Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2019 Sep 1.
Published in final edited form as: AIDS Behav. 2018 Sep;22(9):2815–2829. doi: 10.1007/s10461-018-2147-9

Mood, meth, condom use, and gender: Latent growth curve modeling results from a randomized trial

Eileen V Pitpitan 1, Shirley J Semple 1, Jim Zians 2, Steffanie A Strathdee 1, Thomas L Patterson 3
PMCID: PMC6231985  NIHMSID: NIHMS967504  PMID: 29754267

Abstract

Methamphetamine use poses increased risk for HIV and other sexually transmitted infections. There is robust evidence that methamphetamine use increases sexual risk behavior, like condomless sex, primarily among men who have sex with men but also among heterosexual women and men. Gender differences have been found among women and men who use meth, and there is a high degree of interconnectedness between meth use, depression, and condomless sex. The aims of the current study are to evaluate the efficacy of a theory-based, tri-focal intervention designed to reduce depression, meth use, and condomless sex among women and men, and to examine gender as a moderator of efficacy. A total of 432 HIV-negative women and men who use meth participated in a two-arm randomized controlled trial and completed baseline and follow-up assessments at 4, 8, and 12 months. We used latent growth curve modeling techniques to analyze the data. Results showed that while all participants exhibited reductions in depression, meth use, and condomless sex, the intervention and comparison groups did not differ in changes over time. However, we did find a significant gender moderation effect, such that among men, those in the intervention arm reported greater reductions in meth use relative to those in the comparison group; reduced meth use was associated with reduced condomless sex, but not depression. In contrast, women in the intervention condition did not differ from women in the comparison condition in changes in any of the three outcome variables. Interventions targeting heterosexual women and men who use meth must be gender-specific, and take into account the unique vulnerabilities and experiences of women, including the perceived positive aspects of using meth, gendered power dynamics, higher depression, and violence.

Introduction

Methamphetamine use has long been linked to increased sexual risk behavior and risks for HIV and other sexually transmitted infections (STIs) (1,2). Meth heightens libido, impairs judgment, delays sexual climax (3–5), and therefore is associated with more sexual partners, sexual activity including anal sex, condomless sex, and condom breakage (2,6,7). The evidence for the association between meth use and sexual risks for HIV is robust among men who have sex with men (MSM). Consequently, behavioral HIV prevention interventions have been designed to focus on reducing HIV risk among meth-using MSM (8–11). Although relatively fewer in number compared to relevant research with MSM, research has also demonstrated the important role of meth use in sexual risks for HIV among heterosexual women and men (12).

A recent meta-analysis of studies of heterosexual meth users published since 2000 showed that the odds of engaging in risky sex for heterosexual users is on average between 37% to 72% greater than for non-meth users (13). Similar to MSM, heterosexual women and men experience a heightened sex drive and increased sexual activity while on meth (12). Compared to other drug users, heterosexual women and men who use meth report a greater number of sexual partners, condomless sex acts, and casual or anonymous partners (14). Motivations for using meth appear to differ between heterosexual women and men. Whereas both women and men report a desire to enhance sexual pleasure (12,15,16), women also report feeling greater sexual empowerment and agency while on meth (17,18), suggesting the need to recognize gender differences between meth users. Despite any differences, like men, heterosexual women who use meth also engage in high risk behaviors associated with HIV/STI acquisition, including more sex partners and transactional sex (16,19). Even among drug-using women, women who use meth appear to engage in riskier sexual behavior. Meth-injecting women report more anal sex, condomless anal and vaginal sex, and sex partners relative to non-meth-injecting women (20).

In light of the increased risk for HIV faced by heterosexual meth-using women and men, our research team developed and evaluated a theory-based sexual risk reduction intervention for this population, known as Fastlane. Results of the Fastlane randomized trial showed that participation in the experimental condition (i.e., the Fastlane intervention) was associated with significantly more total condom-protected sex acts at six-month follow-up compared to the attention-control diet and exercise condition (21). However, evidence also showed that regardless of group assignment, participants whose meth use and/or depressive symptoms did not decrease over time also did not reduce their condomless sex. These findings were consistent with the literature that has demonstrated the close links between meth use, depression, and sexual risk behavior.

Individuals who use meth engage in risky sexual behavior, and are also more likely to report depressive symptoms or negative mood (2,22,23). These three variables (depression, meth use, risky sex) co-occur among MSM, and also among heterosexual women and men. It is likely that the three reinforce one another (24); meth may be used to cope or distract oneself from a depressive state, meth use increases sexual activity and risks, and negative life conditions and circumstances stemming from meth use prolong depression. This possible interconnectedness suggests that by intervening upon and reducing one variable, reductions should also be found in the other two. In one study, drug treatment for meth-using MSM was associated with fewer anonymous sex partners and reductions in receptive and insertive anal sex acts at one-year follow-up (25). In another study comparing behavioral treatment approaches for meth dependent MSM, drug abuse treatment resulted in significant reductions in both meth use and sexual risk behaviors (23).

In the current paper, we report findings from Fastlane-II, a follow-up trial from the original Fastlane trial (26), whereby an opportunity to assess the interconnectedness of negative mood, meth use, and risky sex was part of the research design. In Fastlane-II, we conducted a randomized trial to evaluate the efficacy of a tri-focal intervention designed to reduce depressive symptoms, meth use, as well as condomless sex among heterosexual women and men. Participants completed a baseline assessment and three post-tests at four month intervals (4, 8, and 12 months). Our primary aims were: 1) To test the overall efficacy of Fastlane at reducing all three outcomes of interest; and 2) Examine gender as a moderator of efficacy. The extant literature suggests that women and men may differ in their motivations to and experience of using meth (16,17,19). However, Fastlane-II was not tailored to be gender-specific. We therefore hypothesized posthoc that gender would moderate intervention efficacy, in that women would benefit more or less than men from the intervention relative to the comparison group. We used latent growth curve modeling (LGCM), which allows for the analysis of changes over time in parallel processes (e.g., reductions in depressive symptoms, meth use, and condomless sex).

Method

Participants

Participants were eligible for the current study if they: were at least 18 years old; tested negative for HIV at intake; were a self-identified heterosexual woman or man; were sexually active with at least one opposite sex partner in the past two months; and had snorted, smoked, or injected meth at least once during the past two months, and at least once during the past 30 days. Participants were excluded if they: were unwilling to participate in the intervention and follow-up assessments; had current major psychiatric diagnosis accompanied by floridly psychotic symptoms and suicidal ideation within the past two weeks; were not sexually active in the past two months or always used condoms; only had condomless sex with a spouse or steady partner in the past two months; were trying to get pregnant or trying to get a partner pregnant; were currently enrolled in a drug treatment program; and scored 3 or less on the Beck Depression Inventory-Fast Screen (BDI-FS), which indicates mild depressive symptoms. A total of 2,189 individuals were screened for eligibility. Figure 1 summarizes reasons for exclusion.

