Psychology of Addictive Behaviors 2014, Vol. 28, No. 2, 625– 630
© 2014 American Psychological Association 0893-164X/14/$12.00 DOI: 10.1037/a0035417
BRIEF REPORT
Prevalence and Correlates of Transactional Sex Among an Urban Emergency Department Sample: Exploring Substance Use and HIV Risk Rikki Patton
Frederic C. Blow, Amy S. B. Bohnert, Erin E. Bonar, Kristen L. Barry, and Maureen A. Walton
This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
University of Michigan
University of Michigan and Department of Veteran’s Affairs, Health Services Research and Development, Ann Arbor, Michigan Men and women involved in transactional sex (TS) report increased rates of HIV risk behaviors and substance use problems as compared with the general population. When people engaged in TS seek health care, they may be more likely to utilize the emergency department (ED) rather than primary care services. Our goal was to examine the prevalence and correlates of TS involvement among an ED sample of men and women. Adults ages 18 – 60 were recruited from an urban ED, as part of a larger randomized control trial. Participants (n ⫽ 4,575; 3,045 women, 1,530 men) self-administered a screening survey that assessed past 3-month substance use (including alcohol, marijuana, illicit drugs, and prescription drugs) and HIV risk behaviors, including TS (i.e., being paid in exchange of a sexual behavior), inconsistent condom use, multiple partners, and anal sex. Of the sample, 13.3% (n ⫽ 610) reported TS within the past 3 months (64.4% were female). Bivariate analysis showed TS was significantly positively associated with alcohol use severity, marijuana use, and both illicit and prescription drug use, and multiple HIV risk behaviors. These variables (except marijuana) remained significantly positively associated with TS in a binary logistic regression analysis. The prevalence of recent TS involvement among both male and female ED patients is substantial. These individuals were more likely to report higher levels of alcohol/drug use and HIV risk behaviors. The ED may be a prime location to engage both men and women who are involved in TS in behavioral interventions for substance use and sexual risk reduction. Keywords: transactional sex, emergency department, HIV, substance use, gender
as a consequence to being involved in TS for others (Belcher & Herr, 2005; Potterat et al., 1998). Further, substantial proportions of substance users report TS involvement, with 25% of male crack users and over 40% female crack-cocaine users in various samples reporting TS within the past 30 days (Edwards, Halpern, & Wechsberg, 2006; Leukefeld, 1999; Logan & Leukefeld, 2000). Regardless of the motivation for use, these prevalence rates suggest individuals who engage in TS are a population vulnerable to substance abuse problems and engaging TS-involved individuals into substance abuse prevention and intervention programs may provide a crucial avenue for addressing both their substance use and their risky sexual behaviors. TS-involved individuals are considered a difficult-to-reach population (Benoit, Jansson, Millar, & Phillips, 2005). Prior research suggests that substance-abusing women involved in TS tend to seek health care services through the emergency department (ED) compared with other substance-abusing women (Burnette, Lucas, Ilgen, Frayne, Mayo, Weitlauf, 2008). For instance, 39%–56% of women who self-identified as sex workers reported visiting the ED recently or as a result of their TS involvement (Raymond, Hughes, & Gomez, 2001; Shannon, Bright, Duddy, & Tyndall, 2005). These findings suggest that the ED may be a prime location for engaging individuals
Transactional sex (TS) involvement, defined as the exchange of a sexual behavior for money, drugs, or other needs, is associated with increased likelihood of substance use issues (Clarke, Clarke, RoeSepowitz, & Fey, 2012; Wechsberg et al., 2009), and greater risk of contracting sexually transmitted infections (STIs), including HIV (Bobashev, Zule, Osilla, Kline, & Wechsberg, 2009; Shannon et al., 2008). The relationship between TS involvement and substance abuse is complex, with research suggesting that substance abuse may act as the antecedent behavior to TS involvement for some individuals, but
Rikki Patton, The Substance Abuse Research Center, Department of Psychiatry, and School of Social Work, University of Michigan; Frederic C. Blow, Amy S. B. Bohnert, Erin E. Bonar, Kristen L. Barry, and Maureen A. Walton, Department of Psychiatry, University of Michigan. Rikki Patton is now at the University of Akron. This investigation was supported by the National Institutes of Health under Ruth L. Kirschstein National Research Service Award T32 DA007267 and NIDA #026029. Its contents are solely the responsibility of the authors and do not necessarily represent the official views of the NIH. Correspondence concerning this article should be addressed to Rikki Patton, Department of Counseling, 27 South Forge Street, Akron, OH 44325. E-mail: [email protected] 625
PATTON, BLOW, BOHNERT, BONAR, BARRY, AND WALTON
This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
626
involved in TS. Less is known about the degree to which men engaged in TS utilize health care services, although prior reports have suggested that, among a sample of men receiving substance abuse treatment, TS-involved men who were involved in TS reported greater likelihood of use of inpatient mental health services compared with other patients, but not emergency services (Burnette et al., 2008). One study examining the utility of a brief intervention with drug-positive men and women recruited through the ED who reported either cocaine or heroin use stated that approximately 12% of patients reported TS involvement in the past 30 days (Bernstein et al., 2012). Although this study highlights the prevalence of TS involvement within an ED setting among a high-risk substance-using group, findings are limited in their generalizability and applicability to the entire ED patient population.
