specific questions is presented in Figure 4. The intent and rationale for each item on the questionnaire is presented below. 78 Question 1: Would you please describe the health promotion activities which your company offers for its employees. This question was designed to determine whether the company should be classified as an "adopter" or "non-adopter" based on criteria discussed on page ,or whether it should be excluded from the eligible sample. Question 2: Would yOu talk a little bit about the parts of the program which have been particularly successful. This question was included to put people more at ease in talking about their program. It was not designed to test any hypothesis or gather information on any variable, thus criteria for "successful" and "less successful" were purposely not specified. Question 3: Can you describe the involvement of different depart-ments and people in developing the program (or) in planning for a program? This question was used to determine the position of the interviewer in the organizational structure. These data would

PAGE 95

RATIONAL/ECONOMIC PERSPECTIVE Economic Model Primary: Question 11 Secondary: Questions 5 & 6 Structural Model Primary: Question 9 Secondary: Question 3 SYSTEMS 'PERSPECTIVE Marketing Modei BEHAVIORAL PERSPECTIVE Communications Model Primary: Questions 7 & 8 Secondary: Question 2 Champion Model Primary: Question 10 Secondary: Questions 5 & 6 Primary: Question 3 Secondary: Question 5. & 6 Contextual Model Primary: Questions 5 & 6 Secondary: Question 11 Figure 4. Inclusion of Diffusion Perspectives in Interview Guide: Primary and Secondary Focii 79

PAGE 96

allow for comparison of responses by management level or area of responsibility. The question also provided the interviewer with a sense of the extent to which the interviewee had been involved with the adoption decision process and was therefore knowledgeable about the relevant variables. Question 4: When did you first start talking about a program? If appropriate In what year did you actually start the program? This question was designed to focus the ine.rviewee' s attention on the time span prior to program implementation. It also provided information as to the length of time companies had been in the awareness and persuasion stages. Question 5: How did interest in employee health promotion first develop in this company? This question was included as an open-ended question to allow respondents to identify or speculate on the variables that influenced the diffusion process. The focus of the question is on the awareness stage of the diffusion process. Question 6: When the idea. of employee health promotion was pre sented to management or a management committee, what were some of the issues that came up. both "pro" and "con"? This question provides a similar opportunity for respondents to identify or speculate on key variables which arose in the persuasion stage of the diffusion process. 80

PAGE 97

Questions 7 and 8: Can you describe some of the first activities you decided to offer (or) the first activities you would offer? and What were some things that you had considered doing, but backed off from (or) What are some things which you would back away from? 81 These questions provide an Qpen-endedopportunity for respondents to identify product variables considered important in the decision to adopt or not adopt. The information from this question was also used to classify the proposed programs as being simple or complex. Question 9: Once the proposal for a program was written, what was the process for approval in this company (or) would be the process for approval in this company? This question addressed select organizational variables hypothesized to influence the decision process, including the formalization and centralization of the decision process. In addition, information was generated regarding the general management philosophy. Question 10: Can you describe key individuals who strongly supported your program? This question was designed to test the champion model as well as the related variables of staff expertise and top II).anagement support.

PAGE 98

Question 11: What kinds of discussions (if any) did you have regarding potential costs or potential savings for the company? 82 This question assessed the extent to which financial factors had played a role in the decision process. Information was also used to examine corporate financial status and length of return on investment. Question 12: What was your experience when you \V'ere shopping around? Were you able to find programs or activities acceptable to you? This question assessed the importance of program availability as a factor influencing the decision process. Information was also sought on design characteristics which made program components more attractive to prospective adopters. Question 13: Can you describe seminars you attended, articles you read or visits that you received from individuals promoting health promotion? This question focused on the exposure of mmnagement to information on hp/dp programs as a variable in the diffusion process. This included general level of exposure through conferences, written materials and sales visits as well as exposure to any specific model program. Questions 14 and 15: Is there any single factor that we have talked about or not talked about that you think has been the key element in this company? and Do you have any comments you would like to make?

PAGE 99

These questions provide the opportunity for respondents to elaborate on any or all of the variables discussed and identify the single variable they felt most influenced the decision process. The survey instrument was developed and sent to committee members and outside experts for review. They suggested: ea) that fewer, open-ended questions be used, utilizing probes to gain the more specific information", (b) that the sequencing of questions be changed, and (c) that clearer, more straightforward language be used in the questions. These suggestions were subsequently incorporated. The design of the survey instrument followed the model suggested by Patton in Qualitative Evaluation Methods (1980). The instrument was piloted with three colleagues, two of whom were familiar w1th issues in the field of worksite hp/dp programs. 83 Minor changes in wording were made and the instrument was subjected to a second pilot involving five individuals from two companies-one company which had an hp/dp program (adopter) and one company which had not (non-adopter). Only minor revisions were necessary as a result of the second pilot test. Finally, the dialogue preferred to be used in the initial phone contact with companies (see Appendix C) was established and piloted on three persons in three different companies. Data Collection The sample of companies to be included in the survey was based on representatives from companies 250-499 or 500-999 who had

PAGE 100

participated in the 1983 survey and agreed to be contacted again. Each of these persons was contacted by telephone and asked if they would be willing to participate in the study. Of the 24 companies participating in the study, 14 had the person contacted in the 1983 survey. In the remaining 10 companies, upon learning that the original contact person was no longer there, the caller asked to be referred to the person in the company who was most knowledgeable about the development of the hp/dp program in the company or had been involved in promoting or planning the program. These persons were identified as the contact person for this study. Contact persons were then asked to identify two other persons representing other company perspectives (either management, human resources, or health services) who had been involved.in promoting and planning the hp/dp program. Personal interviews were then scheduled. Interviews of all persons identified were conducted by a 84 single interviewer. The interviewer used the survey instrument and a guided interview process. The interviews were tape-recorded and required between 20 to 40 minutes. In all but two instances, the interviews were conducted in person. Persons in one company were interviewed by telephone; the information collected was complete and the quality comparable to that gained in personal interviews. Several items on the questionnaire, specifically probes in questions 11 and 12 (see items marked with IIDII in Appendix B), proved to be confusing to interviewees and did not yield consistent, usable data. To the question IIDid you see the program as costing

PAGE 101

85 or saving you money?" respondents typically answered "Obviously both." They would then enumerate program expenses and the interview wandered off course. To the questions relating to program characteristics, respondents answered that they designed their own programs or they talked about a content area specifically rather than about general program characteristics. After the first'lO interviews these items were excluded from the interview. Similarly, the probes in question 2 designed "to put interviewees at ease, were used selectively. If the contact person was involved in the planning or the initial decision to offer the program, but had no knowledge of current program" activities, probes in question 1 were not used. In the process of interviewing, several new variables and/or refinements on variables were introduced. They are marked with an "A" in Appendix B. These items dealt with two areas--general management philosophy and more specific information on level of staff expertise. After completion of the first round of interviews, telephone calls were made to contact personS in the companies interviewed early ,in the process to gather information on the new variables and/or other missing data items. Data Analysis Information from each taped interview was recorded on an interview record form (Appendix D). interview was then analyzed to determine how the respondent would classify the situation in their company on the identified variables. The interview guide was

PAGE 102

86 structured so that each of the variables was addressed on two or more occasions in the course of the interview (see Table 3). Responses were assessed for consistency among the answers given regarding each variable. In a majority of instances the answers were copsistent primarily due to the use of probes used during the interview process, e. g., if a person said that finances was an important part of the decision process and later in the interview stated that finances were not important, he/she was confronted with the inconsistency and asked to explain it. Information from all interviews within a company was consolidated into a ,single record. Responses from various company representatives were analyzed for consistency. When there was lack of congruence, another person from the company was identified and interviewed with the express purpose of addresing the inconsistency. When inconsistencies could not be explained, the evidence was examined and a judgment was made to resolve the issue. Each company was classified according to each dichotomous variable. A chi-square test was performed with the decision to adopt or not adopt being the dependent variable. Data were then subjected to discriminant analysis and associated procedures which resulted in a Wilks Lambda score, an Eigenvalue and a ,level of statistical significance for each variable. Variables demonstrated to be significant at the .001 level were loaded into a discriminant analysis formula in an attempt to arrive at a model that would predict adoption or non-adoption of hp/dp programs. A second

PAGE 103

Table 3 Source of Information on Variables Variable 1 2 3 4 5 6 Adopter/Non-adopter I Corporate Financial Status I" I Humanism/Accountability I ;; I Staff Expertise I I I Champion Theory I I I Angel Theory I I I Responsibility. for Employees I I Decision Process I Importance of Finances I I I of Benefits I I I Program Availability I I I Exposure to Model Program I I / Complexity of Program I I I Exposure in General I I I Facilities I I I Question Number 7 8 9 10 11 I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I 12 13 I I I I I I I 14 I I I I I I I I I I I I I I I 15 I I I I I I I I I I I I I I I (

PAGE 104

88 discriminant analysis formula was developed using only those variables on which information about a company was readily accessible, e.g., company financial status and presence or absence of facilities. Data from the study as well as the results of the statistical analysis are presented in the following chapter.

PAGE 105

CHAPTER IV RESULTS OF THE STUDY This chapter includes a comprehensive report of find-ings. Qualitative data and observations on the process of diffusion are presented along with the results of statistical analysis of selected variables. Table 4 summarizes the classification of the companies with regard to each of the independent variables examined. These data and other findings are examined in greater detail in text which follows. The information will be presented in three major segments: (a) information on current health promotion/disease prevention (hp/dp) activities being carried out in the 24 companies studied, (b) the history of the hp/dp efforts in these companies from the development of initial interest through the decision to adopt or not adopt, and (c) the reaction of respondents to. diffusion theory models. This order of presentation reflects the organization of the guided inter-view. The names of the companies in the study have been changed to assure confidentiality. The names assigned to the companies are entirely fictitious. Any resemblance to names

PAGE 106

Table 4 Classification of Companies on Variables Examined in Study Case Code Names Airlines Inc. 01 Allied Personal Servo 02 Bicycle Mfg. 03 Mountain Training 04 Family Funds, Inc. 05 Public Benefits 06 Meadowlark Resources 07 Research Assoc. 08 Hilltop Haven 09 Wooden Door Mfg. 10 Northwest Services 11 Engines Lcd. 12 Native Resources 13 Recreation Ltd. 14 Cafeteria Mfg. 15 All Saints 16 Health Marketers 17 Warehouse Co. 18 Charter Health 19 Healthworks South 20 Forklift Inc. 21 Red Brick Finances 22 Security Co. 23 Waverlv Medical 24 .. .. .. ... .. OJ "Variables GJ B >. u .... ta.t.J:! >-In V) C ...... U 4l QI ff CLf... 0 ..... GJ QI C s:: 0 en .-4 C >. CU ""' til ." 0 O'J ..-I aJ o to "CICU""; tI GJ::::I.o ....t c: >. GI OJ In "'"' en..,"" =' II) ........ u U"'UNO.J:! ,D "r4 '" ...-4 til .... .. "" .... "Q, e Ei 0 0 .... 0 GI .... c:t.. to ......,..-4 0 U '"'Btl" ..... I-at-a .... O C QJ-t ... til c.. 0 GI 0.0 GJ aJ .... .... .... GI c> C 0 U 0'\ '" GI U U .J:j GJ ........ Pot' )( C U) .u CD.o C'II'" .., c.. ,..... bl) ..... -< w .u C .... aJaJG.I ..... C .... CUX race cue'"' .... "" QJCW ............ e wcue raQ.l011 .... ,J:!C .............. GJcuQ, GJcao ..... .,... ....t(IJCU UlQJ .c u:J:f""'4cG,.: ftI...c: ;::IU'laJ .. o > > > > > > > > .-4 ..... ..... ,..... ..... "'"' M > > > > > > > > 1 1 1 1 1 1 2 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2 1 1 1 1 2 1 1 1 2 2 1 1 1 1 1 1 2 2 1 1 2 1 2 1 1 1 1 1 1 2 1 1 1 1 2 1 1 2 1 1 1 1 1 1 1 2 1 2 2 1 1 2 1 1 2 1 1 1 2 1 1 1 2 2 1 1 1 1 2 2 2 1 1 2 1 1 1 1 1 2 1 1 1 1 2 2 1 1 2 1 1 1 1 2 1 1 2 1 2 2 1 2 1 1 1 1 2 1 1 1 1 1 1 2 2 1 1 1 1 2 1 1 1 1 1 2 1 1 1 1 2 1 1 1 1 2 i 1 1 1 2 2 1 2 2 2 1 2 1 1 1 1 1 2 1 1 2 1 1 1 1 1 1 1 2 2 1 1 2 1 2 2 1 2 1 2 1 1 1 1 2 1 1 2 2 2 1 1 2 1 1 2 2 2 2 2 2 2 1 2 2 2 1 1 1 2 2 1 2 1 1 1 2 1 2 1 1 2 2 2 2 2 2 2 1 2 1 1 1 1 1 2 1 1 2 2 2 1 1 2 2 2 2 1 2 .2 1 1 2 2 1 2 2 1 1 1 2 2 2 2 1 2 1 2 2 2 2 1 2 2 2 1 1 2 2 1 2 1 2 1 2 2 2 2 2 1 1 2 1 2 2 2 1 2 1 2 1 1 2 2 2 2 2 2 2 2 1 2 1 2 2 1 1 1 2 2 2 2 1 2 2 2 2 1 2 2 2 2 2 2 1 2 2 2 2 1 2 2 2 1 1 2 2 1 2 2 1 1 2 2 2 2 1 2 2 1 1 2 2 2 1 1 2 2 1 1 1 2 2 1 2 2 2 2 2 2 2 2 2 2 1 1 2 \0 a

PAGE 107

. i of actual private businesses or public agencies is entirely coincidental Program Components Current Health Promotion/Disease Prevention Program Activities There was substantial similarity as well as interesting differences in the nature arid extent of hp/dp activities offered by the 24 companies in the study. Physical fitness was the most common component. In adopter companies, this frequently took the form of organized aerobics classes, frequently offered on-site. Proponents of hp/dp programs in non-adopter companies resorted to 91 less formal means including sports 'leagues and competitive sporting events sponsored solely by employees. High blood pressure screen-ing, smoking cessation, nutrition/weight loss, stress management and Employee Assistance programs were offered with moderate frequency, although only six of the companies offered the full complement of all activities listed above. Of the 12 adopter companies, only 4 had a line item for health promotion in the budget. In the remaining 8 companies, programs were part of the human resources budget or were funded simply through an internal 'reallocation of staff time. Activities carried out by persons in.non-adopter companies were most frequently made possible through an internal reallocation of staff time on a limited basis. For example, an occupational nurse supportive of

PAGE 108

92 prevention efforts might suggest and be given time to organize. participation in the Great American Smoke-Out, a campaign of the National Cancer Society. This would require only limited expenditure and frequently involved many volunteer hours on the part of staff. Successful/Unsuccessful Activities Respondents were asked to identify and describe activities which they judged to be most successful. The definition of successful was purposefully not included so as to encourage respondents to apply their own measures of success. In almost all cases, success was interpreted as a high level of employee and satisfaction. The activities identified as most ran the gamut of hp/dp program components. Common attributes of these successful programs were visibility, flexibility, minimal effort, novelty, and enjoyment. A surprising. number of respondents mentioned competition and one-time competitive events as successful undertakings. In a comparison question, respondents were asked to identify least successful activities. The range of response was equivalent to that of successful activities. Logistics, e.g., cancellation of facilities, late notices, was cited as the most common contributor to program failure. Other contributors included poor quality of presenters, inappropriate level of material presented, and program difficulty or complexity. Adopters and non-adopters cited many of the same characteristics as contributing to program success or failure.

PAGE 109

Departments Involved The development of a proposal for an hp/dp program was handled largely through human resources departments in the companies studied. In all of the adopter companies, they. played a central role. Persons from human resources were part of the care planning group in seven of the non-adopter companies but not involved with efforts in the remaining five companies. Respondents from two of these companies spetifically cited the lack of interest and involvement on the part of the human resources department as a major obstacle to program development. 93 Interdepartmental planning committees were formed in 12 of the companies--6 in adopter companies, 6 in non-adopter companies. In the remaining companies, planning and the development of a program proposal was handled by a single department and, in 3 cases, by a single individual. The presence or absence of an interdepartmental planning committee does not appear to be related to the decision to adopt or to not adopt an hp/dp program. Program History Development of Interest A variety of circumstances and events led to the development of interest in hp/dp programs in the 24 companies studied. In the majority of companies, there was no single "trigger event," but rather a convergence of both internal and external factors which heightened awareness and interest in developing programs.