Figure 1.

Figure 1.

Open in a new tab

Diagram summarizing participant flow through the study.

Recruitment

Participants were recruited using the same methods as the original Fastlane trial. Both trials were conducted in San Diego, California. Specifically, community outreach, care and treatment provider networks, participant referrals, posters, “word of mouth,” and walk-ins. Collaboration with community-based agencies that work with substance abuse clients resulted in an effective referral network for potential participants. Providers included substance abuse treatment centers, the needle exchange program, and local case management programs. We also employed a strong social marketing campaign that involved community posters and ads in community newspapers. To emphasize recruitment of ethnic minority participants, we also worked closely with and had the full support of ethnic minority health & social service agencies.

Procedure

Similar to the original trial, Fastlane-II utilized a pre-test, multiple post-test, random assignment, comparison/education-only condition design. Figure 1 summarizes participant flow from screening through follow-up and analysis. After determining eligibility, obtaining written informed consent, and completing a baseline assessment, participants were randomly assigned to one of two conditions/conditions: 1) a behavioral intervention program targeting depressive symptoms, meth use, sexual risk behavior, or 2) an education-only comparison condition focused on information about depression, drug addiction and sexual risk behaviors. In each condition, participants participated in a total of nine hour-long one-on-one counseling sessions. The sessions for the experimental condition are described in detail elsewhere (26). A brief description of the intervention and comparison conditions are described below. Participants completed follow-up assessments at 4, 8, and 12 months. Baseline and follow-up data were gathered using audio computer-assisted self-interviewing technology. In total, 432 eligible women and men were randomized into the Fastlane-II (n=214) or comparison (n=218) condition and were included in data analysis.

Intervention and Comparison-Education Only Conditions

Intervention (Fastlane-II) condition

Individualized counseling sessions formed the core of our intervention approach. Participants in the experimental condition received our nine-session, tri-focal counseling program. In this condition, counselors utilized techniques associated with Cognitive-Behavioral Therapy (CBT) (26,27), Social Cognitive Theory (28), the Theory of Planned Behavior (29), and motivational interviewing (30) to facilitate desired behavior change in three domains (depression/mood, meth use, sexual risk behavior). Each domain included three separate counseling sessions, and each session lasted about one hour. The nine sessions were completed on a weekly basis.

Domain 1: Reducing Depressive Symptoms

Session 1: Developing a treatment plan.

The first three counseling sessions focused on reducing depressive symptoms/negative mood. Counselors employed a structured agenda, with input from participants into their treatment plan. The first session combined a Socratic teaching style with a motivational interviewing approach, which together facilitated the participant’s investment, promoted readiness for change and helped build a client-counselor alliance to support future sessions. First-session discussions focused on building the necessary curriculum-related vocabulary related to the counseling work; hence identifying “hot thoughts,” or triggers of negative mood, learning self-observation skills, and familiarizing oneself about how reactions to past events often resulted in risky behaviors with high habit-strength valences. Participants were taught the “OAC” (observe, assess, and change) technique in relation to negative mood. A list of personal problems related to negative mood was generated and each problem was ranked in terms of treatment goals. Participants were required to complete homework assignments, and track mood related activities between counseling sessions. The latter involved keeping a journal that recorded mood and drug use behaviors between sessions.

Session 2: Skill building.

Once the participant agreed to a counseling agenda, treatment goals, and signed a contract, there was a general shift to a cognitive behavioral style of counseling. This client-centered approach utilized skill building, counselor feedback, and structured exercises to promote behavior change. In session 2 (and all subsequent sessions) motivational interviewing techniques were applied in all cases where clients demonstrated ambivalent feelings related to behavior change. Participants were encouraged to be actively involved in each session and to use their own life experiences in discussions and practice exercises. The counselor and participant conducted exercises that enhanced personal insights into triggers of depressive symptoms. Participants were also taught to reframe their interpretation of a personal event to affect its outcome on mood. The counselor provided skill building and practice in the recognition of new ways to deal with negative mood symptoms.

Session 3: Self-control and self-management of negative mood.

Skill building exercises, counselor feedback, and structured exercises were used to promote self-control and self-management of negative mood. Also, when appropriate, counselors included a motivational interviewing approach to the session. Participants learned the four parts of mood-to-action chain: body arousal, thoughts, feelings, and behaviors. Specific exercises were used to manage negative mood. For example, participants were taught that body arousal can be managed by doing relaxation exercises, or using imagery and symbolism. Additional techniques included talking to a confidant or support buddy and increasing pleasant activities. Depressive symptoms and related problem behaviors were discussed in the context of meth use.

Domain 2: Reducing Meth Use

Session 4: Personal drug history and readiness for change.

In this session, the counselor explored the participant’s current and past history of meth use. The goal of this session was to help participants develop insights regarding motivations and triggers of meth use and to evaluate their readiness to change drug use behavior. Participants were taught to recognize their personal triggers of meth use and related hot thoughts; participants then engaged in exercises that support behavior change (e.g., drug use calendar, thought records, alternative activities list) that helped them to see possible links between their meth use and both negative mood, and high risk sexual behavior. A list of alternative activities regarding trigger situations was generated. The counselor and participant explored ways to employ alternative activities and the participant generated subsequent target goals (e.g., alternative activity instead of using meth, seek a support buddy).

Session 5: Management of drug cravings and urges.

This session focused primarily on management of drug cravings and urges using CBT techniques. Skill building included reframing thoughts, avoiding or leaving a high risk situation, engaging in a distracting activity, and delaying the decision to use. The concept of flawed statements (e.g., I am more fun when I am on meth) and automatic chain sequences were introduced. A list of barriers to reducing/stopping meth use was generated; the counselor and participant worked together to problem-solve each barrier by listing advantages and disadvantages of every solution, and weighing and prioritizing alternatives to select the most promising one. Roleplays were used to problem-solve potential barriers to participants’ target goals.

Session 6: Expanding the CBT “tool box” in pursuit of target goals.

Participants continued to build a “tool box” of resources to help them achieve their target goals in relation to meth use. The “A-B-C-D” method of problem-solving situations was introduced. This method involves talking back to yourself: A is the activating event; B is the belief or thought involved; C is the consequence and D is the way you dispute or talk back to the thought. The counselor used a coaching/teaching style to help participants apply the A-B-C-D method. Participants also reframed faulty interpretations. Eight types of faulty interpretations were discussed and applied to the participant’s own life (e.g., labeling self, all-or-none thinking, emotional reasoning). Participants also learned about assertive communication skills in this session. Assertive communication in the context of meth use focused primarily on ways to deal with negative social influences.

Domain 3: Promoting Safer Sex

Session 7: The context of meth use and unsafe sex.