Present Study The current study aims to address the gaps in the literature by, first, evaluating the prevalence of TS involvement among broad a sample of both men and women seeking care in an ED. In addition, given the prior research regarding increased HIV risk and substance abuse associated with TS involvement (Bobashev et al., 2009; Shannon et al., 2008), we evaluated whether substance abuse and HIV risk indicators differed among individuals involved in TS compared with other ED patients. We hypothesized that those ED patients who reported TS involvement would also report higher rates of substance use and HIV-related risk behaviors than those who did not report TS.
Method Study Design and Setting The current study used data collected as part of a screening survey aimed at identifying patients eligible for a randomized control trial of adult patients (ages 18 – 60) presenting to an ED located in an Academic Level 1 Trauma Center a Midwestern city with similar rates of crime and poverty as other large cities. The study was approved and conducted in compliance with Institutional Review Board (IRB) requirements. A Certificate of Confidentiality was obtained for this study. Data were collected from February, 2011 to March, 2013. Research staff approached participants and described the study; those who were interested in participating provided written informed consent and selfadministered a 15-min computerized screening survey. Participants were compensated for screening with a gift valuing $1 (e.g., playing cards or hand lotion). Patients who presented with acute psychosis, acute sexual assault, medically unstable, or who were in police custody were excluded from screening.
non-TS group and all other respondents were recoded into the group labeled as TS group. Demographic variables. Gender, race, age, household income, marital status, current employment, sexual orientation, and education level were queried using items from validated surveys (e.g., National Survey of Drug Use and Health, Office of Applied Studies, 2009; Psychiatric Outcomes Module: Substance Abuse Outcomes Module, Smith et al., 1996; Global Appraisal of Individual Needs, version 5.4.0., 2006).
Risk Variables Reason for ED visit. Participants were asked a yes/no question regarding whether or not their visit to the ED was injury related, referring to cuts, bruises, broken bones, and so forth. Alcohol and drug use. Alcohol use severity over the past 3 months was assessed with the 10-item Alcohol Use Disorders Identification Test (AUDIT; Saunders, Aasland, Babor, De La Fluente, & Grant, 1993). Questions assessed frequency of alcohol use, number of drinks consumed on a typical drinking day, frequency of having five or more drinks, and negative consequences due to drinking. Participants’ responses on these items were summed to create a single total score reflecting alcohol use severity (Saunders et al., 1993). Cronbach’s alpha for this measure in the current sample was .90. The Alcohol, Smoking and Substance Involvement Screening Test (ASSIST; WHO ASSIST Working Group, 2002) was used to measure frequency of drug use. Respondents were asked if they had used the following drugs at least once in the past 3 months— cocaine, marijuana, methamphetamines, hallucinogens, inhalants, prescription stimulants, prescription sedatives, prescription opioids, street opioids, or any other drugs. Due to the distribution of the data, with low frequencies of use for all illicit drugs, excluding marijuana, and nonmedical use of prescription drugs, these variables were refined into the following groups representing using at least once in the past 3 months: (a) marijuana, (b) other illicit drugs, and (c) nonmedical use of prescription drugs. HIV risk behaviors. Sexual and drug use HIV risk behaviors during the past 3 months were assessed using items selected from the HIV Risk-Taking Behavior Scale (HRBS; Ward, Darke, & Hall, 1990). Specifically, injection drug use, number of sexual partners, inconsistent condom use, and anal sex were queried. Each variable was dichotomized to denote engaging in the risky behavior within the past 3 months. STIs were measured by asking respondents if a doctor or other medical professional ever told the patient that they had an STI, such as chlamydia, gonorrhea, herpes, or syphilis. Respondents answered yes or no to this question (see Substance Abuse & Mental Health Services Administration, 2009).
