Improving implementation of oral cancer screening recommendations

Material Information

Improving implementation of oral cancer screening recommendations
Tilliss, Terri Isaacson
Publication Date:
Physical Description:
xiii, 174 leaves : ; 28 cm

Thesis/Dissertation Information

Doctorate ( Doctor of Philosophy)
Degree Grantor:
University of Colorado Denver
Degree Divisions:
Department of Health and Behavioral Sciences, CU Denver
Degree Disciplines:
Health and behavioral sciences


Subjects / Keywords:
Medical screening ( lcsh )
Mouth -- Cancer -- Diagnosis ( lcsh )
Medical screening ( fast )
Mouth -- Cancer -- Diagnosis ( fast )
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )


Includes bibliographical references (leaves 159-174).
General Note:
Department of Health and Behavioral Sciences
Statement of Responsibility:
by Terri Isaacson Tilliss.

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:
71801621 ( OCLC )
LD1193.L566 2006d T54 ( lcc )

Full Text
Improving Implementation of Oral Cancer
Screening Recommendations
Terri Isaacson Tilliss
B.S., University of Iowa, Iowa City, 1975
M.S., University of Iowa, Iowa City, 1977
M.A., University of Colorado at Denver, 1996
A thesis submitted to the
University of Colorado at Denver and Health Sciences Center
in partial fulfillment
of the requirements for the degree
Doctor of Philosophy
Health and Behavioral Sciences

This thesis for the Doctor of Philosophy
degree by
Terri Isaacson Tilliss
has been approved
Larry Meskin
Barbara Gerbert


Tilliss, Terri Isaacson (PhD, Health and Behavioral Sciences)
Improving Implementation of Oral Cancer Screening Recommendations
Thesis directed by Associate Professor Kitty Corbett
This project was designed and implemented to assess the impact of a tailored,
theory-driven intervention package on adherence to oral cancer screening
recommendations and tobacco counseling practices by dentists and dental
hygienists. The Social Cognitive Theory (Bandura, 1986) was used to guide
development of a written and semi-structured interview conducted with a
convenience sample of 15 dentists and 15 dental hygienists. The intent of these
was to elucidate the environmental, cognitive and behavioral factors influencing
oral cancer screening practices in dental practices. All three types of factors
contribute to adherence to screening recommendations. The information was used
to inform the development of a multi-modal intervention package.

A convenience sample of 6 private dental practices were randomly assigned as
treatment or control. The intervention package introduced into the 3 treatment
practices included; 1) on-site consensus meeting 2) educational workshop
including hands-on practice 3) posters and brochures to serve as reminders for
patients and practitioners 4) diagrammatic forms for recording examination
findings in patient charts. Patient satisfaction surveys with embedded oral cancer
screening questions were mailed to new patients seen by the dentists and all
patients seen by the dental hygienists within a 3-week time period at both the
treatment and control sites prior to the intervention. The same survey was sent to a
different group of patients seen in another 3-week time period sequenced 2 months
after the intervention. Completed surveys (Ti =288, T2=361) resulted in a 34.5%
average return rate.
Response proportions were tabulated and compared pre- to post-intervention for
both control and treatment groups. Tetrachoric correlations were utilized to
examine the relationship between variables and z tests of independent proportions
were used to compare pre-intervention to post-intervention response proportions for
both groups. Following the intervention, trends in the treatment group indicated
that more patients: 1) thought that they had received a screening examination, 2)

thought that a head/neck examination was provided, 3) were told they had been
screened. However, the only statistically significant difference over time was being
told by the practitioner that they had been screened. The trends seen when
comparing pre- and post-intervention screening rates in the treatment group were
encouraging, particularly since control group screening rates remained relatively
constant from pre- to post-intervention. This finding suggested that the tailored
intervention impacted the outcomes. A larger follow-up study utilizing random
sample selection is indicated.
This abstract accurately represents the contents of the candidates thesis. I
recommend its publication.

This thesis is dedicated to my husband, Stacy Pocrass.
His encouragement and assistance were essential and paramount to my
achievement of this educational goal. This endeavor and our marriage began
simultaneously. He has only known me as a PhD student.
I welcome a new identity and our new beginning...

This thesis is the culmination of the continuing support that I have been blessed to
receive from a variety of people in my life in addition to my husband, Stacy
I wish to thank my parents, Dee and Paul Isaacson and my children, Jennifer Tilliss
and Jonathan Tilliss, who have encouraged me in all of my pursuits.
I appreciate the contributions of my committee and particularly chair Dr. Kitty
Corbett whose expertise facilitated the entire project.
The support of my colleagues in the Dental Hygiene Department at the University
of Colorado School of Dentistry is gratefully acknowledged.
Important assistance was also provided by Cheri Burge and Jonathan Weeks.
. A research grant was provided by the Colorado Tobacco Research Program.
A research fellowship was provided by the American Dental Hygiene Association
Institute for Oral Health.

1. INTRODUCTION...............................................1
The Issue..............................................1
2. LITERATURE REVIEW.........................................6
Oral Cancer............................................6
Trends Associated with Oral Cancer..................6
Age and Oral Cancer...............................6
Incidence of Tongue Cancer Under the
Age of Forty......................................7
The Ratio of Males to Females Contracting
Oral Cancer.......................................8
Incidence and Mortality of African Americans.....10
Risk Factors.......................................14

Combined Effects of Alcohol and Smoking............17
Marijuana Use......................................18
Smokeless Tobacco..................................19
Viral Etiology.....................................20
Current Oral Cancer Screening Recommendations,
Practices, Attitudes and Beliefs......................20
Screening Recommendations..........................20
Practices, Attitudes and Beliefs...................31
Risk Behavior Counseling in the Dental Setting........34
Adherence to Screening Guidelines.....................38
Behavior Change Strategies to Encourage Guideline
Change Theory.........................................52
Social Cognitive Theory............................55
3. METHODOLOGY..................................................62
Specific Aims.........................................62
Research Design.......................................64

Phase One
Formative Component...............................66
Treatment and Control Office Recruitment..........67
Phase Two............................................68
Patient Surveys...................................71
Phase Three..........................................77
Data Entry........................................77
4. RESULTS.......................................................79
Qualitative Component...................................79
Environmental Factors................................81
Behavioral Factors...................................84
Cognitive Factors....................................87
Quantitative Component..................................88
5. DISCUSSION................................................. 101
Qualitative Component..................................101
Quantitative Component.................................106

Summary and Outlook
6. CONCLUSION..................................124
B. CHART FORM..........................135
C. POSTER..............................137
D. BROCHURE............................139
F. PATIENT SURVEY......................145
OF PATIENT SURVEY....................150

3.1 Phase Two Study Design.....................................65
3.2 Phase Two Time-Line........................................76

Table 1 Phase One Data Demographics...............................79
Table 2 Phase One Data Environmental Factors......................80
Table 3 Phase One Data,- Behavioral Factors.........................84
Table 4 Phase One Data Cognitive Factors..........................87
Table 5 Numbers of Oral Health Care Providers and surveys
and surveys mailed and returned for treatment and
control offices......................................................89
Table 6 Descriptive Statistics......................................91
Table 7 Comparison of Treatment Group to Control
Group Pre-Intervention and Post-Intervention
Using z Test of Independent Proportions....................94
Table 8 Comparison of Treatment and Control Groups
Pre-Intervention and Post-Intervention
Using z test of Independent Proportions....................96
Table 9 Tetrachoric Correlations....................................98

The Issue
Oral and pharyngeal cancer strikes approximately 30,000 people each year in the
United States, resulting in about 8,000 deaths annually. These statistics have
remained unchanged for many decades (Ries, 1999). Oral/pharyngeal or head/neck
cancer (often referred to as oral cancer) comprises about 3% of all malignant
tumors in the United States (American Cancer Society, 1999). World-wide, oral
cancer is the fifth most prevalent form of cancer (Mignogna et al, 2004). The
world area with the highest incidence is Melanesia (31.5 per 100,000 in men and
20.2 per 100,000 in women). Rates in men are high in western Europe (11.3 per
100,000), southern Europe (9.2 per 100,000), south Asia (12.7 per 100,000),
Southern Africa (11.1 per 100,000), and Australia/New Zealand (10.2 per 100,000)
In females, incidence is relatively high in southern Asia (8.3 per 100,000). These
patterns reflect prevalence of specific risk factors, such as tobacco/alcohol use in
western Europe, southern Europe, and southern Africa, and the chewing of betel
quid in Southcentral Asia and Melanesia. The high rate of oral cancer in Australia

is due to lip cancer, related to solar irradiation (Parkin et al, 2002). In India and
Southeast Asia, the oral cavity is often the first or second most common site for
malignancy (ICMR, 1987). This is largely due to oral exposure to culturally
specific damaging forms of tobacco with additives of areca nut or betel leaf (quid),
often combined with slaked lime or spices (Sciubba, 2001; Llewellyn, 2001). It has
been suggested that non-Westem tobacco practices are becoming more prevalent in
the United States due to the influx of people from diverse cultures (Sciubba, 2001),
which may lead to increasing oral cancer rates in the United States. Other addictive
and damaging forms of tobacco surfacing in the United States include the use of
bidi cigarettes (stronger and more intense than the common cigarette) and kretek
(clove cigarettes).
Oral cancer is a particularly devastating form of cancer and has a relatively poor
prognosis. The fact that the majority of these cancers have nearly always
progressed to later stages at diagnosis accounts for both of these outcomes. Since
the cancer is usually not detected until an advanced stage, lesions are often invasive
or have metastasized, thus drastically reducing the cure rate.

The mainstay of current therapy for oral cancer is surgery and radiation treatment
(Ord, 2001; Neville, 2002). Surgical treatment can be extremely disfiguring,
exerting a major impact on self-image. Additionally, surgery can have deleterious
effects on the functions of eating and speaking, both important hallmarks of social
interaction. Radiation therapy to the head and neck area is destructive to salivary
glands resulting in xerostomia, or extreme dry mouth. A sequela of this condition
can be the development of pervasive dental decay. Just as surgery affects both
eating and speaking, xerostomia also impacts these important components of social
Even more critically, the overall five-year survival rate for oral cancer in the
United States remains at about 50%, despite advances in treatment (Silverman,
2001). Because five-year survival is directly related to stage at diagnosis,
improved early detection rates would have the potential for decreasing the
mortality from this devastating disease. Neither medical nor oral health care
providers have succeeded in improving the stage at which oral cancer is detected.
If early detection rates were to improve, the survival rate would increase
dramatically since individuals with localized tumors have a five year survival rate

of 80%, in comparison to the five-year survival rate of 40% for those with regional
metastases (Ries, 1999). Only half (47%) of oral cavity cancer cases are localized
at time of discovery. This figure is even more dismal for pharyngeal cancer where
only 17% are localized at diagnosis (Swango, 1996). Pharyngeal cancers are nearly
always discovered in later stages since the location impedes early detection by
both patient and practitioner. The irony is that oral cancers are amongst the most
curable of cancers when diagnosed and treated early (Cruz, 2002).
There is a great need to improve the early detection rate of oral cancers by
developing methods of increasing the frequency and regularity of oral cancer
screenings being conducted by oral health professionals. This demonstration
project was designed to assess whether the frequency of oral cancer screening
examinations performed by dentists and dental hygienists in the dental office
setting can be increased through introduction of a multi-modal intervention
package. Initially, semi-structured interviews of dentists and dental hygienists
were conducted to elucidate the issues that impact oral cancer screening practices
of practitioners. This information informed the development of the tailored
intervention, guided by the literature and by Banduras Social Cognitive Theory
(Bandura, 1986). Patients at 6 dental offices who were appointed during a

specified 3-week time period were mailed a survey to evaluate their perceptions of
receiving oral cancer screenings. After an intervention package was implemented
in half of the offices, the patients appointed during another specified 3-week time
period were surveyed by mail. Response proportions were tabulated and compared
from pre-to post-intervention.

