Abstract
This study used a stage-based approach to understand condom use behavior in a representative sample of 309 sexually active homeless youth recruited from shelters, drop-in centers, and street sites in Los Angeles County. Focusing on the youth’s most recent sexual event, the three stages of condom use examined were: (1) whether the partners decided prior to the event about using condoms; (2) whether a condom was available at the event; and (3) whether a condom was used at the event. Logistic regression analysis was used to identify attitudinal, relationship, and contextual correlates of each of these three stages. Deciding ahead of time about condom use was associated with being Hispanic, level of education, condom attitudes, and various relationship characteristics (e.g,. partner type, monogamy, relationship abuse), with the nature of these associations varying depending on the type of decision (i.e., deciding to use, deciding to not use). Condom availability was more likely to be reported by males, if the event was described as being special in some way, or if the event lacked privacy. Condom use was more likely among youth with more positive condom attitudes and among youth who decide ahead of time to use a condom, but less likely among those in monogamous relationships or when hard drugs were used prior to sex. Whether sexual intercourse is protected or unprotected is the end result of a series of decisions and actions by sexual partners. Results from this study illustrate how condom use can be better understood by unpacking the stages and identifying influential factors at each stage. Each stage may, in and of itself, be an important target for intervention with homeless youth.
Keywords: Homeless, youth, condom use, event-level, stages
Condom use is an effective way of stemming the spread of HIV/AIDS among homeless youth, yet a recent study found that only 40% of sexually active homeless youth reported using a condom the last time they had sex (Tucker et al., 2012). While there is ample literature on the correlates of condom use among homeless youth, no study has examined the three distinct, temporally staggered stages of condom use: (1) whether the partners make a decision prior to the sexual event about whether to use a condom; (2) whether a condom is available at the event; and (3) whether a condom is actually used at the event. A better understanding of these three stages of condom use is essential to developing a more complete understanding of this protective behavior and to the development of effective, practical HIV prevention interventions for homeless youth.
Although many attitudinal, relationship and contextual factors have been associated with condom use among homeless youth (Alvarez, Villarruel, Zhou, & Gallegos, 2010; Matson, Adler, Millstein, Tschann, & Ellen, 2011; Outlaw, Naar-King, Janisse, & Parsons, 2010; Steinberg, Grella, Boudov, Kerndt, & Kadrnka, 2011; Tucker, et al., 2012), little is known about whether these factors, many of which precede sexual events in time, are associated with the two stages of condom use that also precede sexual events in time: deciding ahead of time about condom use and having a condom available. Understanding factors that predict decisions made prior to a sexual event are important because prior work suggests that discussions or decisions made prior to sexual events (or the absence of these discussions) may influence whether condoms are used during the event (Bailey et al., 1998; Tucker et al., 2012), and some studies show that an intention to use condoms predicts condom use (Alvarez, et al., 2010; Gu et al., 2009; Janepanish, Dancy, & Park, 2011; Mausbach, Semple, Strathdee, & Patterson, 2009). Further, the availability of a condom may follow a discussion about whether or not to use a condom and may have its own distinct set of predictors. Lack of availability of condoms is cited by youth as one of the primary reasons for not using condoms (Crosby, et al., 2004; Crosby, Sanders, Yarber, & Graham, 2003). With few exceptions (Song et al., 2009), studies have not taken into account either prior decision-making about condom use or their availability at the event.
This study presents a stage-based approach to understanding event-level condom use in a representative sample of homeless youth and poses the following questions: (1) What attitudinal, contextual and relationship factors are associated with deciding ahead of time to use or to not use a condom? (2) What factors are associated with having a condom available at the most recent sexual event? (3) What factors are associated with condom use at the most recent sexual event, taking into account prior decision-making about condom use and condom availability?