The primary goal of this session (and this third domain) was to help participants expand the skills from the six previous sessions and apply them toward safer sex practices. A range of motivators and personal triggers of unsafe sex were explored (e.g., negative attitudes toward condoms, safe sex burnout). In addition, the counselor queried the participant as to what extent s/he perceived his/her sexual risk behaviors as a function of meth use. The counselor discussed the participant’s desire and willingness to change his/her high risk sexual practices.

Session 8: Condom use.

The overall goal of this session was to promote consistent use of condoms by applying skills learned previously. Here the four parts of mood-to-action chain - thoughts, feelings, arousal and behavior—were reviewed in the context safer sex application. Social cognitive strategies, which include increasing knowledge, self-efficacy, and positive outcome expectancies in relation to condom use were utilized. Reasons for not using a condom were identified and a problem-solving approach was used to help the participant find ways to manage risky behavior. Roleplay exercises (including modeling and role reversal) were also used to help the participant rehearse sexual situations that were personally relevant.

Session 9: Negotiation of safer sex practices and enhancement of social supports.

The primary goal of this session was to increase participants’ repertoires of safer sex practices, and to enhance negotiation and assertiveness skills through the use of cognitive and behavioral strategies. Building on previous sessions, problem-solving barriers to safer sex negotiation was placed in the context of teaching the participant steps in the negotiation process. The counselor and participant engaged in roleplay exercises, and together they reviewed successes and failures. The counselor also helped participants enhance positive social supports and minimize negative influences in relation to safer sex goals. As in earlier sessions, this was accomplished by problem-solving barriers to social support and helping participants to identify social network persons who would be supportive of safer sex goals. Participants were provided with information on strategies for managing negative influences (e.g., enhancing the participant’s “power of persuasion” skills) and strategies for developing a new network of people who support safer sex goals.

Comparison/Education-only Condition

Participants in the comparison/education-only condition received nine hour-long educational counseling sessions. The condition was manualized and paralleled the intervention condition by having three domains (negative mood, meth usage, and sexual risk). All counselors were trained to deliver both the intervention and comparison protocols. The negative mood modules focused on education about symptoms of depression and included community resource information. The meth usage modules consisted of education materials about methamphetamine, education about 12-step approaches to substance abuse problems and information about health problems related to meth use. The sexual risk models comprised an overview of biological information about sexual transmitted infections (STIs) and its transmission including some educational information related to HIV/AIDS, and a list of available local resources.

Measures

Sociodemographics.

Participants reported their age, race (White, Black, Hispanic/Latino, Asian/Pacific Islander, American Indian, or Mixed Race/Other), country of birth (U.S. vs other), marital status (married or common law marriage vs. single, never married or divorced), whether they resided with a spouse or steady partner (vs. not), whether they were homeless (vs. not), had children (vs. none), education (higher vs. less than a high school education), employment status (currently employed vs. not), income (annual income higher or less than $10K), and whether they were ever convicted of a felony (vs. not).

Meth use.

In six separate items, participants were asked how often they snorted, smoked, injected, injected in combination with another drug, ate or drank, or used meth in any other way in the past two months. Response options were 0=never, 1=once in a while, 2=fairly often, and 3=very often. The items were summed to create a summary score of “frequency of meth use across different methods in the past two months.” Possible scores ranged from 0 to 18. Participants were also asked about their primary method of meth use.

Other substance use.

Participants were asked about their lifetime and past month use of heroin and cocaine in separate items. They also completed the ten-item AUDIT as a measure of hazardous alcohol use. A score of 8 or higher indicates hazardous drinking. (31,32).

Substance-related sexual risks.

In separate items, participants were asked whether they used alcohol, meth, or any other drug before or during sex in the past two months. They were also asked whether they traded sex for meth, and whether they have had sex in a public venue while high on meth, both in the past two months.

Injection-related risks.

Participants were asked whether they have injected with a borrowed needle and/or shared injection paraphernalia, both in the past two months.

Depression.

To measure depressive symptoms, participants were asked to complete the Beck Depression Inventory (BDI-II) scale. A total of 21 items assess depressive symptoms over the past two weeks (including today), and are summed to create a summary score (alpha = 0.91). We used established cutoffs for minimal (0–13), mild (14–19), moderate (20–28), and severe depressive symptoms (29–63) (33).

Sexual risk behavior.

In separate open-ended items, participants reported their total number of vaginal and anal sex acts in the past two months, and how many times they (for men) or their partner (for women) used a condom for vaginal and anal sex. Items distinguished between insertive (for men) and receptive (for women) sex acts. We computed the total number of sex acts in the past two months, and the total number of condomless sex acts in the past two months.

Self-reported STI.

Participants indicated whether they had any of the following STIs in the past two months: gonorrhea, chlamydia, syphilis, genital or anal warts, genital or anal herpes, chancroid, Hepatitis B, Hepatitis C, Trichomoniasis, or any “other venereal disease or sexually transmitted disease or sores, discharges (in, on, or around the penis or anus) not listed previously.”

Data Analysis

Data were analyzed in stages. We examined descriptive statistics at baseline for sociodemographics, meth use, other substance use, depression, risk behavior, and STI. We examined statistics overall and tested for differences by condition and gender. We tested for group differences using chi-square tests for categorical variables and t-tests for continuous variables. We also conducted latent growth curve modeling (LGCM), a statistical method using structural equation modeling (i.e., of observed and latent variables) that allows for the estimation of inter-individual variability in intra-individual patterns of change, or growth, over time. The method has been used to examine changes over time in depression, meth use, and sexual risk behaviors among MSM (24). In intervention efficacy trials including multiple post-tests, both initial status (i.e., baseline) and change over the course of the study are modeled using latent variables for the “intercept” and the “slope,” respectively (34). Growth trajectories can be modeled in different ways, including linear and quadratic slopes. With parallel process growth models, changes in more than one variable or outcome can be modeled as separate, but co-occurring processes (35). As Fastlane-II was designed to simultaneously reduce depressive symptoms, meth use, and condomless sex, parallel process LGCM is an ideal method to examine intervention efficacy, changes over time, and moderation (36).

We used LGCM to estimate a single latent growth curve that included three parallel processes: one for depression (i.e., BDI score), methamphetamine use (i.e., frequency of meth use across different methods in the past two months), and condomless sex (total number of condomless sex acts in the past two months). For each variable of interest, we estimated a latent variable for the intercept factor, linear slope factor, and a quadratic slope factor, which describes any upturn or downturn over time beyond what is predicted by the linear slope factor. We then tested our hypothesis that there were differences between the two conditions in changes over time for depression, meth use, and condomless sex (i.e., efficacy) by regressing the three linear slope latent factors on condition. Finally, we tested our hypothesis that gender moderated intervention efficacy by regressing the intercept and linear slope latent factors of each of the three outcomes on condition, gender, and the condition by gender interaction. Details about any additional analyses are described in the results. Model fit was assessed by evaluating the Comparative Fit Index (CFI), Root Mean Squared Error Approximation (RMSEA) and the Standardized Root Mean Squared Residual (SRMR). We chose recommended cut off values of CFI ≥ 0.95, RMSEA < 0.05, and SRMR ≤ 0.08 to indicate good model fit (37).