Oral Cancer
Trends Associated With Oral Cancer
Age and Oral Cancer
The most important demographic factor associated with oral cancer is age, which is
strongly associated with cancer incidence in general. The demographics of the
aging population is likely to result in an increase in cancer in the future. Since the
lag time associated with risk reduction is 30 years or more, even behavioral
changes will not manifest a change in cancer incidence for quite some time
(Swango, 1996).
The mean age of onset of oral cancer is in the seventh decade of life, with 90% of
oral cancers occurring in those older than 45 years (Silverman, 2001). Carcinomas
of the oral cavity and oropharynx are rare in patients of age 50 and younger

(Lleewellyn, 2001). However, the incidence of tongue cancer has been reported to
have increased in the past 20 years in those under 40 years of age (Myers, 2000).
Incidence of Tongue Cancer Under the Age of Forty
Generally, oral cancers occur during the sixth to seventh decades of life, probably
due to the cumulative effects of risk factors. It has been considered unusual for oral
cancer to strike those under forty years of age.
Increased incidence and mortality from cancer of the tongue among young males
has been reported in the past two decades (Davis & Severson, 1987; Depue, 1986;
Shemen et al, 1984). It should be noted that where tongue cancer has traditionally
been more prevalent in males, this proportion is changing as more women are being
diagnosed. In fact, the ratio of women to men with tongue cancer has been
reported at 2:1 (Jones, 1989).
The percentage of cancers of the tongue for those younger than forty years has
increased from 3% in 1973 to 6% in 1993 (Myers, 2000). Suggested contributing
factors to this trend are increased smokeless tobacco use in younger Americans,

frequent and prolonged use of marijuana, and perhaps more frequent oral sex, if a
viral connection is determined. Epstein-Barr virus (EBV) and human papilloma
virus (HPV) have been researched (Llewellyn et al, 2001). However, it has been
reported that in patients under forty years of age, particularly females, the usual risk
factors of smoking and excessive drinking do not appear to be as pertinent
(McGregor, et al, 1983; Schantz, 1988). A review of 24 other studies concluded that
the impact of risk factors for young individuals is unclear (Llewellyn, 2001).
Because risk factors do not seem to play as strong a role for younger individuals, it
has been suggested that tongue cancer in this age group may be a distinct entity
from that of older individuals (Atula et al, 1996; Cusumano & Persky, 1988). With
the questionable role of risk factors, the issue of an inherent genetic determinant
has been raised (Lynch, 1995). Differences in the causal factors and clinical course
of oral cancers between younger and older individuals are being investigated
(Myers, et al, 2000; Schantz, 2002; Llewellyn et al, 2001).
The Ratio of Males to Females Contracting Oral Cancer
Males have traditionally exhibited a higher incidence of oral and pharyngeal
cancers than females. The male-to-female ratio in 1950 was 6:1 (Silverman, 1985).

The predilection for males has been largely attributed to a higher incidence of lip
cancer due to sun exposure as a result of male participation in outdoor activities
and occupations. Additionally, males have more exposure to the two main
behavioral risk factors of smoking and drinking habits. In 1980, the male-to-
female ratio for the incidence of oral cancer changed to 2:1 (Silverman, 1985).
The increase in the female incidence has been attributed to changing patterns of
smoking and alcohol use among women. Another factor impacting the trend is that
concomitant with the aging of the population, females in the over 65 age group
outnumber males by more than 40% (U.S. Department of Commerce, 1995). This
change in the demographic profile of the United States population has contributed
to the increase in the proportion of cases among women. Additionally, the
incidence of lip cancers in men has decreased, presumably due to fewer men
engaged in outdoor occupations. This has contributed to the change in the male to
female ratio of oral cancer.
Mortality rates expressed as the number of deaths per 100,000 people was 2.6 for
oral and pharyngeal cancers during the period of 1994-1998 in the U.S. (National
Cancer Institute, 2001). However the mortality rate in men, at 3.9, is more than

twice that of women, which stands at 1.4. A likely reason for this disparity is that
men seek care later than women, when the cancers are in a less curable stage.
Incidence and Mortality in African Americans
The incidence of oral and pharyngeal cancer in African-Americans (12.4 cases per
100,000 population) is higher than that of whites (9.7 cases per 100,000), with
African-American men having the highest incidence at 20.5 cases per 100,000
population (Neville, 2002). The highest incidence and mortality rates for oral
cancers are found among African American men. Among this group, the oral
cavity has been the fourth most frequent site of cancer (Silverman, 2001). African-
American men are the only gender-ethnic group to have an increased annual
percent change for all cancer sites combined (Wingo, 1998).
The oral cancer five-year survival rate for black men prior to 1984 was 25% and
from 1985-1996 was at 27%, about half of that for white men (Shiboski, 2000).
High tobacco use rates are thought to contribute to the high incidence of oral cancer
in African-American men. In 1994, 33.9% of black men were reported to be current
smokers, which was higher than for any other racial/ethnic group except Native

Americans (Parker, 1998; CDC, 1996). Although historically the smoking rate
among African American men has been considerably higher than for White, non-
Hispanic men, these figures have recently reached close approximation. In 2004,
the most recent year that data are available, 24.1% of White, non-Hispanic men and
23.9% of African American men were reported to be smokers (CDC, 2005). It has
also been reported that although African American men smoke fewer cigarettes
than white males, they tend to smoke brands with higher nicotine and tar levels.
They are also more likely to smoke mentholated cigarettes. The menthol produces
a cool sensation in the throat when smoke is inhaled. Consequently, smoking
menthol cigarettes allows one to inhale more deeply or hold the smoke inside the
mouth longer than smokers of non-menthol cigarettes. This may contribute to
understanding why African Americans have higher lung cancer and oral cancer
incidence and higher mortality rates than whites. About 76% of African Americans
smoke menthol cigarettes as compared to 23% of White-non-Hispanics (Oral
Cancer Foundation, 2006).
These declining rates are encouraging, but with a potential 30-year lag time before
the effects of oral cancer become symptomatic, the effects of the lower smoking
rates will not be apparent in oral cancer statistics for quite some time. Alcohol

consumption and the combination of alcohol with smoking are additional risk
factors which may be higher among the African-American male population.
Another factor which may disproportionately affect the cancer rates among the
African-American population is altered immune defenses, due to low social,
economic, and health capital. Additionally, the altered immune status
accompanying HIV/AIDS is prevalent in the African-American population, and
may play a role in cancer incidence.
One preventive practice which is considered a protective measure against oral
cancer is a diet high in fruits and vegetables. Socially and economically deprived
groups, such as African-Americans, would be expected to be low on this practice.
The sexual practices of this and some other ethnic/social/cultural groups may also
increase the risk from this mode of transmission. Viral infection has also been
suggested as a causative factor for cancers, particularly oral cancer. The
association between the human papilloma virus (HPV) and oral cancer is being
investigated, with oral sexual practices seen as the mode of transmission. Sexual
practices are thought to contribute to the spread of the HPV virus from the genital

to oral sites. HPV has been implicated in cervical cancer, and the mucosa of both
sites is similar in nature. Miller & White (1996) have suggested that HPV may be
an effect rather than a cause of oral cancers. A study of black South Africans under
40 years of age (Van Rensberg et al.,1996) concurred with this view. More work is
necessary in this important area to elucidate causative factors.
The most likely factor in the poor survival rate for black men is related to the late
stage at diagnosis. This can be attributed both to less access to medical and dental
care for African-Americans and to less health-care seeking behaviors of men in
African American women have routinely had a lower smoking rate than White non-
Hispanic women. In 2004 these rates were 17.2% and 20.4% respectively..
Smoking by women has not been generally supported socially in the African
American culture. For both genders combined in 2004, the rate for African
Americans (20.2%) is less than that for non-Hispanic whites (22.2%).
Oral cancer incidence and mortality rates for African-American women are higher
than for white women, but do not approximate those for African-American men.
Reasons for the differences between white and African-American women are

probably similar to those for men, primarily less access to health care until lesions
are in advanced stages of spread.
Risk Factors
Some of the risk factors associated with cancers of the head and neck region have
been unequivocally demonstrated as etioiogic agents. These include smoking,
excessive use of alcohol, and the additive effects of these two. Other factors that
have shown some association are marijuana use, smokeless tobacco (chew and
snuff), diet, viral etiology, genetics and gene mutation. While smoking and alcohol
are recognized risk factors, the others remain under investigation.
The role of tobacco in the development of oral cancer is well established with the
risk for developing oral cancer reported at two to five times greater for smokers
than for nonsmokers (Blot et al, 1988; Mashberg,et al.,1993; Spitz, 1998; Marshall,
1992; Schwartz, 1998; Hayes, 1999). The risk increases with both number of
cigarettes smoked daily and years of smoking (Blot et al, 1988; Hayes, 1999; Spitz,

1998; Marshall, 1992). Those with a greater than 40 pack-year history (the smoke
equivalent to 1 pack of cigarettes per day per year) have a seven times greater risk
of developing oral cancer than non-smokers (Marshall, 1992).
Approximately 80% of oral cancer patients smoke, which is three times greater
than for the general population. In addition, oral cancer patients who continue to
smoke run a greater risk of developing an additional malignancy compared to those
who stop smoking (Day et al, 1994). Tobacco contains many carcinogens; the most
putative is thought to be nitrosamine (Gupta et al, 1996). The etiological role of
tobacco in carcinogenesis may be induction of a mutation of the p53 tumor
suppressor gene (Brennan et al, 1995).
One of the national health objectives for the year 2010 (objective 21.1a) is to
reduce the prevalence of cigarette smoking among adults to no more than 12%, and
to eliminate population disparities (National Center for Health Statistics, 2001). In
2004, the percentage of Americans smoking was 20.9%, indicating that nearly one
in five adults smoke cigarettes. It is encouraging that this figure has been declining
by about a percentage point each of the past few years. Extrapolating with the

current rate of annual decrease, the goal of 12% would not be attained by 2010;
with a more dramatic yearly decline it could potentially be reached.
These downward trends indicate that progress is being made, but not enough nor
fast enough. Onset rates of cigarette smoking among the youth have not declined
over the past decade. In 2004, 28% of U.S. high school students were reported as
smokers (CDC, 2004), there is still ample cause for concern.
It can be difficult to separate out the effects of alcohol and smoking on oral cancer
since the majority of oral cancer patients have both smoked and abused alcohol. It
is also difficult to accurately estimate alcohol intake and frequency of use, due to
inaccuracies in subject reporting. In one study, 40% of head and neck cancer
patients were identified as alcoholics (Deleyiannis et al, 1996). Moderate-to-heavy
drinkers have a 3-9 times greater risk of developing oral cancer, depending on type
of alcohol and frequency (Mashberg, et al, 1993; Jovanovic et al, 1993; Blot et al,