Data Analysis Plan Measures Transactional sex involvement. TS involvement was assessed using the participant’s responses to the following question from the HIV Risk-taking Behavior Scale (HRBS; Ward, Darke, & Hall, 1990)—“In the past three months, how often have you used condoms when you have been paid for sex?” Respondents who answered “no paid sex” were recoded into the group labeled
Data were analyzed using SAS version 9.3 (SAS Institute Inc., 2012). First, demographic characteristics of the sample were examined. Variable distributions were examined for normality and appropriate statistics were used. Bivariate analyses, including chisquare and independent samples t tests, were conducted to determine if there were significant differences between the TS group and other participants on demographic characteristics, substance
TRANSACTIONAL SEX INVOLVEMENT IN AN ED SAMPLE
use, and HIV risk behaviors. Finally, a hierarchical logistic regression was conducted to determine the associations of TS involvement with demographic and risk variables. A hierarchical model was used in order to control for demographic characteristics in later steps. Demographic factors were entered on Step 1, followed by substance use variables on Step 2, with sex risk behaviors entered on Step 3. All variables that were significant in bivariate analyses were retained in the regression model. Model fit statistics indicated no evidence of multicollinearity.
the local IRB, we were not able to collect any additional data about characteristics of patients who refused without written informed consent. See Table 1 for sample characteristics.
Bivariate Analysis See Table 1 for a full summary of findings from the bivariate analyses. Chi-square/t test analyses indicated there were several significant differences between the TS group and the non-TS group. Participants in the TS group were more likely to report an educational level of high school completion or below (2 ⫽ 22.9; df ⫽ 1), an annual income more than $20,000, and to be married (2 ⫽ 10.8; df ⫽ 1). The TS group had higher scores on all substance use and HIV risk measures as compared with the non-TS group. There were no significant differences regarding gender, race, age, or sexual orientation between the two groups.
Results This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
627
Sample Characteristics As part of the larger randomized controlled trial (RCT), 6,161 individuals were approached for screening, of whom, 4,575 (74.3%) agreed to complete the screening survey. Respondents completing the screening survey were compared with those who were missed and who refused on gender and race. Males were more likely to be missed (2 ⫽ 94.1; p ⬍ .0001) and to refuse participation (2 ⫽ 30.95; p ⬍ .0001). There were no significant differences by race based on those missed (2 ⫽ 2.576; p ⫽ .11) or refused (2 ⫽ 3.526; p ⫽ .06). Per privacy-related policies of
Regression Analysis The hierarchical logistic regression analysis was conducted in order to examine the relationship between transactional sex involvement and Step 1: demographic characteristics (gender, race, educational level, income, marital status, and ED presentation);
Table 1 Frequencies and Bivariate Analysis for the Whole Sample and Subsamples Transactional sex involvement (n ⫽ 610; 13.3%)