Results

Sample characteristics

Sociodemographics.

Table 1 summarizes baseline descriptive statistics among the entire sample, by intervention condition, and by gender. Separated by condition and gender, a total of 85 men and 105 women were in the control condition, and 102 men and 80 women were in the intervention condition. The sample consisted of relatively equal numbers of men (51.6%) and women (48.4%). Mean age was around 38 years of age (SD = 10.1). A third of the participants were of Caucasian/White race (35.2%), and another third were of African-American/Black race (33.8%). Most of the participants were born in the United States (93.3%). A small proportion (7.6%) were married or in a common-law marriage, although a quarter of participants reported residing with their spouse or steady partner (23.1%). Ten percent of the sample were homeless, a majority had at least one child (65.0%), a quarter completed less than a high school education (25.5%) and were currently employed (22.7%). A majority of the participants reported an annual income less than $10,000 USD (70.1%) and reported being ever convicted of a felony (59.3%).

Table 1.

Descriptive statistics of Fastlane II participants by intervention condition and by gender at baseline (n=432)

Variable Total (n=432) Comparison Condition (n=218) Fastlane II Intervention (n=214) p* Women (n=209) Men (n=223) p*
n % n % n % n % n %
Sociodemographics
Gender 0.50 -
 Male 223 51.6% 110 50.5% 113 52.8% - - - -
 Female 209 48.4% 110 50.5% 99 46.3% - - - -
Age (years) (Mean, SD) 37.8 10.1 37.6 10.1 38.1 10.0 0.58 36.4 9.2 39.2 10.7 .003
Race 0.38 0.36
 White 152 35.2% 86 39.4% 66 30.8% 77 36.8% 73 32.7%
 Hispanic/Latino 86 19.9% 40 18.3% 46 21.5% 44 21.1% 39 17.5%
 Black 146 33.8% 73 33.5% 73 34.1% 56 26.8% 78 35.0%
 Asian/Pacific Islander 9 2.1% 5 2.3% 4 1.9% 5 2.4% 4 1.8%
 American Indian 5 1.2% 1 0.5% 4 1.9% 1 0.5% 4 1.8%
 Mixed Race/Other 19 4.4% 15 6.9% 19 8.9% 26 12.4% 25 11.2%
US-born 403 93.3% 202 92.7% 201 93.9% 0.25 193 92.3% 210 94.2% 0.57
Married 33 7.6% 23 10.6% 10 4.7% 0.03 22 10.5% 11 4.9% 0.03
Resides with spouse or steady partner 100 23.1% 55 25.2% 45 21.0% 0.36 42 20.1% 58 26.0% 0.17
Homeless 44 10.2% 20 9.2% 24 11.2% 0.53 30 14.4% 14 6.3% 0.007
Has children 281 65.0% 136 62.4% 145 67.8% 0.16 151 72.2% 130 58.3% 0.002
Less than a high school education 110 25.5% 55 25.2% 55 25.7% 0.83 60 28.7% 50 22.4% 0.15
Employed 98 22.7% 52 23.9% 46 21.5% 0.65 42 20.1% 56 25.1% 0.25
Annual income less than $10K 303 70.1% 153 70.2% 150 70.1% 0.83 157 75.1% 146 65.5% 0.04
Ever convicted of a felony 256 59.3% 132 60.6% 124 57.9% 0.85 105 50.2% 151 68.0% <.001
Meth Use
Frequency of meth use across different methods in past 2 mo. (Mean, SD) Primary method of meth use 4.2 2.0 4.1 2.1 4.3 2.0 .31 4.2 1.9 4.2 2.1 0.80
 Snort 58 13.4% 21 9.6% 37 17.3% 11 5.3% 47 21.1%
 Smoke 326 75.5% 173 79.4% 153 71.5% 178 85.2% 148 66.4%
 Inject 41 9.5% 21 9.6% 20 9.3% 18 8.6% 23 10.3%
 Eat or drink 4 0.9% 3 1.4% 1 0.5% 1 0.5% 3 1.3%
 Other 3 0.7% 0 0.0% 3 1.4% 1 0.5% 2 0.9%
Smoking as primary method of meth use 326 75.7% 173 79.4% 153 71.5% 0.07 178 85.2% 148 66.4% <.001
Other Substance Use
Ever used heroin 140 32.4% 73 33.5% 67 31.3% 0.76 62 29.7% 78 35.0% 0.26
Ever used cocaine 318 73.6% 154 70.6% 164 76.6% 0.10 147 70.3% 171 76.7% 0.16
Used heroin in past month 48 11.1% 24 11.0% 24 11.2% 1.00 22 10.5% 26 11.7% 0.76
Used cocaine in past month 106 24.5% 52 23.9% 54 25.2% 0.74 45 21.5% 61 27.4% 0.18
Hazardous drinker (≥ 8 on AUDIT) 197 45.6% 96 44.0% 101 47.2% 0.44 85 40.7% 112 50.2% 0.05
Substance-related Sexual Risks
Used alcohol before or during sex in past 2 months 276 63.9% 136 62.4% 140 65.4% 0.37 120 57.4% 156 70.0% 0.007
Used meth before or during sex in past 2 months 405 93.8% 208 95.4% 197 92.1% 0.55 194 92.8% 211 94.6% 0.55
Used other drug before or during sex in past 2 months 233 53.9% 118 54.1% 115 53.7% 0.92 116 55.5% 117 52.5% 0.56
Traded sex for meth in past 2 months 121 28.0% 66 30.3% 55 25.7% 0.39 68 32.5% 53 23.8% 0.05
Had sex in a public venue when high on meth in past 2 months 181 41.9% 83 38.1% 98 45.8% 0.08 84 40.2% 97 43.5% 0.50
Injection-related Risks 0.0%
Injected with a borrowed needle in past 2 months 54 12.5% 22 10.1% 32 15.0% 0.11 27 12.9% 27 12.7% 0.88
Shared injection paraphernalia in past 2 months 75 17.4% 36 16.5% 39 18.2% 0.61 38 18.2% 37 16.6% 0.70
Depression (BDI-II) 0.90 <.001
 Minimal 129 29.9% 69 31.7% 60 28.0% 45 21.5% 84 37.7%
 Mild 78 18.1% 39 17.9% 39 18.2% 34 16.3% 44 19.7%
 Moderate 122 28.2% 62 28.4% 60 28.0% 65 31.1% 57 25.6%
 Severe 103 23.8% 50 22.9% 53 24.8% 65 31.1% 38 17.0%
Sexual risk behavior
Total number of sex acts in past 2 mo (Median, IQR) 22.0 10–50 22.0 8–50 23.0 11–54 0.66 20.0 9–45 25.0 12–58.75 0.012
Total condomless sex acts past 2 mo (Median, IQR) 19.0 7–45.5 19.0 6–44.5 18.0 8–49 0.64 16.0 7–39.5 20.0 6.25–54.75 0.008
Self-reported STI in past 2 mo 87 20.1% 41 18.8% 46 21.5% 0.55 58 27.8% 29 13.0% <.001
Open in a new tab
*

Notes: Chi-square tests for categorical variables and t-tests for continuous variables

Meth use.