Method
Participants
Data come from a larger study of 419 homeless youth in Los Angeles County examining the social context of substance use and sexual risk behavior. Youth were eligible if they: a) were between the ages of 13–24; b) were not currently living with a parent or guardian; c) were not getting most of their support for food and housing from family or a guardian; d) spent the previous night in a shelter, outdoor or public place, hotel or motel room rented with friends (because of no place else to go), or other place not intended as a domicile; and e) were English speaking. The analyses use three overlapping sub-samples of youth who were sexually active in the past 3 months. All 309 sexually active youth were included in analyses of deciding ahead of time about condom use. Analyses of condom availability were based on 279 youth who had not decided with their partner prior to the event to use a condom (deciding ahead of time to use a condom was perfectly predictive of having a condom available at the event, as illustrated in Figure 1). Analyses of condom use were based on 211 youth who had not decided prior to the event to have unprotected sex (deciding ahead of time to have unprotected sex was perfectly predictive of having unprotected sex) and reported that a condom was available.
Figure 1.
Tree diagram of prior decision making about condom use, condom availability, and condom use at most recent event (unweighted; N = 309)
Study Design
To obtain a representative sample of homeless youth from the greater Los Angeles area, we adopted a two-stage design that involved first developing a list/sampling frame of sites used by homeless youth and then drawing a probability sample of youth from each site. We developed two sampling frames of sites: one for eligible service sites (shelters and drop-ins) and the other for street venues in the study area. Our final list of sites consisted of 22 services sites (15 shelters, 7 drop-in centers) and 19 street venues. More details on the sampling design can be found elsewhere (Wenzel, Tucker, Golinelli, Green, & Zhou, 2010).
Study Variables
Outcomes
The three outcome variables were: (1) whether the couple decided about condom use prior to the event (categories: did not decide ahead of time whether to use condoms, decided ahead of time to use condoms, and decided ahead of time to not use condoms); (2) whether condoms were available; and (3) whether a condom was used.
Demographic covariates
These included biological sex, age, race/ethnicity, and education.
Attitudes about condoms, HIV susceptibility, and pregnancy
Positive condom attitudes and perceived susceptibility to HIV were each assessed with four items that were rated on a 4-point scale (strongly disagree to strongly agree; α = .62 and .56, respectively; see Tucker et al., 2012 for more information). Two items assessed attitudes towards pregnancy (Frost, Singh, & Finer, 2007) using a 4-point scale: how they would feel if they found out today that they were pregnant/got a girl pregnant (very upset to very pleased) and thinking about their life right now, how important is it to avoid getting pregnant/getting a girl pregnant (very important to not at all important). These two items correlated r = .38 and were averaged.
Relationship characteristics
These dichotomous variables included whether the most recent partner was a spouse or boyfriend/girlfriend (vs. casual/need-based); whether the relationship was monogamous; whether the partner was same-sex (males only); and whether there was a history of physical, verbal or psychological abuse.
Contextual characteristics
Dichotomous items were developed based on qualitative interviews with a separate sample of 30 homeless youth to identify important contextual features of sexual events associated with having unprotected sex: (1) Was the event a chance meeting or had they planned to meet up; (2) Was there something different or special about this event (e.g., they had not seen each other in a long time, one of them was going away for awhile, one of them had just gotten paid); (3) Did they have sex in a place where they felt a need for more privacy or were concerned that others might see, hear or interrupt them during sex; and (4) Did they talk before or while having sex about “pulling out” before ejaculation. In addition, we assessed the maximum number of drinks consumed by either partner during the event, and used two dummy-coded variables to assess drug use: whether the event involved marijuana use only (vs. no drug use) and whether the event involved any hard drug use (vs. no drug use) by either partner.
Statistical Methods
The use of a disproportionate random sampling technique and differential nonresponse rates require the use of design and nonresponse weights to represent the target population from the sample of respondents. All analyses incorporate these weights and account for the modest design effect that they induce, using the linearization method (Skinner, 1989). We conducted multinomial logistic regression analyses to examine correlates of deciding ahead of time to use a condom and deciding ahead of time not to use a condom (compared with not making any decisions ahead of time), and logistic regression analyses to examine correlates of condom availability and condom use. We first fit bivariate models for each outcome to explore the association between each predictor and the outcome variable, and then modeled a multivariate model for each outcome that included only variables that were associated at p < .10 in the bivariate tests.