The main carcinogenic agent in alcohol is ethanol, although it has not been shown
by itself to be carcinogenic. Rather, it is thought to be other contaminants and
cogenes which are responsible for the association with cancer (Ogden & Wight,
1998), with ethanol acting as a promoter. The metabolism of ethanol results in
acetaldehyde, which is a highly toxic substance suspected to cause tissue damage.
Ethanol metabolism has effects on biological membranes, specifically the
phospholipid bilayer of the cell membrane. In this way, ethanol, acting as a
solvent, may enhance the penetration of tobacco carcinogens. Additionally alcohol
is thought to affect DNA repair mechanisms. Ethanol also enhances liver
metabolic activity, which may activate carcinogens.
Combined Effects of Alcohol and Smoking
Although smoking and alcohol are independent risk factors for the development of
oral cancer, almost complete synergy of the combined exposure to smoking and
alcohol has been demonstrated (Rothman and Keller, 1972).
Heavy smokers and drinkers can have about 100 times greater risk for developing an oral
malignancy (Blot et al, 1988; Mashberg et al, 1993). When viewed independently, it has

been suggested that alcohol may have a greater influence than smoking (Tuyns, 1979),
although the distinction between the effects of smoking and drinking is controversial.
Based upon a review by Llewellyn et al (2001), many researchers have concluded that the
combined.risk for oral cancer is at least greater than the additive effects and is probably
Marijuana Use
Marijuana is the second most commonly smoked substance in the United States after
tobacco. An estimated 31% of the United States population over 12 years old reported
having used marijuana in 1992 (Zhang et al, 1999). A synergistic effect is suggested
between marijuana and tobacco smoking. Marijuana use is associated with increased risk
of head and neck cancer in a dose-dependent fashion, particularly in young individuals,
with marijuana users two to seven times more likely to develop oral cancer than non users
(Zhang et al, 1999). Although a carcinogenic effect of marijuana has been concluded,
direct causation between marijuana and tongue cancer has not yet been proven (Firth,

Smokeless Tobacco
The role of smokeless tobacco in the development of oral cancer has not been
researched to the same degree as smoked tobacco. Smokeless tobacco does appear
to increase the risk of oral cancer and oral leukoplakia (Winn et al, 1981; Grady et
al, 1990) but appears to be associated with a lower cancer risk than with smoked
tobacco (Neville and Day, 2002). The frequent development of visible tissue
changes can serve as a motivator to quit if used as an educational tool. Oral
cancers frequently develop in smokeless tobacco users at the site of tobacco
The use of smokeless tobacco can result in a tobacco addiction, which can progress
to the use of cigarettes or other sources of nicotine.
According to Llewellyn et al (2001), Research with large sample numbers has
uniformly shown that frequent consumption of vegetables, citrus fruit, fish and
vegetable oils are the major features of a low-risk diet for cancer of the oral cavity

adjusting for smoking and alcohol intake (page 411). The role of micronutrient
ingestion with an associated antioxidant effect is being researched. A diet high in
fruits and vegetables is being referred to as cancer protective relative to oral
Viral Etiology
Both the Epstein-Barr virus (EBV) and the human papilloma virus (HPV) have
been implicated in oral carcinogenesis (Llewellyn et al, 2001). It is unclear
whether EBV plays a causative or coincidental role in oral cancers. The role of
HPV is also somewhat controversial; the potential source of viral colonization in
the oral cavity is sexual transmission between genital and oral mucosal surfaces.
Current Oral Cancer Screening Recommendations,
Practices, Attitudes and Beliefs
Screening Recommendations
The standard for an oral cancer screening includes a visual/tactile examination of
the mouth, full protrusion and lateral views of the tongue with the aid of a gauze

wipe, and palpation of the tongue, floor of the mouth and lymph nodes of the neck
(U.S. Dept. Health and Human Services, 2000). However, there is a lack of
evidence-based support that this type of examination decreases the oral cancer
mortality rate.
The most recent comprehensive review of screening programs for the early
detection and prevention of oral cancer Was conducted by the Cochrane
Collaboration (Cochrane Library 2005). The search strategy included electronic
data bases from 1966-2002 and also included bibliographies, hand searching of
specific journals and published and unpublished data supplied by authors.
Randomized controlled trials of screening for oral cancer or potentially
premalignant oral lesions using an array of screening techniques were included.
One-hundred potentially relevant articles were selected for review. Of the 100
citations selected for review, 48 described uncontrolled or non-experimental
studies, 8 were observational, 37 were narrative reviews or commentaries, 4
described controlled clinical trials and 3 citations described randomized controlled
trials. However, these randomized controlled trials did not study screening
techniques. Consequently, there was only one randomized control trial included in
the Cochrane evaluation; the focus of this study was on visual screening for the

detection and prevention of oral cancer (Sankaranarayanan, et al, 2005). This study
was conducted in Kerala, India where oral cancer is the most common form of
cancer and of cancer-related death in men (Ferlay de al, 2002; Parkin et al, 2002).
This study was the first systematic review of the impact of screening strategies on
oral cancer reported in the literature based on randomized controlled trials. Data
were reported on oral cancer incidence, disease specific mortality, and stage at
diagnosis after 6 years of follow up. The Cochrane group determined that there
were some methodological weaknesses in the study, mostly concerning the method
for assigning the 13 clusters of residents into intervention and control arms since
the smoking and drinking distribution was uneven for the two groups. However, it
was acknowledged that such differences in the baseline variables might be expected
to occur more often in cluster randomized studies. In both groups, subjects were
interviewed to extract information concerning social and personal habits such as
tobacco and alcohol use, and dietary supplements. In the intervention group, extra-
and intra-oral examinations including cervical lymph node palpations were
conducted. For the control group, there were no visual oral inspections conducted.

The fact that there was not a significant difference in overall death rate from oral
cancer between the intervention group and the control group led the Cochrane
review authors to conclude:
There is insufficient evidence to support or refute
the use of visual examination as a method of screening for
oral cancer in the general population. Systematic
examination of the oral cavity should remain an integral
part of their routine daily work. Particular attention should
be paid to high risk individuals (Kujan et al, 2005, p.8).
They do acknowledge that the sensitivity and specificity of visual examination to
detect oral lesions was over 80%. This is similar to findings from other
investigations of the discriminatory ability and consistency of oral cancer screening
(Wamakulasuriya and Pindborg 1990; Jullien et al, 1995; Downer et al, 2004).
Their explanation for the apparent lack of correspondence between the high
specificity and sensitivity of the visual screening and the lack of effect on the
mortality rate is that there is a dearth of understanding of the natural history of oral
cancer and the effectiveness of treatment.
Patton (2003) analyzed 23 studies related to the effectiveness of visual examination
which met her review criteria. Her review was particularly geared toward the

effectiveness of visual screening in community-based settings and agreed with
conclusions of other reviews that:
There is insufficient evidence to draw a conclusion
regarding the effectiveness of oral cancer visual screening
in community settings. The actual value of the screening in
identifying new oral cancer cases may be minimal when
low risk populations are screened. Due mostly to higher
population prevalence, premalignant lesions are more
readily detected by community-based screenings than are
invasive oral cancers. Overall effectiveness of oral cancer
screening programmes in controlling oral cancer would be
enhanced if detection of premalignant oral lesions resulted
in access to effective treatments to prevent malignant
transformation. A secondary benefit of community-based
oral cancer screening programmes is the potential to
improve patient oral health awareness and to motivate
individuals to reduce oral cancer risk behaviours by
introducing them to and engaging them in tobacco and
alcohol cessation activities (page 711).
Another review of the evidence on the prevention of oral cancer mortality was
conducted by the Canadian Task Force on Preventive Health Care (Hawkins 1999).
It also focused on visual clinical examination studies, but did not indicate the
review methodology used. The results of that review led the Task force to
conclude that there is fair evidence to exclude population screening for oral cancer
by clinical examination, but there is insufficient evidence to include or exclude
opportunistic screening for oral cancer by clinical examination of asymptomatic
patients. (Hawkins RJ et al, 1999).

The previous policy of the US Preventive Health Services Task Force (USPHSTF,
1966) was that,
There is insufficient evidence to recommend for or
against routine screening of asymptomatic persons for oral
cancer by primary care clinicians... clinicians may wish to
include an examination for cancerous and precancerous
lesions of the oral cavity in the periodic health examination
of persons who chew or smoke tobacco (or did so
previously), older persons who drink regularly, and anyone
with suspicious symptoms or lesions detected through self-
examination. ... Appropriate counseling should be offered
to those persons who smoke cigarettes, pipes, or cigars,
those who use chewing tobacco or snuff, and those who
demonstrate evidence of alcohol abuse (section 16).
Their most recently updated recommendation (2004), states that the evidence is
insufficient to recommend for or against routinely screening adults for oral cancer.
Their rationale for this recommendation is:
The USPSTF found no new good-quality evidence
that screening for oral cancer leads to improved health
outcomes for either high-risk adults (i.e., those over the age
of 50 who use tobacco) or for average-risk adults in the
general population. It is unlikely that controlled trials of
screening for oral cancer will ever be conducted in the
general population because of the very low incidence of
oral cancer in the United States. There is also no new
evidence for the harms of screening. As a result, the
USPSTF could not determine the balance between benefits
and harms of screening for oral cancer (page 1).

The American Cancer Society recommends a cancer-related check-up annually for
all individuals aged 40 and older and every three years for those between the ages
of 20 and 39, which includes examinations of cancers of the thyroid, oral cavity,
skin, lymph nodes, testes and ovaries (Smith et al, 2002). Further, the American
Cancer Society (2005) recommends that,
Primary care doctors and dentists examine the mouth and
throat as part of a routine cancer-related checkup. Dentists and
primary care doctors have the opportunity, during regular
checkups to see abnormal tissue changes and to detect cancer at
an early, curable stage. Many doctors and dentists also
recommend that people, especially those at higher risk, take an
active role in the early detection of theses cancers by doing
monthly self-examinations. This means using a mirror to check
for any of the signs and symptoms of cancer in the mouth and
throat (page 1).
The American Cancer Society (2005) further provides what they refer to as the
'bottom line. Their bottom line statement is that,
Most oral cancers could be prevented if people did
not use tobacco or drink heavily. Quitting tobacco and
limiting alcohol use sharply reduces any risk of developing
oral cancer, even after many years of use. Many oral
cancers may be found early by a combination of routine
screening examinations by a doctor or dentist and self-
examination ( page 3).