PAGE 110

Respondents were asked' to identify key events, circumstances or people which led to t'he development of interest. Responses gener-ally fell into three categories: financial factors, the human factors, and factors related to the general wellness movement in society. Responses categorized as examples of financial factors include: We thought we might ,be able to market it. (Waverly Medical, 1985) Our Workmen's Compensation insurers have been increasingly concerned over the past years about our high rate of disability. (Public Benefits, 1985) We knew what we wanted in general--the grant funds made it possible. (Hilltop Haven, 1985) We became self-insured and needed to look more carefully at our health care costs. (Mountain Training, 1985) Responses categorized as related to human factors include: We got a new general manager--one of the first programs she wanted to get started was a wellness program. (Bicycle Mfg., 1985) Original interest came from the chairman who was on the Board of Directors of Hospital. (Meadowlark Resources, 1985) Program probably developed because of the type of personnel we have here they're more educated they expect it. (Bicycle Mfg., 1985) Statements categorized as examples of factors related to the general we11ness movement in society include: They got religion about wellness. (Family Funds, Inc., 1985) We11ness is just such a topic everywhere. (Airlines, Inc., 1985) 94

PAGE 111

We must give cognizance to was being said in the mid to late 70's about lifestyle change and about health promotion as an issue. Because we are in touch with such things, it turned us on. The basis for the belief is out there. (Charter Health, 1985) Twenty-two of the companies studied cited human factors as con-95 tributing to the development of interest. Fifteen repor'ted finances as a consideration and 13 alluded to the general we11ness movement in society as a factor in their developing a program. Factors Contributing to Initial Interest Human factors 22 Financial factors 15 Societal we11ness movement 13 Although multiple factors were often mentioned, not all were given equal weight by the respondents. One or two factors tended to dominate. ',An attempt was made to classify the factors as dominantly external or dominantly internal. In only a small number of cases were the contributing factors only external or only internal. The line between internal and external factor was often unclear. Therefore, the attempt at categorization was abandoned. One respondent aptly described the interplay of both internal and external factors. It was the lucky congruence of rich and fertile ground and fertilizer to feed it. (Angel Security Co., 1985) Factors contributing to the development of interest in hp/dp programs were then categorized as being of a positive or a negative nature. Factors were judged to be negative if they

PAGE 112

96 presented a problem or a crisis to which the company must respond. They were considered positive if they represented an opportunity for improvement or growth. Rising health care and worker's compen-sation costs would be considered negative financial factors; a perceived opportunity to market the program to the community would be considered a positive financial factor. A heart attack of someone in top management could be considered a negative human trigger, while the chance to create a more caring environment for workers would be considered to be ofa positive nature. Improving a poor public image would be a negative factor related to environment; promoting an already positive image of innovativeness would be seen as a positive action. Nature of Factors Cited Positive only Positive dominant Negative dominant Negative only Adopter 6 5 I o Non-Adopter 7 4 I o The overwhelming majority of factors which stimulated initial interest in hp/dp programs were positive in nature. Only two companies began exploring hp/dp programs out of a problem-oriented mode. The majority of companies, both adopters and non-adopters, pursued the programs initially out of a growth mentality. In contrast, it should be noted that Employee Assistance Programs (EAP),when discussed separately from hp/dp programs, were almost always traced to a negative trigger event. High suicide rates, loss of valued employees, a forklift driven through a wall, or high absenteeism and injury rates were all given as

PAGE 113

examples of problems which spurred companies to take action in the EAP area. Information on trigger events leading to the development of separate EAP programs was not systematically collected as a part of this study. The information was volunteered and recorded in only six instances. In five of these six cases negative factors dominated. Decision Criteria and Processes As companies moved from the awareness to thepersuasian phase of the diffusion process, proposals for hp/dp programs were developed and presented to management. Merits of the proposed programs were more closely examined. A review of issues raised in discussions with management provides a general idea of the criteria used to approve or disapprove a proposal. These criteria generally fell into five categories: (a) finances, (b) productivity and morale, (c) corporate responsibility for health of employees, (d) public image, and (e) liability issues. Finances. The impact of an hp/dp program on the finances 97 of a company could be viewed either positively or negatively. It could be seen positively as a vehicle to cut health care costs or negatively as an additional program expense. Companies in which financial issues were a major consideration were less likely to adopt an hp/dp program than were companies in which financial issues were considered less important. Statements reflecting a low level of concern for finances included:

PAGE 114

Bottom line was never a factor at all. (Family Funds, Inc., 1985) It's real interesting .. for the first. year and a half it was absolutely not a factor. (Research Assn., 1985) Statements reflecting a high level of concern for finance. included: Give them any program that doesn't cost money. They would love it. They really would. When it costs money, they do a Jekyll and Hyde. (Warehouse Co., 1985) We have to balance the budget every year .... can't see it (funds or saving generated by hp/dp) coming through the coffers. (Recreation Ltd., 1985) Finances were a major issue in only 2 of the 12 companies adopting programs. In contrast, it was a major issue in 11 of the 12 non-adopter companies. A cross tabulation of the distribution is presented in Table 5. Using a chi-square test, this difference between adopters and non-adopters is statistically significant at the .001 level. Discriminant analysis (see Table 6) revealed 98 that the variable "Importance of Finances" correctly predicted group classification in 87.50% of the cases. Benefits. Employee benefits, productivity and employee morale were a second set of issues which came up for discussion as hp/dp proposals were presented to management. In most instances, the programs were viewed as a mechanism to increase benefits and productivity and to improve morale. In some instances, they were viewed as cutting into productive time on' the job. Companies were classified as to whether or not employee benefits, including productivity and morale, were important factors in the decision

PAGE 115

Table 5 Association Between "Importance of Finances" Variable and the Decision to Adopt or Not Adopt a Worksite Health Promotion! Disease Prevention Program Decision Adopt Non-Adopt Count Row Pct Col Pct 1 2 Column Total Importance High 1 2 16.7 15.4 11 91. 7 84.6 l3 54.2 of Finances Low 2 10 83.3 90.9 1 8.3 9.1 11 45.3 Chi-square = 10.74126 with 1 degree of freedom Significance. = .0011 Row Total 12 50.0 12 50.0 24 100.0 99

PAGE 116

Table 6 "Importance of Finances" Variable as a Predictor of Adoption/ Non-Adoption of Worksite Health Promotion/Disease Prevention Program Classification Results 100 No. of Predicted Group Hembership Actual Group Group Adopt Group Non-Adopt 1 2 Cases 1 2 12 12 10 83.3 1 8.3 2 16.7 11 91. 7 Percentage of Grouped Cases Correctly Clasiified 87.50 Note. Eigenvalues Canonical Correlation Wilks Lambda = Significance = 1.30645 .7526178 .4335664 .0000

PAGE 117

process. In llcompanies concern for employee benefits was high; in 13 companies, it was low. A cross-tabulation of the distribution is presented in Table 7. Companies in which employee benefits were of major importance were more likely to adopt hp/dp programs than were companies in which employee benefits was a minor consideration. Using thechi-square test this difference between adopters and non-adopters is significant at the .001 level. Discriminant analysis (see Table 8) revealed that the variable "Importance of Benefits" correctly predicted group classification in 87.5% of the cases. Corporate responsibility for health of employees. Corporate responsibility for the health of employees surfaced as a third major issue considered by management in deciding whether or not to adopt a program. Management in a number of companies expressed the philosophy that business has only a limited responsibility for employee health. Statements to this effect include: We pay men good wages, they can buy their own car seat. (Warehouse, Inc., 1985) Don't really feel it is a corporate function at this time; (Cafeteria Mfg., 1985) Employee health service is beginning encroachment of socialized medicine. The Joint Budget Committee said this (more health services for public employees) is not something we should be doing. (Hea1thworks South, 1985) In contrast, other companies promoted an expanded role for their company in the health of employees. Statements to this effect include:

PAGE 118

. Table 7 Association Between "Importance of Benefits" Variables and the Decision to Adopt or Not Adopt a Worksite Health Promotion/ Disease Prevention Program Decision Adopt Non-Adopt Count Row Pct Col Pct 1 2 Column Total Importance High 1 10 83.3 90.9 1 8.3 9.1 11 45.8 of BEnefits Low 2 2 16.7 15.4 11 91. 7 84.6 13 54.2 Chi-square = 10.74126 with 1 degree of freedom Significance = .0011 Row Total 12 50.0 12 50.0 24 100.0 102

PAGE 119

Table 8 "Importance of Benefits" Variable as"a Predictor of Adoption/ Non-Adoption of Worksite Health Promotion/Disease-Prevention Program Classification Results -103 No. of Predicted Group Membership Actual Group Cases 1 2 Group Adopt Group Non-Adopt 1 2 12 12 10 83.3 1 8.3 Percentage of Grouped Cases Correctly Classified Note. Eigenvalues Canonical Correlation = Lambda = 1.30645 .7526178 .4335664 Significance = .0000 87.50 2 16.7 11 91. 7

PAGE 120

104 If the employee feels better, he's going to do a better job, have a better life. (Wooden Door Mfg., 1985) This company is centered around health. (Bicycle Mfg., 1985) We have a corporate goal to promote good health and physical fitness of employees. (Airlines, Inc., 1985) Not all persons in the-companies agreed with the expanded role. Even in adopter companies this view was frequently challenged. But \.;rhere this philosophy dominated top management, a company was classified as promoting an expanded responsibility for employees and employee health. When management saw a more circumscribed role they were classified as promoting a limited responsibility for employees and employee health. Companies promoting an expanded responsibility for health were more likely to adopt hp/dp programs than were companies subscribing to a more limited role. Fourteen companies expressed a philosophy of expanded responsibility for employees; 10 expressed a philosophy of limited responsibility. A cross-tabulation of the distribution is presented in Table 9. Using a chi-suare test, the difference between adopters and non-adopers is significant at the .004 level. Discriminant analysis (see Table 10) revealed the variable "Responsibility for Employee" correctly predicted group classifi-cation in 83.3% of the cases. Corporate image/liability. Corporate image and liability were other issues raised with less frequency. Hp/dp programs in general were considered to have a positive effect on corporate image:

PAGE 121

Table 9. Association Between "Responsibility for EmployeeH Variable and the Decision to Adopt or Not Adopt a Worksite Health Promotion/ Disease Prevention Program Decision Adopt Non-Adopt, Count Row Pct Col Pet 1 2 Column Total Responsibility High 1 11 91.7 78.6 3 25.0 21.4 14 58.3 for Employee Low 2 1 8.3 10.0 9 75.0 90.0 10 41. 7 Chi-square = 8.40000 with 1 degree of freedom Significance = .0038 Row Total 12 50.0 12 50.0 24 100.0 105

PAGE 122

Table 10 "Responsibility for Employees" Variable as a Predicto'r of Adoption/Non-Adoption of Worksite Health Promotion/Disease Prevention Program Classification Results 106 No. of Predicted Group Hembership Actual Group Group Adopt Group Non-Adopt 1 2 Cases 1 2 12 12 11 91. 7 3 25.0 1 8.3 9 75.0 Percentage of Grouped Cases Correctly Classified = 83.33 Note. Eigenvalues Canonical Correlation Wilks Lambda Significance .84211 .6761234 .5428571 .0003

PAGE 123

[hp/dp programs] ... would help maintain the structure and the favoreod reputation that [Company X] has achieved as an employee-oriented company. (Airlines, Inc., 1985) It would increase our credibility when selling programs to businesses. (Health Marketers,1985) Occasionally, programs were viewed as detrimental to public image: If the public got hold of the fact that we were building showers for employees, they would get outraged. (Recreation Ltd., 1985) 107 This type of comment was made almost exclusively by public agencies conscious of taxpayer scrutiny. Liability surfaced as a management concern in five companies. Issues ranged from potential injuries on outdoor exercise courses, to heart attacks during exercise, to physician skepticism of holistic medicine practices which they feared might become part of an hp/dp program. Although image and liability issues were points of discus-sion in a number of companies, they did not appear ,"ith the same frequency and intensity as other issues. No statistical analysis was performed. Activities proposed/activities deferred. Respondents were asked to identify the first activities they proposed to management and the reasons why they selected these particular aCotivities. Some companies began several activities simultaneously. Ten companies included high blood pressure screening as one of the first activities to be initiated. The remainder of first activities

PAGE 124

ran the gamut from safety and seatbelt programs (3 companies) to health fairs (4 companies) to exercise and fitness competitions (7 companies) to smoking cessation (3 companies), health risk appraisal survey (16 companies), stress management (5 companies). Although initial programs chosen were diverse, the reasons given for choosing the activities were very similar. The most common reasons given were: simplicity, staff or other resources available, small investment required in time and money, enjoyable activity perceived as something that could be successfully done by employees, compatibility with services currently offered, employee interest and identified risk, e.g., smoking. Employee interest was a more dominant factor than was health risk. Respondents in 17 of the 24 companies studied indicated that their choice of first activities was based more on employee interest than health risks in the employee population. Seven companies considered a health risk a more important factor than employee interest. Respondents were asked what activities they would defer or "put on the back burner." Most responses fell into four categories: (a) programs requiring special facilities, (b) programs requiring a large capital outlay, (c) controversial programs such as company-wide smoking cessation, and (d) programs that carried potential liabilities such as open PAR courses. The first activ ities proposed and the activities deferred were similar among adopter and non-adopter companies.

PAGE 125

.109 Complexity of proposed programs. Further analysis was done of the programs proposed. Each program was reviewed and categorized as complex or simple. Programs were considered simple if: (a) they could be done with existing resources, (b) they began with only one or two activities, and (c) if they targeted only a portion of the workforce. They were considered complex if (a) they required additional approval, (b) they began with three or more activities as part of the initial effort, and (c) they targeted the general workforce. Programs were classified on the basis of possessing two of the three characteristics. Of the programs proposed, 12 were judged to be simple in nature, 12 were judged to be complex. A cross-tabulation of the distribution is presented in Table 11. The chi-square test reveals no significant association of program proposed and the decision of a company to adopt or not adopt a program. Decision process. The approval process for hp/dp programs varied greatly among the companies. In some companies; the ability to develop new programs and to commit large amounts of money rested in autonomous units; in others, the process was very centralized. Respondents were asked to describe the approval process and classify it as formal or informal, centralized or decentralized. In 14 companies the process was designated as centralized. Comments to this effect include:

PAGE 126

Table 11 Association Between "Program Proposed" Variable and the Decision to Adopt or Not Adopt a Worksite Health Promotion/'Disease Prevention Program Decision Adopt Non-Adopt Count Row Pct Col Pct 1 2 Column Total Program Proposed High Low 1 4 8 33.3 66.7 33.3 66.7 8 4 66.7 33.3 66.7 33.3 12 12 50.0 50.0 Chi-square = 1.50000 with 1 degree of freedom Significance = .2207 Row Total 12 50.0 12 50.0 24 100.0 110

PAGE 127

If it were really to fly, it would need corporate support from all departments concerned. (Angel Security, 1985) He [CEO] mandated it. He wanted things done and done by this time. Definitely top drawer. (Engines, Ltd., 1985) Very complex process. A lot of people have to sign off. (Northwest Services, 1985) In the remaining 10 companies the approval process was less centralized. 1.Jhen I started this, I didn't even ask permission. I just did it. (Allied Personal Services, 1985) I've really been given freedom to do whatever I want. (Airlines, Inc., 1985). We have a lot of autonomy at this site. (Cafeteria Mfg., 1985) Many respondents noted that the formality and the centrality of the decision process was expected to increase as the amount of money required for the program increased. The decision process was centralized in 14 companies and decentra1ized in 10. A cross tabulation of the data is presented in Table 12. The chi-square test reveals there is no significant association between the centralization of the decision process and the decision made to adopt or not adopt an hp/dp program. General Company Environment Management philosophy. Respondents were asked to charac-terize management philosophy, placing it on a continuum between an accounting mindset and a humanistic mindset. If persons identified as top management were not all of one philosophy, they were asked

PAGE 128

Table 12 Association Between "Decision Process" Variable and the Decision to Adopt or Not Adopt a Worksite Health Promotion/Disease Prevention Program Count "" Row Pct Col Pct Decision 1 Adopt 2 Non-Adopt Column Total Chi-square .17143 with 1 Significance = .6788 Decision Process High Low 1 2 6 6 50.0 50.0 42.9 60.0 8 4 66.7 33.3 57.1 40.0 14 10 58.3 41.7 degree of freedom Row Total 12 50.0 12 50.0 24 100.0

PAGE 129

to characterize the dominant philosophy. Management in 10 companies were characterized as having a humanistic philosophy. Comments to this effect included: The attitude here is incredibly better than anywhere I've ever worked. The managers here are cheerleaders. Our general manager is such people person. (Airlines, Inc., 1985) concern with growing number of emp1oyees--1ack of intimacy, community wanted through the we11ness program to try to enhance spirit. (Family Funds, Inc., 1985) Nanagement is very humanistic--we are a family. Services, 1985) In contrast, management in 14 companies was characterized as having more of an accounting philosophy. Comments to this effect include: The dollar is the bottom line. (Health Marketers, 1985) If I'm going to have to pay you every penny, you're going to have to give me every minute of your time. It's hard to go back and say let's do something nice for employees. (Warehouse, Inc., 1985) very far away from family. (Forklift, Inc., 1985) Ten companies operated with a philosophy characterized as humanistic; 14 companies operated with a philosophy characterized as an accounting mindset. Companies with a humanistic mindset were more likely to offer programs than companies with an account-ing mindset. A cross-tabulation of this distribution is presented in Table 13. Using the chi-square test, the difference between adopters and non-adopters is significant at the .0002 level. Discriminant analysis (see Table 14) test revealed that the variable "Management Philosophy" correctly predicted the group classification in 91.67% of the cases.