Demographic, attitudinal, and relationship characteristics were included in models for each outcome variable. Contextual factors that could have preceded the event in time were included in the condom availability models and those factors, plus contextual factors that occurred concurrently with the sexual event, were included in the condom use models. In addition, our condom availability models included an indicator of whether the youth decided ahead of time not to use a condom (vs. did not decide ahead of time), and our condom use models included an indicator of whether the youth decided ahead of time to use a condom (vs. did not decide ahead of time). Recall that all youth who decided ahead of time to use a condom had a condom available (and therefore could not be included in the condom availability models) and that all youth who decided not to use a condom did not use a condom (and therefore could not be included in the condom use models).
Results
Descriptive Statistics
The majority of youth (72.5%) reported not making a decision about condom use prior to the event, 10.7% decided ahead of time to use condoms during the event, and 16.9% decided ahead of time not to use condoms (weighted percentages). As shown in Figure 1, of the 30 youth who made a decision ahead of time to use condoms, all of them reported having a condom available and 25 of them actually used a condom during the event. Of the 44 youth who decided ahead of time not to use a condom, 20 reported having a condom available, but none of those who had a condom available actually used a condom during the event (unweighted ns).
Table 1 presents descriptive statistics for the study variables. The sample was 64% male and primarily White (37%), African American (23%), or Hispanic (18%). Half of the youth did not have a high school diploma. On average, youth had fairly positive attitudes towards condoms (mean = 3.09, range = 1.0 to 4.0). Perceived vulnerability to HIV scores ranged from 1.0 to 4.0, with a mean of 2.33. More than half of the youth (55%) reported on an event with a primary partner and 54% of youth indicated they were in a monogamous relationship. Over one-third (40%) of the youth had a history of abuse with their partner. In describing the most recent sexual event, 35% of youth indicated that it was a “chance meeting,” 42% that it was special in some way, and 38% that it took place in a location that lacked privacy. The mean number of drinks consumed by either partner during the event was 2.17 (SD 6.69). Drug use before the event was reported by 34% of youth in the case of any drug use and 8% in the case of hard drug use (e.g., crack, cocaine, methamphetamine, hallucinogens).
Table 1.
Descriptive Statistics for Study Variables (Weighted, N = 309)
| Variable | Mean (SD) or Percent |
|---|---|
| Demographic Covariates | |
| Biological sex = male | 63.67 |
| Race/ethnicity: | |
| Non-Hispanic white | 37.09 |
| African American | 23.16 |
| Hispanic | 18.34 |
| Asian/Other | 21.40 |
| Education: | |
| Less than high school graduate/GED | 49.80 |
| High school graduate/GED | 29.06 |
| Some college or higher | 21.13 |
| Age (in years) | 20.40 (2.27) |
| Attitudinal Characteristics | |
| Positive condom attitudes (range = 1.0 to 4.0) | 3.09 (0.66) |
| Perceived HIV susceptibility (range = 1.0 to 4.0) | 2.33 (0.95) |
| Pregnancy concerns (range = 1.0 to 4.0) | 3.08 (0.91) |
| Relationship Characteristics | |
| Type of partner: | |
| Primary | 55.27 |
| Casual or need-based | 44.73 |
| Monogamous relationship | 54.44 |
| History of relationship abuse | 39.71 |
| Same sex partner | 11.47 |
| Contextual Factors | |
| Chance meeting | 35.25 |
| Event was special in some way | 41.55 |
| Sex was in a non-private place | 38.16 |
| Maximum number of drinks consumed during event | 2.17 (4.71) |
| Marijuana use only (not hard drugs) during event | 25.98 |
| Any hard drug use during event | 7.60 |
| Did not decide ahead of time about condom use (n = 235) | 72.46 |
| Decided ahead of time to use condoms (n = 30) | 10.69 |
| Decided ahead of time to not use condoms (n = 44) | 16.86 |
Three-Stage Multivariate Models
Correlates of deciding ahead of time about condom use
Table 2 shows results from the multinomial logistic regression analyses examining correlates of deciding to use or not use condoms (vs. making no decision ahead of time). Youth were more likely to report that they decided ahead of time to use condoms (compared to making no decision) if they held more positive attitudes about condoms and were in a monogamous relationship, but less likely if they were Hispanic (vs. White) and lacked a high school education. Deciding ahead of time to not use condoms (compared to no decision) was more likely among youth in a monogamous relationship or one with a history of relationship abuse, but less likely among youth with more positive attitudes towards condoms and who had sex with a casual partner.