The mean frequency of meth use across different methods in the past two months was 4.2 (SD = 2.0).

Other substance use and substance-related risk behavior.

In terms of substance use apart from methamphetamine, a third reported lifetime use of heroin (32.4%), whereas three-quarters reported lifetime use of cocaine (73.6%). A small proportion used heroin in the past month (11.1%) and a quarter used cocaine in the past month (24.5%). Almost half of the sample screened positive for hazardous alcohol use (45.6%). Substance-related sexual risk behavior was reported, with almost all (93.8%) reporting using meth before or during sex, 63.9% using alcohol before or during sex, 53.9% reporting using a different drug before or during sex, 28.0% reporting trading sex for meth, and 41.9% reporting having sex in a public venue while high on meth, all in the past two months. Injected-related risk behavior was less common, with 12.5% reporting injecting with a borrowed needle and 17.4% sharing injection paraphernalia, both in the past two months.

Depression and sexual risk behavior.

In terms of depressive symptoms, 29.9% had minimal symptoms, 18.1% mild, 28.2% moderate, and 23.8% severe. In the past two months, average total number of sex acts was 56.6 (SD = 197.5), and average total of condomless sex acts was 48.1 (SD=194.7). One-fifth (20.1%) self-reported having an STI in the past 2 months.

Differences by intervention condition and by gender

A significantly greater proportion of married participants were allocated to the comparison condition (10.6%), compared to the Fastlane II condition (4.7%). The two conditions were equivalent to each other across all the other variables.

There were significant differences between women and men at baseline. Women were more likely than men to report being married (10.5% vs. 4.9%), homeless (14.4% vs. 6.3%), have children (72.2% vs. 58.3%), have an annual income less than $10,000 (75.1% vs. 65.5%), and trade sex for meth in the past two months (32.5% vs. 23.8%). Women were also more likely than men to report smoking as their primary method of meth use (85.2% vs. 66.4%), and report severe depressive symptoms (31.1% vs. 17.0%). Men were more likely than women to report being convicted of a felony (68.0% vs. 50.2%), hazardous alcohol use (50.2% vs. 40.7%), and report using alcohol before or during sex in the past 2 months (70.0% vs. 57.4%). Although men reported a greater number of sex acts and condomless sex acts relative to women (Median = 25.0, IQR = 12–58.75 vs. Median = 20.0, IQR = 9–45, and Median = 20.0, IQR = 6.25–54.75 vs. Median = 16.0, IQR = 7–39.5, respectively), women were more likely than men to report having a recent STI (27.8% vs. 13.0%).

Latent growth curve modeling

Parallel processes model.

We estimated a single latent growth curve model with 3 parallel processes for depression, methamphetamine use, and condomless sex. Table 2 summarizes the results from this parallel process model. The means for the intercept, linear slope, and quadratic slope latent factors were all statistically significant (p < .001). For the intercept factor, results showed that levels of depression, meth use, and condomless sex were significantly different from zero at baseline. For the linear slope factor, results showed that on average, depression, meth use, and condomless sex decreased from baseline to 12-month follow-up. For the quadratic slope factor, results showed that on average, there was a significant upturn in depression, meth use, and condomless sex beyond the decline showed by the linear slope factor. Figure 2 displays the trajectories from baseline through 12-month follow-up in the three outcome variables by intervention condition. We examined associations between the linear slope factors for the three variables. The slope of depression was significantly associated with the slope of meth use (B = 8.61, p < .001), but not with the slope of unprotected sex (B = 90.49, p =0.33). The slope of meth use was significantly associated with the slope of unprotected sex (B = 44.5, p = 0.01). The model fit the data well (χ2 (36) = 69.83, p <.001, CFI =.96, RMSEA = 047, SRMR = .046).

Table 2.

Latent growth curve model results including three parallel processes for depression, meth use, and condomless sex among heterosexual men and women (n=432).

Means Variances
M t p D t p
Intercept Latent Factor
 Depression 20.13 35.16 < .001 38.91 1.77 0.08
 Methamphetamine Use 4.15 43.56 < .001 −0.42 −0.49 0.63
 Condomless Sex 70.10 15.36 < .001 1101.34 0.88 0.38
Linear Slope Latent Factor
 Depression −7.39 −9.73 < .001 9.13 0.29 0.77
 Methamphetamine Use −1.56 −9.78 < .001 −1.70 −1.29 0.20
 Condomless Sex −44.45 −7.43 < .001 386.36 0.20 0.84
Quadratic Slope Latent Factor
 Depression 1.58 6.90 < .001 1.39 0.58 0.57
 Methamphetamine Use 0.40 7.23 < .001 0.03 0.27 0.79
 Condomless Sex 11.68 6.62 < .001 161.92 1.14 0.26
Open in a new tab
Figure 2.

Figure 2.

Open in a new tab

Growth (change) trajectories from baseline through 12-month follow-up in the three outcome variables by intervention condition.

Main effect of intervention condition on changes over time.

Next, we tested intervention condition as a predictor of the three slope latent factors. Overall, on average there were no significant differences between the comparison condition and Fastlane II intervention condition in changes over time in depression (b = −0.03, p = 0.95), meth use (b = −0.02, p=0.81), or condomless sex (b= 0.44, p=0.85).

Gender moderating intervention efficacy.

We then tested our hypothesis that gender moderated intervention efficacy. We regressed the intercept and linear slope latent factors of each of the three outcomes on condition, gender, and the condition by gender interaction. This model fit the data well (χ2 (60) = 129.41, p <.001, CFI =.92, RMSEA =.052, SRMR = .054). Results are displayed in Table 3. There were no significant effects on the intercept or linear slope latent factors for condomless sex. For the intercept latent factor for depression, there was a significant gender effect such that at baseline, men reported lower levels of depression than women. There were no significant effects on the linear slope factor for depression. For methamphetamine use, there was a significant of effect of gender on the slope latent factor such that on average, men showed greater declines in meth use compared to women. There was also a significant condition by gender interaction on the slope latent factor for meth use.

Table 3.

Latent growth curve modeling results of condition, gender, and the condition by gender interaction effects on the intercept and slope latent factors for depression, meth use, and condomless sex (n=432).