PAGE 130

Table 13 Association Between "Management Philosophy" Variable and the Decision to Adopt or Not Adopt a Worksite Health Promotion/ Disease Prevention Program Count Row Pct Col Pct Decision 1 Adopt 2 Non-Adopt Column Total Chi-square = 13.85571 Significance = .0002 Management Philosophy High Low 1 2 10 2 83.3 16.7 100.0 14.3 0 12 0 100.0 0 85.7 10 14 41.7 53.3 with 1 degree of freedom Row Total 12 50.0 12 50.0 24 100.0 114

PAGE 131

Table 14 "Management Philosophy" Variable as a Predictor of Adoption/ Non-Adoption of Worksite Health Promotion/Disease Prevention Program Classification Results .115 No. of Predicted GrouE HembershiE Actual Group Cases 1 Group 1 12 10 Adopt 83.3 Group 2 12 0 Non-Adopt 0 Percentage of Grouped Cases Correctly Classified Note. Eigenvalues Canonical Correlation Wilks Lambda Significance 2.5000 .8451543 .2857143 .0000 2 2 16.7 12 100.0 91-.67

PAGE 132

,116 Respondents in two of the adopter companies noted that at the time of the decision to adopt an hp/dp program, the management philosophy had been one of humanism, but that subsequently several key managers had been replaced by persons with more of an account-ing mindset so ,that the dominant philosophy at the current time was more in line with an accounting mindset. Facilities. In the course of the ,interview respondents described the type and location of facilities where hp/dp program activities were or could be carried out. The size of facilities ranged from small conference rooms to larger exercise facilities complete with equipment. Stress management and smoking cessation programs could be offered in conference rooms, while exercise programs required larger areas with shower facilities. Thirteen companies reported easy access to 11 companies had difficulty in arranging appropriate facilities. A cross-tabulation of the distribution is presented in Table 15. Using a chi-square test, the difference between adopters and non-adopters is significant at the .001 level. Discriminant analysis (see Table 16) reveals that the variable "Facilities" correctly predicted group classification in 87.5% of the cases. Of the 13 companies which had facilities available, only 2 / had areas specifically designed and designated for an hp/dp program. In the other companies, programs were given access to existing facilities that served other purposes for the company, e.g., showers available for workers who might cpme in contact with

PAGE 133

Table 15 Association Between "Facilities" Variable and the Decision to Adopt or Not Adopt a Worksite Health Promotion/Disease Prevention Program Decision Adopt Non-Adopt Count Row Pct Col Pct 1 2 Column Total High 1 11 91. 7 84.6 2 16.7 15.4 13 54.2 Facilities Low 2 1 8.3 9.1 10 83.3 90.9 11 45.8 Chi-square = 10.74126 with 1 degree of freedom Significance = .0011 Row Total 12 50.0 12 50.0 24 100.0 .117

PAGE 134

Table 16 "Facilities" Variable as a Predictor of Adoption/Non-Adoption of Worksite Health Promotion/Disease Prevention Program Classification Results .118 No. of Predicted Group Hembership Actual Group Group 1 Adopt Group 2 Non-Adopt Cases 12 12 1 2 11 91.7 2 16.7 1 8.3 10 83.3 Percentage of Grouped Cases Correctli Classified 87.50 Note. Eigenvalues Canonical Correlation Wilks Lambda Significance = 1.30645 .7526178 .4335664 = .0000

PAGE 135

hazardous substances. Almost every company had conference rooms of sufficient size to hold educational seminars. In some cases, however, these rooms were so heavily booked that hp/dp activities were in constant competition with other company activitie.s. In summary, the issue appears to be one of availability rather than merely existence of facilities. Companies with existing, underutilized facilities are more likely to offer hp/dp programs than companies which have no facilities or in which hp/dp programs would be in competition for limited resources. Hypothesis Testing A series of questions in the final section of the interview were designed to examine the relative of the human element, financial factors and environmental factors. In a general sense, they tested and compared components from the rationa1/ economic, the human relations, and the systems models. The following is a summary and analysis of the responses. Human Elements Champions/angels. Respondents were informed that opinion varied regarding the importance of the human element in a company's decision to adopt a program (Question #10). They were then asked to describe the experience in their company in this regard. Respondents were frequently able to identify a person who strongly supported the program and actively pushed it through the company.

PAGE 136

.120 Actions taken in support of the program were beyond normal job responsibilities. If the person identified was in mid-level management or below, they were called a "champion." If a person with these same characteristics was in top management, i.e., that group involved in the final decision process, they were called an "ange1." Statements used to describe the champion and the angel include: She's on the white.horse leading the charge. (Wooden Door Mfg 1985) We knew we were going to have a wellness program come hell or high water. (Hilltop Haven, 1985) It was the pet project of [CEO]. (Northwest Services, 1985) Anhp/dp champion, or champions, were identified by respond-ents in 13 of the 24 companies studied. A cross-tabulation of the distribution of champions between adopter and non-adopter companies is displayed in Table 17. Champions were almost equally distributed between adopter and non-adopter companies. The chi-square test reveals no signif-icant association between the presence or absence of champions and adoption or non-adoption of programs by companies. In contrast, there is an association between the "angels" and the adoption or non-adoption of a program. A cross-tabulation of the distribution of angels is presented in Table 18. Using the chi-square test, the association is significant at the .004 level. Discriminant analysis (see Table 19) revealed that the variable predicted the group classification correctly in 83.3% of the cases.

PAGE 137

Table 17 Association Between "Champion" Variable and the Decision to Adopt or Not Adopt a Worksite Health Promotion/Disease Prevention Program Champion Count Row Pct Col Pct High Low Row Total Decision 1 Adopt 2 Non-Adopt Column Total Chi-square = () with 1 Significance = 1.0000 1 .6 50.0 46.2 7 58.3 53.8 13 54.2 degree of freedom 2 6 50.0 54.5 5 41. 7 45.5 11 45.8 12 50.0 12 50.0 24 100.0 121

PAGE 138

Table 18 Association Between "Angel" Variable and the Decision to Adopt or Not Adopt a Worksite Health Promotion/Disease Prevention Program Decision Adopt Non-Adopt Count Row Pct Col Pct 1 2 Column Total High 1 9 75.0 90.0 1 8.3 10.0 10 41.7 Angel Low 2 3 25.0 21.4 11 91. 7 73.6 14 53.3 Chi-square = 8.40000 with 1 degree of freedom Significance = .0038 Row Total 12 50.0 12 50.0 24 100.0 .122

PAGE 139

Table 19 "Angel" Variable as a Predictor of Adoption/Non-Adoption of Worksite Health Promotion/Disease Prevention Program Classification Results 123 No. of Predicted GrouE HembershiE Actual Group Cases 1 2 Group 1 12 9 3 Adopt 75.0 25.0 Group 2 12 1 11 Non-Adopt 8.3 91. 7 Percentage of Grouped Cases Correctly Classified = Note. Eigenvalues = .84211 Canonical Correlation = .6761234 Wilks Lambda = .5428571 Significance = .0003

PAGE 140

Respondents in a number of adopter companies commented on the presence of a certain chemistry individuals supporting an hp/dp program. They also alluded to the need for a "critical mass" of key persons who must demonstrate interest in a for the program to advance within a company. The formula for the critical mass varied from company to company. Respondents were then asked to judge whether the basis of the champions' support was dominantly personal, or dominantly professional. It was almost an even split. Eight were judged to have personal interest and commitment; 11 were judged to support hp/dp programs primarily for professional reasons; the basis of interest of the remaining two champions was not sufficiently clear to judge. Although the majority of "angels" appeared to have a professional interest supported by a healthy personal lifestyle, insufficient data were gathered to make accurate judgments. Staff expertise. The presence of staff with expertise to deliver various hp/dp program components was suggested as a factor which might favorably influence a company's decision. If 50% or more of the proposed program could be delivered by existing staff, the company was considered to have a high level of staff expertise. If less than 50% could be delivered by existing staff, the company was categorized as low in staff expertise. Staff expertise was evenly distributed between the adopter and non-adopter agencies. A cross-tabulation of the distribution is presented in Table 20. A chi-square test reveals no significant association between level

PAGE 141

Table, 20 Association Between "Staff Expertise" Variable and the Decision to Adopt or Not Adopt a Worksite Health Promotion/Disease Prevention Program Decision Adopt Non-Adopt Count' Row Pct Col Pct 1 2 Column Total Staff Expertise High 1 8 66.7 50.0 8 66.7 50.0 16 66.7 Low 2 4 33.3 50.0 4 33.3 50.0 8 33.3 Chi-square = 0 with 1 degree of freedom Significance 1.0000 Row Total 12 50.0 12 50.0 24 100.0 125

PAGE 142

126 of staff expertise and the decision to adopt or to not adopt an hp/dp program in the 24 companies studied. In summary. it appears that the presence of a person in top management.who is supportive of the program is key to adoption. Expertise and support among staff or lower levels of management does not appear to substantially influence the initial program decision. Economic Issues Respondents were informed that opinion varied as to the importance of finances in a company's decision to adopt or to not adopt an hp/dp program (Question #11). They were then asked to describetheir experience. As noted earlier in this chapter, the importance of finances was negatively associated with a company's decision to adopt a program. Support this is evidenced by comments from respondents in adopter companies. including: I was given a blank check think about it and let me know what you need for a wellness program. (Bicycle Mfg 1985). No questions. The bottom line was never a factor at all. (Family Funds. Inc 1985) We're not in it for the dollars. (Wooden Door 1985) One respondent succinctly summarized the attitude of many companies: The point is. your top management group is either going to believe wellness is a good thing. and let you do something. Or all the numbers in the world aren't going to do anything for you. (Meadowlark Resources. 1985)

PAGE 143

.127 The sentiment was expressed by a number of respondents that finances would be important in selling the program to other companies. Yet when asked about their own experience, they would typically reply "You know it's really interesting--I was never asked for any hard facts or figures." Financial status. In addition to the overall importance of finance, several other financial factors were examined. Respondents were asked to categorize the overall financial status of their organization at the time the decisions were made to adopt or. to not adopt a program. The categories given were declining, stable, or -growing. All but one of the adopter agencies was in a stable or growing environment, while only four non-adopters were in a favorable financial situation. A cross-tabulation of the distribution is presented in Table 21. Using a chi-square test, the difference between adopters and non-adopters is significant at the .01 level. Discriminant analysis (see Table 22) revealed that the variable "Financial Status" correctly predicted the group classification in 79.2% of the cases studied. Two of the adopter companies were currently experiencing financial difficulty, but tlie program decision--made in more solvent times--was still in effect although implementation had slowed. Return on investment. Return on investment was considered as a second financial factor which might be related to adoption or

PAGE 144

. Tab1e 21 Association Between "Financial Status" Variable and the Decision to Adopt or Not Adopt a Health Promotion/Disease Prevention Program Count Row Pct Col Pct Decision 1 Adopt 2 Non-Adopt Column Total Chi-square = 6.40000 Significance = .0114 Financial Status High Low 1 2 11 1 91.7 8.3 73.3 11.1 4 8 33.3 66.7 26.7 88.9 15 9 62.5 37.5 with 1 degree of freedom Row Total 12 50.0 12 50.0 24 100.0 128

PAGE 145

Table 22 "Financial Status" Variable as a Predictor of Adoption/Non-. Adoption of Work site Health Promotion/Disease Prevention Program Classification Results No. of Predicted Group Hembership Actual Group Group Adopt Group Non-Adopt 1 2 Cases 1 2 12 12 11 91. 7 4 33.3 1 8.3 8 66.7 Percentage of Grouped Cases Correctly Classified = 79.17 Note. Eigenvalues = .56977 Canonical Correlation = .6024641 Wilks Lambda = .6370370 Significance = .0019

PAGE 146

non-adoption. Respondents were asked whether they say hp/dp as saving the company money or as costing the company money. The question failed to produce interpretable and consistent results. It was abandoned in favor of a second question in which respond-ents were asked more specifically about the length of time a program would be given to show a return on investment. They were asked to describe the situation and categorize management's view as short-term or long-term oriented. The change in question format was made approximately one-third of the way through the' study process, thus the short-term/10ng-term orientation of the first companies interviewed had to be inferred from comments from the taped interviews. Comments taken to 'reflect a short-term investment perspective include: We had to have answers to present to the legislature in six months. (Healthworks South, 1985) Price Waterhouse came in to do cost accounting because of DRGs--How can you do this philosophy with one thing and not look at we11ness in the same way. (Health Marketers, 1985) They are immediate production oriented.. We can show programs save money in the long run, but not the short tun. (Cafeteria Corp., 1985) ,130 Comments which indicated a long-term approach to investment included: You invest a certain amount in that person and then you lose them--then you're down. (Engines, Ltd., 1985) Our philosophy ... is that we're willing to do this for several years and not see a real behavioral change. I think you have to be patient. (Public Benefits, 1985)

PAGE 147

It should be noted that the advent of DRGs was cited by respondents in two of the four hospitals studied as a major reason for the delay in developing an hp/dp program. Because the collection of information was not consistent across the sample, and not sufficiently in-depth in some cases to make consistent, accurate judgments, no statistical analysis was done on this variable. In summary, it appears that certain factors such as the financial status of an organization and the importance of financial criteria in the decision process have a direct bearing on the decision to adopt .or not adopt an hp/dp program. Other factors such as length of time allowed for return on investment merit further study. Marketing/Access Issues Program availability. Program availability had been suggested as a possible factor affect a company's decision to start anhp/dp program. It was thought that smaller companies were at a disadvantage--being unable to find programs appropriate to their size. This does not appear to be the case. Only four companies reported that there were lack of programs available to meet their needs. In two of these cases, the companies were geographically isolated making the issue more one of access than general availability of such programs. In the remaining 20 companies, availability of programs was not seen as an issue. Comments to this effect included:

PAGE 148

We have a lot of resources, professional people in the county not concerned with that at all. (Mountain Training, 1985) We know the resources are the-re, we just don't have the money. (Hea1thworks South, 1985) -132 A cross-tabulation is in Table 23. A chi-square test revealed no significant association between "Program Avai1-ability" and the decision to adopt or not adopt a worksite health promotion/disease prevention program. Almost all respondents considered the availability or non-availability of programs to have no or only marginal influence on the initial decision as to whether or not to initiate a program. General level of exposure. In a related question, respond-ents were informed that opinion varied as to the importance of promotional seminars, personal contracts, and written materials in prompting a company to begin a program. They were then asked to describe the level of exposure in-their company to these types of promotional activities, and rate the exposure level as high or low. They were asked to support their ratings. The categorization was not based on a tally of numbers and types 6f exposure but rather on the respondent's perception of level of exposure. Fourteen companies had a higher level of exposure; 10 companies reported a low level of exposure. A cross-tabulation of distribution is presented in Table 24. While adopter companies in the sample showed a slightly higher rate of high exposure using the chi-square test, thedifference is not statistically significant at the .05 level.