Full sample (n ⫽ 4,575) Variable Demographics Female gender Caucasian race High school/GED education or lessⴱⴱ Age (mean/SD) Household income less than $20k/yrⴱ Nonheterosexual orientation† Married/living togetherⴱⴱ Correlates ED presentation for injuryⴱ Alcohol and/or drug use (past 3 months) Alcohol use severity (mean/SD)ⴱⴱ Any marijuana useⴱⴱ Any other illicit drugⴱⴱ (cocaine, heroin, methamphetamines, inhalants, hallucinogens) Cocaineⴱⴱ Heroinⴱⴱ Methamphetamineⴱⴱⴱ Inhalantsⴱⴱⴱ Hallucinogensⴱⴱⴱ Any prescription drug useⴱⴱ Stimulantsⴱ Sedativesⴱⴱ Opioidsⴱⴱ HIV risk behaviors Ever inject drugsⴱⴱ Diagnosed with STI in past 3 monthsⴱⴱ More than 1 sex partner in past 3 monthsⴱⴱ Inconsistent condom use in past 3 monthsⴱⴱ Engaged in anal sex past in three monthsⴱⴱ † ⴱ
No transactional sex involvement (n ⫽ 3,965)
Frequency
%
Frequency
%
Frequency
3045 2033 2645 34.3 2378 323 1458
66.6 44.4 57.8 12.1 52.0 8.1 31.9
393 249 407 33.9 291 58 230
64.4 40.8 66.7 11.5 47.7 10.0 37.7
2652 1784 2238 34.3 2087 265 1228
66.9 45.0 56.4 12.2 52.6 7.8 31.0
1330
29.1
198
32.5
1132
28.6
2.7 1124
5.2 24.6
3.9 185
7.0 30.3
2.5 939
4.9 23.7
184 149 51 11 4 17 206 24 103 139
4.0 3.3 1.1 0.2 0.1 0.4 4.5 0.5 2.2 3.0
62 47 22 3 2 4 52 7 35 37
10.2 7.7 3.6 0.5 0.3 0.7 8.5 1.1 5.7 6.1
122 102 29 8 2 13 154 17 68 102
140 243 684 2733 398
3.1 5.3 15.0 59.7 11.4
36 51 131 560 104
5.9 8.4 21.5 91.8 17.0
104 192 553 2173 294
n ⫽ 584 reported selves celibate, did not select any orientation. p ⬍ .05. ⴱⴱ p ⬍ 0.001. ⴱⴱⴱ Fisher’s exact test, due to small cell frequencies, nonsignificant finding.
%
3.1 2.6 0.7 0.2 0.05 0.3 3.9 0.4 1.7 2.6 2.6 4.8 14.0 54.8 10.3
PATTON, BLOW, BOHNERT, BONAR, BARRY, AND WALTON
628
This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
Table 2 Hierarchical Binary Logistic Regression Assessing Correlates of Transactional Sex Involvement Variable
Model 1
Model 2
Model 3
Step 1 Male gender Caucasian race High school/GED education or less Household income less than $20k Married or living together ED presentation for injury Step 2 Alcohol use severity (past 3 months) Marijuana use (past 3 months) Other illicit drug use (past 3 months) Any prescription drug use (past 3 months) Step 3 Ever inject drugs? Diagnosed with STI (past 3 months) # sexual partners ⬎ 1 (past 3 months) Inconsistent condom use (past 3 months) Engaged in anal sex (past 3 months) Fit Statistics LR 2 (df) Generalized R2 Hosmer-Lemeshow goodness of fit 2
AOR [95%CI] 1.02 [0.85, 1.23] 0.80ⴱ [0.67, 0.95] 1.59ⴱⴱⴱ [1.32, 1.91] 0.83ⴱ [0.70, 0.98] 1.43ⴱⴱ [1.19, 1.72] 1.18 [0.97, 1.42]
AOR [95%CI] 0.88 [0.72, 1.06] 0.78ⴱⴱ [0.65, 0.93] 1.58ⴱⴱⴱ [1.31, 1.89] 0.79ⴱⴱ [0.67, 0.95] 1.53ⴱⴱⴱ [1.27, 1.84] 1.14 [0.94, 1.38]
AOR [95%CI] 0.94 [0.76, 1.15] 0.75ⴱⴱ [0.63, 0.91] 1.65ⴱⴱⴱ [1.36, 2.00] 0.68ⴱⴱⴱ [0.56, 0.83] 1.08 [0.89, 1.32] 1.15 [0.94, 1.41]
— — — —
1.03ⴱⴱⴱ [1.01, 1.04] 1.11 [0.90, 1.36] 2.62ⴱⴱⴱ [1.81, 3.81] 1.53ⴱ [1.06, 2.22]
1.03ⴱⴱⴱ [1.01, 1.05] 0.97 [0.78, 1.20] 2.19ⴱⴱ [1.44, 3.33] 1.60ⴱ [1.07, 2.41]
— — — — —
— — — — —
1.70ⴱ [1.05, 2.76] 0.58ⴱⴱⴱ [0.45, 0.73] 1.20 [0.94, 1.53] 9.75ⴱⴱⴱ [7.17, 13.24] 1.77ⴱⴱⴱ [1.36, 2.30]
51.01 (6) 0.02 6.67 (8)
127.79 (10) 0.05 10.41 (8)
523.48 (15) 0.20 11.85 (8)
ⴱ
p ⬍ .05.