Table 2.
Multivariate Multinomial Logistic Regression Model Predicting Deciding Ahead of Time About Condom Use (Full sample; N = 309)
| Variable | Decided ahead of time to use a condom a | Decided ahead of time to not use a condom a | ||||||
|---|---|---|---|---|---|---|---|---|
|
| ||||||||
| OR | 95% CI | p = | OR | 95% CI | p = | |||
| Demographic Covariates | ||||||||
| Biological sex: male | 0.97 | 0.34 | 2.82 | 0.96 | 0.88 | 0.36 | 2.18 | 0.79 |
| Black (vs. white) | 1.01 | 0.30 | 3.43 | 0.99 | 0.43 | 0.12 | 1.57 | 0.20 |
| Hispanic (vs. white) | 0.19 | 0.04 | 0.90 | 0.04 | 0.91 | 0.31 | 2.73 | 0.87 |
| Other/Mixed (vs. white) | 0.83 | 0.22 | 3.10 | 0.78 | 0.76 | 0.26 | 2.27 | 0.62 |
| Age (in years) | 0.88 | 0.70 | 1.10 | 0.25 | 1.03 | 0.86 | 1.24 | 0.74 |
| Less than HS graduate/GED (vs. any college) | 0.19 | 0.05 | 0.65 | 0.01 | 0.66 | 0.22 | 1.98 | 0.46 |
| High school graduate/GED (vs. any college) | 0.48 | 0.14 | 1.59 | 0.23 | 1.20 | 0.39 | 3.72 | 0.75 |
| Attitudinal Characteristics | ||||||||
| Positive condom attitudes | 2.29 | 1.01 | 5.16 | 0.046 | 0.52 | 0.28 | 0.99 | 0.045 |
| Perceived HIV susceptibility | 1.09 | 0.69 | 1.71 | 0.72 | 0.77 | 0.50 | 1.20 | 0.25 |
| Pregnancy concern | 0.88 | 0.53 | 1.48 | 0.63 | 0.66 | 0.42 | 1.03 | 0.07 |
| Relationship Characteristics | ||||||||
| Casual partner | 1.66 | 0.57 | 4.82 | 0.35 | 0.23 | 0.07 | 0.76 | 0.02 |
| Monogamous relationship | 2.60 | 1.01 | 6.70 | 0.047 | 4.53 | 1.42 | 14.47 | 0.01 |
| History of relationship abuse | 1.34 | 0.50 | 3.60 | 0.57 | 3.27 | 1.41 | 7.60 | 0.01 |
| Same sex partner | n/a | n/a | n/a | n/a | n/a | n/a | n/a | n/a |
Note. n/a = Not significant in bivariate model.
Reference group: did not decide ahead of time about condom use.
Correlates of condom availability
As shown in Table 3, youth were more likely to report having a condom available during the sexual event if they were male or the event was special in some way, but marginally less likely if sex occurred in a non-private location (p = .05). We conducted post-hoc analyses to determine whether the type of special event had any bearing on whether a condom was available. Having a condom available was associated with not seeing each other in a long while (OR = 2.67, 95% CI = 1.08 – 6.63) and one of them going away (OR = 7.96, 95% CI = 2.30 – 27.58), but not with just having gotten paid or some other type of special event.
Table 3.