PAGE 149

Table 23 Association Between "Program Availability" Variable and the Decision to Adopt or Not Adopt a Health Promotion/Disease Prevention Program Program Availability Count Row Pct Col Pct High Low Row Total Decision Adopt Non-Adopt 1 2 Column Total 1 10 83.3 50.0 10 83.3 50.0 20 83.3 Chi-square = 0 with 1 degree of freedom Significance = 1.0000 2 2 .16.7 50.0 2 16.7 50.0 4 16.7 12 50.0 12 50.0 24 100.0 '133

PAGE 150

Table 24 Association Between "General Exposure" Variable and the Decision to AdoEt or Not AdoEt a Worksite Health Promotion/Disease Prevention Program General Exposure Count Row Pct High Low Col Pct Decision 1 2 1 9 3 Adopt 75.0 25.0 64.3 30.0 2 5 7 Non-Adopt 41. 7 58.3 35.7 70.0 Column 14 10 Total 58.3 41. 7 Chi-square = 1.54236 with 1 degree of freedom Significance = .2142 Rml1 Total 12 50.0 12 50.0 24 100.0 134

PAGE 151

The sources of exposure were multiple and varied from company to company, but they were overwhelmingly along professional lines. Occupational nurses referred to conferences and journals in their field; personnel managers sought information thr.ough personnel conferences and benefits publications; persons in medical care cited health related and medical sources. Although popular literature and other sources were sometimes cited, the dominant information seeking pattern was most frequently focused on conferences and written materials related to the respondent's professional area. Exposure to model program. Level of general exposure to hp/dp information did not discriminate between adopters and nonadopters. Exposure to a model program operating in another setting, however, is positively related to the decision to adopt a program. Fourteen companies had been exposed to model programs; 10 companies had not. A cross-tabulation of the distribution is presented in Table 25. Using a chi-square test, the difference between adopters and non-adopters is statistically significant at the .01 level. Discriminant analysis (see Table 26) revealed that the variable "Exposure to Model" correctly predicted the group classification in 75% of the cases. In summary, program availability and general level of exposure did not vary significantly between adopters and nonadopters. Exposure to a successful, operational program, on the

PAGE 152

Table 25 Association Between "ExEosure to Model" Variable and the Decision to AdoEt or Not AdoEt a Worksite Health Promotion/Disease Prevention Program Exposure to Model Count Row Pct High Low Col Pct Decision 1 2 1 10 2 Adopt 91.7 8.3 73.3 11.1 2 4 8 Non-Adopt 33.3 66.7 26.7 83.9 Column 14 10 Total 62.5 37.5 Chi-square = 6.40000 with 1 degree of freedom Significance = .0114 Row Total 12 50.0 12 50.0 24 100.0

PAGE 153

Table 26 "Exposure to Model" Variable as Predictor of Adoption/NonAdoption of Worksite Health Promotion/Disease Prevention Program Classification Results 137 No. of Predicted Group Nembership Actual Group Group Adopt Group Non-Adopt 1 2 Cases 1 2 12 12 10 83.3 4 33.3 2 16.6 8 66.7 Percentage of Grouped Cases Correctly Classified = 75.00 Note. Eigenvalues Canonical Correlation Wilks Lambda = .56977 .6024641 = .6370370 .0019

PAGE 154

other hand, appears to be a factor in the decision to adopt or to not adopt an hp/dp program. Discriminant Analysis Model The variables shown to be significant at the .001 level were loaded into a discriminant analysis model. The .purpose was to establish whether a combination of variables could predict with accuracy whether a company would adopt or would not adopt an hp/dp program. The variables included were: management philos-ophy; importance of finances, facilities, presence of "angel" in company, responsibility for employee, and financial status. Discriminant analysis revealed that this combination of variables correctly predicted group classification in 100% of the cases (see Table 27). Responses reflecting a high level of concern for employee benefits include: If this is what makes people happy, he'll be very supportive. (Airlines, Inc., 1985) Employee interest was a major concern. (Waverly, Inc., 1985) Responses reflecting a low level of concern for employee benefits include: They [management] are against incentives of any kind. (Warehouse Co., 1985) They're not into individual development here. (Red Brick Finances, 1985)

PAGE 155

Table 27 Discriminant Analysis Model Using Multiple Variables F R E Q u E N C Y 8 + 6 + 4 + 2 + All-Groups Stacked Histogram Discriminant Function 1-1 1 1 1 1 2 1 2 1 1 2 1 1 2 1 11 1 111 2 2 1 11 1 111 2 2 + + 2 + 2 2 2 2 2 + 2 2 22' 2 2 22 2 2 Ou t ........ + ........ -+ ........ -+ ......... +....... -+........ + ......... + ........ Out Classification Group Centroids -6 -2 0 2 4 6 11111111111111111111111111111111111111111222222222222222222222222222222222222222 1 2 ..... W \0

PAGE 156

Table 28 Multiple Variables as Predictors of Adoption/Non-Adoption of Worksite Health Promotion/Disease Prevention Program Classification Results .140 No. of Predicted Group Hembership Actual Group Group Adopt Group Non-Adopt 1 2 Cases 1 2 12 12 12 100.0 o o o o 12 100.0 Percentage of Grouped Cases Correctly Classified =100.00 Note. Eigenvalues Canonical Correlation Wilks Lambda Significance = 8.19133 .944039 .1087981 .0000

PAGE 157

In a second model only variables were used which were significant at the .001 level and which were based on objective not on the perceptions of respondents. These variables included: facilities and financial status. Discriminant analysis (see Table 29) revealed that this combination of variables correctly predicted group classification in 87.5% of the cases. None of the perspectives discussed in Chapter II exclusively explained the pattern of adoption of worksite hp/dp programs. A combination of variables representing mUltiple perspectives however, provides a model for predicting adoption/ non-adoption of program. They are summarized in Tables 31 and 32. Summary A number of variables are related to the decision by companies to adopt or to not adopt an hp/dp program. These variables represent components of several of the major diffusion perspectives including the rational/economic perspective, the human relations/communications perspective and the market/systems perspective. Several variables hypothesized to influence the decision process did not prove to discriminate between adopter and non-adopter companies. In addition, a number of observations were ma4e about conditions, attitudes, and practices which occurred in companies "141

PAGE 158

Table 29 Analysis Using Accessible Variable F R E Q U E N C Y 16 + 12 + 8 + 4 + All-Groups Stacked Histogram -Canonical Discriminant Function 1-2 1 1 1 1 1 1 1 2 1 2 1 2 2 I 1 1 + + + 2 2 2 2 + 2 2 2 Out -+ ....... -+_ -+-...... -+-.... -+_ -+_ -+-... -Out Classification Group Centroids -6 -4 -2 0 2 4 6 ill111lll111111111l111l1l11111111111111112222222222222222222222222222222222222222 1 2 ....... N

PAGE 159

Table 30 Accessible Variables and Predictor of Adoption/Non-Ado tion of Worksite Health Promotion Disease Prevention Program Classification Results 143 No. of Predicted Group Nembership Actual Group Group 1 Adopt Group 2 Non-Adopt Cases 12 12 1 2 11 91. 7 2 16.7 1 8.3 10 83.3 Percentage of Grouped Cases Correctly Classified = 87.50 Note. Eigenvalues = 1. 74522 Canonical Correlation = .7974106 Wilks Lambda = .3641364 Significance = .0000

PAGE 160

Table 31 Variables Associated with the Decision to Adopt or Not Adopt a Worksite Health Promotion/ Disease Prevention Program Primary Diffusion Perspective Adopters Non-Adopters Represented Importance of Finances Low High Rational/Economic Importances of Benefits High Low Behavioral Responsibility for Employees Expanded Limited Behavioral Management Philosophy Humanism Accounting Behavioral Facilities Availalbe Not Available Rational/Economic "Angel" in Top Management Present Absent Behavioral Financial Status Growing/Stable Declining Systems Exposure to Model Program Present Absent Systems .po. .po.

PAGE 161

Table 32 Variable Not Associated With the Decision to Adopt or Not Adopt a Worksite Health Promotion/ Disease Prevention Program Decision Process Centralized vs. Decentralized Program Proposed Simple vs. Complex "Champion" in Middle Management Present vs. Absent Staff Expertise High vs. Low Program Availability High vs. General Exposure High vs. Low Primary Diffusion Perspective Represented Rational/Economic Behavioral Behavioral Behavioral Systems Systems ...... +:'I..n

PAGE 162

146 as they progressed from the awareness of possible programs, through the discussion stage, to the decision whether or not to develop a company-sponsored hp/dp program. The implications of these findings are discussed in the following chapter.

PAGE 163

CHAPTER V CONCLUSIONS AND RECOMMENDATIONS This chapter (a) limitations of the study, (b) discussion of implication of findings, (c) discussion of the role of government in the diffusion of worksite hp/dp programs, (d) future research issues, and (e). a chapter summary. Limitations of the Study Retrospective Bias Rogers (1984, pp. 11-12) notes that many diffusion studies suffer from a "recall problem," Le., errors introduced when "respondents are asked to remember the time at which they adopted a new idea." In several companies in this study, the decision to adopt a program had been made over 3 years ago. More subtle contributing factors and events might not have been reported simply due to lack of recall. Or as Eveland (1983), succinctly states, organizations are notorious for 'editing the past' and failing to remember how it was that they got where they are'! (p. 431). Focus on Selected Variables As Michael Quinn Patton notes (1980, pp. 46-47), there is a tendency in holistic-inductive research to move from a "discovery mode" to a "verification mode" (Guba, 1978). Initially, a researcher

PAGE 164

148 is open to whatever responses are given. As patterns emerge from the data, there is the tendency to focus on verifying hypotheses, giving less attention and credence to responses which do not fit the pattern. The process of continuing to add and new factors with equal vigor was considered desirable, but would have required a cyclical process beyond the scope of this study. Variables such as "level of education" and "time span in return on investment" suggested later in the interview process may be of equal power in explaining the pattern of diffusion as are other variables. These variables should be examined in future studies. Classification Bias As in all research using a guided interview technique, there /' is the potential for error in interpretation of responses to open-ended questions. Care was taken to structure the questionnaire so that key variables were addressed in two or more sections of the questionnaire. Potential for bias and misinterpretation, however, still exists. Use of a dichotomy to classify responses, likewise, imposes limitations. It obscures what may turn out to be important differences which would only be elucidated.through the use of more subtle scales and measures. For example, a simple classification of level of staff expertise as high or low may aggregate companies who have no staff to deliver programs, with those who have staff exper-tise to deliver 40% of programs but who fall below the 50% cut-off.

PAGE 165

Decision Focus This study focused only on the company's decision to offer an hp/dp program and, to a lesser extent, the' steps leading up to that decision. The variables associated with a company's. decision to adopt a program are not necessarily those related to effective program implementation. For example, while the presence of an "angel" is important in the decision process, the presence of a champion or champions may be important in the implementation process. Non-Random Sample The sample for this study was not a random sample but rather drawn from a list of agencies which had participated in an earlier study in worksite hp/dp and which had agreed to be contacted again. There may be some biases introduced by the fact that they agreed to be contacted again. In addition, the number of companies, 12 adopters and 12 non-adopters, is too small to make any inferences about the total population of businesses considering hp/dp programs. Discussion of Findings Limitations of the study notwithstanding, there are several general observations which can be made regarding the process and the variables related to the adoption and non-adoption of worksite hp/dp programs. Targetting Diffusion Efforts Certain attitudes, practices, and conditions are positively associated with the adoption of worksite health promotion/disease

PAGE 166

150 prevention programs. These include: growing or stable financial status; a humanistic management philosophy; the presence of an "angel" in top management who actively supports company involvement in hp/dp; an on employee benefits, productivity,. and morale in the decision process; a deemphasis on financial issues; the existence of facilities to carry out program activities; and exposure to a model program. If a company at the decision stage of the diffusion process possesses all three characteristics, it can be predicted with a high degree of accuracy, that they will choose to adopt an hp/dp program. Conversely, if a company at the decision stage of the diffusion process is in a period of financial decline; operates with an overall accounting philosophy; does not have an hp/dp "angel" in top management; emphasizes financial consideration in the decision process; deemphasizes employee benefits, morale, and productivity; does not have facilities available; and has not been exposed to a model program it can be predicted with a high degree of accuracy that they will not adopt a worksite hp/dp program. In the 24 companies studied, each of these variables independently predicted adoption or non-adoption with at least 70% accuracy. If proponents of worksite hp/dp programs could determine the status of companies vis-a-vis any or all of their variables, they could more effectively target companies likely to adopt programs. It should be noted that all companies in the study had a high level of awareness of hp/dp and had actively discussed such

PAGE 167

151 programs. They were all at the decision stage of the diffusion process. It is not clear that the model has the same power to predict adoption/non-adoption in companies who are just beginning or have not yet begun to consider hp/dp programs. Developing Positive Triggers Early work on change hypothesized that change and innovation occurred in response to a "performance gap." i.e when individuals perceive a gap between their actual and their desired level of performance (Downs, 1967). A performance gap can be created in one of two ways: standards can be raised, or performance can decline. When the gap is large enough, individuals and organizations will act to reduce the gap. A number of planning models, including the dominant hp/dp planning model (Green et al., 1980). are built with the identifi-cation of the performance gap as the necessary first step. As applied to worksite, hp/dp planners are urged to demonstrate declining performance using data such as increased health care cost. health risks in the population. and absenteeism rates (Everly & Feldman, 1985; O'Donnell & Atkinson, 1984; Parkinson et al., 1979). Health promotion/disease prevention programs are then introduced as a solution. The overall approach is problem-oriented. Findings from this study indicate that such a negative model for worksite programs may be misguided. In the companies

PAGE 168

studied, interest most frequently stemmed from positive triggers. Programs were more likely to develop in response to a perceived opportunity than in reaction to a perceived threat. The raising 152 of standards, e.g., emphasis on what other companies are poing, discussion of marketing/public relations potential, appears to be initially more engaging to management than do the more scientific but more dour analyses of health care costs and health risks in the Most companies attributed the development of interest in hp/dp to multiple financial, human, and environmental factors. The case of the single stimulus-response was non-existent. In all cases it was a convergence of internal and external factors. There may indeed be some sort of "threshold level" which needs to be reached before a company becomes actively interested in hp/dp. Findings from this study suggest that positive, personal elements were central to the development of interest. Little is known, however, about the most efficient mix of triggers or the formula most appropriate to particular types of organizations. Addressing Management Criteria Finances, personnel benefits and the company position vis-a.-vis its general responsibility for employees were the most common criteria used by companies to decide whether or not to adopt an hp/dp program. When finances became the major or sole criterion, it was very unlikely that a company would adopt a program. Thus hp/dp program components should not focus their

PAGE 169

153 presentation. solely on financial issues. They would do well to focus on benefits, general corporate responsibility for health or the other positive triggers which originally caused the company to investigate hp/dp programs. Another option would be to. better understand the philosophy of management and to develop an approach which would address the major criteria used by management. In companies where humanism dominates, approaches include an emphasis on benefits, morale, and corporate responsibility in the community. Conversely, in companies where an accounting mentality dominates, approaches should focus on cost containment, absent.eeism and related issues. Given the limited data on the cost/benefit of hp/dp programs, this latter approach should be taken with caution. In addition, there is no evidence from this study to suggest that well-formulated financial arguments are the key to the decision to adopt. A word of caution may be appropriate here. The adopter companies in this sample may indeed represent the innovators described by Rogers (1983, p. 260). Innovators are generally more educated, of higher social status, less dogmatic, better able to deal with abstractions, more cosmopolite, and more positive toward change in general. The benefits and humanism approaches which appeal to this group may not hold the same appeal to later adopters who are more concrete and skeptical by nature. Reaching Key Actors Top management plays a key role in the awareness, pers1lasion and decision stages of the diffusion of worksite hp/dp programs.