Independent Variable: Condition Gender Condition X Gender Interaction
B t p B t p B t p
Intercept Latent Factor
 Depression −0.57 −0.39 0.70 −5.41 −3.78 <.001 0.24 0.12 0.91
 Methamphetamine Use −0.05 −0.18 0.86 −0.32 −1.21 0.22 0.32 0.84 0.40
 Condomless Sex −8.20 −0.97 0.33 4.85 0.58 0.56 7.68 0.64 0.52
Linear Slope Latent Factor
 Depression −0.15 −0.28 0.82 0.24 0.37 0.71 0.60 0.66 0.51
 Methamphetamine Use 0.17 1.27 0.20 0.39 2.88 0.004 −0.49 −2.58 0.01
 Condomless Sex 2.65 0.66 0.51 4.93 1.25 0.21 −3.17 −0.57 0.57
Open in a new tab

We tested an equivalent model that controlled for potential confounding variables. Since women and men significantly differed at baseline on a number of characteristics (i.e., marital status, homelessness, having children, income, felony conviction, smoking is primary method of meth use, hazardous drinking, alcohol use during sex, trading sex for meth, and STI) that may account for the results, we included paths from these variables to the intercept and linear slope latent factors for meth use. Even after controlling for these variables, the interaction effect between condition and gender on the linear slope factor for meth use remained statistically significant (B = −0.42, t = −2.31, p = 0.02).

Post-hoc analyses

We conducted stratified analyses between women and men to probe the condition by gender interaction. Specifically, we tested the effect of condition on the linear slope latent factor for methamphetamine use separately for women and for men. Figure 3 displays the trajectories from baseline to 12-month follow-up in meth use by condition, separately for women and for men.

Figure 3.

Figure 3.

Open in a new tab

Growth (change) trajectories from baseline to 12-month follow-up in methamphetamine use by condition, separately for women and for men.

Women only.

Among women, condition did not significantly affect changes in meth use over time (b= 0.13, p=0.16).

Men only.

Among men, there was a significant effect of condition on changes in meth use over time (b= −0.32 p=0.01), indicating that men in the Fastlane II intervention condition showed greater decreases in meth use compared to men in the comparison condition. In the same model, we also found that among men, decreases in meth use were significantly correlated with decreases in condomless sex (b = 5.27, p= 0.01), but not with decreases in depression scores (b= 0.20, p = 0.43). We also found a marginal trend such that the upturn in meth use appeared to differ by condition (b= −0.09, p= 0.06), with those men in the Fastlane II intervention showing less of an upturn in meth use at 8- and 12-month follow-up compared to men in the comparison condition.

Discussion

Individuals who use meth are at an increased risk for HIV and other STIs. Apart from the association between meth use and risky sexual behavior, research has also demonstrated the co-occurrence of meth use with depression. The overall goal of the current paper was to evaluate the efficacy of Fastlane-II, a tri-focal behavioral intervention designed to reduce depressive symptoms, meth use, and condomless sex among meth-using heterosexual women and men. Another aim of the paper was to test whether women and men differed in their response to the intervention, in other words whether gender moderated efficacy. Using LGCM techniques, we found that the intervention and comparison groups did not differ in changes over time in depression, meth use, or condomless sex. We did find that gender moderated efficacy with respect to meth use, such that among men, those randomized into the Fastlane-II intervention exhibited greater reductions in meth use compared to those in the comparison condition. Reductions in meth use among men was associated with concomitant reductions in condomless sex, but not depression. Among women, there were no differences in changes over time in any of the three outcomes between the two conditions.

Analysis using LGCM revealed that participants in both the intervention and comparison conditions exhibited improvements in health outcomes over the course of the study, including reductions in depressive symptoms, meth use, and condomless sex. Previous studies in the area of health research have found similar patterns of improvements among control group participants (38). While one reason may be measurement reactivity, a type of Hawthorne effect in which the mere act of completing measures can lead to changes in the behaviors being measured (39), one would assume that educational conversations between a client and the counselor on topics related to the study’s outcomes would only enhance the possible Hawthorne effect. This effect may be especially pronounced among vulnerable populations like the one studied here. A majority of the participants reported a low annual income (i.e., less than $10,000, a history of incarceration (i.e., being convicted of a felony), and were currently experiencing moderate to severe depression. Counselors for both the intervention and comparison groups may have been perceived as a source of social support. Indeed, all participants completed nine sessions with their counselor during which the counselor exhibited obvious and explicit interest in the participant’s health and well-being. In sum, the current efficacy trial employed rigorous standards by including an education-only comparison group, which may have led to the lack of difference between arms.

In spite of the rigor, we were able to demonstrate efficacy at reducing meth use among men, which was related to reductions in condomless sex (but not depression). Unlike men, the health outcomes of women did not benefit from the intervention. In this study, women were more likely than men to be homeless, have dependent children, have a lower income, have more moderate and severe depression, and trade sex for meth. Although this was not measured, women may have also been more likely than men to trade sex for money or shelter. Thus, the women in this study appeared to experience greater social and structural vulnerabilities, and perhaps needed greater or different intervention attention compared to the men.

Previous research with women who use meth conducted by Lorvick (17) has demonstrated that women do not often concern themselves with the sexual risks involved when using meth. Instead, they emphasize how meth increases their sexual desire, allowing them to experience more intense and prolonged sexual activity. They also describe more sexual pleasure while on meth, with some describing how achieving orgasm was only possible while on meth. Finally, women describe how meth allows them to feel uninhibited, empowered, and agentic. While these experiences can also be found among men who use meth, the experience likely differs when one considers how women are generally stereotyped, treated as, and are expected to be sexually submissive, lacking desire and agency in sex compared to men (40–43). In order to effectively reduce meth use among women, interventions must recognize women’s desire or need to maintain the pleasure, empowerment, and agency they feel when using meth that they may not otherwise experience.

Not addressed and therefore a limitation of the current study - women also experience a more unique vulnerability to intimate partner violence compared to men. Although both women and men can experience violence as both perpetrators and targets, violence against women poses increased and exacerbated risks for HIV (44,45). In addition, compared to men, heterosexual women are also more likely to experience co-morbid substance use, violence, depression, and sexual risks (46). Women who use meth are more likely to experience violence while high or during withdrawal (47). Gender-specific interventions targeting meth use and/or HIV risk behavior should take into account the role of violence (48). Another limitation of the study that is worth noting is that the findings are limited to a two-month recall period, thus failing to capture the totality of behaviors in the four months between assessments.

Researchers have argued that HIV prevention interventions for substance-using women must take into account the individual and environmental factors that promote both substance use and HIV risk behavior for women. These factors include gendered power dynamics within heterosexual partnerships, sex work, and criminalization (49). While taking such factors into account, interventions can also be developed that employ female-initiated prevention methods. In a qualitative study of women with a history of methamphetamine use and partner violence (50), women were not interested in using female condoms, but were interested in vaginal microbicides. The latter was viewed as convenient, and also offered a sense of control and empowerment. Future research should examine whether other biomedical HIV prevention strategies, including pre-exposure prophylaxis, or PrEP, are feasible and acceptable to use among women who use meth. Overall, the current research helps to point to the different needs of heterosexual women and men when reducing HIV risks, depression, and meth use.