Multivariate Logistic Regression Models Predicting Condom Availability and Condom Use
| Variable | Condom availability a | Condom use b | ||||||
|---|---|---|---|---|---|---|---|---|
|
| ||||||||
| OR | 95% CI | p = | OR | 95% CI | p = | |||
| Demographic Characteristics | ||||||||
| Biological sex: male | 3.03 | 1.43 | 6.40 | 0.004 | 1.14 | 0.49 | 2.64 | 0.76 |
| Black (vs. white) | 2.31 | 0.84 | 6.31 | 0.10 | 1.34 | 0.48 | 3.75 | 0.58 |
| Hispanic (vs. white) | 0.70 | 0.26 | 1.88 | 0.48 | 1.16 | 0.40 | 3.36 | 0.79 |
| Other/Mixed (vs. white) | 0.69 | 0.27 | 1.81 | 0.45 | 1.10 | 0.37 | 3.27 | 0.87 |
| Age (in years) | 0.86 | 0.74 | 1.01 | 0.07 | 1.03 | 0.87 | 1.23 | 0.74 |
| Less than HS graduate/GED (vs. any college) | 0.81 | 0.28 | 2.30 | 0.69 | 0.73 | 0.23 | 2.28 | 0.59 |
| HS graduate/GED (vs. any college) | 0.40 | 0.13 | 1.19 | 0.10 | 0.84 | 0.26 | 2.75 | 0.77 |
| Attitudinal Characteristics | ||||||||
| Positive condom attitudes | 1.43 | 0.81 | 2.50 | 0.22 | 1.98 | 1.09 | 3.59 | 0.03 |
| Perceived HIV susceptibility | n/a | n/a | n/a | n/a | n/a | n/a | n/a | n/a |
| Pregnancy concern | 1.24 | 0.85 | 1.81 | 0.27 | 1.44 | 0.92 | 2.25 | 0.11 |
| Relationship Characteristics | ||||||||
| Casual partner | 1.98 | 0.87 | 4.49 | 0.10 | n/a | n/a | n/a | n/a |
| Monogamous relationship | n/a | n/a | n/a | n/a | 0.28 | 0.12 | 0.67 | 0.004 |
| History of relationship abuse | n/a | n/a | n/a | n/a | 0.53 | 0.23 | 1.23 | 0.14 |
| Same sex partner | n/a | n/a | n/a | n/a | n/a | n/a | n/a | n/a |
| Contextual Factors | ||||||||
| Event was special in some way | 3.43 | 1.63 | 7.19 | 0.001 | n/a | n/a | n/a | n/a |
| Sex was in a non-private place | 0.49 | 0.24 | 1.01 | 0.05 | 0.80 | 0.34 | 1.86 | 0.61 |
| Decided to not use a condom | 0.60 | 0.24 | 1.51 | 0.28 | ||||
| Decided to use a condom | 5.26 | 1.47 | 18.89 | 0.01 | ||||
| Maximum number of drinks | 0.97 | 0.86 | 1.04 | 0.23 | ||||
| Marijuana use only | 0.67 | 0.25 | 1.75 | 0.41 | ||||
| Hard drug use | 0.14 | 0.04 | 0.59 | 0.01 | ||||
Note. n/a = Not significant in bivariate model.
Among youth who did not decide ahead of time to use a condom (n = 279).
Among youth who both had a condom available and did not decide ahead of time to not use a condom (n = 211).
Correlates of condom use
Table 3 also shows the correlates of condom use at the most recent event, excluding youth who decided ahead of time not to use a condom and those who did not have a condom available at the event. Youth were more likely to report that a condom was used if they held more positive attitudes toward condoms and decided ahead of time to use a condom, but less likely if the couple was in a monogamous relationship or used hard drugs prior to having sex.
Discussion
Whether an act of sexual intercourse is protected or unprotected is the end result of a series of decisions and actions by the partners. Results from this study illustrate how this behavior can be better understood by unpacking the stages of condom use and identifying influential factors at each of these stages. For example, decisions made ahead of time to use or not use a condom have a strong effect on whether condoms are available and whether they are used during sexual events. Encouraging homeless youth to decide ahead of time to use a condom may go a long way towards increasing actual condom use in this population. However, our results suggest that certain subgroups may need assistance at this stage, including youth who are Hispanic, have less education, feel less positive about condoms, or have a history of relationship abuse. Negotiating condom use may be especially challenging for young women in abusive relationships (Wingood et al., 2002), particularly if they rely on their male sex partner for shelter and other needed resources. There is some evidence that HIV prevention programs can encourage condom use among young women in abusive relationships if they address gender- and power-related issues such as low self-esteem, lack of perceived control over safe sex, and partner-related barriers associated with unprotected sex (Wingood et al., 2006).