PAGE 170

154 In a common scenario for adopter companies, the CEO is exposed to a program in another setting. He/she brings it back to the company, personally endorses it and sets in motion the management mechanisms to facilitate its adoption. In select cases mid-to managers had the initial exposure and commitment. They, however, identified and recruited individuals in top .management to support their cause. Enthusiastic mid-level managers and committees may have been important to adoption in some companies, but as a general rule, their presence did not seem to increase the likelihood of a company adopting a program. These findings suggest that upper level and top management could be.the target of promotion activities in the early stages of the diffusion process. Without their active support, decisions to adopt are not likely to occur. Information-seeking behavior of mid-level managers along professional lines. Human r.esource managers attended personnel conferences and read hp/dp articles in benefits journals; occupational nurses, education, marketing specialists and others, likewise, had their own professional sources of information. Health sources were rarely cited by non-health personnel and vice versa. This suggests that proponents of hp/dp programs should enlist the cooperation of key professional groups and associations. They should use existing channels which are most frequently consulted by mid-level managers.

PAGE 171

Unfortunately, the information seeking behavior of top management and CEOs was not systematically documented., A general impression is that this group is much more cosmopolite, using 155 more numerous and more varied sources of information, e.g., running with the Governor and serving on boards of progressive health care institutions. This is fertile ground for future study. Avenues for Change The variables in this study identified as positively related to the adoption/non-adoption of hp/dp programs include: sound financial status, humanistic management mindset, existence of an "angel" in top management, a company philosophy of expanded responsibility to employees, limited concern with finances, available facilities, and exposure to a model program. The vast majority of these are non-tractable, likely to change only over a long period of time. The exception is "exposure to a model program'" a factor which can be changed with relative ease. The variable, "General Level of Exposure" did not differentiat'e between adopters and non-adopters in this study. Attendance at seminars in which hp/dp is included as a topic, articles in professional journals, and personal contacts appear necessary but not sufficient stimuli for change. A more intense, personal exposure of top management to successful programs seems indicated. These findings are consistent with a statement by Rogers (1984): "Person-to-person imitation and social modeling is thus

PAGE 172

I 156 an essential element in the diffusion process" (p. 7). They are further supported by the work of Bandura (1977) on social learning theory which emphasizes the importance of modeling and suggests specific factors which facilitate the adoption of a new behavior. These factors include: (a) direct observation of desirable behavior, (b) use of pleasant and familiar role models, (c) assistance in interpretation of experience, and Cd) opportunity for application of desired behavior. Proponents of might do well to give more cognizance to the principles of modeling by (a) arranging visits to successful programs where information is pre:sen ted by credible role models, (b) assisting top management to interpret their observations, and (c) encouraging and assisting top management in the application of the-behavior in their own organization. Finally, while effective exposure to appropriate models may increase the likelihood-of program adoption, it is not a "magic _bullet." Management in 4 of the 12 non-adopter companies has been exposed to successful model programs. Other factors such as declining financial status, lack of an "angel," and accounting mindset appear to have combined and outweighed the effects of exposure to a model program. Government Role This study did not specifically focus on the role of governmental agencies in the diffusion of hp/dp programs.

PAGE 173

The findings from the research, however, may be instructive to public administrators charged with the formulation of worksite hp/dp policy and program efforts. 157 The government appears to have played only a limited role in the diffusion of worksite hp/dp programs. Initial interest in many companies resulted from an awareness of what was happening in society-at-large coupled with internal events and circumstances. Seminars attended and publications read were frequently along professional and business lines. Model programs visited and experts consulted were most often associated with private industry. The government played a role in generating the data used in many of the professional articles and presentations or worked to cooperatively fund and sponsor Yet, positive action by governmental agencies was mentioned only by a few companies. Some government actions may indeed have had a negative effect. In several instances, publicly-funded agencies cited the negative attitude of the state legislature toward prevention as a deterrent to program development. Several of the health care institutions reported that the introduction of a reimbursement system based on diagnosis-related groups (DRGs) caused financial turmoil and focused attention on programs that provide an immediate return on investment. Both of these conditions may hinder the introduction of long-term prevention programs more difficult. The proper role of government in the diffusion of worksite hp/dp programs is a complex issue, one that can be raised, but not

PAGE 174

158 resolved in this short space. It hinges on at least three factors: (a) the contribution of the innovation to the public's general welfare, (b) the extent to which the variables influencing diffusion are amenable to government influence or intervention, and (c) the extent to which the innovation would diffuse without government. assistance. There is substantial evidence that worksite hp/dp programs can be effective in reducing personal health risks and thereby health risks in the total population (see Chapter I). The economic benefits of such programs are not adequately documented. While some program components such as hypertension control and smoking cessation demonstrate favorable cost/benefit ratios, cost/benefit data in other program areas such as nutrition and stress management are weak. Further, some analysts argue that prevention programs may well resuLt in increased longevity, with all the attendant societal costs (Kristein, 1982). In sum, the effects of hp/dp programs on the general welfare of the population appear to be positive, but the issue is far from being reached. The second issue deals with the extent to which the variables influencing the diffusion process are amenable to government intervention. As discussed earlier in this chapter, most of the variables identified in this study are related to general company characteristics and circumstances that will be difficult to change. There are two exceptions: exposure to a model and return-on-investment. Through programmatic efforts, the government

PAGE 175

could facilitate visits by managers in select companies to effective, operational hp/dp programs in other sites. Through policy levers such as tax benefits and regulation, the government could increase or otherwise favorably alter the return-on-investment, making hp/dp programs appear more attractive to companies. Other factors such as "staff expertise," while not related to the decision to adopt, may be related to successful program implementation. In sum, it appears that while overall conditions may be difficult to change, there are variables which are amenable to government intervention. A difficult but important question about the role of government is the extent to which hp/dp programs would diffuse without government action. Information gathered from the companies in this study indicates that the government has not been a major factor in either the development of interest or the decision to adopt. Programs have resulted more from events in society at large, from actions by private businesses and professional groups or from circumstances internal to the company. On the basis of these findings, one might assume that the government has had no major function in the diffusion of hp/dp programs although its important role in the development of a data base for prevention is clear. Rather, market forces will determine the demand, supply, and ultimately the diffusion of these programs. There may be, however, a number of groups for whom market forces operate differently. For example, industries in which there is a high turnover rate will not experience the benefits of long-term cardiovascular

PAGE 176

risk-reduction programs. Likewise, smaller companies may not develop such programs because there are no" economies of scale; nor are they profitable targets of private enterpreneurs who "160 sell hp/dp programs. Certain segments of the business community will seek out and implement programs with minimal government assistance, while other segments may require government assistance. In summary, the government may be justified in taking action to support the diffusion of proven programs such as smoking cessation, seatbelts, and EAP programs that have demonstrated benefits to the general public. The mechanisms used should address variables known to influence the diffusion process. Government efforts should be directed at segments of the population with demonstrated need and who are unlikely targets for the private sector. Further Research Issues "The findings presented in this study describe the process and the variables related to the diffusion of worksite hp/dp in 24 private businesses and public agnecies in Colorado. The study should be replicated with a larger, random sample of businesses. Such a sample would allow for inference to a larger population. It would also allow for examination of subsets within the business connnunity. Several variables which had little or no predictive power such as "program availabilityll--lIgeneral level of exposurell--

PAGE 177

.161 can be excluded from future studies. Other variables such as "importance of and "return on investment" merit clearer operational definitions and further exploration. In some instances, where two variables appear to address the same concept, e.g., "management philosophy" and "responsibility for employees," they could be collapsed into one. Multiple value measures should be established for each variable. More emphasis should be given in future studies to identifying predictive variables for which information is easily accessible from company records. Targeting of companies likely to adopt programs could then be done without going through a personal interview process. Likewise, more emphasis might be given to more tractable variables. Company financial status, management mindset, and the presence of an angel, while highly predictive, are not amenable to rapid change. Potential interventions such as exposure to model need to be identified and tested. In addition to identifying the variables related to the decision to adopt or to not adopt a program, work should continued on understanding the entire process of diffusion of worksite hp/dp programs. Each stage from awareness through confirmation operates differently. An understanding of each stage is vital to the planning and management of the entire process. Summary The purpose of this study was to examine the process and the variables related to the adoption of innovative health programs

PAGE 178

by organizations. The study utilized diffusion theory as the conceptual framework. and focused specifically on the adoption/ non-adoption of worksite health promotion/disease prevention programs. A number of variables suggested by diffusion t?eory demonstrated the power to predict adoption/non-adoption of worksite hp/dp programs in the companies studied. Variables positively associated with adoption of programs include: sound status; humanistic management philosophy; emphasis on employee benefits, morale and productivity; limited emphasis on finances in the decision process; expanded responsibility for employees; the presence of an hp/dp "angel" in top management; availability of facilities; and explosure to a model hp/dp program. These variables represent the rational/economic, the behavioral, as well as the systems perspectives in diffusion. Other variables suggested by diffusion theory did not demonstrate a significant association to the decision to adopt or not adopt a program. These variables include: level of staff expertise; presence or absence of an hp/dp "champion" in middle management position; centralization or decentralization of the decision process; complexity or simp1.icity of the proposed program and general level of exposure to hp/dp information. In the majorit.y of companies, there was no single "trigger event" responsible for the development of initial interest in hp/dp. Rather interest resulted form a convergence of both internal and external factors which heightened awareness and stimulated action.

PAGE 179

Programs were more likely to develop in response to a. perceived opportunity than in response to a perceived threat. Exposure to successful programs, discussion of employee benefits and/or marketing potential of hp/dp programs appeared to be more engaging to management than did discussions of health risks and health care costs. 163 Finances, employee benefits/morale/productivity, and company philosophy vis-a-vis responsibility for employees were the most common criteria used by companies to decide whether or not to adopt an hp/dp program. When finances became the major or sole criterion, it was very unlikely that a company would adopt a program. Without the active support of top management, decisions to adopt a program are not likely to occur. Programs were more likely to flourish in companies where someone in top management was actively supportive, where management operated with a humanistic philosophy and espoused expanded responsibility for employees. The majority of factors positively associated with adoption of worksite hp/dp programs are non-tractable, e.g., management philosophy, financial status, they are likely to change only over a long period of time .. Exposure to model programs is an exception to this rule and may provide a mechanism to affect the diffusion process. In summary, the findings of this study suggest that companies likely to adopt hp/dp programs can be effectively

PAGE 180

164 identified and targeted. The approach to top management in these companies should be balanced and positive, emphasizing employee benefits/morale/productivity as well as addressing health risk and financial concerns. Although many of the variables influencing the decision to adopt or not adopt an hp/dp program are beyond the control of the proponents of hp/dp, exposure to model programs may provide a mechanism for effective intervention. Limitations of the study were examined and a number of hypotheses were identified for further study.

PAGE 181

BIBLIOGRAPHY Airline, Inc. Personal interview. April 1985. Alderman, M., Green, L. W., & Flynn, B. S. (1980, March/April). Hypertension control programs in occupational settings. Public Health Reports, 75(2), 158-163. Allied Personal Service. Personal interview. March 1985. Angel Security Co. Personal interview. February 1985. Bauer, K. B. (1980). Improving the changes for health: Lifestyle change and health evaluation. San Francisco: National Center for Health Education. Baumgarten, S. (1975, February). The innovative communicator in the diffusion process. Journal of Marketing Research, 12. Beck, R. N. (1982, January 22). Health promotion and health protection: The role of industry. Second Annual Lester Breslow Distinguished Lecture. Los Angeles: UCLA School of Public Health. Becker, M. H. (Ed.). (1974). The health belief model and personal health behavior. Thorofare, NJ: Charles B. Slack. Behrens, R. (1983, August). Worksite health promotion: Some questions and answers _.to help you get started. Washington, DC: DREW, Public Health Service. Bell, W. (1963). Consumer innovators: A unique market for newness. Proceedings of the American Market Association, 85-95. Berry,. C. (1981). Good health for employees and reduced health care costs for industry. Washington, DC: Health Insurance Association of America. Bicycle Mfg. Personal interview. April 1985. Bingham, R. D., Freeman, P. K., & Fe1binger, C. L .. (1982). Toward an understanding of innovation adoption: An empirical application of the theoretical contributions of Downs and Mohr. Milwaukee, WI: Report to the National Science Foundation, University of Wisconsin-Milwaukee.

PAGE 182

166 Brown, L. A. (1981). Innovation diffusion: A new perspective. New York: Methuen. Cafeteria Mfg. Personal interview. May 1985. Campbell, D. T., & Stanley, J. C. (1966). Experimental and quasiexperimental designs for research. Chicago: Rand McNB;lly. Carlson, R. o. (1968). Summary and critique of educational diffusion research. Paper presented at the National Conference on the Diffuson of Education Ideas, East Lansing, MI. Cassuto,J. (1967). New health education programs. Journal of Occupational Medicine, 620-623. Chakrabarti, A. K. (1974). The role of champion in product innovation. California Management Review, !I(2), 58-62. Chakrabarti, A. W., & Rubenstein, A. H. (1976, February). Interorganizationa1 transfer of technology: A study of adoption of NASA innovations. lEE Transactions on Engineering Management, EM-23(1),20-34. Charter Health. Personal intervie1.;r. Harch 1985. Coleman. J., Katz, E., & Menzel, H. (1957). The diffusion of an innovation among physicians. 253-270. Cummings,K. M., Becker, M. H., & Maile, M. C. (1980). Bringing the models together: An empirical approach to combining variables used to explain health actions. Journal of Behavioral Medium, 1(2), 1980. Cunningham, R. M. (1982). We11ness at work: A report on health and fitness programs for employees of business and industry. Chicago: Blue Cross Association and Blue Shield Association. Danaker, B. G. (1980, March/April). Smoking cessation programs in occupational settings. Public Health Report, 149-157. Davis, M., Rosenberg, K., Iverson, D., Vernon, T., & Bauer, J. (1984). Worksite health promotion in Colorado. Public Health Reports. Dickman, F., & Emener, W. G. (1982, August). Employee assistance programs: Basic attributes, and an evaluation. Personnel Administrator, 55-62. Donoghue, S. (1977. May/June). The correlation between physical fitness, absenteeism, and work performance. Canadian Journal of Public Health, 201-203.

PAGE 183

.167 Downs, A. (1967). Inside bureaucracy. Boston: Little, Brown. Downs, G. w. (1978). Complexity and innovation research. In M. Radnor, I. Feller, & E. M. Rogers (Eds.), The diffusion of innovations: An assessment. Evanston, IL: Northwestern Univer sity, :Final Report to National Science Foundation, Grant No. PRA-7680388. Downs, G. W., & Mohr, L. B. (1976). Conceptual issues in the study of innovation. Administrative Science Quarterly, 700-714. Dunkin, W. S. (1980). The employee assistance manual. National Council on Alcoholism. DuPont, R. L., & Basen, M. M. (1980, MarchI April) Control of alcohol and drug abuse in industry--a literature review. Public Health Reports, 95(2), 137-147. Durbeck, D. C. et al. (1972, November 20). The national aeronautics and space administration--U.S. public health service health evaluation and enhancement program. The American Journal of Cardiology, 784-790. Engines Ltd. Personal interview. May 1985. Evaluating employee assistance policy adoption in organizations served by an HMO-based occupational alcoholism project. (1984). Washington, DC: National Institute of Alcohol and Alcohol Abuse, DREW, Grant #AA022l0. Eveland, J. D. (1981). Some themes in the interactions of technology and administration Policy Studies Journal, 409-418. Everly, G. S., Jr., & Feldman, R. R. L. (1982). Occupational health promotion. Nelv York: John Wiley & Sons. Family Funds, Inc. Personal interview. March 1985. Feller, I. (1978). The applicability of economic analysis to the adoption of innovations by public sector organizations. In M. Radnor, I Feller, & E. M. Rogers (Eds.), The diffusion of innovations: An assessment. Evanston, IL: Final Report, Northwestern University, Grant Number PRA-7680388, National Science Foundation. Feller, I., & Menzel, D. C. (1977). Diffusion milieus as a focus of research on innovations in the public sector. Policy Sciences, 8(1),49-68.