Mignogna and colleagues (2004), provide a perspective on oral cancer screening
based upon on the World Cancer Report of the World Health Organization (IARC-
WHO, 2002) and recent oral cancer screening reviews:
Nevertheless, the real benefit and cost-effectiveness
of oral cancer screening is still controversial. The majority
of available studies have underlined there is no definitive
evidence that routine examination of asymptomatic and
symptomatic patients can reduce mortality from oral
cancer. These poor outcomes were mainly related to high
rates of false-positive referrals, a low yield of oral cancer in
the screened population, and low compliance, mainly
among high-risk subjects. However, it is our opinion that
these data need further critical evaluation. First, it should
be stressed that cancer screening is also a strongly debated
issue in other oncological fields: the reduction in mortality
due to screening for lung (via chest radiography, sputum
cytology or spiral computed tomography) and breast cancer
(via mammography) has been questioned by several
authors (Olsen and Gotzsche, 2001; Manser et al., 2003)
who reported that the currently available reliable evidence
does not support the introduction of these screening
methods into clinical practice (page 140).
They further state,
It should be underlined that two recent systematic
reviews have reported that there is no evidence to support
or refute the use of a visual examination as a method of
screening for oral cancer in the general population (Kujan
et al., 2003; Patton, 2003), highlighting that there is
insufficient evidence to draw a definitive negative or
positive conclusion about this issue and suggesting that
absence of evidence is not evidence of absence (Altman
and Bland, 1955) Although the efficacy and cost
effectiveness of organized and systematic population-based

oral screening in reducing the incidence and mortality from
oral cancer have yet to be established, we stress there is no
other oncological specialty in which the WCR preventive
guidelines could be applied in such an easy and effective
manner, enhancing efficiency and maximizing outcomes
with a substantial reduction of cost, as in the field of oral
cancer (page 141).
Another approach to screening for oral cancer which has been suggested as an
alternative to population screening is opportunistic screening. Opportunistic
screening is undertaken when patients access a health care professional for some
other reason. In this way, high risk groups can be targeted within the general
population for oral cancer screening, since mass screening may not be cost-
effective or uniformly recommended. One study focused upon this concept within
18 dental practices in Britain where risk habit data were combined with
identification of oral lesions. Data on 2,265 patients demonstrated that there was a
significant association between positive lesions, heavy smoking, and heavy alcohol
use in males; both risk behaviors are more prevalent in Britain than in the United
States. The researchers suggested that patients attending general dental practices
are representative of the general population both in terms of lesion prevalence and
high risk habits such as smoking and drinking, suggesting that opportunistic
screening may be a realistic alternative to population screening (Lim et al. 2003).

The oral cancer screening study which has recently emerged as the most definitive
and often quoted is that of Sankaranarayanan et al (2005). This is the same
investigation that The Cochrane Review (Cochrane Library, 2005) cited as the only
randomized controlled trial of screening techniques.
Methodology of the study involved 13 clusters; seven were randomized to three
rounds of oral visual inspection by trained health workers at 3-year intervals and
six to a control group during 1996-2004, in Trivandrum district, Kerala, India.
Healthy participants aged 35 years and older were eligible for the study. Screen-
positive people were referred for clinical examination by doctors, biopsy, and
treatment. Outcome measures were survival, case fatality, and oral cancer
mortality. Oral cancer mortality in the study groups was analyzed and compared
by use of cluster analysis. Analysis was by intention to treat.
The authors summarized the findings as follows:
Of the 96 517 eligible participants in the
intervention group, 87 655 (91%) were screened at least
once, 53 312 (55%) twice, and 29 102 (30%) three times.
Of the 5145 individuals who screened positive, 3218 (63%)
complied with referral. 95 356 eligible participants in the
control group received standard care. 205 oral cancer cases

and 77 oral cancer deaths were recorded in the intervention
group compared with 158 cases and 87 deaths in the control
group (mortality rate ratio 0-79 [95% Cl 0-51-1-22]). 70
oral cancer deaths took place in users of tobacco or alcohol,
or both, in the intervention group, compared with 85 in
controls (0-66 [0-45-0-95]). The mortality rate ratio was
0-57 (0-35-0-93) in male tobacco or alcohol users and 0-78
(0-43-T42) in female users (page 1927).
They go on to say:
Our results showed that overall, the rate of oral
cancer deaths in the intervention group (that was screened
for cancer) was non-significantly lower than those in the
control group, 9 years after initiation of screening.
However, in users of tobacco or alcohol, or both, this value
was significantly lower in the intervention group than in
controls. Mortality rates were also reduced in users of
tobacco or alcohol, or both, in the intervention group
compared with controls, although this difference was only
significant in male users (p.1930).
Our findings support the routine use of oral visual
screening in the reduction of oral cancer mortality in the
high-risk group of users of tobacco or alcohol, or both.
This study indicates that oral screening could be restricted
to high-risk individuals and organised visual screening is a
worthwhile initiative of control for oral cancer in addition
to primary prevention efforts to reduce tobacco and alcohol
use (p. 1932).
The authors concluded:
Oral visual screening can reduce mortality in high-
risk individuals and has he potential of preventing at least
37 000 oral cancer deaths worldwide (page 1932).

Until there is concordance concerning a guideline for oral cancer screening, one
approach is to consider the recommendation of the American Cancer Society as the
standard. The ACS prescribes a cancer check-up annually for all individuals aged
40 and older to include examinations oral cavity along with five other structures.
This provides consistency and ease of recall until definitive evidence suggests
establishment of a different guideline.
Practices, Attitudes, Beliefs
The two main avenues for assessing current oral cancer screening practices are
surveying patients following a dental visit, and surveying the practitioners. Such
self-reported information is rarely accurate due to over-estimation. Despite the
proclivity of practitioners to over-estimate their compliance with screening
recommendations, compliance is still low.
The 1998 National Health Interview Survey Adult Prevention Supplement,
conducted by the Centers for Disease Control and Prevention, surveyed over
12,000 individuals by asking, Have you ever had a test for oral cancer in which

the doctor or dentist pulls on your tongue, sometimes with gauze wrapped around it
and feels under the tongue and inside the cheeks? Only 16% of the respondents
reported that they had ever had such an examination. This investigation also
assessed whether cigarette smoking and alcohol use were associated with receipt of
an oral cancer examination. Current smokers were not more likely to have received
an examination than were never smokers. Dentate current and former alcohol users
were more likely than abstainers to have received an examination, while there was
not a statistically significant association between alcohol use and receipt of an oral
cancer examination among edentulous adults (Macek et al. 2003).
A survey conducted in 2002 of 1096 adults in North Carolina found that 29%
indicated that they had an examination for mouth cancer where, the doctor or
dentist pulled on your tongue, sometimes with gauze wrapped around it, and felt
under the tongue and inside the mouth. Of particular concern is that smokers were
2.3 times less likely to recall having an examination, and only 23% of those over 40
years reported having a screening within the last year (Patton, LL, 2004). These
low figures suggest that often clinicians are not following ACS or USPHSTF
recommendations. Even when examinations are performed, professional/diagnostic
delays can be associated with disease stage. A study of 188 subjects with upper

aerodigestive carcinoma showed that a professional delay of greater than 1 month
can have a marked increase in occurrence of a late stage diagnosis (Allison, 1998).
Most of the studies conducted with practitioners have evaluated both the oral
cancer knowledge base and screening practices of dentists and dental hygienists.
Two conclusions of nearly all of these researchers were consistent. First, a
substantial or systematic relationship between respondents knowledge about oral
cancer and oral cancer detection behaviors was not found; beliefs and attitudes
influence behavior more strongly than knowledge; both the knowledge base and
oral screening practice habits need to be increased (Yellowitz and Goodman, 1995;
Sadowsky et al, 1988; Horowitz et al, 2001; Forrest et al, 2001; Horowitz et al,
Secondly, most of the literature on this topic consistently concludes that there is a
pervasive need for systemic updates and interventions in order to increase the
competence and confidence of practitioners in performing the oral cancer
examination if the outcome of early detection is to be achieved (Horowitz et al,
2002; Howowitz et al, 2000; Forrest et al, 2001; Alonge and Narendran, 2003;
Siriphant et al, 2001; Forrest et al, 2001).

Risk Behavior Counseling in the Dental Setting
Tobacco cessation programs in dental offices have been promoted for
approximately the past 20 years (Christen et al, 1984). Although many initiatives
have supported the involvement of dentists in tobacco cessation activities, the
large-scale integration of tobacco cessation programs in dental office settings has
not been successful.
The First National Dental Symposium on Smoking Cessation was held at the
American Dental Association headquarters in 1989. The National Dental Tobacco-
free Steering Committee of the National Cancer Institute (NCI) supports oral health
teams and dental organizations in smoking cessation efforts. U.S. dentists were
involved in the NCIs Community Intervention Trial for Smoking Cessation
(COMMIT) program at the community level (Jones et al, 1993).
Documents such as Clinical Practice Guidelines published by the Department of
Health and Human Services (Fiore et al, 2000), a monograph published by NCI
(Monograph #5,1996), and the CDCs Best Practices for Comprehensive Tobacco

Control Programs (CDC, 1999) contain information describing how to conduct
tobacco cessation efforts in dental offices. A meta-analysis of 29 studies contained
within the Clinical Practice Guidelines (Fiore et al, 2000) suggests that a variety of
health professionals, including dentists and dental hygienists, can be successful
tobacco cessation counselors for their patients. The number of studies involving
dental settings, however, has been sparse.
A recent large-scale U.S. national survey, with 1,746 dentists responding, suggests
that 33% ask most patients whether they smoke, 66% advised smoking patients to
quit, and 29% provided some tobacco use cessation service (Dolan et al, 1997). A
review of several dental setting-based studies demonstrates that the quit rate has
varied from 2.5% to 44% (Wamakulasuriya, 2002). This variability in the quit rate
is usually dependent on the intensity of the cessation approach.
The smoking cessation intervention most described in the literature, shown to be
effective in the dental office, and yet not requiring excessive time or interruption to
the visit, is termed the Oral Screening Brief Counseling Model, which was adapted
from the 4As (Ask, Advise, Assist, Arrange) of the NCI.

The Oral Screening/Brief Counseling Model consists of the following five steps
(Gordon and Severson, 2001):
1) Check the tobacco use status of all patients
2) Relate oral health findings to tobacco use
3) Urge the patient to quit all tobacco use (includes setting a quit date)
4) Supply self-help materials to patients, (includes nicotine replacement
5) Encourage patients via follow-up.
Dental hygienists are the most appropriate individual in the office to conduct such
programs (Christen, 2001). The consensus from a variety of studies indicates that
that training should involve the entire dental team, that training should take place in
the office setting rather than in large continuing education venues, and that prompts
and chart reminders can be a helpful cue to action (Cohen et al, 1987; Cohen et al,

Meta-analyses and literature reviews also indicate that Nicotine Replacement
Therapies (NRT) can contribute to the success of smoking cessation. In a review of
94 trials, Silagy et al. (1999) concluded that a variety of commercially available
forms of NRT (nicotine gum, transdermal patch, nicotine nasal spray, nicotine
inhaler, and nicotine sublingual tablets) can serve an adjunctive role during
smoking cessation efforts. In fact, the Clinical Practice Guidelines for Treating
Tobacco Use and Dependence support the use of NRT along with counseling to
maximize effectiveness. Some types of NRT (i.e. chewing gum) are available over
the counter, while others can be prescribed by a physician or dentist. Occasionally,
when NRT and counseling have not been effective, other forms of
pharmacotherapy, such as Bupropion (Zyban) can be prescribed by physicians, but
not by dentists.
Similar to tobacco, the use of alcohol contributes to many health disorders and
social problems which may affect both the individual and the community, and is a
major risk factor for oral cancer and potentially malignant lesions. Counseling by
physicians and dentists can increase a users motivation to stop alcohol use but it is
not often applied in a systematic or frequent manner to people presenting with
potential for development of malignant oral lesions (Petti and Scully, 2005).