Acknowledgments

Compliance with Ethical Standards:

This study was funded by National Institute of Mental Health grant R01MH061146. Preparation of this manuscript was supported by a Mentored Career Development Award from the National Institute on Drug Abuse to the first author (K01DA036447).

Footnotes

All authors have no conflicts of interest to declare.

All procedures performed in this study involving human participants were in accordance with the ethical standards of the institutional research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. This article does not contain any studies with animals performed by any of the authors.

References

  • 1.Colfax G, Shoptaw S. The methamphetamine epidemic: Implications for HIV prevention and treatment. Curr HIV/AIDS Rep. 2005. December 1;2(4): 194–9. [DOI] [PubMed] [Google Scholar]
  • 2.Halkitis PN, Parsons JT, Stirratt MJ. A double epidemic: crystal methamphetamine drug use in relation to HIV transmission. J Homosex. 2001;41(2): 17–35. [DOI] [PubMed] [Google Scholar]
  • 3.Chesney MA, Barrett DC, Stall R. Histories of substance use and risk behavior: precursors to HIV seroconversion in homosexual men. Am J Public Health. 1998;88(1): 113–116. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Cohen J Life in the Fast Lane: HIV and Meth. Science. 2012. July 13;337(6091): 176—7. [DOI] [PubMed] [Google Scholar]
  • 5.Plankey MW, Ostrow DG, Stall R, Cox C, Li X, Peck JA, et al. The relationship between methamphetamine and popper use and risk of HIV seroconversion in the multicenter AIDS cohort study. J Acquir Immune Defic Syndr 1999. 2007. May 1;45(1):85—92. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Semple SJ, Zians J, Strathdee SA, Patterson TL. Sexual Marathons and Methamphetamine Use among HIV-Positive Men Who Have Sex with Men. Arch Sex Behav. 2009. August 1;38(4):583. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Stone E, Heagerty P, Vittinghoff E, Douglas JM, Koblin BA, Mayer KH, et al. Correlates of condom failure in a sexually active cohort of men who have sex with men. JAIDS J Acquir Immune Defic Syndr. 1999;20(5):495–501. [DOI] [PubMed] [Google Scholar]
  • 8.Mausbach BT, Semple SJ, Strathdee SA, Zians J, Patterson TL. Efficacy of a behavioral intervention for increasing safer sex behaviors in HIV-positive MSM methamphetamine users: results from the EDGE study. Drug Alcohol Depend. 2007;87(2):249–257. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Reback CJ, Grant DL, Fletcher JB, Branson CM, Shoptaw S, Bowers JR, et al. Text messaging reduces HIV risk behaviors among methamphetamine-using men who have sex with men. AIDS Behav. 2012. October; 16(7): 1993–2002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Shoptaw S, Reback CJ. Associations between methamphetamine use and HIV among men who have sex with men: a model for guiding public policy. J Urban Health. 2006;83(6): 1151–1157. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Shoptaw S, Reback CJ. Methamphetamine use and infectious disease-related behaviors in men who have sex with men: implications for interventions. Addiction. 2007;102(s1):130–135. [DOI] [PubMed] [Google Scholar]
  • 12.Corsi KF, Booth RE. HIV Sex Risk Behaviors Among Heterosexual Methamphetamine Users: Literature Review from 2000 to Present. Curr Drug Abuse Rev. 2008. November 1;1 (3):292–6. [DOI] [PubMed] [Google Scholar]
  • 13.Hittner JB. Meta-analysis of the association between methamphetamine use and high-risk sexual behavior among heterosexuals. Psychol Addict Behav. 2016;30(2): 147–57. [DOI] [PubMed] [Google Scholar]
  • 14.Molitor F, Truax SR, Ruiz JD, Sun RK. Association of methamphetamine use during sex with risky sexual behaviors and HIV infection among non-injection drug users. West J Med. 1998. February;168(2):93–7. [PMC free article] [PubMed] [Google Scholar]
  • 15.Melis MR, Argiolas A. Dopamine and sexual behavior. Neurosci Biobehav Rev. 1995; 19(1): 19–38. [DOI] [PubMed] [Google Scholar]
  • 16.Semple SJ, Grant I, Patterson TL. Female methamphetamine users: social characteristics and sexual risk behavior. Women Health. 2004;40(3):35–50. [DOI] [PubMed] [Google Scholar]
  • 17.Lorvick J, Bourgois P, Wenger LD, Arreola SG, Lutnick A, Wechsberg WM, et al. Sexual pleasure and sexual risk among women who use methamphetamine: A mixed methods study. Int J Drug Policy. 2012. September 1;23(5):385–92. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Maher L, Phlong P, Mooney-Somers J, Keo S, Stein E, Couture MC, et al. Amphetamine-type stimulant use and HIV/STI risk behaviour among young female sex workers in Phnom Penh, Cambodia. Int J Drug Policy. 2011. May 1;22(3):203–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Brecht M- L, O’Brien A, Von Mayrhauser C, Anglin MD. Methamphetamine use behaviors and gender differences. Addict Behav. 2004;29(1):89–106. [DOI] [PubMed] [Google Scholar]
  • 20.Lorvick J, Martinez A, Gee L, Kral AH. Sexual and Injection Risk among Women who Inject Methamphetamine in San Francisco. J Urban Health. 2006. May 1; 83(3):497—505. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Mausbach BT, Semple SJ, Zians J, Patterson TL, Strathdee SA. Efficacy of a behavioral intervention for increasing safer sex behaviors in HIV-negative, heterosexual methamphetamine users: results from the Fast-Lane Study. Ann Behav Med. 2007;34(3):263–274. [DOI] [PubMed] [Google Scholar]
  • 22.Mimiaga MJ, Reisner SL, Pantalone DW, O’Cleirigh C, Mayer KH, Safren SA. A pilot trial of integrated behavioral activation and sexual risk reduction counseling for HIV-uninfected men who have sex with men abusing crystal methamphetamine. AIDS Patient Care STDs. 2012. November;26(11):681—93. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Shoptaw S, Reback CJ, Peck JA, Yang X, Rotheram-Fuller E, Larkins S, et al. Behavioral treatment approaches for methamphetamine dependence and HIV-related sexual risk behaviors among urban gay and bisexual men. Drug Alcohol Depend. 2005. May 9;78(2): 125–34. [DOI] [PubMed] [Google Scholar]