ⴱⴱ
p ⬍ .01.
ⴱⴱⴱ
p ⬍ .001.
Step 2: substance use variables (alcohol use severity, any marijuana use, any other illicit drug use, and any prescription drug use); and Step 3: sexual risk behaviors (inject drugs, diagnosed with an STI, number of sexual partners, inconsistent condom use, and anal sex). Although gender and race were not significant in the bivariate analysis, these variables were included in the multivariate analysis due to the strong evidence in prior literature regarding the higher probability of women and racial minorities engaging in transactional sex behaviors. Model fit statistics indicated that each step improved the model (see Table 2). In Step 1, being Caucasian and income (making less than $20,000) were negatively associated with TS involvement (AOR ⫽ 0.80 and 0.83, respectively). Having a high school education or less (AOR ⫽ 1.59) and being married or living together as married (AOR ⫽ 1.43) were positively associated with TS involvement. No other demographic variables were significant. With the addition of substance abuse variables in Step 2 the following variables were positively associated with TS involvement: alcohol use severity (AOR ⫽ 1.03), illicit drug use (AOR ⫽ 2.62), prescription drug use (AOR ⫽ 1.53), educational status (AOR ⫽ 1.58), and marital status (AOR ⫽ 1.53). Race (being Caucasian) and income were negatively associated with TS involvement (AOR ⫽ 0.78 and 0.79, respectively). In the final model including HIV-risk variables, the following variables were positively associated with TS involvement: alcohol use severity (AOR ⫽ 1.03), illicit drug use (AOR ⫽ 2.19), prescription drug use (AOR ⫽ 1.60), ever injecting drugs (AOR ⫽ 1.70), inconsistent condom use (AOR ⫽ 9.75), and engaging in anal sex (AOR ⫽ 1.77). Individuals involved in TS were less likely to be Caucasian (AOR ⫽ 0.75), earn less than $20,000 annually (AOR ⫽ 0.68), and to ever be diagnosed with an STI (AOR ⫽ 0.58), and were more likely to report an education level of high school diploma or less (AOR ⫽ 1.65). Nonsignificant
variables in the final model included gender, marital status, reason for presenting to the ED, marijuana use, and having multiple sexual partners in the past 3 months.
Discussion This study presents novel findings regarding the prevalence and correlates of TS involvement within an urban ED sample of adult men and women. Findings indicated that 13.3% of patients sampled had engaged in TS within the past 3 months and, of those, gender distribution was similar to the larger sample. Findings also showed that patients involved in TS were more likely to report substance use and HIV-risk behaviors compared with other patients. That more than one in 10 patients presenting to this urban ED reported recent TS involvement has several implications for engagement and treatment. Most research examining TS involvement includes substance-abusing samples only, with rates of TS involvement ranging from 12%– 44% (e.g., Bernstein et al., 2012; Burnette et al., 2008). The current sample included individuals with and without substance abuse problems and the prevalence was still within the range of substance-abusing only samples. Thus, the ED may be a useful venue for engaging individuals involved in transactional sex in brief interventions without limiting services to only those individuals who also present with substance abuse problems. Additionally, the current findings showed that TS involvement was not associated with gender, suggesting that both men and women involved in TS may be reached in one location—in the ED. The present findings for ED patients also support previous reports of increased alcohol and illicit drug use among individuals involved in TS in the general population (Clarke et al., 2012; Wechsberg et al., 2009). A unique finding from this study is that TS involvement was related to increased odds of misusing pre-
This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
TRANSACTIONAL SEX INVOLVEMENT IN AN ED SAMPLE
scription drugs, including opioids and sedative/hypnotics. Given these findings and the recent increases of prescription drug misuse throughout the United States (Gu, Dillon, & Burt, 2010; SAMHSA, 2009), additional research is needed to understand this association in order to appropriately tailor intervention strategies to address this emerging pattern of substance use. Prior research suggests that individuals who engage in TS may have higher levels of substance use for multiple reasons, including coping with the stresses related to TS involvement (Belcher & Herr, 2005; Burnette et al., 2008; Wechsberg et al., 2009). Alternatively, TS may be partially a result of greater involvement in substance use among impoverished populations as a means to maintain substance use (Clarke et al., 2012; Potterat et al., 1998). Although the causal nature of this relationship requires future study, our findings suggest that the ED may be an appropriate venue for interventions addressing substance use and TS, both of which are associated with risk for STIs and HIV infection, among this vulnerable population. Finally, the present findings also showed a significant association between TS and HIV-risk behaviors, including injecting drugs, STI diagnosis, inconsistent condom use, and anal sex among patients recruited from the ED, which are consistent with prior work with substance-using samples (Bobashev et al., 2009; Shannon et al., 2008). Interestingly, current findings indicate that men and women who engage in TS are less likely to have received a formal diagnosis of a STI, despite also being more likely to report inconsistent condom use. This may not necessarily reflect a true decreased risk of STI in this population, however, because it is possible that individuals from this vulnerable population may have more barriers to STI screening and treatment services (Kurtz, Surratt, Kiley, Marion, & Inciardi, 2005). Overall, current findings suggest that the ED may be a relevant site for intervention efforts focused on sexual risk reduction among men and women, particularly those who engage in TS given the high prevalence (91.8%) of inconsistent condom use, one of the most potent predictors of HIV/STI.
Limitations and Future Directions Although the present study augments current literature by examining the prevalence and correlates of TS involvement among men and women in the ED, several limitations should be noted. The criterion we used to define TS was limited by the use of a single item assessment regarding condom use with partners who paid for sex. Although this question asked about recent involvement in exchanging sex for money, the depth of involvement in sex work for these participants is unknown. In addition, this question only inquired about being paid for sex, and may not have identified those who trade sex for drugs or other goods. Further, one of the exclusion criteria for the original study was presenting to the ED for acute sexual assault. Individuals involved in exchanging sex for money experience repeated sexual assaults (Dalla, Xia, & Kennedy, 2003; Karandikar & Prospero, 2010) and these exclusion criteria may have biased the sample, producing a conservative estimate of the prevalence of TS involvement among ED patients. Despite these limitations, the results showed that 13.3% of individuals attending the ED, including both men and women, report engaging in TS. Additionally, although our sample was racially diverse, these results may not be generalizable to other populations
629
or racial/ethnic groups (e.g., Hispanics, Asians), or to other geographic locales due to recruitment from a single site within an ED located in an economically depressed urban area. Further, the analysis was cross-sectional in design, thereby limiting our ability to understand causal relationships among TS and substance use or other factors. To address these limitations, future research should explore the relationship between TS involvement and use of the ED longitudinally in order to determine how these individuals use the ED to meet their needs and their connection to other services, such as substance abuse treatment. More in-depth assessment of TS involvement is warranted to allow further understanding of how interventions can be developed to effectively intervene with individuals in the ED who engage in high-risk sexual behaviors and substance use. Additionally, given that so many men receiving services in the ED also reported TS involvement, it may be helpful to explore gender differences in TS to inform prevention, behavioral intervention, and medical screening and treatment of STIs. Focusing on these individuals, defined as a difficult-to-reach population, for intervention services in the ED may prove to be a fruitful area for research and practice. Future research is needed to develop and implement combined HIV risk and substance use intervention programs tailored for patients in the ED engaged in both behaviors.