Adherence to Screening Guidelines
Health care providers do not optimally provide nationally recommended preventive
services for their patients. Despite the fact that approximately three quarters of
Americans visit a physician annually, they receive disease prevention services at
levels lower than recommended by national advisory groups.
Recent emphasis on guideline development was based on the hypothesis that many
healthcare decisions were unsupported by strong evidence, and that unnecessary
testing practices would be discouraged. The concept was that unexplained and
unnecessary variations in practice would decrease. However, this scenario is not
what generally materializes. Instead, when expensive technological diagnostic and
treatment advances are validated as effective, expenditures often increase.
Consequently, in some cases, clinical guidelines have been utilized by insurers for
rationing healthcare services (Browman, 2005).
Despite wide dissemination to physicians, uptake of clinical practice guidelines has
generally been less than was expected, and screening guidelines have produced
limited effects in changing practice patterns and outcomes (Browman, 2005).

Consequently, the gap between prevention guidelines and practice is considerable,
resulting in millions of Americans who fail to obtain screening in accordance with
recommendations (Cohen, et al, 1994). Physicians perform, on the average, only
20 to 60% of the tests and procedures recommended by major consensus groups
(Pommerenke and Weed, 1991). Even when guideline recommendations are
considered to be non-controversial and evidence-based, nearly one third of the time
they are not followed by primary care providers (Grol et al, 1998). Additionally,
much of this data is based upon self- reports by physicians who are likely to
overestimate the degree of adherence with the guidelines (Main et al, 1995). One
study found that the number of cancer screening activities performed, as measured
by chart audits, was only one-tenth to one-half of the physician self-estimates
(McPhee et al, 1989).
Consequently, according to Browman (2005):
The focus is shifting from getting the evidence
straight to research transferidentifying reasons for the
gap between evidence and clinical practiceand designing
interventions to address the gap. To be useful, clinical
practice guidelines need to inform the clinician about
appropriate practice. Recommendations can be made more
accessible within a clinical environment by improving the
format or mode of presentation. They should be better
contextualized, so that recommendations can be fitted to

specific local circumstances, with sufficient flexibility to
allow legitimate differences in interpretation (page 480).
Referring to clinical practice guidelines, Browman (2005) further posits that:
They should also be made more sensitive to the
social culture of medical practice, by fostering a sense of
ownership and accountability in those intended to use them,
resulting from participatory processes (page 480).
In fact, the study of strategies to encourage guideline implementation has become
so important that two conferences were convened solely to discuss this issue (Gross
et al, 2001; Heffner et al, 2000), and an entire 73 page issue of the journal Chest
(introductory article by Heffner, 2000) and a 92-page issue of the journal Medical
Care (introductory article by Gross and Romano, 2001) were devoted to issues of
guideline implementation.
A systematic review of the article that identified barriers to guideline adherence
analyzed 76 papers in the literature (Cabana et al., 1999). Identified as the most
consistent barriers to guideline adherence were: lack of awareness, lack of
familiarity, lack of agreement, lack of self-efficacy, lack of outcome expectancy,
the inertia of private practice, and external barriers.

Improving physician compliance with preventive medicine recommendations may
save thousands of lives each year (Pommerenke and Weed, 1991). Physicians have
contact with at least two-thirds of all smokers annually. Therefore the majority of
the 5 million current adult smokers in the U.S. could potentially be reached by
physicians during the course of on-going medical care. This high patient-physician
contact rate, even if coupled with only a small absolute effect on smoking
prevention could produce substantial changes in the smoking behavior of the
general population (Goldberg et al, 1994).
Although dentists have a prime opportunity to carry out routine oral examinations,
it has been emphasized that physicians bear responsibility in detecting and
preventing oral malignancies since high-risk patients utilize medical services more
often than dental services (Mignogna et al, 2001; Carpenter, 1993).
Zakrzewska (1999) posits that physicians ought to systematically check the oral
mucosa of heavy smokers and drinkers, especially those older than 45 years of age.
According to Smart (1993) physicians find that conducting pelvic examinations and
pap smears is more acceptable than looking in the mouth, although the incidence
and mortality rate for this malignancy is nearly double that of cervical cancer. The

attitudes and practices of physicians concerning oral cancer prevention and
detection have not been adequately investigated (Mignona et al, 2001).
One study found that integrative screening within curative visits to health care
providers is the model most likely to reach at-risk women (Klassen, 2003).
However, a study of 51 patients with newly diagnosed oral or oropharyngeal
squamous cell carcinoma found that the lesions were largely discovered by dentists,
dental hygienists, and oral/ maxillofacial surgeons. Overall, detection of oral and
oropharyngeal squamous cell carcinomas during a non-symptom driven
examination was associated with a lower stage at diagnosis, and this was most
likely to occur in a dental office (Holmes et al, 2003).
Health care providers clearly hold a strategic position in disease prevention and
early detection. While national recommendations for disease prevention and health
promotion have been published, they have not resulted in the necessary adherence
by physicians to those recommendations.
It may be unrealistic to expect unequivocal adherence to guidelines all of the time.
Gross and colleagues (2001) convened a conference in Leeds Castle, England in

1999 to address the issue of how to encourage implementation of best practices for
guideline implementation. As an opening premise to the meeting, they observed
Most practitioners apply the critical parts of
guidelines much of the time but almost never all of the
time. Although all of the time may be too much to
expect, most of the time should be our goal. How to
encourage practitioners to implement best practices most of
the time is the challenge (page 11-86).
Behavior Change Strategies to Encourage Guideline
In an effort to develop strategies for preventing and controlling oral cancer in the
United States, the Centers for Disease Control and Prevention sponsored a national
conference supported by the National Institute of Dental Research of the National
Institutes of Health and the American Dental Association (ADA) in 1996. Inherent
in this conference was an attempt to create cooperative collaborations with
commercial firms. Such a collaboration was formed between the ADA and the
company that developed and marketed an oral brush biopsy product. In 2001, they
collaborated to sponsor a 7-month national oral cancer public awareness campaign
to highlight the importance of the early detection of oral cancer. Eleven U.S. cities

were selected for placement of educational marketing utilizing billboards, bus
shelter signage, commuter bulletins and taxi tops. This was accompanied by a
campaign directed toward dentists with spots in ADA publications, and provision
of in-office materials for patients. The goal was distribution to approximately
100,000 dentists.
Four months after the campaign ended, a study was conducted to assess the
outcomes of the campaign both among the professionals and the public (Stahl et al,
2004). A professional marketing research firm conducted a random survey of
1,270 consumer adults by telephone and the ADA Survey Center surveyed a
random sample of dentists by mail to assess the impact of the oral cancer awareness
Findings of the consumer component found that 61% of respondents were not
aware that dentists look for oral cancer during a routine dental examination, but that
67% felt that they would feel that they were getting the best possible care if they
knew that their dentist was routinely conducted an oral cancer examination.

Among the dentists, 73.2% were aware of the campaign. The majority
(83.4 %) believed that the campaign helped to raise public awareness of oral
cancer. Ten percent more dentists who were aware of the campaign had
discussions with their patients about oral cancer than those who were not aware of
the campaign. In fact, the increased likelihood of informing the patient that an oral
cancer examination was being performed was the most common change in practice
made by the dentists as a result of the campaign. These findings demonstrate a
positive impact of the campaign, although they would have been more noteworthy
if the study design had included an assessment both before and after the campaign.
An extensive literature describes research efforts to evaluate the most effective
means of changing the behavior of health care providers. This information is
primarily found in the medical rather than the dental literature. A partial
explanation of .the reasons that efforts to change the behavior of health
professionals have been met with limited success may be because very few
interventions throughout the years have been based upon conceptual theory-based
models of behavior change. Beginning in the 1990s, various interventions were
developed that were designed to change the behavior of physicians; these met with
varying degrees of success. Cohen (1994) suggested that behavior change

strategies should be derived from the use of theory rather than merely elimination
of barriers. He supported use of the transtheoretical model of readiness for change
in addressing the behaviors of physicians.
Most of the efforts to change physician behavior, and more importantly to impact
health outcomes, have occurred through continuing medical education (CME),
which is broadly defined as all ways by which physician learning and clinical
practice may be altered by educational or persuasive means (Davis et al, 1992). The
objective of CME, commonly configured as the short-course model or
conference, is usually to motivate physicians by changing their knowledge,
attitudes, and beliefs. Such efforts generally have not been effective in eliciting or
sustaining changes in physician behavior, although occasionally they do produce a
minimal behavior change (Cohen et al, 1994). There is little evidence to indicate
that physician performance and patient outcome are improved by the knowledge
gained in traditional CME settings (Haynes et al, 1984). In fact, education alone
seldom leads to lasting change (Davis et al, 1999). Research demonstrates that even
when knowledge of recommendations is the objective of the educational effort,
increased compliance is not often the outcome. (Lomas et al, 1989)

Passive education such as conferences or the publication of clinical practice
guidelines and recommendations has consistently been shown to be ineffective
(Davis et al, 1995). More active strategies to implement guidelines, such as
educational outreach (academic detailing), feedback, reminder systems, and
continuous quality improvement, offer greater promise and have captured the
interest of physicians, health systems, hospitals, managed care plans, and quality
improvement organizations (Chassin and Halvin, 1998).
Although many methods have shown some success on their own, interventions that
rely on more than one method appear to be the most successful (Greco and
Eisenberg, 1993). According to Grol (2001), A multi-faceted approach with
written (scientific journal, support materials) and personal approaches, local
consensus discussions, contact with colleagues, outreach visits by peers seems to be
effective in the dissemination (page 11-46).
It is somewhat difficult to make comparisons of interventions across studies since
there are several thousand studies assessing the outcomes of clinical education of
physicians. Most comprehensive reviews select a certain number of studies to
review which meet certain criteria. Demonstrating the difficulty in drawing

conclusions from the research, one review of 58 studies of strategies for improving
preventive care found that most interventions were effective in some studies but not
in other studies (Hulscher et al, 1999). A similar observation was made by Oxman
et al, (1995) in stating that, Almost all approaches work at least some of the time,
but none works all the time (page 1429). It is generally regarded that a
multifaceted approach which combines more than one motivational technique,
rather than the traditional one-shot lecture-style educational approach, is most
productive in motivating physicians, teaching them new skills, and helping them to
change their practice environments (Cohen et al, 1994; Gross and Pujat, 2001;
Cretin et al, 2001; Hulscher et al, 1999).
One systematic review of the effect of continuing medical education strategies
reviewed 99 trials that met two criteria: 1) randomized controlled trials of
educational strategies or interventions that objectively assessed physician
performance and/or health care outcomes; and 2) 50% of subjects being either
physicians or medical residents (Davis et al, 1995). The intervention strategies
included (alone or in combination) educational materials, formal CME activities,
outreach visits such as academic detailing, use of opinion leaders, patient-mediated
strategies, audit with feedback, and reminders. Of the 81 single intervention