PAGE 184

Fielding, J. E. bottom line. (1979, February). Preventive medicine and the Journalof Occupational Medicine, 21(2), 79-88. Fielding, J. E. (1982, November). Effectiveness of employee health improvement programs. Jourrial of Occupational Medicine, 906-916. Fielding, J. E. (1984). Health promotion and disease prevention at the worksite. Annual Review of Public Health, i, 237-265. Fielding, J. E., & Breslow, L. (1983, May). Health promotion programs sponsored by California employers. American Journal of Public Health, 1(5), 538-543. Follman, J. F., Jr. (1978). Economics of industrial health. New York: American Management. Foote, A., & Erfurt, J. C. (1983). Hypertension control at the worksite. New England Journal of Medicine, 308(14). Forklift, Inc. Personal interview. May 1985. Green, L. W., Kreuter, M., Deeds, S., & Patridge, K. (1980). Health education planning: A diagnostic approach. Palo Alto, CA: Mayfield. Griliches, Z. (1957, October). Hybrid corn: An exploration in the economics of technological change. 501-572 Guba, E. G. (1978). Toward a methodology of naturalistic inquiry in educational evaluation. CSE Monograph Series in Evaluation No. 8. Los Angeles: Center for the Study of Evaluation, Univer sity of California, Los Angeles. Hage, J., & Dewar, R. (1973). Elite values versus organizational structure in predicting innovation. Administrative Science Quarterly, 18. Haskell, W., & Blair, S. N. (1980). The physical component of health promotion in occupational setti.ngs. Public Health Reports, 21(2), 109-118. Health Marketers. Personal interview. February 1985. Healthworks South. Personal interview. March 1985. Healthy people: the surgeon general's report on health promotion and disease prevention. (1979). Washington, DC: U.S. Department of Health, Education, and Welfare.

PAGE 185

169 Hilltop Haven. Personal interview. April 1985. Hochbaum, G. H. (1958). Public participation in medical screening programs: A sociopsychological study. Public Health Services, Publ. #572. Washington, DC: Government Printing Office. Hypertension detection and follow-up program cooperative .study. (1979). Journal of the American Hedical Association, 242, 2562-2577 Kaluzny, A. D., & Riordan, J. (1984, August 10-11). DRGs as an organizational innovation: Research issues. Boston, MA: Health Services Administration Doctoral Consortium. Katz, B., & Lazarsfeld, P. F. (1955). Personal influence. New York: Free Press. Kegeler, S. S. dental care. (1963, July). Some motives for seeking preventive Journal American Dental Association, &2, 90-98. Keller, R. T., & Holland, W. E. (1978). Technical information flows and innovation and processes. Houston, TX: Final Report, University of Houston, Grant No. PRA-76l844l, National Science Foundation. Kiefhaber, A., Weinberg, A., & Goldebeck, W. (1979). A survey of industry sponsored health promotion, prevention, and education programs. Washington, DC: Washington Business Group on Health. Kimberly, J. R., & Hiles, R. H. (Eds.). (1980). The organizational life cycle. San Francisco: Jossey-Bass. Kristein, H. H. (1982, Fall). The economics of health promotion at the worksite. Health Education Quarterly (special supplement), 27-36. Lalonde, H. (1974). A new perspective on the health of Canadians. Ottawa, Canada: Canadian Ministry of National Health and Welfare. Leventhal, H., Hochbaum, G., & Rosenstock, I. (1960). Epidemic impact on the general population in two cities. In The impact of Asian influenza on community life: A study in fine cities. Washington, DC: DHEW, PHS, Pub. 11.766. Lusterman, S. (1974). Industry roles in health care (Vol. 67). New York: The Conference Board. Mansfield, B. (1968). Industrial research and technological innovation. New York: Norton.

PAGE 186

McMahon, B., Sajewski, C., & Graff, W. (1984). Health promotion programs in small businesses. Seattle, WA: Health Works Northwest. Meadowlark Resources. Personal interview. April 1985. Merkle, J. A. (1980). Management and ideology. University of California Press. Midgley, D. F. (1977). Innovation and new product marketing. New York: John Wiley & Sons. Minnesota Department of Health. (1982). Workplace health promotion survey. Minneapolis, MN. Mohr, L. B. (1969). The determinants of innovation in organizations. American Political Science Review, 111-126. Mort, P. R., & Connell, F. G. (1938). Adoptability of public school systems. New York: Teachers College, Columbia Univer sity. Mountain Training. Personal interview. May 1985. Native Resources. Personal interview. April 1985. 170 Nickerson, H. H. (1967). An evaluation of health education programs {n occupational settings. Health Education Monographs, 11, 16-31.' Northwest Services. Personal interview. March 1985. O'Donnell, M. P., & Ainsworth, T. H. (1984). Health promotion in the workplace. New York: John Wiley & Sons. Paffenberger, R. S., Wing, A. L., & Hyde, R. T. (1978) .. Physical activity as an index of heart attack risk in college alumni. American Journal of Epidemiology, 108, 161-175. Parkinson, R. S., & Associates. (1982). Managing health promotion in the workplace. Palo Alto, CA: Mayfield. Pat.ton, M. Q. (1980). Qualitative evaluation methods. Beverly Hills: Sage Publications.' Pelz, D. C., & Munson, D. (1980). The innovating process: A conceptual framework. Ann Arbor, MI: Working Paper, Center f0r Research on Utilization of Scientific Knowledge, University of Michigan.

PAGE 187

Promoting health/preventing disease: Objective for the nation. (1980). Washington, DC: U.S. Department of Health and Human Services. Public Benefits. Personal interviews. April 1985. Public Health Service. (1958). Small plant health and medical programs. Washington, DC: U.S. Department of Health, Education, and Welfare. Recreation, Ltd. Personal interview. May 1985. Red Brick Finance. Personal interview. May 1985. Research Assoc. Personal interview. May 1985. Rhodes, E. C., & Dunwoody, D. (1980, September/October). Physi ological and attitudinal changes in those involved in an employee fitness program. Canadian Journal of Public Health, ]!, 331-336. Roberts, E. B. (1969). Entrepreneurship and technology. In W. A. Gruber & D. G. Marquis (Eds.), Factors in the transfer of technology. Cambridge, MA: MIT. Rogers, E. M. (1983). Diffusion of innovations (3rd ed.). New York: Free Press. Rogers, E. M. (1984, June 18-20). The diffusion of preventive innovations. Paper presented at Working Conference on SelfProtective Behavior, Rutgers University, New Brunswick, NJ Rogers, E. M., & Shoemaker, F. F. (1971). Communication of innovations. Ne'll York: Free Press. Romeo, A. A. (1975). Interindustry and interfirm differences in the rate of diffusion of an innovation. Review of Economics and Statistics, 22, 311-319. Ruchlin, H. C., & Alderman, M. H. (1980). Cost of hypertension control at the workplace. Journal of Occupational Medicine, 795-800. Ryan, B., & Gross, N. C. (1943, March). The diffusion of hybrid seed corn in two Iowa communities. Rural Sociology, 15-24. Sehnert, K. W., & Tillotson, J. K. (1978). How business can promote good health for employees and their families: A national health care strategy. Washington, DC: National Chamber Foundation.

PAGE 188

"172 Tarde, G. (1903). The laws of imitation, trans. E. C. Parsons. New York: Holt. Third special report to n.S. Congress on alcohol and health. (n.d.). Washington, DC: U.S. Department of Health, Education, and Welfare. Tillotson, J. K., & Rosala, J. C. (1978). How business can use specific techniques to control health care costs. Washington, DC: National Chamber Foundation. Tornatzky, L. G., Eveland, J. D., Boylan, M. G., Hetzner, W. A., Johnson, E. C., Roitman, D., & Schneider, J. (1983) The process of technological innovation: Reviewing the literature. National Services Foundation. Trice, H. M. (1979, January). Drug use and abuse in industry. Washington, DC: Office of Drug Abuse Policy. Udell, G. G., Baker, K. G., & Albaum, G. S. (1976). Creativity: Necessary, but" not sufficient. Journal of Creative Behavior, 10(2), 92-103. Veterans administration cooperative study group on antihypertensive agents. (1967). Journal of the American Medical Association, 202, 1028-1034. Ware, B. (1982, Fall). History has a message. supplement), 2... Health education in occupational settings: Health Education Quarterly (special Warehouse Co. Personal interview. April 1985. Warner, K. A., & Murt, H. A. (1984). Economic for health. Annual Review of Public Health, 2, 107-133. Warner, K. E. (1974). The need for some innovative concepts of innovation: An examination of research on the diffusion of innovation. Policy Sciences, 2, 433-451. Warshmv, L. J., Weingarten, J., Barr, J., & Lucas, T. (1984). Small business project. New York: The New York Business Group on Health. Waverly Medical. Personal interview. April 1985. Whyte, W. H. (1954, November). The web of word-of-mouth. Fortune, 140.

PAGE 189

Wooden Door Mfg. Personal interview. April 1985. Wright, C. C. (1982, December). Cost containment through health promotion programs. Journal of Occupational Medicine, 24(12), 965-968. Yenney, S. L. (1984). Health promotion and business coalitions: Current "activities and prospects for the future. Washington, DC: U.S. Department of Health and Human Services. Za1tman, G., & Duncan, R. (1977). Strategies for planned change. New York: John Wiley & Sons. Za1tman, G., Duncan, R., & Ho1bik, J. (1973). Innovations and organizations. New York: John Wiley & Sons. Za1tman, G., & Wa11endorf, M. (1983). Consumer behavior. New York: John Wiley & Sons. 173

PAGE 190

APPENDIX A BUSINESS AND INDUSTRY SURVEYS

PAGE 191

o ;HEALTH PROMOTION AT THE WORKSITE BUSINESS & INDUSTRY SURVEY 175 This form is for your use. Please fill in the reBponses in preparation for the phone interview. A member of the State Department staff will contact you and record your responses at this time. Definitions you may find helpful in answering the survey questibns: ... HEALTH PROHOTION: Health promotion begins with people who are basically healthy and seeks to support the development of lifestyles aimed at maintaining and enhancing the state of well being. DISEASE PREVENTION: Disease prevention focuses on screening for, and identification of disease which if left untreated may lead to acute or chronic health problems. HEALTH PROHOTING/DISEASE PREVENTING ACTIVITIES: Exercise and fitness; improved nutrition; reducing misuse of alcohol and drugs; smoking cessation; hypertension screening; accident prevention; minimizing health risks at work and home; etc. DO NOT return completed survey to the State Health Department. You may wish to keep a copy of these responses for your reference. Colorado State Health Department --Health Promotion & Education Section 4210 East 11th Avenue Denver, Colorado 80220 (303) 320-6137 (x 237)

PAGE 192

STATEMENT OF CONFIDENTIALITY The information which you provide through this survey will be confidential. No information will be released regarding your individual responses to the survey. .. Responses given by you to survey questions will be added to that received from several hundred other companies. It will be reported as aggregate information only. Only project staff will have access to completed surveys. The name of your company will be noted in the survey report only in an alphabetical list of survey respondents. Only with your written permission will the name of a contact person from your company be released. A Contact Consent Form is attached to the survey. 176

PAGE 193

,177 Your company does not currently have a formal health promotion/disease prevention program. However, we understand there is some interest in developing this type'of program in the future. 1. The following are common reasons for a health promotion/disease prevention program. How important would the following reasons be in initiating your program? (Check (V) one answer for each item.) VERY NOT IHPORTANT IMPORTANT UIPORTANT a. To improve employee health/reduce _/ -/ _/ -/ _/ -/ health problems. b. To improve employee morale. _/ -/ _/-/ _/ -/ c. To reduce turnover and/or absenteeism. _/ -/ _/ -/ _I-I d. To be part of an innovative trend in _/ -/ _/ -/ _/ -/ employee health care. e. To improve the public image of company. _/-/ _/ I _/-/ f. Response to employee demand/interest. _I-I _1/ _/ -/ g. To reduce health care costs (including '-/-/ _/-/ _/ -/ dissbility claims and insurance premiums). h. To improve productivity. _/I _/ -/ _/ -/ i. Other (specify). _/ -/ _/ -/ _/ I 2. If your company were to provide an employee health promotion/disease prevention program, which individual(s) would be responsible for making the decision to offer this program? (Describe by title of position no names please.) Title: _____________ ________________________________ 3. If your company were to provide an employee health promotion/disease prevention program. which individual(s) would have responsibility for management of this program? (Describe by title of position --no names please.) Title: -1-

PAGE 194

178 4. A health promotion'disease prevention program can be categorized under three major headings: screening, information program and service. Place a check (/) next to the activities your company would be most interested in including in its program. a. SCREENING 1) Periodic administration of: a) General health or cardiovascular risk appraisal., _' ___ I b) Pulmonary (lung) function test. _' ___ I c) Height'weight screening. _' ___ I d) High blood pressure screening. _' ___ I e) Colon'rectal cancer screening. _' ___ I f) Diabetes screening. _' ___ I 'g) Screening for work-related health problems. _' ___ I h) Cervical cancer screening. _' ___ I 2) Pre-employment medical examination. _' ___ I 3) Annual medical examination. _' ___ I 4) Other (specify). _'_I b. INFORMATION PROGRA}IS -(includes speakers, materials aryd exhibits to provide educational information.) 1) Benefits of exercise. 2) Harmful effects of smoking. 3) Harmful effects of alcohol'drug abuse. 4) Prevention of low back pain. 5) Health effects of stress. 6), Benefits of improved nutrition. 7) Benefits of seat belt use. 8) Cancer prevention and early detection. 9) Benefits of breast self-examination. 10) High blood pressure. -2--I __ I __ I __ I __ I __ I __ I __ I __ I __ I __ I __ I __ I __ I __ I __ I __ I __ I __ I __ I CONTINUED

PAGE 195

11) Prevention of work-related injury & illness. 12) Other (specify). c. SERVICE 1) Group and'or individual instruction in: a) Weight management. b) Exercise. c) Low back pain prevention. d) Stress management. e) Self-defense for women. f) Smoking cessation. 2) Employee assistance counseling program (which includes family, marit;>.l, personal and alcoholldrug counseling.) 3) Industrial alcoholism program. 4) Referrals to community health promotion classes' activities. 5) Employee health service. 6) Car poolinglvan poqling to improve air quality. 7) Other (specify). 179 _I _I _'_I _'_I _I_I _I _I _1_' _'_I _'_I _1_' _1-. I _'_I _1_1 _1_1 _I _, _1_1 5. If your company were to offer a health promotionldisease prevention program, how helpful would the following services be to initiation andlor development of your program? (Check answer for each item.) a. Assessment. of employee interest in health promotion activities. b. Insurance data review for use in program planning. c. Health promotion awareness sessions for top executives. d. Health promotion awareness sessions for employees. e. Information on community resources. f. Assistance to design behavior change strategies. g. Materials to market the program to workforce. -)NOT HELPFUL HELPFUL _t_ _I-I _1, _I-I _I-I _I-I _I _I -', _I-I -,I _I I -,I _I-I _I-I CONTINUED

PAGE 196

h. Information on new developments in worksite health promotion. i. Assistance to develop evaluation tools. j. Assistance to design employee health risk assessments. k. Insurance data review for use in designing health care cost containment options. 1. Training for in-house staff m. Assistance to develop comprehensive hypertension screening services. n. Assistance to develop "employee assistance counseling or industrial alcoholism services. o. Assistance to develop cancer education -early detection screening services. p. Assistance to develop methods to prevent work-related injury and illness. 180 NOT HELPFUL HELPFUL -', -', _1-, -'-" -', -', -', -', _'_I -', -'-"I -', -', -', -', -', -', -', 6. Is there any service not listed in Question 05 which would be helpful to you? "If so, please explain. 7. Referring to the items listed in Question OS, place an asterisk (*) next to the services for which you would consider paying. 8. If your company has an Employee Assist"ance Counseling Program or an" Industrial Alcoholism Program, check which of the services listed below are provided: a. Problem assessment. _' __ I b.Short-term counseling. _' __ I c. ReferraL _'_I d. Follow-up. e. Other (specify). -'-' _'_I 9. If your company has an Employee Assistance Counseling Program or an Industrial Alcoholism Program, check which of the following positions are included on the staff: CONTINUED: -4-

PAGE 197

181 a. Psychologist. d. Addictions counselor. _'_I b. Social worker. _'_I _'_I _'_I e. Other (specify). _______ '_1 c. Nurse. 10. When one of your employees is found to have high blood pressure, does your company have a system continued monitoring of the employee's blood pressure? Yes ___ No ___ 11. Does your company have a policy to control or eliminate smoking at work? Yes ___ No ___ 12. If your company does a smoking policy, which of the following best describes why your company policy was adopted: (Check many as apply.)' a. Necessary because of the product'we produce. _1-, b. Necessary because of safety factors. -,I c. Necessary because of sensitive or, special equipment. -', d. The promotion of good health for employees. -,I e. To enhance our pubiic image. -,I f. To better meet the needs of non-smoking employees. -', g. Other (specify) -,I 13. If your company were to develop a program, how would this be carried out? (Check (......,. the most appropriate option(s).) a. In-house staff. b. Purchase services from individual agencies for specific programs. c. Arrange for free-of-charge services from individual agencies for specific programs. d. Join with a group of businesses with similar interests. e. Purchase a complete program package from a proprietary agency. f. Other (specify). -5_'_I _'_I _'_I _'_I _'_I _'_I

PAGE 198

182 DEMOGRAPHIC INFORMATION 14. a. Is your firm one branch of a larger company? Yes ______ __ No --------b. If YES, would the health promotion/disease prevention program be managed from this office or the headquarters/central office of the company? (Check (./) one answer.) This of f ice ______ __ Headquarters/Central office ---"IS. How many people are employed by the company at this site? 16. Describe the make-up of your workforce (approximate percentages). Blue Collar _/ ___ / White Collar _/_/ Total: 100% 17. As you complete this questionnaire, you may feel we have missed issues important to your company's interest in health promotion/disease prevention activities. Your comments on these issues would be helpful to us. THIS CONCLUDES "THE SURVEY. TIIANK YOU FOR YOUR COOPERATION. -6-

PAGE 199

183 CONTACT CONSENT FORM Occasionally we receive requests from voluntary health agencies, businesses, individuals or community groups for the names of health promotion contact persons within business and. industry. If you are interested in discussing your company's health promotion program with others, and are willing to serve as the contact person for your company, please sign the form below and return it to the Health Promotion & Education Section, Colorado Department of Health. I give permission to' the Colorado Department of Health to release my name as a contact person for health promotion/disease prevention within this company. Signature: Company: Thank you. Date: RETURN TO: Health Promotion & Education Section Colorado Department of Health 4210 E. 11th Avenue Denver, CO B0220

PAGE 200

o HEALTH PROMOTION AT THE WORKSITE BUSINESS & INDUSTRY SURVEY -184 This form is for your use. Please fill in the responses in preparation for the phone interview. A member of the State Health Department staff will contact you and record your responses at this time. Definitions you may find helpful in answering the survey questions: HEALTH PRotl0TION: Health promotion begins with people who are basically healthy and seeks to support the development of lifestyles aimed at maintaining and enhancing the state of well being. DISEASE PREVENTION: Dt'sease prevention focuses on screening for. and identification of disease which if left untreated may lead to acute or chronic health problems. HEALTH PRONOTING/DISEASE PREVENTING ACTIVITIES: Exercise and fitness; improved nutrition; reducing misuse of alcohol and drugs; smoking cessation; hypertension screening; accident prevention; minimizing health risks at work and home; etc. DO NOT return survey to the State Health Department. You may wish to keep a copy of these responses for your reference. Colorado State Health Department --Health Promotion (, Education Section 4210 East 11th Avenue Denver. Colorado 80220 (303) 320-6137 (x 237)

PAGE 201

.185 STATEHENT OF CONFIDENTIALITY The information which you provide through this survey will be confidential. .. .. .. .. No information will be released regarding your individual responses to the survey. Responses given by you to survey questions will be added to that received from several hundred other companies. It will be reported as aggregate information only. Only project staff l,;ill have access to completed surveys. The name of your company will be noted in the survey report only in an alphabetica.l list of survey respondents. Only with your written permission will the name of a contact person from your company be released. A Contact Consent Form is attached to the survey.

PAGE 202

186 A. PROGRAM DESCRIPTION 1. The following are common reasons for
PAGE 203

2) 3) .4) e) Colon/rectal cancer screening. f) Diabetes screening. _I _/ _I _I g) Screening for work-related health problems. _1 __ 1 h) Cervical cancer screening. Pre-employment medical examination. Annual medical examination. Other (specify). _/_1 _/_1 _I _I _I _I 187 b. INFORNATION PROGRANS -(includes speakers, materials and exhibits to provide educational information.) 1) 2) 3) 4) 5) 6) 7) 8) 9) Benefits of exercise. Harmful effects of smoking. Harmful effects of al.cohol/drug abuse. Prevention of low back pain. Health effects of stress. Benefits of improved nutrition. Benefits of seat belt use. Cancer prevention and early detection. Benefits of breast self-examination. High blood pressure. 10) 11) 12) Prevention of work-related {nj ury .& illness. _I_I _1_/ _1_1 _1_1 _1_1 _I_I _I_I _I _I _I _I _I _I _I _I _I _I Other (specify). c. SERVICE 1) Group and/or individual instruction in: a) Height management. b) Exercise. c) Low back pain prevention. d) Stress management. e) Self-defense for women. f) Smoking cessation. -2-_I _I _I _I _I _I _/_1 '_1 _I _I _I CONTINUED:

PAGE 204

2) Employee assistance counseling program (which includes family, marital, personal and alcohol/drug counseling.) 3) Industrial alcoholism program. 4) Referrals to community health promotion classes/activities. 5). Employee health service. 6) Car pooling/van pooling to improve air quality. 7) Other (specify). 188 _/_/ _/_/ _/_/ _/_1 "_/_/ _/_/ 4. In general, are these activities available to ALL your employees? Yes _____ No ______ If "No," please indicate eligible employee group(s). 5. If your company has an Employee Assistance Coun!;eling Program or an Industrial Alcoholism Program, check (.,;) which of the services listed below are provided: a. Problem assessment. _/ __ / d. Follow-up. b. Short-term counseling. _1 __ / e. Other (specify). _/_/ ______ 1_1 c. Referral. _1_/ 6. If your company has an Employee Assistance Counseling Program or Industrial Alcoholism Program, check of the following positions are included on the staff: a. Psychologist. b. Social worker. c. Nurse. _/_1 _/_1 _I_I d. Addictions counselor. e. Other 7. When one of your employees is found to have high blood pressure, does _/_1 _1_1 your company have a system for continued monitoring of the employee's blood pressure? Yes ____ __ No ___ __ 8. Does your company have a policy to control or eliminate smoking at work? Yes ____ No ___ -3-

PAGE 205

189 9. If your company does have a smoking policy, which of the following best describes why your company policy was adopted? (Check many as apply.) a. Necessary because of the product we produce. b. Necessary because of safety factors. c. Necessary because of sensitive or special equipment. d. The promotion of good health for employees. e. To enhance our public image. f. To better meet the needs of non-smoking employees. g. (specify). a. PROGRAM MANAGEMENT _1_1 _1_1 _1_1 _1_1 _1_1 _1_1 _1_1 In. IVhlch individual in the company was responsible for the decision to offer health promotion/disease prevention activities? (Describe by title of position no names please). Title 11. Which individual in the company is currently responsible for management of the health promotion/disease prevention activities? (Describe by title of position no names please.) Title ___________ ________________ ___ 12. Is therea line item for the health promotion/disease prevention program in the company's budget? Yes ____ No ________ 1]. In general, which of the following policies apply to your employees for the pro gram you offer? (Check many as apply.) a. Company time available for participation. b. A "flex-time" policy for participants. c. Cost for on-site activity paid by the company. d: Some form of "incentive" is attached to participation (e.g., gifts, money, award.) -4_/ --, _/ --, _/ _/ _/ _/ CONTINUED:

PAGE 206

190 e. Reimbursement for activities off-site. _/ ___ / f. Other (specify.) _/_/ 14. In making a decision to continue your health promotion/disease prevention prograr.l, how important are the following? (Check (0 one answer' for each item.) VERY NOT DON'T IMPORTANT INPORTANT INPORTANT KNOW a. Adequate level of employee __ I __ / __ I __ / __ I __ 1 --I __ 1 participation. b. Improved employee health/reduced __ I __ / __ I __ / __ I __ / __ I __ / health problems. c. Reduced health care costs (including __ I __ / __ I __ / '--/ -/ __ I __ / disability claims and insurance premiums. ) d. Reduced turnover a'f).d/or' absenteeism. __ I __ / __ I __ / __ I __ / __ I __ / e. Improved public image of company. __ I __ / __ I __ / __ I __ / __ I __ / f. Improved employee knowledge of __ I __ / __ I __ / __ I __ / __I __ / healthy lifestyles. g. Improved productivity. __ I __ / __ I __ / __ I __ / __ I __ / h. Other (specify) __ I __ / __ I __ / __ I __ / __ I __ / 15. l.fhich of the following records do you keep [or your program? (Check (v'f as many as apply.) a. Health care costs of participants (includes disability claims and insurance premiums).' b. Level of employee participation. c. Neasure participants' knowledge before and immediately after the program. d. Measure participants' health practices before and immediatelY,after the program. e. Heasure of participants' health practices before and following end of program (3-mo., 6-mo., 1 year, etc.) f. Turnover and/or absenteeism data of participants. -5-__ I __ I __ I __ I __ I __ I __ I __ I __ I __ I __ I __ I CONTINUED:

PAGE 207

191 g. Positive feedback from community (inquiries from prospective employees, requests for information about activities, etc.) h. Aggregate data on employee health status (e.g., health risk appraisa 1. ) i. Measure participants' productivity before and following end of program (3-mo., 6-mo., 1 year, etc.) j. Other (specify). _I _I _I _I _I _I _I _I 16. In which of the following areas do you perceive the introduction of employee health promotion/disease prevention activities have had a positive effect? (Check as many as apply.) a. Improved employee morale. _I-I f. Reduced and i11-_I-ness on the job. b. Reduced rate of rise of health _I-I care costs (inc ludes disability g. Attracted better caliber i_ claims and insurance premiums.) app lican ts .. I I c. Improved employee healthlreduced _I-I h. Improved productivity. _I-I health problems. i. Other (specify). _II d. Reduced hospital and medical _I-I utilization. e. Reduced employee turnover andl _II or absenteeism. 17. How helpful would the following services be to improve andlor expand your program? (Check (V'J one answer for each item.) a. Assessment of employee interest in health promotion activities. b. Insurance data review for use in program planning. c. Health promotion awareness sessions for top executives. (I. Heal th promotion awareness sessions for employees. e. Information on community resources. f. Assistance to design behavior change strategies. g. Nateria1s to market the program to workforce. -6NOT HELPFUL II=:LPFln. _I I _I I I I _I I _I I _I-I _I-I _II _II _II _I I _I I _I-I _II CONTINUED:

PAGE 208

h. Information on new developments in worksite health promotion. i. Assistance to develop evaluation tools. j. Assistance to design employee health risk assessments. k. Insurance data review for use in designing health care cost containment options. 1. Training for in-house staff. m. Assistance to develop comprehensive hypertension screening services. n. Assistance to develop employee assistance counseling or industrial services. o. Assistanc. to develop cancer education -early detection screening services. p. Assistance to develop methods to prevent work-related injury and illness. HELPFUL _I I _I I ._1-I _I-I _I-I I I _1_1 _I-I _I-I .192 NOT HELPFUL _I-I _I I _I-I _I-I ._1 I _I I _I-I _I I _I-I 18. Is there any service not listed in Question #17 which would be helpful to you? Ifso, please explain. 19. Referring to the items listed in Question#17. place an asterisk (*) next to the services for which you would consider paying. 20. The following organizations, both voluntary and for profit. offer materials and services for use in health promotion programs. \olhich organization(s) has your company used, or has plans to use, in the development or implementation of its program? (Check (y') as many as apply.) a. American Cancer Society. b. Colorado Heart Association. c. American Lung Association. d. American Red Cross. e. f. Jewish Cpmmunity Centers (JCC). _I-I _I-I _I-I _I I _I-I _I-I g. Commercial vendors (weight management. smoking cessation,. etc.) h. Alcoholics Anonymous. i. State Health Deportment. _I _I _I _I _I_I j. Locallcounty health department._I __ 1 k. Local hospitals. 1. Institute for Health. -7_I _I _I _I CONTINUED:

PAGE 209

m. Dairy Council n. Private consultant(s) _/_/ _/_/ _/_/ o. Other (Hpecify). C. DEMOGRAPHIC INFORMATION" 21. a. Is your firm one branch of a larger company? Yes ----193 No ___ b. If YES, is the promotion program managed from this office or the headquarters/central office of the company? (Check one answer.) This office ______ __ Headquarters/Central office ______ __ 22. How many people are employed by the company at this site? 23. Describe the make-up of your (approximate percentages). Blue Collar _/_/ l.fhite Collar _/_/ Total: 100% 24. As you complete this questionnaire, you may feel we have missed issues" important to your company's health promotion/disease prevention activities. Your comments on these issues would be helpful to us. THIS CONCLUDES TilE SURVEY. YOU FOR YOUR COOPERATION. -8-

PAGE 210

194 CONTACT CONSENT FOIU-I Occasionally we receive requests from voluntary health agencies, businesses, individuals or community groups for the names of health promotion contact .persons within business and industry. If you are interested in discussing your company's health promotion program with others, and are willing to serve as the contact person for your company, .please sign the form below and return it to the Health Promotion & Education Section, Colorado Department of Health. I give permission to the Colorado Department of Health to release my name as a contact person for health promotion/disease prevention within this company. Signature: Company: Thank you. Date: RETURN TO: Health Promotion & Education Section Colorado Department of Health 4210 E. 11th Avenue Denver, CO 80220

PAGE 211

APPENDIX B GUIDE

PAGE 212

,196 INTERVIEW GUIDE CURRENT PROGRAM INFORMATION: I'd like to start by asking you a few questions about your current program activities. 1. Would you please describe the health promotion activities which your company offers for its employees. -how often are these activities offered? -do you provide classes, screenings, or educational programs on on-going basis? -does management recognize and/or support your activities? -how many employees do you have? -is there a line item for health promotion in the budget? 2. Would you talk a little bit about the parts of the program which have been particularly successful. -what do you think makes x such a success? -what parts of the program have been less successful? -what do you think makes x less successful? 3. Can you describe the involvement of different departments and people in developing the program (or) in planning for a program? -what has been your relationship to the program? -what is your position in the organization? I think I have a' fair sense of your current activities. I would like to shift now and'talk for a few minutes about how the program got started (or) how interest in health promotion developed. PROGRAM HISTORY 4. When did you first start talking about having a program? If appropriate In what year did you actually start the program?

PAGE 213

s. How did interest in employee health promotion first develop in this company? -can you identify key events, circumstances, or people which triggered this initial interest? 6. When the idea of employee health promotion was presented to management or a management committee, ,,,hat were some of the issues that came up, both "pro" and "con"? -what were things that encouraged you to start a program? -what were the things that discouraged you from starting a program? 7. Can you describe some of the first activities you decided to offer (or) the first activities you would offer? -why did you choose these pa.rticular activities or programs? .197 8. What were some things that you had considered doing, but backed off from (or) What are some things which you would back away from? -what was it that made you put off, qr decide against these activities or programs? 9. Once the proposal for a. program was written, what was the process for approval in this company (or) would be the process for approval in this company? -how many levels of management were involved? -how many signatures/sign-offs/and endorsements were required? -on an accountability/humanism continuum when would you place top management in this company? (A) -how does this company view its role and responsibility vis-a-vis the health of employees? (A) I'd like to move now to some thoughts about how companies in general do or don't get into the business of health. promotion.