Interestingly, the factors associated with deciding ahead of time about condoms were generally distinct from those related to condom availability and condom use. Condom availability was reported less by females than males. It was also less likely when the event occurred in a non-private place, but more likely when it was described as special in some way. These contextual factors may reflect the degree to which the youth anticipated that they would be having sex and thus planned ahead to have a condom available (indeed, when controlling for condom availability, these contextual factors were unrelated to actual use). Given that homeless youth tend to live fairly unstable lives (Tucker et al., in press), and thus may engage in relatively furtive sexual encounters, increasing accessibility to free condoms at public venues frequented by homeless youth, and encouraging youth (particularly females) to routinely carry them, may help ensure that a condom is available at both planned and unplanned sexual events. When a condom was available at the event, our results suggest that it was less likely to be used if the youth had been using hard drugs, thought that condoms would make sex less enjoyable, or believed that the partner was “safe” (i.e., not having sex with other people). Even for homeless youth who may be somewhat favorably disposed to use a condom, these may be relatively “in-the-moment” risk factors that tip the scales towards deciding to have unprotected sex.
Results from this study suggest a number of important directions for future research and interventions with this population. Advance decision-making to use condoms significantly increases the likelihood that a couple will actually use a condom when the time comes, but relatively little attention has focused on the importance of advance decision-making and how to encourage it. Also needed is a better understanding of the effects of one partner’s advance decision-making (rather than a discussion between two partners) on the couple’s condom use. In the case of condom availability, homeless youth often have access to free condoms through service agencies, but a better understanding is needed of how to address barriers to having a condom available during sex, particularly among females. Finally, HIV prevention program for homeless youth should focus on “in the moment” behavioral and attitudinal factors that can discourage condom use. Fostering more positive attitudes towards condoms may be particularly important as it appears to play a significant role in youth’s decision-making about condom use both “in the moment” and in advance of the sexual event. In the long-term, helping homeless youth achieve more stable living situations, although not a focus of this study, may itself be an effective strategy for reducing their risk of contracting and transmitting HIV (Wolitski, Kidder, & Fenton, 2007).
Acknowledgments
This research was supported by Grant R01DA020351 from the National Institute on Drug Abuse. We thank the youth who shared their experiences with us, the service agencies that collaborated in the study, and the RAND Survey Research Group for their assistance in data collection.
Contributor Information
Joan S. Tucker, RAND Corporation
Allison Ober, RAND Corporation
Gery Ryan, RAND Corporation
Daniela Golinelli, RAND Corporation
Brett Ewing, RAND Corporation
Suzanne L. Wenzel, School of Social Work, University of Southern California
References
- Albarracin D, Johnson BT, Fishbein M, Muellerleile PA. Theories of reasoned action and planned behavior as models of condom use: a meta-analysis. Psychological Bulletin. 2001;127(1):142–161. doi: 10.1037/0033-2909.127.1.142. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Alvarez C, Villarruel AM, Zhou Y, Gallegos E. Predictors of condom use among Mexican adolescents. Research and Theory for Nursing Practice. 2010;24(3):187–196. doi: 10.1891/1541-6577.24.3.187. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Bailey SL, Camlin CS, Ennett ST. Substance use and risky sexual behavior among homeless and runaway youth. Journal of Adolescent Health. 1998;23(6):378–388. doi: 10.1016/s1054-139x(98)00033-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Crosby R, Salazar LF, Diclemente RJ. Lack of recent condom use among detained adolescent males: a multilevel investigation. Sexually Transmitted Infections. 2004;80(6):425–429. doi: 10.1136/sti.2004.009639. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Crosby R, Sanders S, Yarber WL, Graham CA. Condom-use errors and problems: A neglected aspect of studies assessing condom effectiveness. American Journal of Preventive Medicine. 2003;24(4):367–370. doi: 10.1016/s0749-3797(03)00015-1. [DOI] [PubMed] [Google Scholar]