strategies, 60% demonstrated change in at least one major outcome measure. Of 39
two-method interventions, 64% were positive. Of 39 multifaceted interventions,
79% were positive. The interventions in these studies targeted two broad domains
of behavior change: roughly 1/3 were in the area of health promotion and disease
prevention and 2/3 were in clinical disease management.
Hulscher and colleagues (1999) suggest that more detailed studies with better
research methodology are needed to show which elements of interventions work,
why they work, and at what cost. It has also been suggested that:
Cumulative progress in designing and refining
successful interventions will be difficult to achieve until
researchers begin looking inside the black box, measuring
and analyzing the provider's beliefs, attitudes, reactions and
judgments in some detail (Kanouse et al, 1995, page 190).
Grimshaw et al (2001) systematically reviewed professional behavior change
interventions published between 1996 and 1998. As a result of analyzing 41
reviews, they concluded:
Passive approaches are generally ineffective and
unlikely to result in behavior change. Most other
interventions are effective under some circumstances; none
are effective under all. Promising approaches include
educational outreach (for prescribing) and reminders.
Multi-faceted interventions targeting different barriers to

change are more likely to be effective than single
interventions (page II-2).
A comprehensive review was also conducted by the Cochrane Effective Practice
and Organization of Care Review Group in an effort to evaluate how well research
findings are being translated into practice (Bero et al, 1998). They found that the
passive dissemination of information (for example, the publication of consensus
conferences in professional journals or the mailing of educational materials) was
generally ineffective and, at best, results only in small changes in practice.
However, ironically, these passive approaches probably represent the most
common behavior change approaches utilized by researchers, professional bodies,
and health care organizations. The Cochrane Group classified behavioral change
interventions among health professionals into three categories. Consistently
effective interventions included:
Educational outreach visits
Reminders (manual or computerized)
Multifaceted interventions (a combination that includes two or more of the
following audit and feedback, reminders, local consensus processes, or
Interactive educational meetings (participation of healthcare providers in
workshops that included discussion or practice)

Interventions of variable effectiveness included:
Audit and feedback (or any summary of clinical performance)
The use of local opinion leaders (practitioners identified by their colleagues
as influential)
Local consensus processes (inclusion of participating practitioners in
discussions to ensure that they agree that the chosen clinical problem is
important and the approach to managing the problem is appropriate)
Patient mediated interventions (any intervention aimed at changing the
performance of healthcare providers for which specific information was
sought from or given to patients)
Interventions that had little or no effect included:
Educational materials (distribution of recommendations for clinical care,
including clinical practice guidelines, audiovisual materials, and electronic
Didactic educational meeting (such as lectures)
Bero and colleagues of the Cochrane Group state that "it is striking how little is
known about the effectiveness and cost effectiveness of interventions that aim to

change the practice or delivery of health care (page 467). In summarizing their
findings, they further state that:
Given the paucity of evidence it is vital that
dissemination and implementation activities should be
rigorously evaluated whenever possible. Studies evaluating
a single intervention provide little new information about
the relative effectiveness and cost effectiveness of different,
interventions in different settings. Greater emphasis should
be given to conducting studies that evaluate two or more
interventions in a specific setting or help clarify the
circumstances that are likely to modify the effectiveness of
an intervention (page 467).
Finally, Bero and colleagues also point out, as has been emphasized previously in
this literature review, that few of the systematic reviews that have been done have
explicitly attempted to link the findings to theories of behavior change.
Change Theory
Methods and strategies for guideline implementation can be better understood and
applied in the context of change theory which provides a useful perspective in
changing provider behavior in specific clinical situations. In assessing theory and
strategy, the characteristics of the target group, the practitioners perceptions, the
health care setting, the desired behavior change, and the identified barriers to

implementation all need to be considered (Grimshaw and Russell, 1993; Davis et al,
1995; Palmer and Hargraves, 1996; Greco and Eisenberg, 1993; Soumerai et al,
1998; Wensing and Grol, 1994).
In their paper entitled Optimal Methods for Guideline Implementation, in which
they summarized the conclusions from the Leeds Castle conference on guideline
implementation, Gross and colleagues (2001) described a variety of theories of
facilitating change. There is probably not a single, superior theory-based strategy.
Different strategies will be effective under different conditions, and much of the
time a combination of theoretical approaches will be indicated.
Of all interventions, reviews have consistently demonstrated that reminders show
the best evidence of consistent effectiveness (Haynes and Walker, 1987; Johnston et
al, 1994; Shea et al, 1996). However, Smith (2000) states:
Trials have not yet measured some aspects of when
reminders work, including when physicians do not agree
with what theyre being reminded to do. And few studies
evaluate how long a behavior response lasts after the
reminder stimulus has been discontinued (pagel5S).
Cook and colleagues (1997) in analyzing the relation between systematic reviews
and practice guidelines conclude that guideline development and implementation

methods should be theory-driven and evidence-based. And Smith (2000) provides
a reminder that there is no one easy answer to changing professional behavior,
In the language of clinical medicine we must
diagnose the lesion (why change is not adopted) before
prescribing therapy (a change strategy). In practical
implementations of physician performance improvement,
multiple tools will likely be necessary and should be
chosen carefully (page!6S).
Finally, Grol and Grimshaw (2003), who have done much of the research and
writing in this arena, conclude their article describing how to get from best
evidence to best practice with an almost tongue-in-cheek explanation of the
immensity of the task. They write,
If you would like to start tomorrow to change
practice and implement evidence, prepare well: Involve the
relevant people, develop a proposal for change that is
evidence based, feasible, and attractive; study the main
difficulties in achieving the change, and select a set of
strategies and measures at different levels linked to that
problem; of course, within your budget and possibilities.
Define indicators for measurement of success and monitor
progress continuously or at regular intervals. And, finally,
enjoy working on making patients care more effective,
efficient, safe, and friendly (pagel229).

Social Cognitive Theory
A theoretical construct that seems to correlate with the myriad of factors that
impact oral cancer screening practices in the dental setting is the Social Cognitive
Theory of Albert Bandura (1986). In basic terms, this theory purports that human
action can be understood through the triadic interaction of: (1) personal factors
(including cognitions), (2) environmental influences, and (3) behavior. The
personal factors as described by Bandura (1997) include the individuals capacity
to symbolize behavior, to anticipate the outcomes of behavior; to leam by
observing others, to have confidence in performing a behavior (including
overcoming the problems in performing the behavior), to self-determine or self-
regulate behavior, and to reflect on and analyze experience. The environment
refers to those objective factors which are physically external to the person, but can
affect their behavior. The behavioral component is sometimes referred to as
behavioral capability and involves acquisition of the knowledge and skill to
perform a given behavior, often through use of skills training. In essence, the
Social Cognitive Theory amalgamates the earlier notion that behavior change was
created solely by cognition, behavior, or environment. The interplay and
interaction of these became known as the concept of reciprocal determinism within

the Social Cognitive Theory, in which environment, person, and behavior all
interact and influence one another and the eventual outcome.
The Social Cognitive Theory (Bandura, 1986) evolved from social learning theory
(Bandura (1962). The constructs of social learning theory contrasted with the
prevailing operant learning theory of that time, which suggested that learning
occurred only if associated with an immediate reward. It was also previously
postulated that the principal causes of behavior evolved from forces or impulses
within the individual. Such internal motivators, could not, however, explain the
marked variation in behaviors exhibited by the same individual in different
situations, toward different people, or at different times.
Social learning theory purported that children, for example, could learn a new
behavior by observing the behavior of others (modeling) and similarly by observing
the rewards that others received (vicarious reinforcement). As explained by
Bandura (1986),
The capacity to learn by observation enables people
to acquire large, integrated patterns of behavior without
having to form them gradually by tedious trial and error
(page 12).

The development of Banduras cognitive concept of self-efficacy, defined as a
persons confidence in performing a particular behavior and uncovering barriers to
that behavior (Bandura 1977a, 1977b), was formative for the later development of
Social Cognitive Theory. Self-efficacy remained an important construct within
Social Cognitive Theory, and supported the idea that the individual functions as an
agent in control of his/her own life. As stated by Bandura (1986),
In the social cognitive model of interactive agency,
persons are neither autonomous agents nor mechanical
conveyers of animating environmental forces. Rather, they
serve as a reciprocally contributing influence to their own
motivation and behavior within a system of reciprocal
causation involving personal determinants, action and
environmental factors. These sets of determinants affect
each other bi-directionally rather than uni-directionally
(page 12).
The relative influence exerted by each of the three interacting factors varies
according to the circumstances. When situational constraints are weak, the
personal factors may predominate. When cognitive beliefs are very powerful, they
may over-ride personal and environmental effects.

The nature of individual behavior is defined within five basic capabilities (Bandura,
1) Symbolizing Capability with the use of symbols, transient experiences
are processed and transformed into internal models that guide them through
future action. It is through symbols that meaning and form is ascribed to
past experiences. The combination of cognition coupled with
symbolization enables people to create ideas that transcend their sensory
2) Forethought Capability most behavior is not simply a reaction to the
immediate environment nor an exact replication of the past. Most behavior,
being purposive, is regulated by forethought. As a result of forethought,
individuals motivate themselves and anticipate and guide their actions.
Forethought is reflective and is translated into action with the assistance of
self-regulating mechanisms.
3) Vicarious Capability learning that results from direct experience occurs
vicariously by observing other peoples behavior and its consequences.

Learning by observation allows people to generate and regulate behavioral
patterns without the requirement of tedious trial and error.
4) Self-Regulatory Capability rather than behaving just to suit the
preferences of others, behavior is motivated by and regulated by internal
standards and self-evaluative reactions to ones own actions. Discrepancies
between a performance and the standard against which it is measured
activate evaluative self-reactions, which influence subsequent behavior. By
arranging facilitative environmental conditions, recruiting cognitive guides,
and creating incentives for their own efforts, causal contributions to ones
own motivation and actions occur. Consequently, external influences, as
well as self-influence govern ones behaviors.
5) Self-Reflective Capability reflective self-consciousness allows
individuals to analyze their experiences and to think about their own
thought processes. In gaining understanding through reflection, thinking
can be evaluated and altered. A type of thought that is central to ones
actions is the perception of their capabilities to deal effectively with
different realities. Self-efficacy perceptions determine both how invested

one becomes in an activity and the attitude that is brought to a challenging
In producing socio-cultural change, Bandura (1985) suggests that instructional and
motivational factors are required to overcome the initially unfavorable conditions
associated with adopting new ways. Such a diffusion process has four phases:
selecting an optimal setting for introducing innovations; creating the necessary
preconditions for change; implementing a demonstrably effective program; and
dispersing the innovations to other areas through the aid of successful examples.
Innovations are best introduced in settings where members are willing to try them
on a provisional basis. The preconditions for change are created by increasing
ones awareness and knowledge of the innovations. They need to be provided with
information about the purpose of the new practices, the relative advantages, and
how adopting them is likely to affect ones life.
Persuasion alone is not enough to promote adoptive behavior. To ensure social
change, one must also create optimal conditions for learning new ways to provide
positive incentives for adoption, and build supports into the social system to sustain

Modeling behavior change through guided mastery involves three facets, modeling,
guided enactment, and self-directed application of acquired skills (Goldstein, 1973;
Rosenthal and Bandura, 1978). Complex skills must be broken down into
constituent subskills and organized hierarchically to ensure optimal progress in
learning. The activities are then modeled in easily mastered steps. Videotaped
modeling and other symbolic media serve as convenient aids to actual
demonstrations in teaching new practices (Bandura, 1986).
Additionally, diffusion theory should be considered while designing tailored
interventions. Diffusion theory is attributed to Rogers, who in 1962 described the
science inherent in adoption of a new idea (Rogers, 1995). Failure to tailor
information about the innovation to the particular desires and cognitive capabilities
of would-be adopters can hamper a diffusion program at the outset (Rogers and
Adhikarya 1979).

The objective of this doctoral dissertation project was to improve the oral cancer
screening behaviors of dentists and dental hygienists in office-based practice
settings in order to increase compliance with American Cancer Society oral cancer
screening recommendations.
Specific Aims
The Specific Aims of the proj ect were:
1. To assess the practice milieu of selected dental office settings to determine
the needs and barriers which impact adherence to oral cancer screening
2. To identify, develop, and implement tailored, multi-faceted, theory-driven
intervention strategies to encourage oral cancer screening examinations and
discussions of tobacco use.

3. To evaluate process and outcomes (comparing an intervention with a
control group) to determine whether this approach has merit for replication
in other settings or on a larger scale.
The hypothesis of this project is that implementation of a tailored intervention will
change the oral cancer screening behaviors of dentists and dental hygienists in
dental office-based practice settings to better comply with the oral cancer screening
recommendations of the American Cancer Society.
The study hypothesis can be expressed as Ho: Ho=^i versus Hi: Ho the mean change in the proportion of patients in the intervention group who report
being screened, contrasted with the proportion of patients in the control group. A
power analysis suggests that a minimum of 216 subjects is necessary (across
periods) to detect an effect size of 0.25 at a 0.05 significance level. This means that
if the minimum N is met, the test statistic will accurately detect Hi 95% of the time
when there is a difference of at least 0.25 standard deviations between Ho and /X;.

Research Design
This study was a Demonstration Project, i.e., a project of limited scope and
duration, to carry out and evaluate a new method, new target, and/or new setting to
see if it later merits testing or replication elsewhere on a larger scale. The project
entailed a two-stage, multi-method demonstration project with an intervention (i.e.
treatment) group and a control group. Phase One was comprised of a qualitative
component to assess the personal and professional barriers faced by dentists and
dental hygienists that impact compliance with oral cancer screening guidelines.
This assessment informed the design and tailoring of a theory-based intervention
package to increase oral cancer screening in dental offices. Phase Two was the
implementation and evaluation of the intervention in dental practices, compared
with control sites.
The study was designed based on findings from the initial phase. Patient
satisfaction surveys with embedded oral cancer and risk behavior screening
questions were mailed 2 months post-appointment. New patients of the dentists
and all new and returning patients of the dental hygienists that were seen within a
3-week time period in all practices were surveyed. The multi-faceted, tailored,

theory-driven intervention package was implemented only in the treatment
practices. Two months after the intervention was implemented, the same patient
satisfaction surveys with embedded oral cancer and risk behavior screening
questions were mailed to new patients of the dentists and all new and returning
patients of the dental hygienists seen within a different 3-week period. Figures 3.1
and 3.2 summarize the research design.
Grouo N Practice Sites N Patients 1 survev N Patients 2"d survey Intervention Elements
1. Treatment group 3 New patients seen by the dentists. New and recall patients seen by dental hygienists. 3 week time period New patients seen by the dentists. New and recall patients seen by dental hygienists. 3 week time period Impatient survey On-site consensus meeting Interactive workshop on head and neck oral cancer screening examination technique In-office patient chart screening forms In-office posters, brochures 2nd patient survey
2. Control group 3 New patients seen by the dentists. New and recall patients seen by dental hygienists. 3 week time period New patients seen by the dentists. New and recall patients seen by dental hygienists. 3 week time period 1st patient survey 2nd patient survey
TOTAL 6 288 361
Figure 3.1 Phase Two Study Design

Phase One
Formative Component (June-July 2004)
A written questionnaire immediately followed by a personal interview was
conducted with 15 dentists and 15 dental hygienists to gain an understanding of the
contextual factors influencing oral cancer screening and counseling performance
(Appendix A). The dentists and dental hygienists were not necessarily employed
within the same practices.
Each practitioner was interviewed privately to encourage candid responses and
eliminate response bias which could have ensued from answering in the presence of
peers. Respondents initially answered the questions in a written format; the
researcher/interviewer then personally interviewed each individual to further
amplify the responses using the semi-structured interview format. Each interview
session lasted approximately 30 minutes.
Knowledge, attitudes, and beliefs concerning oral cancer and oral cancer screening
issues, as well as barriers and needs that impact oral cancer screening by oral health

professionals, were explored. The Social Cognitive Theory of Bandura served as
the framework for constructing the questions in the categories of environmental,
cognitive, and behavioral factors, as designated within the theory.
The offices where these individuals were employed did not serve as either control
or experimental sites in the Phase Two quantitative portion of the project, to avoid
the interview serving as an unintentional intervention. Formative data were
transcribed and coded. Responses were analyzed thematically using the Social
Cognitive Theory as a framework. Responses were also enumerated whenever
possible using content analysis (Tables 1-4).
Treatment and Control Office Recruitment:
(June-Oct. 2004).
A list of dentists who registered for a continuing education course on Oral
Pathology was obtained from the director of continuing education at the University
of Colorado School of Dentistry. The course was held in March of 2004 at the
School of Dentistry. Dentists who attended were contacted by letter, follow-up
phone call, and, for those interested, a personal visit until 6 dentists agreed to have
their office participate in the study. When the list had been exhausted, with no

willing participants, the list was broadened to include the offices of dental
hygienists attending a continuing education course on oral cancer held at the annual
session of the Colorado Dental Association in June of 2004. When this approach
still did not result in agreement, personal contacts combined with cold calling from
the telephone directory eventually produced a total of 6 dental practices willing to
participate in the study. Three of the practices were randomly designated as the
experimental sites and 3 as the control sites. The control sites did not receive any
intervention elements.
Phase Two
Intervention (Oct-Nov. 2005)
A multi-method interventional approach was selected, since the superiority of this
approach has been repeatedly demonstrated in the literature. The intervention,
using the elements of the Social Cognitive Theory and the findings of the
qualitative component of the project as a guide, included the following core
components: a local consensus process to facilitate team-building and buy-in to the
project, including discussion about how the interventions could best be applied

within the office context, and multi-modal reminders for patients and practitioners.
The in-office interventions were also designed to have the added benefit of
facilitating communication concerning tobacco risk factors between clinician and
The professional reminder was a form prepared for the patient chart for recording
oral cancer screening findings (APPENDIX B). The patient reminders were
examination room posters and reception room brochures which also served as
visual cues to practitioners and patients about the importance of oral cancer
screening and issues around tobacco use. The poster (APPENDIX C) and brochure
(APPENDIX D) were designed by the researcher and developed and produced by a
graphic artist.
The interactive educational workshop for the clinicians consisted of a power-point
presentation/discussion as well as hands-on practice performing the oral cancer
screening examinations. Videotaped testimonials of oral cancer survivors were
viewed. Participants practiced performing extra-oral (head and neck) and intra-oral
examinations on each other until they determined that they felt comfortable and
competent performing the examinations. An extensive take-home packet of

reading materials/manuals was also provided. More specifically, the intervention
package was designed within the Social Cognitive Theory in the following ways:
Personal/Cognitive: 1) a meeting of the dentist(s) and dental hygienist(s) from
each intervention office to discuss the project and to work toward consensus and
voluntary participation, 2) a VA hour multi-media presentation consisting of a
power-point enhanced presentation that covered the statistical relevance of oral
cancer, risk factors with special emphasis on tobacco and alcohol use, demographic
and cultural components, lesion review, biopsy and sampling procedures, a step-by-
step explanation of how to perform a thorough extra- (head and neck) and intra-
(within the mouth/throat) oral cancer screening and 2 taped segments of
testimonials of oral cancer survivors, 3) provision of extensive reading materials
for each practitioner. Behavioral: 4) a 1-hour session for practicing the technique
with a staff partner until a self-determined comfort level was verbalized,
Environmental: 5) production/distribution of professionally prepared posters and
brochures to educate/remind patients and practitioners about oral cancer screening,
6) provision of a chart form for recording oral lesions, 7) consensus building with
all office staff concerning the importance of oral cancer screening.

Patient Surveys
All adults over 18 who visited the office for a new patient examination or a recall
(cleaning) visit during the each of the 3-week data collection time periods were
eligible to participate, and were sent a mailed survey. It would have been too
cumbersome to have asked the office staff to designate the ages of all patients in
order to survey only those 40 years of age and over, as recommended for oral
cancer screenings by the American Cancer Society. Instead, they eliminated
known child patients from the mailing list.
To assess patient perceptions of having received an oral cancer screening, a mailed
written survey was sent to all patients with appointments during a pre-determined
3-week time period. The questionnaire contained approximately 42 questions

categorized as follows (number of questions is approximate because practices could
include patient satisfaction questions of their choosing):
24 questions (approximate) practice satisfaction issues
8 questions receipt of oral cancer examination and risk behavior
6 questions assessment of past and current smoking and drinking
4 questions demographics
For the first mailing, patients from both control and treatment offices who were
seen in their respective offices within the same 3 week time period (Nov 1-19,
2004) were sent a survey 2 months following their visit to the dental office. These
surveys were mailed January 12, 2005. The 3-week period (Nov 1-19) was later in
the calendar year than was ideal. However the time required for acquiring 6
participating offices was much longer than anticipated. Since it took 4 months
(June-Oct.) to acquire the 6 participating offices, and because grant monies needed
to be spent within one year, it was essential to get the initial mailing out as soon as
the 6 offices were on board. Since the 3-week time period fell just before the major
winter holiday season, it was deemed risky to send out the surveys during the
holiday mail rush and expect an adequate return rate. In order to encourage a

suitable response rate, the surveys were sent out just after the first of the year when
normalcy has been restored to the mail system and to prospective respondents
lives after the holidays.
The consensus building session and interventions for the treatment offices occurred
February 21-23, 2005. In order to measure longer-term, rather than immediate
retention and application of the material presented in the interventional workshop,
the follow-up surveys were sent to patients seen 2 months following the
intervention. These surveys were mailed to all patients with appointments in both
control and treatment offices during a 3-week period of time (April 4-22, 2005).
For consistency with the initial mailing, these were mailed 2 months after the
appointment (June 13, 2005). The patient surveys were professionally formatted
and printed to obtain an aesthetically pleasing Took.
The surveys were sent with a cover letter on the dentists letterhead and signed by
both the researcher and the dentist, explaining that the office is participating in a
University study. Addressed envelopes were generated at the offices so that patient
address lists were not provided to the researcher in order to obviate the HIPAA
issues of patient identifiers. A staff member from each office was solicited to

access the patient identification information from the computer and facilitate
printing the envelopes. This staff person was remunerated for their time.
The mailing consisted of a cover letter (APPENDIX E) and a general patient
satisfaction survey (APPENDIX F) with each office having the opportunity to
include additional pertinent questions of their choosing which pertained specifically
to the office and not to oral cancer screening. The oral cancer screening questions
were embedded within the rest of the survey. Data were collected regarding the
following dependent variables: 1) patients perception of having received an oral
cancer screening, 2) whether the practitioner examined the neck/throat/tongue, 3)
whether the patient was told they were being screened, 4) using/used tobacco, 5)
using/used alcohol, and 6) age of over 40 years.
By composing the survey with the inclusion of patient satisfaction questions, it was
easier to get buy-in from the dentists since they would gain information about
their patients satisfaction levels with the practice. The intent of embedding the
oral cancer screening questions within the patient satisfaction survey was to
encourage the patients to answer the questions honestly, and without concern about
incriminating their dentist and dental hygienist practitioners. In an additional effort

to encourage candid, valid responses, patients were provided a stamped envelope
addressed to the University investigator, rather than to their dental office, with the
assurance that the respondents would not be identifiable to the practitioners.
Figure 3.1 Phase Two Study Design At the completion of the project, the control
offices received a large food/ffuit basket to be shared by the office staff. The
individual dentists and dental hygienists who participated in the treatment offices
each received a $100 restaurant gift certificate.

NOV 04
DEC 04
JAN 05
FEB 05
MAR 05
APR 05
MAY 05
(Nov 1-19)
(Jan 12) 0O
(Feb 21-23)
(Apr 4-22)
(Jun 13)
Figure 3.2 Phase Two Time-line
The time-line for the Phase Two, the quantitative aspect of the study, is depicted in
Figure 3.2 above.

Phase Three
Data entry (July-Aug. 2005)
Data from the qualitative interviews conducted with the 15 dentists and 15 dental
hygienists were transcribed, coded and analyzed. Classical Content Analysis
(Carley, 1990:726; Patton, 2002:250) was used to enumerate responses for the
open-ended questions. With Classical Content Analysis, categories are established
for the data and then the number of times the category appears is ascertained. For
thematic analysis, Analytical Deductive Coding was used in which codes from
theory/ literature are established prior to the data analysis, and the text searched for
the presence of the codes (Patton, 2002:494; Taylor and Bogdan, 1984:127). The
direct multiple-choice responses were tabulated mathematically for the number and
percentage of respondents answering each question with each choice (Appendix G).
Analysis (Sept. 2005)
Using SPSS software, descriptive and inferential statistics were employed to assess
the quantitative data. To evaluate the screening outcomes, data from the first and

second administration of the surveys were compared. Descriptive statistics, z tests
of independent proportions (Ferguson, 1976), and tetrachoric correlations (Roscoe,
1969) were used to analyze the relationships between the variables.
For the purposes of this report, data were combined for the two types of
practitioner, and dependent variables were condensed to: 1)1 received an oral
cancer screening examination; 2) The practitioner felt the outside of my head/neck
with his/her fingers/hands and/or pulled my tongue out with a piece of gauze and
felt around my tongue and inside my mouth; 3) I was told that I was being screened
for oral cancer. The independent variables were: 1) Control/Treatment group; 2)
Ever/never tobacco user; 3) Ever/never weekly alcohol use; 4) Age at less
than/greater than or equal to 40 years.
The overall purpose of this demonstration project was to compare the intervention
group over time to determine if this intervention approach has merit for replication
in additional settings or on a larger scale. The control group was also compared
over time as a validation for the effects of the intervention.

Qualitative Component
The data from the Phase One questionnaire and follow-up interview conducted with
15 dentists and 15 dental hygienists that could be enumerated with content analysis
are provided in Tables 1-4. Utilizing the Social Cognitive theory as a theoretical
construct, respondents were asked about environmental, behavioral and cognitive
factors related to providing oral cancer screening examinations.
Table 1 Demographics
Dentists n=15 Dental Hveienists n=15
Age Range 27-62 26-54
Mode 53 44
Median 42 45
Gender 8m/7f Om/ 15 f
Years in practice 19 Years 18.6 Years
Ethnicity 15 Caucasian 15 Caucasian 2 Hispanic

Immediately following administration of the written questionnaire, follow-up
interviews were completed with 15 dentists and 15 dental hygienists. There were
nearly equal numbers of female (7) and male (8) dentists, with all dental hygienists
(15) being female. The ages of the dentists ranged from 35-62, while the age range
of the dental hygienists was from 26-58. The median age for the two groups of
practitioners was similar at 42 for the dentists and 45 for the dental hygienists.
Years in practice for the dentists ranged from a low of 2 to a high of 30 (mode=19).
For the dental hygienists this range was from 6 months to 37 years (mode=18.5).
This was predominantly a mature and experienced group of respondents.
Table 2 Environmental Factors
Dentists n=15 Dental Hveienists n=15
(n - %) (n - %) (n - %) (n - %)
OC examination barriers? 12 - 80% 3 - 20% 11 - 73% 4 - 17%
Is time a barrier? 11 - 73% 4 - 17% 9 - 60% 6 - 40%
Medical form: Alcohol questions? 12 - 80% 3 - 20% 14 - 93% 1 - 7%
Medical form: Tobacco questions? 3 - 20% 12 - 80% 10 - 67% 5 - 33%
OC examination policy? 1 - 7% 14 - 93% 7 - 47% 8 - 53%

Environmental factors
Most of the dentists and about 2/3 of the dental hygienists believed that their office
did have a policy about querying patients concerning their smoking habits. In
contrast, most of both practitioner groups did not believe that there was a policy in
place concerning alcohol usage; only 3 dentists and 1 dental hygienist believed that
there was such a policy in place. While nearly all the dentists (14/15) believed that
there was an office policy about conducting an oral cancer examination, the dental
hygienists were more divided on this issue, with 8 of the 15 believing that there
was a policy in place.
When asked if there were barriers to performing the oral/head & neck cancer
screening examination, most dentists (12/15), and most dental hygienists (11/15)
did not feel that there were barriers. The next question posed was, When asked
about office-based barriers, people sometimes say that not having enough time is a
factor; do you agree? Most dentists (11/15) and most dental hygienists (9/15) did
not agree that lack of time during the appointment was a barrier. When asked to
eliminate the factor of time, and encouraged to consider what other office-based or
personal factors might impact routinely performing the head and neck/intraoral

cancer screening examination, the dental hygiene respondents were more easily
able to identify behavioral factors that were office-based, than personal.
Environmental/office-based responses that they were able to identify included:
having continuing education information about doing the examination, setting aside
a designated time to do it, not knowing whether the other dental hygienists in the
office are doing it, not being trained enough about what to look for, uneducated
practitioners, mentioning things to patients that they are uncomfortable about.
The responses of the dentists were similar, with a greater focus on behavioral rather
than environmental/office-based factors being identified. Office-based barriers
included: making it a priority, changing the routine of practice, not making it a
routine and lack of proper training.
When the practitioners were asked how they believed that patients would react to
having the examination performed, the consistent themes were: favorably,
accepting, and appreciative. When asked to identify which environmental,
cognitive/personal or behavioral characteristics would need to change so that
routine head and neck/oral cancer screening examinations could take place, the
practitioners were not able to directly differentiate between these 3 categories of the

Social Cognitive Theory, but instead expressed the consistent theme that the main
thing that would need to change was making it a habit in their routine. This theme
would be categorized as a mostly behavioral, rather than an environmental or
cognitive barrier.

Table 3 Behavioral Factors
Dentists n=15 Dental Hveienists n=15
NO (n %) YES (n %) NO (n %) YES (n %)
Ever found a suspicious lesion? 0 0% 15 100% 1 7% 14 93%
Reluctant to perform OC exam? 14 93% 1 7% 12 80% 3 20%
Comfortable discussing tobacco use? 0 0% 15 100% 0 0% 15 100%
Comfortable discussion alcohol use? 12 80% 3 20% 13 87% 2 13%
Reluctant to perform screening? 14 93% 1 7% 12 80% 3 20%
How prepared before graduation?
Very prepared 8 - 62% 4 - 17%
Somewhat prepared 3 - 23% 8 - 53%
Unprepared 2 - 15% 3 - 20%
How prepared now?
Very prepared 8 - 62% 6 - 43%
Somewhat prepared 5 - 38% 7 - 50%
Unprepared 0 - 0% 1 - 7%
Behavioral Factors
All of the dentists and dental hygienists said that they personally felt comfortable
discussing tobacco use with the patients. However, when discussing alcohol use,
only 2 of the dentists and 3 of the dental hygienists expressed a comfort level.

Behavioral responses offered by both groups of providers included: not feeling
confident or comfortable with performing the examination, how I respond to the
patients personality, apathy, not realizing the importance, not in the habit, things
that break my routine since I save the examination for the end of the appointment.
Behavioral factors stated by the dentists included: motivation, commitment to
comprehensive care, apathy, getting side-tracked discussing dental treatment needs,
education of practitioners, comfort level, need to have it as a routine and habit,
organization and putting the exam in your routine, discipline, laziness, and more
training needed. Consistent themes that emerged concerning the barriers to
performing the examination were: lack of proper training, making it a routine, and
not thinking about the importance of the examination.
Both groups were asked how prepared they felt upon graduation to perform the
head and neck/oral cancer screening examination. Only 8 of 13 dentists who
responded 4/15 the dental hygienists responded that they felt very prepared to do so
at that time. When asked how prepared they felt at this point in their career, 8 of 13
dentists who responded and 6 of 14 dental hygienists who responded that they felt
very prepared. Current reluctance to performing the head and neck/oral cancer
screening was expressed by 1/15 of the dentists and 3/15 of the dental hygienists.

When asked why they feel reluctant, the answers of the 3 dental hygienists were: a
combination of needing to be refreshed on what Im doing and making it a habit,
not used to doing them even though it is quick and easy, only uncomfortable with
the extra-oral exam. The one dentist who was reluctant also expressed that it is so
only with the extra-oral (head and neck) examination, but not the intra-oral
A question that received a unanimous yes response from all 15 dentists and 15
dental hygienists asked whether when examining patients, they had ever found
something that they were suspicious about being oral cancer. The question was
posed in a generic manner without asking or assuming that they purposefully or
regularly conduct head and neck/mouth examinations. That is, the practitioner
could have noticed a suspicious area while examining for decay or periodontal

Table 4 Cognitive Factors
Dentists Dental Hveienists
n=15 n=15
(n %) (n %)
Know 3 % of cancers are oral cancers in the United States? 4 17% 4 17%
Know 50% survive 5 years? 4 17% 3 20%
Know the OC screening guidelines of the American Cancer Society? 1 7% 0 0%
Know where to refer for smoking cessation? 9 60% 9 60%
How likely to ever detect OC?
Very likely 5 33% 5 33%
Somewhat likely 6 40% 5 33%
Somewhat unlikely 1 7% 5 33%
Very unlikely 1 7% 0 0%
Cognitive Factors
The cognitive factors indicated that neither professional group was very
knowledgeable about the prevalence and mortality of oral cancer in the United
States. When asked in a multiple choice format, only lA of both groups knew that
3% of the cancers in the United States are oral cancers. Each of the other choices,
10%, 25%, 40%, and 60%, were selected about evenly by some dentists and dental
hygienists. When asked about the 5-year survival rate, just 25% of the dentists and
20% of the dental hygienists knew that the survival rate is 50%. All of the other