  • 24.Jaffe A, Shoptaw S, Stein JA, Reback CJ, Rotheram-Fuller E. Depression ratings, reported sexual risk behaviors, and methamphetamine use: Latent growth curve models of positive change among gay and bisexual men in an outpatient treatment program. Exp Clin Psychopharmacol. 2007;15(3):301–7. [DOI] [PubMed] [Google Scholar]
  • 25.Reback CJ, Larkins S, Shoptaw S. Changes in the meaning of sexual risk behaviors among gay and bisexual male methamphetamine abusers before and after drug treatment. AIDS Behav. 2004. March;8(1):87–98. [DOI] [PubMed] [Google Scholar]
  • 26.Pitpitan EV, Semple SJ, Zians J, Strathdee SA, Patterson TL. Cognitive Behavioral Therapy for HIV-Negative Heterosexual Meth Users: The FASTLANE II Intervention. Cogn Behav Pract. 2014;21(2): 176–190. [Google Scholar]
  • 27.Beck JS. Cognitive-behavioral therapy. Prim Psychiatr. 2006; 13(4):31–34. [Google Scholar]
  • 28.Bandura A Social cognitive theory: An agentic perspective. Annu Rev Psychol. 2001;52(1): 1– 26. [DOI] [PubMed] [Google Scholar]
  • 29.Ajzen I, Fishbein M. Understanding attitudes and predicting social behavior. Vol. 278 Prentice-Hall; 1980. [Google Scholar]
  • 30.Miller WR, Rollnick S. Motivational interviewing: Preparing people to change addictive behavior. [Internet]. Guilford press; 1991. [cited 2013 Sep 27]. Available from: http://psycnet.apa.org/psycinfo/1991-98398-000 [Google Scholar]
  • 31.Conigrave KM, Hall WD, Saunders JB. The AUDIT questionnaire: choosing a cut-off score. Addiction. 1995;90(10): 1349–1356. [DOI] [PubMed] [Google Scholar]
  • 32.Saunders JB, Aasland OG, Babor TF, de la Fuente JR, Grant M. Development of the Alcohol Use Disorders Identification Test (AUDIT): WHO Collaborative Project on Early Detection of Persons with Harmful Alcohol Consumption--II. Addict Abingdon Engl. 1993. June;88(6):791–804. [DOI] [PubMed] [Google Scholar]
  • 33.Beck AT, Steer RA, Brown GK. Manual for the BDI-II. San Antonio, TX: Psychological Corporation; 1996. [Google Scholar]
  • 34.Wang J, Siegal HA, Falck RS, Carlson RG, Rahman A. Evaluation of Hiv Risk Reduction Intervention Programs Via Latent Growth Model. Eval Rev. 1999. December 1;23(6):648–62. [DOI] [PubMed] [Google Scholar]
  • 35.Roesch SC, Norman GJ, Adams MA, Kerr J, Sallis JF, Ryan S, et al. Latent Growth Curve Modeling of Adolescent Physical Activity: Testing Parallel Process and Mediation Models. J Health Psychol. 2009. March 1; 14(2):313–25. [DOI] [PubMed] [Google Scholar]
  • 36.Beauchaine TP, Webster-Stratton C, Reid MJ. Mediators, moderators, and predictors of 1-year outcomes among children treated for early-onset conduct problems: a latent growth curve analysis. J Consult Clin Psychol. 2005;73(3):371. [DOI] [PubMed] [Google Scholar]
  • 37.Hu L, Bentler PM. Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Struct Equ Model Multidiscip J. 1999;6(1): 1—55. [Google Scholar]
  • 38.Becker H, Roberts G, Voelmeck W. Explanations for Improvement in Both Experimental and Control Groups. West J Nurs Res. 2003. October;25(6):746–55. [DOI] [PubMed] [Google Scholar]
  • 39.Capellan J, Wilde MH, Zhang F. Measurement Reactivity in a Randomized Clinical Trial Using Self-Reported Data. J Nurs Scholarsh. 2017. January 1;49(1): 111—9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Hundhammer T, Mussweiler T. How sex puts you in gendered shoes: Sexuality-priming leads to gender-based self-perception and behavior. J Pers Soc Psychol. 2012;103(1): 176–93. [DOI] [PubMed] [Google Scholar]
  • 41.Kiefer AK, Sanchez DT, Kalinka CJ, Ybarra O. How Women’s Nonconscious Association of Sex with Submission Relates to Their Subjective Sexual Arousability and Ability to Reach Orgasm. Sex Roles. 2006. July 1;55(1–2):83–94. [Google Scholar]
  • 42.Rudman LA, Glick P. Prescriptive Gender Stereotypes and Backlash Toward Agentic Women. J Soc Issues. 2001. January;57:743–62. [Google Scholar]
  • 43.Sanchez DT, Kiefer AK, Ybarra O. Sexual Submissiveness in Women: Costs for Sexual Autonomy and Arousal. Pers Soc Psychol Bull. 2006. April 1;32(4):512–24. [DOI] [PubMed] [Google Scholar]
  • 44.Dunkle KL, Jewkes RK, Brown HC, Gray GE, McIntryre JA, Harlow SD. Gender-based violence, relationship power, and risk of HIV infection in women attending antenatal clinics in South Africa. The lancet. 2004;363(9419):1415–1421. [DOI] [PubMed] [Google Scholar]
  • 45.Jewkes RK, Dunkle K, Nduna M, Shai N. Intimate partner violence, relationship power inequity, and incidence of HIV infection in young women in South Africa: a cohort study. Lancet. 2010. July 3;376(9734):41–8. [DOI] [PubMed] [Google Scholar]
  • 46.Tsuyuki K, Pitpitan EV, Levi-Minzi MA, Urada LA, Kurtz SP, Stockman JK, et al. Substance Use Disorders, Violence, Mental Health, and HIV: Differentiating a Syndemic Factor by Gender and Sexuality. AIDS Behav. 2017. August;21(8):2270–82. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47.Abdul-Khabir W, Hall T, Swanson A- N, Shoptaw S. Intimate Partner Violence and Reproductive Health Among Methamphetamine-Using Women in Los Angeles: A Qualitative Pilot Study. J Psychoactive Drugs. 2014. August 8;46(4):310–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48.Stockman JK, Ludwig-Barron N, Hoffman MA, Ulibarri MD, Dyer TVP. Prevention interventions for human immunodeficiency virus in drug-using women with a history of partner violence. Subst Abuse Rehabil. 2012. February 1;3(Suppl 1):45—57. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49.Auerbach JD, Smith LR. Theoretical Foundations of Research Focused on HIV Prevention Among Substance-involved Women: A Review of Observational and Intervention Studies. J Acquir Immune Defic Syndr 1999. 2015. June 1;69(Suppl 2):S146–54. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 50.Stockman JK, Syvertsen JL, Robertson AM, Ludwig-Barron NT, Bergmann JN, Palinkas LA. Women’s Perspectives on Female-Initiated Barrier Methods for the Prevention of HIV in the Context of Methamphetamine Use and Partner Violence. Womens Health Issues. 2014. July 1;24(4):e397–405. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES