Citation
The role of mental health in HIV prevention with Latinas

Material Information

Title:
The role of mental health in HIV prevention with Latinas
Creator:
Granato, Hollie Fay
Publication Date:
Language:
English
Physical Description:
vi, 41 leaves : ; 28 cm

Subjects

Subjects / Keywords:
HIV infections -- Prevention ( lcsh )
Hispanic American women -- Mental health ( lcsh )
Hispanic American women -- Mental health ( fast )
HIV infections -- Prevention ( fast )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Bibliography:
Includes bibliographical references (leaves 39-41).
General Note:
Department of Psychology
Statement of Responsibility:
by Hollie Fay Granato.

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:
464224323 ( OCLC )
ocn464224323
Classification:
LD1193.L645 2009m G72 ( lcc )

Full Text
THE ROLE OF MENTAL HEALTH IN HIV PREVENTION WITH LATINAS
by
Hollie Fay Granato
B.A., Rollins College, 2007
A thesis submitted to the
University of Colorado Denver
in partial fulfillment
of the requirements for the degree of
Master of Arts
Psychology
2009


This thesis for the Psychology Master of Arts
degree by
Hollie Granato
has been approved
by
Eric Benotsch
Date
2


Granato, Hollie A. (M.A., Clinical Psychology)
The Role of Mental Health in HIV Prevention with Latinas
Thesis directed by Associate Professor Elizabeth Allen, Ph.D.
ABSTRACT
In the United States, women account for approximately one third of all new HIV infections
(CDC, 2008). Latinas specifically are one group that has been particularly affected by the
AIDS epidemic. Research has shown that Latinas account for 16% of all new HIV
infections, and that the AIDS rate among Latinas is five times higher than it is for white
women (CDC, 2008). This study will investigate the effectiveness of a peer administered,
community based HIV prevention program designed for Latinas. This prevention program
specifically targeted the unique risk factors facing this population and general factors of
mental health. The primary question to be explored is whether substance use, depression
and anxiety link to sexual risk behaviors for these women, and whether the intervention can
have an impact in these areas. Specifically, this study will examine whether (1) anxiety,
depression, drug use, and alcohol use are linked to safer sex practices with a primary
partner (i.e., condom use, and drug and alcohol use prior to intercourse) prior to the
intervention; (2) the intervention shows effects on increasing safer sex practices with the
primary partner, increasing knowledge about HIV and AIDS, and decreasing drug and
alcohol use; (3) a history of mental health moderates the effects of the intervention; and (4)
examine the relationship between changes in drug and alcohol use, changes in HIV
knowledge, and changes in safer sex practices with the primary partner over the course of
the intervention.
This abstract accurately represents the content of the candidates thesis. I recommend its
publication.
Signed.
Elizabeth Allen
3


TABLE OF CONTENTS
Figures ..........................................................v
Tables ..........................................................vi
CHAPTER
1. INTRODUCTION......................................................1
Purpose.....................................................2
Research Questions..........................................2
2. LITERATURE REVIEW.................................................5
Anxiety, Depression, and Sexual Risk........................5
Substance Use and Sexual Risk...............................8
Latinas.....................................................9
HIV Prevention with Latinas................................10
3. METHODS .........................................................14
Participants...............................................14
Measures...................................................14
Data Collection............................................23
4. RESULTS .........................................................24
5. DISCUSSION.......................................................30
WORKS CITED ..........................................................39
IV


FIGURES
Figure
1 INTERACTION EFFECT FOR HYPOTHESIS 3
v


TABLES
Table
1 HIV KNOWLEDGE SUBSCALE..................................16
2 SEXUAL RISK VARIABLES...................................18
3 ANXIETY OR DEPRESSION CRITERIA..........................20
4 ALCOHOL ABUSE SUBSCALE..................................21
4 DRUG ABUSE SUBSCALE.....................................22
4 MEANS AND STANDARD DEVIATIONS...........................26
vi


CHAPTER 1
INTRODUCTION
In the United States, women account for approximately one third of all new HIV infections
(CDC, 2008). The most common route of transmission for women is through heterosexual
intercourse, and the rates for transmission through heterosexual intercourse have been
found to be higher to women than to men. Latinas specifically are one group that has been
particularly affected by the AIDS epidemic. Research has shown that Latinas account for
16% of all new HIV infections, and that the AIDS rate among Latinas is five times higher
than it is for white women (CDC, 2008). Research has also suggested that Latinas are less
likely to have partners that use condoms and are more likely to have misconceptions about
HIV transmission than Caucasian women (Gomez & Martin, 1996). It has been suggested
that cultural factors such as lack of information, traditional gender roles, and cultural
attitudes towards condom use could place Latinas at an increased risk for contracting HIV
(Moreno, 2007). These findings emphasize the need for increased research into the factors
influencing sexual risk amongst Latinas, as well as the need for HIV prevention programs
targeting this population to increase awareness about sexual risk behaviors to promote
condom use. In addition to the risk factors that may be particularly pronounced for Latinas,
more general risk factors are also important to consider. For example, mental health
problems could also influence sexual risk behaviors (Crepaz & Marks, 2001). In fact,
Latinos and women within samples of mentally ill persons may be at a particularly high risk
for contracting HIV (Loue, 2007).
1


Purpose
Given these risk factors, any intervention targeting Latinas should address both specific
risk factors (e.g., negative attitudes toward condoms) and more generic factors (e.g.,
mental health). For this study, I investigated the effectiveness of a peer administered,
community based HIV prevention program designed for Latinas. This prevention program
specifically targeted the unique risk factors facing this population and general factors of
mental health. The primary question to be explored is whether substance use, depression
and anxiety link to sexual risk behaviors for these women, and whether the intervention
can have an impact in these areas.
Research questions
This study examines the relationship between mental health, substance abuse, and sexual
risk behaviors within a primary relationship, as well as the effects of a peer delivered, skill
building intervention program on knowledge related to HIV and sexual risk behaviors within
a primary relationship.
First, this study examines the relationship between anxiety, depression, drug use, and
alcohol with safer sex practices with a primary partner (i.e., condom use, and drug and
alcohol use prior to intercourse) prior to the intervention. It is hypothesized that a history of
mental health problems at the pre-test will be significantly correlated with condom use with
the primary partner during the past 12 months, as well as condom use during the most
recent intercourse with the primary partner. It is also hypothesized that alcohol use at the
pre-test will be significantly correlated with alcohol use prior to intercourse with the primary
2