PAGE 214

.198 GENERAL HYPOTHESES 10. Some people say that most health promotion programs are started because of one person who strongly supports the idea and pushes : it through the company. Other people say "no" that the role of the individual in getting a company to start a program is greatly overrated. I'd like to' know what your experience has been. Canyou describe key individuals who strongly supported your program? -what position did they hold in the company? -were there persons who were s'ignificant roadblocks? -why were these people so committed/opposed? Was it personal or professional? -was there a single "champion" who took it as his/her mission to get the company to offer a program? -what was the level of in-house staff expertise? Was it sufficient to provide programs for employees or would you have to hire someone? (A) -on a scale of 1-5" how important were champions. (D) 11. A lot of companies say that they start health promotion programs, and looked carefully at potential costs/potential savings. Others don't appear to be terribly concerned with the financial aspects when they are considering whether or not to start a program. What kinds of discussions (if any) did you have regarding potential costs or potential savings for the' company? -what kind of data did you look at? -did you do a cost/benefit analysis? -did you see the program as costing or saving your money? (D) -would you say that your company was growing, stable or declining at the time management made the decision regarding an hp/dp program? A Additions D Deletions

PAGE 215

. I -what length of time would. you have to demonstrate return on investment? (A) -On a scale of 1 to 5 how important were finances? (D) 199 12. Another school of thought says that if companies could find the IIrightll program the right exercise program, the right risk appraisal they would be more likely to start something for employees. Others think that the availability of good programs and activities really has very little to do with a company's decision to start a program. What was your experience when you were shopping around? Were you able to find programs or activities acceptable to you? -what were the ideal characteristics you were looking for? (D) -what features made a program unattractive to you? (D) -when you asked other companies about specific actitivies or programs they were implementing, what kinds of questions did you :ask them? (D) -on a scale of 1 to 5, how important was'program availability? (D) 13. Recently there have been a lot of articles, seminars, and personal contacts made to individuals like yourself by persons promoting health promotion programs. For some companies these activities seem to be the thing that triggers them to start a program. Others don't appear to be much influenced by them. Can you describe seminars you attended, articles you read or visits that you received from individuals promoting health promotion? -who or what were the major sources? -what was your reaction to these activities? --did people in the company have a lot or a little exposure? -on a scale of 1 to 5, how important was exposure? (D) A Additions D = Deletions

PAGE 216

14. We have talked about a lot of different factors that are important in a company's decision to offer or to not offer a program. Is there any single factor that we have talked about or not talked about that you think has been the key element in this company? 15. That concludes my questions. Do you have any comments you would like to make? Thank you for your time and assistance. ASK FOR PROPOSAL--BROCHURE -200

PAGE 217

APPENDIX C DIALOGUE USED WITH CONTACT PERSONS

PAGE 218

DIl'.l.OGUE: COHPAlUES WITH PROGRAMS/ORIGINAL CONTACT PERSON. AI. IIi, my na.me is ______ I'm doing a follow-up to the worksite health promotion survey conducted about two years ago by the State Health Deoartment. Your company p,articipated in the survey. You gave your name as a person who was willing to be contacted again about your program. Do you have a few minutes for me to explain the survey and the follow-up project? YP.$ to A2 No go to A9, A2. You may remember when we did the original survey, we interviewed about 300 businesses of ail sizes in Colorado to find out the type of health l'l:uwuL.i.uII acLivities they offered-, and what we found was that a lot of companies were interested in health promotion, usually only large companies of a thousand or more employees actually have programs. But your company was an exception to this.' In the follow-up study, we want to go back to a small of companies like yours to find how you were able to do it--things that helped you most, and the kind of obstacles you had to overcome. This information would really be useful to companies that are struggling to get something going" but don't know where to start. Would you be willing to participate in this project? Not sure or yes go to A3 No go to AIO A3. Let me explain what this follow-up project involves. I would like to interview you and two other persons from your someone from personnel and someone from top management were involved in setting up the program. 1 would come to your agency at a time

PAGE 219

2 convenient to you. The interviews would be done individually and would take about 30 minutes apiece. wuu1" 11.:.. .. Lu LiSP" L.,CUL" Li,,, iut.,rvi.,ws. as I wouldn't: have t:o iake, so many notes, but the information would be confidential. Neither your name, nor the name of your company, 'would appear anywhere in the report. Would you be willing to help? Yes continue to A4 No Thank you for your time. Terminate. A4. Good. I would like you to identify two other people--one from top management and one from personnel (or health service) who were most :!.nvn'vi:>d :I.n l5ett:l.ne '.II' th., I''!'oeram. Are there that co!!!e to !!lind right now? (Negotiate choices if persons from ideal. ca.tegories are not available.) AS. It would help a lot if you could talk to these two other people and schedule a day when I could all three of you. What day do you think would be best? (Negotiate time) A6. I will be sending you a letter explaining the follow-up project. It will include a list of topics to be covered. Perhaps you could use this letter to explain the project to the two other people and schedule a time with them. A7. Do you have any questions? AB. Good. I will check back with you on the ________________ ___ to find out the Bchedule for interviews. Thanks again f.or your help. 203

PAGE 220

3 A9. Is there a more conven1ent time when I can call and expla1n the project? If no Thank you for your t1me. Terminate. It yes Record t1me _______ Good. I wHl call you on the _______ at _______ Thank you. AIO. Is there another person 1n your company who m1ght be more appropr1ate for me to talk to about th1s project? Yes Record name, phone number ___________ No Thank you for your. t1me. Term1nate. 204

PAGE 221

DIALOGUE: COMPANIES WITH PROGRAM/NEW CONTACT PERSON Call to original contact person--no longer there--ask for referral to Chief Executive's Office. BI. Hello. My name is Perhapsyou can help me. I'm doing a follow-up to the worksite health promotion survey conducted about two years ago by the State Health Department. Your company participated in the survey and gave her (his) name as a person who was willing to be contacted again regarding your company's health promotion program. Since ________________ is no longer with the company, I wonder if you could give me the name of the personwho is currently in charge of health promotion. (If hesitating, say ". By this, I mean such activities as physical fitness, smoking cessation, or high blood pressure screening.) (Record name, phone II) Thank you very much. To newly identified contact: B2. Hi. My name is I'm doing a follow-up to a worksite health promotion survey conducted two years ago by the State Health Department. Your company participated in the survey and ________________ who was the person originally interviewed, agreed to be cont.acted again regarding your company's health promotion activities. Since ___________________ is no longer with the company, I was referred to you by _______________________ in as the person who is currently responsible for health promotion. (If hesitating 205 say By that I mean such activities as physical fitness, smoking cessation, high blood pressure screening.) Are you the person I should speak with about this type of program? Yes go to B3 No go B12

PAGE 222

B3. Do you have a few minutes for me to talk about the survey and the the follow-up project? No to Bll Yes continue to B4 You may .remember when we did the original survey, we interviewed about 300 businesses of all sizes in Colorado to find out the type of promotion activities they offered, and what we found was that 2 a lot of companies were interested in health promotion, usually only companies of a thousand or more employees actually have programs. But your company was an exception to thifi. In the follow-up study. we want to go back to a small number of companies like yours to find how you were able to do .it--things that helped you most. and the kind of obstacles you had to overcome. This infuLluai.iul. wuultl ." .. lly ut! ul;t![ul tu cumpanies t:nat: are st:ruggling to get something going. but don't know where to start. Would you be willing to participate in this project? Not sure or yes go to No go to 206 B5. Let me explain what this follow-up project involves. I would like to interview you and two other persons from your company--preferably someone from personnel and someone from top management who were most involved in setting up the program. I would come to your agency at a time convenient to you. The interviews would be done individually and would take about 30 minutes apiece. I would like to tape record the interviews, as I wouldn't have to take so many notes. but the information would be confidential. Neither your name. nor the name of your company. would appear anywhere in the report.

PAGE 223

Would you be willing to help1 Yes continue to ,.,. .. ... .,LC;a.U4.1oUO""c;. B6. Good. I would like you to identify two other people--one from top management and one from personnel (or health service) who were most involved in setting up the program. Are there two that come to mind right now? (Negotiate choices if persons from ideal categories are not available.) B7. It would help a lot if you could talk to these two other people and schedule a day when I could interview all three of you. do you think would be best? (Negotiate time) 207 3 Ba. I will be sending you a letter explaining the follow-up project. It will include a list of topics to be covered. Perhaps you could use this letter to explain the project ,to the two other people and schedule a time with them. B9. Do you have any questions? no. Good. I will check back with you on the __________________ __ to find out the schedule for interviews. Thanks again for your help. Bll. Is there a more convenient time when I can call and explain the project? If no Thank you for your time. Terminate. If yes Record time _____________

PAGE 224

208 4 Good. I will cali you on the at --------------------------Thank you. for me to talk to about this project? Yes Record name, phone number ___________________ No Thank you for your time. Terminate.

PAGE 225

. 209 DIALOGUE: COMPANIES INTERESTED IN CONTACT PERSON Cl. Hi, my name is __ ....,.-______ I'm doing a follow-up to the worksite health promotion survey conducted about two years ago by the State neaitn Department. Your company participated in the survey. You gave your name as a. person who was willing to be contacted again about your program. Do you have a few minutes for me to explain the survey and the follow-up project? Yes continue to No go to 300 businesses of all sizes in Colorado to find out the of health program activities they offered. And what we found was that while a lot of companies were interested in health promotion, usually only the large companies of 1,000 or more employees actually have programs. In the follow-up study, we want to Co back to small companies like yours who have expressed interest in health programs but who have been unable to get a program started who have been able to start only a limited program. need to find out more about the obstacles you face lind the.things you believe would help you most. This infonnation would be useful to other small businesses who are considering starting a program. Would you be willing to participate in this project? No go to Yes continue to C3

PAGE 226

210 C3. Let me explain what the follow-up project involves. I would like to interview you and two other people in your company--preferably someone 'from personnel and someone from top management who have been most I would come to your agency, at a time convenient to you. The interviews would be done individually and would take about 30 minutes apiece. 2 I would like to tape record the interviews, as I wouldn't have to take so many notes, but the information would be confidential" Neither your name, nor the name of your company, would appear anywhere ,in the report. Would you be willing to help? Yes continue fo C4 No Thank you for your time. Terminate. C4. 'Good. I would like you to identify two other people--one from top management and one from personnel (health service) who have been most involved in discussions about whether or not to start a program. Are there two that come to' mind right now? (Negotiate choices if persons from ideal categories are not available.) C5. It would help a lot if you could talk to these two other people and ,schedule a day when I could interview all three of you. What day do you think would be best? (Negotiate time) C6. I will be sending you a letter explaining the follow-up project. It will include a list of topics to be covered. Perhaps you could use this letter to explain the project to the two other people and schedule a time with them.

PAGE 227

211 3 C7 Do you have any questions? CS. Good. I will check back with you on the _________ to find __ .... _. __ ........ """ ......... \:0. .. ..,. Thanks again for your help C9. Is there a more convenient time when I can call and explain the project? If no Thank you for your time. Terminate. If yes Record _________ Good. I will call you on the _________ at _______ -"-_ Thank you. C10. Is there another person in your company who might be more appropriate for me to talk to about this project? Yes Record name, phone number ____________ __ No Thank you for your time. Terminate.

PAGE 228

DIALOGUE: COMPANIES INTERESTED IN PROGRA}I/NEW CONTACT PERSON Call to original contact person no longer there. Ask for referral to Chief Executive's Office. DI. Hi. My name is ______ Perhaps you can help me. I'm doing a follow-up to the worksite health promotion survey conducted about two years ago by the State Health Department. Your company participated in the survey and gave his (her) name as a person who was willing to be contacted again regArding your company's interest in health promotion. (By health promotion I mean activities such as physical fitneRs, smok1ne 212 cessation, and high blood pressure screening.) Since _________ ___ is no longer with the company, I wonder if you could give me the name of another person who shared interest in health promotion and would be involved in the decision of whether or not to Record name, phone number Thank you. To newly identified contact person. 02. Hi. My name is I'm doing a follow-up to a worksite health promotion survey conducted two years ago by the State Health Department. Your company participated in the survey. and who the person originally interviewed, agreed to be contacted again regarding your company's interest in health promotion. Since is no longer with the company, I was referred to you by ________ ___ in as a person interested in health promotion. By that I mean activities such as physical fitness,smoking cessation, high blood pressure screening.

PAGE 229

Are you the person I should speak with about this type of program? Yea proceed to .. ,---... .., 6"" '"'" 213 2 D3. Do you have a few minutes for me to explain survey and 'the follow-up project? Yes go to No go to D4. When we did ,the original survey, we about' 300 businesses of all sizes 1n Colorado to find out the type of health program activities they offered. And what we found was that while a lot of companies were interested in he:tlth prnmotion, usually only the large companies of 1,000 or more employees actually have programs. In the follow-up study, we want to go back to small companies like yours who have expressed interest in health programs but who have been unable to get a' program started or who have been able to start only a limited program. We need to find out more about the obstacles you face and the things you believe would help you most. This information would be useful to other small businesses who are considering starting a program. Would you be willing to participate in this project? No go to Yes continue to D5 D5. Let me explain what the follow-up project involves. I would like to interview you and two other people in your company, preferably someone from personnel and someone from top management who have been most involved in the discussions about whether or not to start a program. I would come to your agency, at a time convenient to you. The interviews would be done individually and would take about 30 minutes apiece.

PAGE 230

214 3 I would like to tape record the interviews, as I wouldn't have to take so many notes, but the information would be confidential. Neither your name, nor the name of your company, would appear anywhere in the report. Would you be willing to help? Yes continue to D6 No Thank you for your time. Terminate. D6. Good. I would like you to identify two other people--one from top management and one from personnel (health service) who have been most involved illdis'::u$sions about whether UL' nol Lu l:iLarL a prugram. Are there two that come to mind right now? (Negotiate choices if persons from ideal categories are. not available.) D7. It would help a lot if you could talk to these two other people and schedule a day when I could interview all three of you. What day do you think would be best? (Negotiate time) DB. I will be sending you a letter explaining the follow-up project. It include a list of topics to be covered. Perhaps you could use this letter to explain the project to the two other people and schedule a time with them. D9. Do you have any questions? D10. Good. I tdll check back with you on the ________ to find out the schedule for interviews. 'Thanks again for your help. will

PAGE 231

4 DII. Is there a more convenient time when I can call and explain the project? If no Thank you for your time. Terminate If yes Record time ________ Good. I will call you on the _____ -'____ at ____ -,__ Thank you. D12. Is there another person in your company who might be more appropriate for me to talk to about this project? Yes Record name, phone number No Thank you for your time. Terminate. 215

PAGE 232

APPENDIX D INTERVIEW RECORD FORM

PAGE 233

.217 INTERVIEW RECORD FORM NAME OF COMPANY/AGENCY ------------------------------------BASIC CATEGORIES: Size: 250-499 Sector: Private LEVEL OF PROGRAMMING: None or Minimal Moderate Sporadic Management Approval/Support: Comments: SUCCESSFUL: Characteristics Conunents: 500-999 Position: Public Moderate Formal Management Health Human Resources ____ Comprehensive Formal UNSUCCESSFUL: Characteristics

PAGE 234

DEPARTMENT/COMMITTEE STRUCTURE: ____ No planning group Single department/Single focus Multiple departments/Single focus Multiple departments/Multiple focii Comments: DEVELOPMENT OF INTEREST: Internal dominant Internal/External Comment: Pe()ple Finances Comments: Trigger event: External/Internal External dominant Program availability Exposure/Promotion

PAGE 235

i I ISSUES: PRO Comments: FIRST ACTIVITIES CHOSEN: Characteristics: Comments: APPROVAL PROCESS: __ __ Very informal ____ Moderately informal Comments: 219 ISSUES: CON ACTIVITIES DEFERRED: Characteristics: ____ Moderately formal Very formal

PAGE 236

I \ \ I I \ I CORPORATION CULTURE Humanism/Accountability: Responsibility for employee: CHAMPION HYPOTHESES: \ \ I \ \1 Ii 'I, I \ 'II, \,' I' I' I I, 'i 'I Staff expertise: Champion: Comments: CEO: \ I I I I \ \ 'I I 1\ 'I \\ ____ personal only personal dominant "-, "I \ r, i I' f II ,,, I' I" II II I 'I \" i, I I \ I \ II \ '\ \ ____ professional dominant ____ professional \ 'I \. dominant \ '\ ,'"

PAGE 237

\ \ \. CORPORATION CULTURE Humanism/Accountability: Responsibility for employee: CHAMPION HYPOTHESES: Staff expertise: Champion: Comments: CEO: Comm(:nts: / The BOARD:' 1 ( Importance: ____ personal only personal dominant personal only personal dominant professional dominant 220 professional only professional dominant professional only

PAGE 238

FINANCIAL ASPECTS: Little or no discussion High moderate discussion Little moderate discussion Extensive discussion Short-term Return or Investment Long-term Return or Investment Cost/benefit analysis Comments: Importance: PROGRAM DESIGN: Very difficult time ____ Moderately difficult time Conunents: Ideal characteristics: Importance: ____ Moderately easy time ____ Very time locating programs Unattractive characteristics: .221

PAGE 239

PROMOTION/EXPOSURE Little or no exposure Low moderate Comment: Seminars/conferences: Written materials: Sales visits: Comments: Importance: KEY ELEMENT: OTHER COMMENTS: High moderate. Extensive