- Frost JJ, Singh S, Finer LB. Factors associated with contraceptive use and nonuse, United States, 2004. Perspectives on Sexual and Reproductive and Health. 2007;39(2):90–99. doi: 10.1363/3909007. [DOI] [PubMed] [Google Scholar]
- Gu J, Lau JT, Chen X, Liu C, Liu J, Chen H, Li Z. Using the Theory of Planned Behavior to investigate condom use behaviors among female injecting drug users who are also sex workers in China. AIDS Care. 2009;21(8):967–975. doi: 10.1080/09540120802657548. [DOI] [PubMed] [Google Scholar]
- Janepanish P, Dancy BL, Park C. Consistent condom use among Thai heterosexual adult males in Bangkok, Thailand. AIDS Care. 2011;23(4):460–466. doi: 10.1080/09540121.2010.516336. [DOI] [PubMed] [Google Scholar]
- Matson PA, Adler NE, Millstein SG, Tschann JM, Ellen JM. Developmental changes in condom use among urban adolescent females: influence of partner context. Journal of Adolescent Health. 2011;48(4):386–390. doi: 10.1016/j.jadohealth.2010.07.016. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Mausbach BT, Semple SJ, Strathdee SA, Patterson TL. Predictors of safer sex intentions and protected sex among heterosexual HIV-negative methamphetamine users: an expanded model of the Theory of Planned Behavior. AIDS Care. 2009;21(1):17–24. doi: 10.1080/09540120802017628. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Outlaw AY, Naar-King S, Janisse H, Parsons JT. Predictors of condom use in a multisite study of high-risk youth living with HIV. AIDS Education and Prevention. 2010;22(1):1–14. doi: 10.1521/aeap.2010.22.1.1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Skinner CJ. Domain means, regression, and multivariate analyses. In: Skinner CJ, Holt D, Smith TMF, editors. Analysis of Complex Surveys. New York: Wiley; 1989. pp. 59–88. [Google Scholar]
- Song YS, Calsyn DA, Doyle SR, Dierst-Davies R, Chen T, Sorensen JL. Predictors of condom use among men enrolled in drug treatment programs. AIDS Education and Prevention. 2009;21(5):460–473. doi: 10.1521/aeap.2009.21.5.460. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Steinberg JK, Grella CE, Boudov MR, Kerndt PR, Kadrnka CM. Methamphetamine use and high-risk sexual behaviors among incarcerated female adolescents with a diagnosed STD. Journal of Urban Health. 2011;88(2):352–364. doi: 10.1007/s11524-011-9557-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Tucker JS, Ryan GW, Golinelli D, Munjas B, Wenzel SL, Kennedy DP, Zhou A. Substance use and other risk factors for unprotected sex: Results from an event-based study of homeless youth. AIDS and Behavior. 2012;16(6):1699–1707. doi: 10.1007/s10461-011-0017-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Tucker JS, Sussell J, Golinelli D, Zhou A, Kennedy DP, Wenzel SL. Understanding pregnancy-related attitudes and behaviors: A mixed method study of homeless youth. Perspectives on Sexual and Reproductive Health. doi: 10.1363/4425212. in press. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Wenzel SL, Tucker JS, Golinelli D, Green HDJ, Zhou A. Personal network correlates of alcohol, cigarette, and marijuana use among homeless youth. Drug and Alcohol Dependence. 2010;112(1–2):140–149. doi: 10.1016/j.drugalcdep.2010.06.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Wingood GM, DiClemente RJ, Hubbard McCree D, Harrington K, Davies S. Dating violence and African-American adolescent females’ sexual health. Pediatrics. 2002;107:E72–E75. doi: 10.1542/peds.107.5.e72. [DOI] [PubMed] [Google Scholar]
- Wingood GM, DiClemente RJ, Harrington KF, Lang DL, Davies SL, Hook EW, 3rd, Hardin JW. Efficacy of an HIV prevention program among female adolescents experiencing gender-based violence. American Journal of Public Health. 2006;96(6):1085–1090. doi: 10.2105/AJPH.2004.053595. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Wolitski RJ, Kidder DP, Fenton KA. HIV, homelessness, and public health:critical issues and a call for increased action. AIDS & Behavior. 2007;11(Suppl 6):167–171. doi: 10.1007/s10461-007-9277-9. [DOI] [PubMed] [Google Scholar]