partner at the pre-test, and that drug use at the pre-test will be significantly correlated with
drug use prior to intercourse with the primary partner at the pre-test. It is further
hypothesized that drug and alcohol use at the pre-test will both be significantly correlated
with condom use with the primary partner over the past 12 months as well as during the
most recent intercourse at the pre-test.
Second, this study investigates intervention effects. Specifically, if there is a difference
between pre-test and post-test scores on HIV knowledge, alcohol abuse, drug abuse, and
sexual risk behavior with the primary partner. It is hypothesized that the post-test scores
will be significantly different from the pre-test scores following the implementation of the
HIV prevention program.
Although it would also be desirable to examine changes in mental health over the course of
the intervention, the measure of mental health used in this study is not sensitive to
changes over time. However, this study will examine how a history of mental health at the
pre-test influences these hypothesized intervention effects in HIV knowledge and condom
use. It is hypothesized that a history of anxiety or depression at the pre-test will moderate
changes in HIV knowledge and sexual risk behavior with the primary partner, such that
persons with a history of mental health problems will show more improvements on these
variables over time relative to persons without such a history.
Finally, the relationship between changes in HIV knowledge, drug use, alcohol use, and
sexual risk behaviors with the primary partner will be examined. It is hypothesized that
3


increases in HIV knowledge and decreases in alcohol and drug use will be associated with
improvements in sexual risk behavior with the primary partner.
4


CHAPTER 2
LITERATURE REVIEW
Anxiety, Depression, and Sexual Risk Taking
A meta-analysis conducted by Crepaz and Marks (2001) focused on the impact of negative
affect, including depressive symptoms, anxiety, and anger on sexual risk behavior. This
study reflected inconclusive and mixed findings concerning the influence that negative
affect plays in risky sexual behaviors. Anger specifically was found to be the most highly
correlated with sexual risk behaviors. However, one limitation to studies included in the
meta-analysis was that they have been primarily cross-sectional and retrospective when
examining negative affect. Crepaz and Marks (2001) suggested that research into this area
using different methodology (e.g., longitudinal) could provide more information into the
potentially significant role that negative affect could have on sexual risk behavior. Although
this meta-analysis reported mixed findings specifically regarding negative affect and sexual
risk behaviors, research has shown that HIV is overall more prevalent in individuals with
mental illness such as major depressive disorder, bipolar disorder, and schizophrenia, and
that Latinos and women within samples of mentally ill persons may be at a particularly high
risk for contracting HIV (Loue, 2007). This may be due to a lower overall perception of
personal risk for persons with mental illness, as well as little fear of HIV or a vulnerability to
coercion due to their illness.
5


A study conducted by Surratt et al. (2005) found that among female street sex workers,
mental health issues associated with anxiety and depressive symptoms were found to
have significant associations with engagement in unprotected intercourse and a lack of
condom negotiation. Among adolescents, research has also suggested that depressed
mood in particular may be a factor influencing risky sexual behavior. A study conducted by
Brown et al. (2006) found that when focusing specifically on condom use, young people
between the age of 15 and 24 were five times more likely to use a condom if they had
scored below the cutoff for depressive symptoms than the participants that scored above
the cutoff (Brown et. al, 2006). Another study conducted by Shrier, Harris, Sternberg, and
Beardslee (2001) found that both female and male adolescents who expressed more
depressive symptoms were more likely to have had an STD in the past. Thus, although
the meta-analysis conducted by Crepaz and Marks regarding negative affect was
inconclusive, other research focusing on depression and anxiety has found some links to
condom use.
Depressive symptoms may be associated with an increase in sexual risk behavior for a
number of reasons. A study conducted by Sterk (2005) found that the overall frequency of
HIV risk taking behaviors was higher in African American women that expressed low self-
esteem. Additionally, research has suggested that depressive symptoms can contribute to
feeling a loss of control or a lower level of self-efficacy (Shier et. al, 2001). Therefore
individuals that express symptoms of depression may not feel able to engage in safer
6


sexual behavior, and may have a lower personal sense of control over preventing
themselves from sexual risk.
In addition to depressive symptoms, symptoms associated with anxiety may also play a
role in sexual risk behavior. Research has suggested that social anxiety is related to
unprotected sex in adolescents as well as college-aged women (Leary & Dobbins, 1983;
Higgins et. al, 1993). Social anxiety may be related to less discussion about the use of
contraceptives, as well as discomfort in negotiating the use of condoms and a decreased
ability to refuse sex without a condom (Hart & Heimberg, 2005). Anxiety in general
surrounding sex may also influence an individuals ability to negotiate condom use. For
Latinas, research has suggested that engaging in behaviors that are not culturally
sanctioned can lead to an increase in anxiety. (Zambrana et al., 2004). This research has
focused on substance use, but it could also apply to condom use, which may not be a
culturally encouraged practice, particularly among married couples.
This research suggests that symptoms of anxiety and depression could have a significant
impact on sexual risk behavior. Additionally, symptoms of anxiety and depression may be
related to other factors that research has consistently found to be related to sexual risk
behaviors, such as drug and alcohol use. Numerous studies have linked substance use to
symptoms of anxiety and depression (Tomlinson & Stein, 2003), and previous research
has consistently linked the use of drugs and alcohol with an increase in high-risk sexual
7


behaviors (Leigh & Stall, 1993). Therefore, it is also important to examine the role of
substance use in mental health.
Substance Use and Sexual Risk Taking
At noted above, drug and alcohol use is consistently associated with high-risk sexual
behavior (Leigh & Stall, 1993). Research has also suggested that Latinas may use alcohol
more frequently as a part of their sexual behavior (Flores-Oritz, 1994). Additionally,
research has shown that Latinas who are more highly acculturated are more likely to have
a history of alcohol and substance usage (Gossage, 1998). Research has also suggested
that Latinas relate their inability to negotiate alcohol abstinence during sex to their inability
to negotiate safe sexual practices with their partners (Moreno, 2007). Therefore, they may
feel coerced into risky sexual behaviors, such as alcohol usage during sexual intercourse
and the failure to negotiate condom use. However, a study conducted by Marin and Flores
(1994) found that while increased alcohol use for Latina women was associated with an
increased number of sexual partners, it was not associated with a lower use of condoms
with either primary or secondary partners. Thus, the research is not entirely clear on this
point. Acculturation may play a factor, as other researchers have found that less
acculturated Latinas may have a lower ability to address issues of condom use with their
partners (Romero, Arguelles, & Rivero, 1993)
8


Latinas
Researchers often refer to Latinas ability or inability to negotiate condom use with their
sexual partners when discussing cultural factors influencing sexual risk behaviors. As
previously mentioned, less acculturated Latinas may have a lower ability to address these
issues with their partners (Romero, Arguelles, & Rivero, 1993). Therefore, acculturation is
closely tied into the ability that Latinas have for engaging in safe sexual behaviors within
their primary relationship. Research has suggested that overall, Latinas have been found
to use condoms less and perceive their partners as less sexually risky when compared to
African-American women (Moreno, El-Bassel, & Morrill, 2007). Additionally, research has
also shown that individuals in a primary relationship have been found to report a lower
frequency of condom use compared to individuals not in a primary relationship (Dixon,
Peters, & Saul, 2003). Therefore, although acculturation can result in increased sexual risk
taking behaviors, so too can relationship status.
One primary reason why Latinas within a relationship may report a lower frequency of
condom use and a high number of sexual risk behaviors may be due to an unspoken trust
within their relationship. However, as previously mentioned, cultural roles can complicate
sexual risk by challenging the reality of these perceptions. Research has indicated that
although Latinas within a relationship have fewer sexual partners, they may be placed at a
greater risk for sexual transmitted diseases by their partners behavior than by their own
behavior (Marin, Tschann, Gomez, & Kegeles, 1993). Machismo, or a sense of heightened
masculinity, is part of many Latino cultures for males. However, machismo may place both
Latinos and Latinas at a greater risk for sexually transmitted diseases. Research has
9


suggested that machismo has been found to be related to multiple sex partners in males,
less condom use, and less intimacy in relationships (VanOss Marin, 2003).
Latinas are an extremely heterogenous group in the United States, and research has
suggested that there are numerous complex factors placing this group at an elevated risk
for HIV. Limited research has focused specifically on the role of negative affect in sexual
risk with Latinas. However, research does suggest that negative affect can play a role in
overall sexual risk, and that there are complex factors related to negative affect for Latinas.
Therefore, when attempting to understand the role of negative affect in sexual risk
behaviors it is also important to consider other complex factors that may be closely
connected with symptoms of anxiety or depression for Latinas.
HIV Prevention with Latinas
It is important to consider the different and complex factors that can place Latinas at a
heightened risk for HIV when developing HIV prevention programming for this population.
Research has shown that information delivered by demographically and behaviorally
similar interventionists is the most effective in producing behavioral change (Durantini et
al., 2006). Therefore, one way to effectively communicate information to Latinas is to
provide the information from Latinas. One major barrier to behavior change for Latinas,
particularly who are less acculturated, is their perception of risk (Newcomb et al., 1998).
Previous research has indicated that individuals with common misconceptions about
contracting HIV, for example that it can be contracting through a kiss on the cheek, also
have a heightened anxiety associated with contracting HIV (Ritieni, Moskowitz, & Tholandi,
10


2006). This anxiety can hinder prevention efforts. Due to the advantage shown in peer
delivered information in affecting behavior change, an intervention delivered by Latinas
regarding the realistic risks associated with HIV may have more impact on behavior. That
is, by providing information from individuals who are familiar with or a part of their culture,
Latinas may begin to more seriously consider their potential for risk in their relationships
and through their behavior.
Research has also shown that interventions with Latinas are more effective if they target
environmental, social, economic, and political factors (Weeks et al., 1996). This further
emphasizes the importance of delivering information from an understanding peer group, as
well as takes into consideration the complex risk factors discussed that could potentially be
contributing to sexual risk behaviors. This can specifically include factors associated with
culture. For example, acknowledging that Latinas are often in a teaching role about health
with family provides an environment in which HIV prevention and condom use does not
have the potential for being considered contradictory to the family or traditional gender
roles. This could include discussions about how sexual partners might potentially react to
the suggestion of using a condom. Dialogue as well as role-playing can be useful for these
issues.
Role-playing also speaks to the effectiveness of skill building versus information giving in
interventions. Research has suggested that in intervention programs, an increase in
knowledge does not necessarily result in an actual change in behaviors (Weeks et al.,
1996). Although Latinas might gain more information about risk for sexually transmitted
11


diseases, as previously discussed, there may be other barriers such as a lack of partner
communication that will limit how much they incorporate this new information into their
sexual behavior. However, by having discussions about these issues and role-playing
exercises, the women can leam skills for tactfully dealing with situations that may have
previously been problematic for them.
Additionally, by taking culture into account when designing these interventions, it may also
be important to avoid an empowerment approach to HIV prevention, which has commonly
been used with other groups, and instead help the women to merge their beliefs and
attitudes into approaching safe sexual behaviors in a way that does not compromise their
culture. Latinas may traditionally feel a discomfort towards discussing sexual issues, and
this would be a culturally sensitive way to help Latinas feel a sense of comfort with the
intervention and information being provided. Taking the complex set of risk factors
associated with Latinas and acculturation into account when developing interventions can
therefore not only help to encourage the success they will have in increasing knowledge
about HIV, but also facilitating safer sexual behaviors for this population.
The current study examines improvements in mental health and sexual risk variables for a
group of Latinas who participated in an intervention guided by the above principles.
Specifically, this was a peer delivered intervention which incorporated role playing as a
means of skill building. Thus, the current study examines the effects of a culturally
sensitive and empirically informed intervention on general mental health outcomes and
specific sexual risk outcomes. Also, the current study examines the influence mental health
12


and substance abuse has on sexual risk, and whether improvements in sexual risk as a
result of the program are in part explained by improvements in mental health and
substance abuse.
13


CHAPTER 3
METHODS
Participants
The participants were a group of 112 female, heterosexual participants that self-identify as
Latina. The average age of the participants was 35.88 years, with the youngest participant
being 18-years-old and the oldest participant being 65-years-old. For level of education, 40
percent of the participants had not completed high school, 31 percent of the participants
had obtained a high school diploma or equivalent, and 29 percent had attended some
college or obtained a college degree. The participants were recruited through word of
mouth at a community-based center for Latinas in Denver County.
Measures
Demographics
The questionnaire gathered demographic information about the participants including
their age, level of education, marital status, ethnicity, number of children, and number
of brothers and sisters.
HIV Knowledge
This subscale included 10 statements concerning the participants knowledge about HIV
that the participants rated their agreement with on a four point Likert-scale ranging from
14


strongly disagree to strongly agree, with a fifth option for unsure. Questions focused on
HIV knowledge, for example AIDS is not a fatal disease (see Table 1). This subscale also
dealt with stereotypes surrounding AIDS, such as if AIDS only affects homosexuals and if a
cure exists. Cronbach's alpha was .78. Higher scores on this subscale indicated a higher
knowledge about HIV.
15


Table 1
HIV Knowledge Subscale
Item Score Reverse Scored
Strongly Agree Strongly Disagree
1. AIDS is a diseased caused by a virus. 1-4 no
2. There is a difference between having HIV and having AIDS. 1-4 no
3. AIDS is a disease which only affects homosexual men. 1-4 yes
4. There is no known cure for AIDS at the present time. 1-4 no
5. AIDS is not a fatal disease. 1-4 yes
6. AIDS cripples the bodys natural protection against diseases. 1-4 no
7. Any person who has HIV can pass it on to someone else during intercourse. 1-4 no
8. A pregnant women with AIDS/HIV can pass it on to her baby. 1-4 no
9. The risk of getting HIV is higher if a person has a lot of sexual partners. 1-4 no
10. Condoms are not effective in prevention the transmission of HIV. 1-4 yes
Total Score = 10-40
Alpha = .78
16


Sexual Risk Variables
Four sexual risk items were included in the study (see Table 3). Item one dealt with
condom use with the primary partner in the past 12 months. This was scored on a five
point likert scale ranging from never' to always. The second item dealt with condom use
during that participant's most recent sexual intercourse with the primary partner, and was
scored as either yes or no. Items three and four dealt with alcohol use and drug use
prior to sexual intercourse with the primary partner in the last six months. These were
scored on a five point Likert scale ranging from never to always. The reliability was low
for the items when they were scored together as a scale; Cronbachs alpha was .50.
Therefore the items will each be evaluated as separate risk variables.
17


Table 2
Sexual Risk Variables
Item
1. When you had vaginal intercourse with your
primary partner during the last 12 months, how
often did you use condoms?
2. When you had vaginal intercourse most recently
with your primary partner, did you use a
condom?
3. In the last 6 months, how often did you drink
alcohol before having vaginal or anal intercourse
with your primary partner?
4. In the last 6 months, how often did you use drugs
like marijuana, cocaine, crack, or heroin before
having vaginal or anal intercourse with your primary
partner?
Original Score
1-5 (Always-Never)
yes/no
1-5 (Never-Always)
1-5 (Never-Always)
18


Anxiety and Depression
Two questions administered at the pre-test asked about past experiences of extreme
feelings of sadness or depression, of if they had ever experienced extreme feelings of
anxiety (see Table 3). Subsequent questions asked about duration of these feelings
(responses ranging from less than a week to more than a month) and how long the
respondent felt sadness and depression at least once a week (response options ranging
from less than six months to more than five years). Persons who denied having depression
or anxiety were ranked as never on these items. Thus, total mental health scores were a
composite score of these four follow up questions with the initial screener questions coded
as zero if they never experienced such feelings (see table 3). Cronbachs alpha was .75.
For moderation analyses, participants were categorized into two mutually exclusive groups:
persons with history of mental health (82 participants) as indicating yes to either question
about ever experiencing depression and anxiety, versus those who answered no to both of
these questions (30 participants).
19


Table 3
Anxiety or Depression Criteria
Item Score
1. If you have ever experienced extreme feelings week to 0-6 (ranging from less than a
of sadness and depression, how long did you continuously feel sadness and depression? greater than a month)
2. If you experienced feelings of sadness and depression, how long have you felt sadness and depression at least once a week? 0-6 (ranging from less than six months to greater than five years)
3. If you have ever experienced extreme feelings of anxiety, how long did you continuously feel anxious? 0-6 (ranging from less than a week to greater than a month)
4. If you experienced feelings of anxiety, how long have you felt anxious at least once a week? Total Score = 0-24 Alpha = .78 0-6 (ranging from less than six months to greater than five years)
A yes to either of the following questions qualified the participant for the mental health
group in the moderation analyses:
1.. Have you ever experienced extreme feelings sadness or depression? yes/no
2. Have you ever experienced extreme feelings of anxiety? yes/no
20


Alcohol Abuse
The participants were asked to indicate how many days in the most recent 30 days they
had used alcohol. The participants scored ranged from 0-6, based on the frequency of
days they indicated (see Table 4).
Table 4
Alcohol Abuse Subscale
Item Score
How many times in the past 30 days have you used:
Alcohol 0-6
21


Drug Abuse
Participants were asked to indicate how many days in past 30 days they had used
marijuana, cocaine, crack, inhalants, heroin, and hallucinogens (see Table 5). Participants
that indicated they had used the substances at least one day in the prior 30 days were
given a score of one for that item, and participants that indicated not using any of these
substances were given a score of zero for that item. The scores for each item were then
totaled into a composite score for drug use. Cronbachs alpha was .96.
Table 5
Drug Abuse Subscale
How many times in the past 30 days have you used:
Marijuana or hashish 0-1
Cocaine 0-1
Crack 0-1
Inhalants 0-1
Heroin 0-1
Hallucinogens 0-1
Total Score = 0-6
Alpha = .96
22


Data Collection
The program was implemented at a community-based center for Latinas. Latina volunteers
from the community were trained in the material and administration of the program. The
program included the presentation of HIV information, discussion groups about sexual risk,
and role-playing how to discuss condom use and sexual risk with their partners. Prior to
implementation of the program material, the questionnaire was administered to the
participants. The program material was presented over an 18-week period in either one or
two hour sessions on a weekly basis depending on the participants availability. The same
questionnaire was administered following the implementation of the program material.
23


CHAPTER 4
RESULTS
The first hypothesis was that pre intervention mental health, drug use, alcohol use, condom
use, and drug and alcohol use prior to intercourse would be related. Pearson and point
biserial correlations were conducted to evaluate the links among these variables. Contrary
to the hypothesis, there were no significant correlations between scores on the mental
health scale at the pre-test and either condom use variable at the pre-test. There was a
significant correlation found between having a history of mental health at the pre-test and
alcohol use at the pre-test, r{104) = .30, p < .05. As predicted, alcohol and drug use were
mostly related to alcohol and drug use prior to intercourse. Specifically, alcohol use at the
pre-test was significantly correlated with alcohol use prior to intercourse at the pre-test,
/f99,| = .62, p < .01. Alcohol use at the pre-test was also significantly correlated with drug
use prior to intercourse at the pre-test, r(104J = .40, p < .01. Drug use at the pre-test was
not significantly correlated with drug use prior to intercourse at the pre-test. Neither alcohol
use or drug use were related to condom use in the past 12 months, and neither alcohol or
drug use were significantly related to condom use at last intercourse. As expected, drug
use at the pre-test was significantly correlated with alcohol use at the pre-test, r( 112) = .39,
p < .01, and drug use prior to sexual intercourse at the pre-test was significantly correlated
with alcohol use prior to intercourse at the pre-test, r(99) = .62, p < .01. Thus, depression
and anxiety were not linked to sexual risk behavior, but drug and alcohol use often were.
24


The second hypothesis was that post-test scores on alcohol use, drug use, HIV
knowledge, and the sexual risk variables would be significantly improved relative to the
pre-test scores. To test this, paired sample t-tests were conducted between the pre- and
post-test scores on these subscales. A significant difference between pre- to post- test
scores on scores of HIV knowledge, #(92) = -2.47, p < .01 was found, with HIV knowledge
increasing from pre- to post-test scores. The results also showed a significant difference
between pre- to post-test condom use over the past 12 months, #(87) = -13.39, p < .001,
with condom use increasing over the past 12 months. There was a significant difference
between pre- to post-test condom use during the most recent sexual encounter, #(88) = -
4.92, p < .001, with condom use increasing. Therefore, the hypothesis that HIV knowledge
and condom use would be significantly improved was supported. There was not a
significant different between pre- to post-test scores on scores of overall alcohol use,
although the means suggest that alcohol use did decrease slightly from pre to post test.
Therefore the hypothesis that alcohol use would decrease significantly was not supported,
but the means indicate that alcohol use did decrease as the hypothesis predicted.
Unexpectedly, there was a significant different between pre to post test scores on scores of
overall drug use, #(111) = -5.97, p < .001, with drug use increasing from pre- to post-test
scores. There was also an startling significant difference between pre- to post-test alcohol
use prior to the most recent intercourse, f(77) = -13.13, p < .001, with alcohol use
increasing prior to intercourse. Additionally, a significant difference was found between pre-
to post-test drug use prior to most recent intercourse as well, f(84) = -3.40, p < .001, with
25


drug use prior to intercourse increasing. These findings were unpredicted and contrary to
the hypothesis.
Table 5
The Means and Standard Deviations for Subscales and Variables
Variable Deviation Mean Standard
Pre-Test Post-Test Pre-Test Post-Test
HIV Knowledge 24.22 26.91 3.75 8.05
Alcohol Abuse .02 0 .12 0
Drug Abuse .59 1.65 2.40 2.41
Condom Use: Past 12 Months 1.80 3.94 1.54 .24
Condom Use: Most Recently .03 .38 .17 .49
Alcohol Use Prior to Intercourse .11 2.17 .31 1.41
Drug Use Prior to Intercourse .02 .14 .12 .39
26


It was also hypothesized that pre-scores on the mental health subscale would moderate
improvements in sexual risk. To test this hypothesis, repeated measures ANOVAs were
conducted in which time (pre versus post) was a within subjects factor and if a history of
mental health was present or not at the pre-test was a between subjects factor. As
hypothesized, there was a significant interaction effect found F( 1, 91) = 4.02, p < .05) in
which individuals with a history of mental health problems showed greater improvements
over time in HIV knowledge relative to those without a history of mental health problems.
However, contrary to hypotheses, no interaction effects were found for changes in condom
use over the last 12 months or changes in condom use during the most recent intercourse.
27


EstfauUd Mvginal Mnni
Tim* (1 Pr e-Tett; 2 P*t-T*t)
Green Mental Health
problems
Blue No Mental Health
Problems
Figure 1: INTERACTION EFFECT FOR HYPOTHESIS 3
28


Lastly, it was hypothesized that change scores in HIV knowledge, alcohol use, and drug
use would be related to positive changes in condom use during the past 12 months,
condom use during the most recent intercourse, alcohol use prior to intercourse, and drug
use prior to intercourse. Since the results from the second hypothesis found that overall
drug use, drug use prior to intercourse, and alcohol use prior to intercourse unexpectedly
increased from pre- to post-test, this analysis focused only on the relationship between
change scores in HIV knowledge, changes in condom use over the past 12 months, and
condom use during the most recent intercourse. To these this, change scores for each of
these variables were calculated and then correlated. The results showed that no significant
correlations were found between changes in HIV knowledge, changes in condom use over
the past 12 months, or condom use during the most recent intercourse. Therefore, the
hypothesis was not confirmed.
29


CHAPTER 5
DISCUSSION
This study first sought to examine the relationship that having a history of mental
health at the pre-test would have with substance use, condom use, and substance use
prior to sexual intercourse at the pre-test. As the results indicated, the hypothesis that a
history of mental health would be significantly correlated with condom use was not
supported. However, there were several limitations that may have influenced this finding.
Firstly, mental health history was measured by the participants endorsement of ever
having experienced extreme feelings of depression, sadness, or anxiety. This measure
does not address the causes of these feelings or how long ago they may have occurred. In
fact, the research suggesting depression and anxiety may be related to sexual risk has has
indicated that the symptoms must be currently present to have an influence (Brown et. al,
2006; Shrier, Harris, Sternberg, and Beardsle, 2001; Surratt et al., 2005). However, this
measure of mental health history was not able to predict which participants were currently
experiencing symptoms of depression or anxiety.
Additionally, the research that has indicated that anxiety and depression may have
an impact on sexual risk has found that not all symptoms of anxiety or depression have the
same impact. For example, low self-esteem may be specifically related to riskier sexual
behavior, whereas feelings of sadness are less related (Sterk, 2005). This would make
sense given the theoretical links between certain aspects of anxiety and depression and
sexual risk behavior. For example, low self-esteem might undermine ability to negotiate
30


condom use whereas sadness would not (Lykins, Janssen, & Graham, 2006). The
measure of mental health history used in this study was therefore limited in that it did not
provide any information on specific symptoms of depressed or anxious feelings the clients
might have experienced, and how these specific symptoms may be tied to condom use.
Future research could seek to further examine more thorough aspects of mental health
history, as well as specific symptoms associated with depression and anxiety to more
thoroughly ascertain the influence that mental health has on condom use and risk
behaviors.
Another factor potentially influencing these findings is that the variables assessing
condom use and substance use prior to intercourse asked the participants about these
behaviors specifically in regards to their primary partner. Research has indicated that
individuals in primary relationships report a less frequent condom use compared to
individuals not in a primary relationship (Dixon, Peters, & Saul, 2003). Therefore, if the
participants had been asked about their overall condom use rather than specifically with
their primary partner, a clearer link may have been found between condom use and mental
health. A relationship may not have been found between mental health at the pre-test and
condom use at the pre-test because the participants condom use specifically with their
primary partner may have been related to other factors such as the length of the
relationship, rather than specifically mental health.
Research has further suggested that factors such as acculturation can play a role
in condom use within committed relationship for Latinas, and that this may also be linked to
feelings of depression (Sanchez, 2004). Although this study did not assess acculturation,
the targeted participants for this prevention program were less acculturated Latinas.
31


Therefore, this study was also unable to assess the role of acculturation in the participants
condom use with their primary partner, and how feelings of anxiety and depression may be
linked to condom with their primary partners.
Although the results did not confirm a link between condom use and a history of
mental health at the pre-test, they did show that higher alcohol use at the pre-test was
correlated with a history of mental health at the pre-test. This coincides with research
emphasizing the relationship between substance use and mental health (Tomlinson &
Stein, 2003). Thus, the broad index of mental health used in this study may have indirect
links with sexual risk behavior through an increased frequency of alcohol use.
Furthermore, the results did support the hypothesis that alcohol use at the pre-test would
be significantly correlated with alcohol use prior to sexual intercourse, and that drug use at
the pre-test would be significantly correlated with drug use prior to intercourse at the pre-
test. These correlations were positive, suggesting that as alcohol and drug use increased,
so did alcohol and drug use prior to sexual intercourse. Furthermore, higher drug use was
correlated with higher alcohol use. This suggests that alcohol and drug use are related,
and that both are potentially linked to sexual behavior. Previous research has consistently
linked substance use to risky sexual behavior (Leigh & Stall, 1993). These results indicate
that the participants may be using substances specifically in conjunction with their sexual
behavior, which could ultimately result in risky sexual behavior. This is consistent with
research indicated that Latinas may use alcohol more frequently as a part of their sexual
behavior (Flores-Oritz, 1994). Therefore, it was also important for this study to examine
32


how the implementation of the HIV program may have had a positive impact on
participants to reduce risk variables and increase condom use.
For this reason, the second aspect of this study was to determine changes in HIV
knowledge, condom use, alcohol and drug use, and alcohol and drug use prior to
intercourse prior to and following the implementation of the described HIV prevention
program. Significant changes found in the results indicate that this hypothesis was
supported for each area except alcohol use. Therefore, the program was successful in
increasing knowledge about HIV, increasing condom usage over the past 12 months, and
increasing condom usage during the most recent intercourse. This suggests that a large
priority of the program was successful.
However, the results also indicated that drug use significantly increased following
the implementation of the program, and that both alcohol and drug use prior to sexual
intercourse increased following the implementation of the program. Interestingly, although
alcohol and drug use increased prior to intercourse, condom use was found to increase. As
previously mentioned, research has suggested that while alcohol use for Latina women
has been associated with an increased number of sexual partners, it has not been
associated with a lower use of condoms with either primary or secondary partners (Marin
and Flore, 1994). Therefore, their increased use of drugs overall, and their increased use
of drugs and alcohol prior to intercourse may not be related to their condom use.
This also suggests that the program was not successful at targeting these risk
behaviors. One reason for this may be due to this particular group of participants, and their
unique drug habits that the program was unable to target. Research has suggested that
33


Latinas have a decreased ability to negotiate alcohol abstinence during sex as well as safe
sexual behavior (Moreno, 2007). Therefore, the program may have increased the
participants ability to more effectively increase condom use through negotiation at the
expense of decreasing ability to negotiate substance use prior to intercourse.
Additionally, exploratory analyses found that marijuana was significantly increased
following the implementation of the program. Therefore, the participants increased use of
alcohol and drugs prior to intercourse, and the overall increased use of marijuana could
potentially be linked to anxiety surrounding sexual behavior. Research has repeatedly
linked substance use to symptoms of anxiety and depression, and this may suggest an
overall increase in these symptoms following the implementation of the program
(Tomlinson & Stein, 2003). Therefore, their increased use of these substances prior to
intercourse may be suggestive of increases in stress or anxiety they are experiencing
towards sex. Therefore, this suggests that mental health may in fact play a role in these
findings.
Having determined the changes in HIV knowledge and condom use following the
implementation of the program, this study next sought to examine how having a history of
mental health at the pre-test may have influenced these changes. The results did not
support the hypotheses that having a history of mental health at the pre-test would have an
interaction with condom use changes following the implementation of the program, but the
hypothesis that a history of mental health at the pre-test would have an interaction on
changes in HIV knowledge was supported. This suggests that regardless of mental health
at the pre-test, the program was effectively in increasing condom use. However, the group
34


that reported a history of mental health at the pre-test was found to have more drastic
improvements in their HIV knowledge than the group without a history of mental health at
the pre-test. There were no significant differences between the pre-test and post-test
scores of HIV knowledge, and this may have been due to limitations in power. These
findings suggest the program may have impacted the participants that reported a history of
mental health at the pre-test more than the participants who did not.
Although both groups seemed to start out with approximately the same score on
the subscale of HIV knowledge, perhaps the social support aspects of the program allowed
the mental health groups to more effectively retain the information provided. Research has
suggested that prevention programs are particularly effective if they target environmental,
social, economic, and political factors (Weeks et al., 1996). The program sought to target
these factors, and addressing these factors while providing information about HIV may
have particularly benefited the group with a history of mental health. Additionally, research
has emphasized that an increase in knowledge does not necessarily translate into
corresponding behavior change (Weeks et al., 1996). This could further explain why this
difference was seen for increases in HIV knowledge and not for increases in condom use.
This suggests that having a history of mental health may play a role in the effectiveness of
prevention efforts. Furthermore, this also indicates that this program was effective in not
only increasing knowledge about HIV, but also increasing corresponding behavior change
as indexed by condom use over the past 12 months and condom use during the most
recent intercourse.
35


For these reasons, the final aim of this study was to examine how changes seen from the
pre-test to the post-test related to each other. Specifically, this study sought to examine
how changes in HIV knowledge, alcohol use, and drug use were related to variables of
behavior change, specifically condom use and drug and alcohol use prior to sexual
intercourse. As the results indicated, the hypothesis that HIV knowledge would be
significantly correlated to condom use changes and substance use prior to sexual
intercourse changes was not found. This may suggest that although both condom use and
HIV knowledge increased from pre-test to post-test, there may have been unique aspects
of the program that increased condom use unrelated to increases in knowledge.
Additionally, this also suggests that the program fell short of linking increases in knowledge
to decreases in substance use prior to sexual intercourse. This may further emphasis the
difficulties that HIV prevention efforts have in address substance use and sexual behavior
with this population. Research has indicated that one large barrier for increasing behavior
change with Latinas is their low perception of risk (Newcomb et al., 1998). It may be that
the program was not able to effectively increase the participants perception of risk related
to substance use and sexual behavior.
Additionally, the hypothesis that drug use changes would be significantly
correlated to drug use prior to sexual intercourse and condom use was not supported.
However, the results also indicated that drug use significantly increased from the pre-test
to the post-test, and that drug use prior to sexual intercourse significantly increased from
the pre-test to the post-test. This might suggest that there may have been a group of
participants who increased in their overall drug use without necessarily linking this increase
36


to sexual intercourse, and that there may have been another group of participants who
increased their drug use specifically prior to sexual intercourse. However, there was a
significant relationship between increases in alcohol use and increases in drug use prior to
sexual intercourse. This relationship supports the previously discussed relationship
between alcohol, drug use, and sexual behavior.
One limitation of this study was that the measure of both drug and alcohol use
measured the participant's use in number of days, and asked them to estimate this use
during the previous 30 days. Specifically for alcohol use, the participants may have only
been able to endorse a few days of drinking, but this may of characterized episodes of
binge drinking. Therefore, more changes may have been seen specifically in alcohol use if
the study had more specifically examined the participants drinking behavior. Future
research should more thoroughly examine the link between substance use and sexual
behavior for the participants. This study was limited in that it only asked participants how
frequently they had used alcohol or drugs prior to sexual intercourse in the past six
months. Since this study did seem to find that substance use increased from pre-test to
post-test, it might also be interesting to examine how the use of substances may have
changed following the implementation of the program. It might also be useful to examine
the role that substance prior to sexual intercourse plays in reducing anxiety about sex.
The findings suggest that mental health history may play a role in sexual risk, as
well as the effectiveness of prevention efforts targeting Latinas. Future research should
tease apart symptoms of anxiety and depression that may be specifically related to sexual
risk, and how the targeting of these symptoms may increase the effectiveness of HIV
37


prevention programming. Additionally, future research should examine changes in mental
health that arise from participating in prevention efforts.
It is also important for future research to gain an increased understanding of
alcohol and drug use, and the role this use plays in sexual behavior for Latinas. The results
of this study indicate that while some positive behavior change and increases in knowledge
may have occurred through the implementation of this program, the program still fell short
of decreasing overall drug use or substance use prior to sexual intercourse. Therefore, it
would also be important to gain an understanding of how to more effectively target
substance use and its role in sexual behavior through prevention efforts. As previously
mentioned, understanding more thoroughly the role that substance use plays in sexual
behavior and how this may be linked to feelings of anxiety or depression would also be
important to understand for the purposes of prevention efforts.
Although this study was limited in its measurement of mental health, the results
from this study emphasize the complex factors, including mental health, that complicate
prevention efforts targeting Latinas. It is crucial to gain an understanding of how mental
health, substance use, and sexual risk are related in order to gain insight into how positive
behavior change can be the most efficiently increased with this population. The limitations
of this study highlight the need for further research into teasing out how these factors can
be most effectively addressed to increase positive sexual behavior with Latinas.
38


WORKS CITED
Adams, K. B., & Sanders, S. (2004). Alzheimers caregiver differences in experience of
loss, grief reactions and depressive symptoms across stage of disease. Dementia,
3(2), 195-210.
Brown, A., Yung, A., Cosgrave, E., Killackey, E., Buckby, J., Standord, C., Godfrey, K., &
McGorry, P. (2006). Depressed mood as a risk factor for unprotected sex in young
people. Austrailian Psychiatry, 14(3), 310-312.
Caceres-Dalmau, M. (2004). Maternal marianismo and anxious somatic depression among
Hispanic female college students. Dissertation Abstracts International: Section B:
The Sciences and Engineering, 64,4026-4038.
CDC (2008). HIV/AIDS Surveillance Report, 18.
Crepaz, N. & Marks, G. (2001). Are negative affective states associated with sexual risk
behaviors. Health Psychology, 20, 291-299.
Dixon, D. Peters, M., & Saul, J. (2003). HIV sexual risk behaviors among Puerto Rican
women. Health Care for Women International, 24(6), 529-543.
Durantini, M. R., Albarracln, D., Mitchell,. L., Earl, A. N., & Gillette, J. C. (2006).
Conceptualizing the influence of social agents of behavior change: A meta-
analysis of the effectiveness of HIV-prevention interventionists for different
groups. Psychological Bulletin, 132(2), 212-248
Flores-Oritz, Y. G. (1994). The role of cultural and gender values in alcohol use patterns
among Chicana/Latina high school and university students: Implications for AIDS
prevention. International Journal of the Addictions, 29(9), 1149-1171.
Ford, A. E. & Norris, K. (1994). Condom beliefs in urban, low income, african american and
hispanic youth. Health Education Quarterly, 21, 39-53.Marin, B. V.,
Gomez, C. Z. & Martin, B. V. (1996). Gender, culture, and power: Barriers to HIV
prevention strategies for women. Journal of Sex Research, 33, 355-362.
Gossage, J. P. (1998). An examination of the relationship of acculturation to high-risk
behaviors among Hispanic women of childbearing age in New Mexico. Dissertation
Abstracts International Section A: Humanities and Social Sciences, 59, 967-977.
Jimenez, J., Potts, M. K., & Jimenez, D. (2002). Reproductive attitudes and behaviors
among Latina adolescents. Journal of Ethnic and Cultural Diversity in Social Work,
11(3), 221-249.
39


Kalichman, S. C. & Weinhardt, L. (2001). Negative affect and sexual risk behavior:
Comment on crepaz and marks. Health Psychology, 20(4), 300-301.
Loue, S. (2007). HIV research among severely mentally ill latinas. Acta Bioethica, 13(1),
79-95.
Lykins, A. D., Janssen, E., & Graham, C.A. (2006). The relationship between negative
mood and sexuality in heterosexual college men and women. Journal of Sex
Research, 43(2), 136- 143.
Marin, B. V. (2003). HIV prevention in the Hispanic community. Sex, culture, and
empowerment. Journal of Transcultural Nursing, 14(3), 186-192.
Marin, B. V. & Flores, E. (1994). Acculturation, sexual behavior, and alcohol use among
Latinas. International Journal of the Addictions, 29(9), 1101-1114.
Marin, B. V., Tschann, J. M., Gomez, C. A. & Kegeles, S. M. (1993). Acculturation and
gender differences in sexual attitudes and behaviors: Hispanic vs. non-hispanic
white unmarried adults. American Journal of Public Health, 83,1759-1761.
Marks, G., Cantero, P. J., & Simoni, J. M. (1998). Is acculturation associated with sexual
risk behaviors? An investigation of HIV-positive latino men and women. AIDS
Care, 10, 283-295.
Moreno, C. L., El-Bassel, N., Morrill, A.C. (2007). Heterosexual women of color and HIV
risk: sexual risk factors for HIV among latina and african american women.
Womens Health, 45(3), 1-15.
Newcomb, D., Wyatt, G. E., Romero, G. J., Tucker, M. B., Wayment, H. A., Carmona, J. V.,
Solis, B., & Mitchell-Kernan, C. (1998). Acculturation, sexual risk taking, and HIV
health promotion among Latinas. Journal of Counseling Psychology, 45(4), 454-
467.
Nyamathi, A. M., Bennett, C., Leake, B., Lewis, C. & Flaskerud, J. (1993). AIDS related
knowledge, perceptions, and behaviors among improvised minority women.
American Journal of Public Health, 83, 65-71.
Ritieni, A., Moskowitz, J., & Tholandi, M. (2006). HIV/AIDS misconceptions among Latinos:
Findings from a population-based survey of California adults. Health, Education,
and Behavior, 20(10), 1-15.
Rogler, L. H., Malgady, R. G., Costantino, G. & Blumenthal, R. (1987). What do culturally
sensitive mental health services mean? The case of hispanics. American
Psychologist, 42, 565-570.
40


Rojas-Guyler, L., Ellis, N.t & Sanders, S. (2005). Acculturation, health protective sexual
communication, and HIV/AIDS risk behavior among Hispanic/Latino women in a
large midwestern city. Health Education & Behavior, 32, 767 779.
Romero, G. J., Arguelles, L., & Rivero, A. (1993). Latinas and HIV infection/AIDS:
Reflections on impacts, dilemmas and struggles. In B. Blair & S. Cayleff (Eds.),
Wings of gauze: women and the experience of illness. Wayne State University
Press: Detroit, Ml.
Padilla, A. M. & Perez, W. (2003). Acculturation, social identity, and social cognition: A new
perspective. Hispanic Journal of Behavioral Sciences, 25(1), 35-55.
Patton, M. Q. (2002). Qualitative Research and Evaluation Methods. Washington, D. C.:
Sage.
Petrocelli, J. V. (2003). Hierarchal multiple regression in counseling research: Common
problems and possible remedies. Measurement and Evaluation in Counseling and
Development, 36, 9-22.
Sanchez, M. M. (2004). Latinas experience of depression. Dissertation Abstracts
International: Section B: The Sciences and Engineering, 64, 5232-5248.
Shrier, L. A., Harris, S. K., Sternberg, M., & Beardslee, W. R. (2001). Associations of
depression, self-esteem, and substance use with sexual risk among adolescents.
Preventive Medicine: An International Journal Devoted to Practice and Theory,
33(3), 179-189.
Sterk, C. (2005). Self-esteem and at risk women. Women and Health, 40(4), 75-92.
Tschann, J. M., Gomez, C. A. & Kegeles, S. M. (1993). Acculturation and gender
differences in sexual attitudes and behaviors: Hispanic vs. non-hispanic white
unmarried adults. American Journal of Public Health, 83,1759-1761.
Weeks, M. R., Singer, M Grier, M., & Schensul, J. J. (1996). Gender relationships,
sexuality, and AIDS risk among African American and Latina Women. In C. F.
Sargent, & C. B. Brettell (Eds.), Gender and health: An international perspective.
Prentice-Hall: Upper Saddle River, NJ.
Zambrana, R. E., Cornelius, L. J., Boykin, S. S., & Lopez, D. S. (2004). Latinas and
HIV/AIDS risk factors: Implications for harm reduction strategies. American Journal
of Public Health, 94(7), 1152-1158.
41