References Belcher, J. R., & Herr, S. (2005). Development of grounded theory: Moving towards a theory of the pathways into street prostitution among low-income women. Journal of Addictions Nursing, 16, 117–124. doi: 10.1080/10884600500196651 Benoit, C., Jansson, M., Millar, A., & Phillips, R. (2005). CommunityAcademic research on hard-to-reach populations: Benefits and challenges. Qualitative Health Research, 15, 263–282. doi:10.1177/ 1049732304267752 Bernstein, E., Ashong, D., Heeren, T., Winter, M., Bliss, C., Madico, G., & Bernstein, J. (2012). The impact of a brief motivational intervention on unprotected sex and sex while high among drug-positive emergency department patients who receive STI/HIV VC/T and drug treatment referral as standard of care. AIDS and Behavior, 16, 1203–1216. doi:10.1007/s10461012-0134-0 Bobashev, G. V., Zule, W. A., Osilla, K. C., Kline, T. C., & Wechsberg, W. M. (2009). Transactional sex among men and women in the South at high risk for HIV and other STIs. Journal of Urban Health, 86, 32– 47. doi:10.1007/s11524-009-9368-1 Burnette, M. L., Lucas, E., Ilgen, M., Frayne, S. M., Mayo, J., & Weitlauf, J. C. (2008). Prevalence and health correlates of prostitution among patients entering treatment for substance use disorders. Archives of General Psychiatry, 65, 337–344. doi:10.1001/archpsyc.65.3.337 Clarke, R. J., Clarke, E. A., Roe-Sepowitz, D., & Fey, R. (2012). Age at entry into prostitution: Relationship to drug use, race, suicide, education level, childhood abuse, and family experiences. Journal of Human Behavior in the Social Environment, 22, 270 –289. doi:10.1080/ 10911359.2012.655583 Dalla, R. L., Xia, Y., & Kennedy, H. (2003). “You just give them what they want and pray they don’t kill you”: Street-level sex workers’ reports of victimization. Violence Against Women, 9, 1367–1394. doi:10.1177/ 1077801203255679 Edwards, J. M., Halpern, C. T., & Wechsberg, W. M. (2006). Correlates of exchanging sex for drugs or money among women who use crack cocaine. AIDS Education and Prevention, 18, 420 – 429. doi:10.1521/ aeap.2006.18.5.420
This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
630
PATTON, BLOW, BOHNERT, BONAR, BARRY, AND WALTON
Gu, Q., Dillon, C. F., & Burt, V. L. (2010). Prescription drug use continues to increase: U. S. prescription drug data for 2007–2008. NCHS Data Brief, 42. Karandikar, S., & Prospero, M. (2010). From client to pimp: Male violence against female sex workers. Journal of Interpersonal Violence, 25, 257–273. doi:10.1177/0886260509334393 Kurtz, S. P., Surratt, H. L., Kiley, M. C., & Inciardi, J. A. (2005). Barriers to health and social services for street-based sex workers. Journal of Health Care for the Poor and Underserved, 16, 345–361. doi:10.1353/ hpu.2005.0038 Leukefeld, C. (1999). HIV risk behavior and drug use among heterosexual male crack users by prostitution involvement. Population Research and Policy Review, 18, 23–38. doi:10.1023/A:1006192426092 Logan, T. K., & Leukefeld, C. (2000). Sexual and drug use behaviors among female crack users: A multi-site study. Drug and Alcohol Dependence, 58, 237–245. doi:10.1016/S0376-8716(99)00096-4 Office of Applied Studies. (2009). 2010 national survey on drug use and health: CAI specifications for programming English version. Rockville, MD: Substance Abuse and Mental Health Services Administration. Potterat, J. J., Rothenberg, R. B., Muth, S. Q., Darrow, W. M., & PhillipsPlummer, L. (1998). Pathways to prostitution: The chronology of sexual and drug abuse milestones. Journal of Sex Research, 35, 333–340. doi:10.1080/00224499809551951 Raymond, J. G., Hughes, D. M., & Gomez, C. J. (2001). Sex trafficking of women in the United States: International and domestic trends. Washington, DC: U.S. Department of Justice. doi:10.1037/e624522007-001 SAS Institute Inc. (2012). What’s new in SAS® 9.3. Cary, NC: SAS Institute Inc. Saunders, J. B., Aasland, O. G., Babor, T. F., De La Fuente, J. R., & Grant, M. (1993). Development of the Alcohol Use Disorders Identification Test (AUDIT): WHO Collaborative project on early detections of persons with harmful alcohol consumption-II. Addiction, 88, 791– 804. doi:10.1111/j.1360-0443.1993.tb02093.x
Shannon, K., Bright, V., Duddy, J., & Tyndall, M. W. (2005). Access and utilization of HIV treatment and services among women sex workers in Vancouver’s downtown eastside. Journal of Urban Health, 82, 488 – 497. doi:10.1093/jurban/jti076 Shannon, K., Kerr, T., Allinott, S., Chettiar, J., Shoveller, J., & Tyndall, M. W. (2008). Social and structural violence and power relationship in mitigating HIV risk of drug-using women in survival sex work. Social Science & Medicine, 66, 911–921. doi:10.1016/j.socscimed.2007.11.008 Smith, G. R., Kramer, T., Babor, T., Burnam, M. A., Mosley, C. L., & Rost, K., Burns, B. (1996). Substance and outcomes module–User’s manual. 1996 –2007. Little Rock, AR: University of Arkansas for Medical Sciences. Substance Abuse and Mental Health Services Administration, Office of Applied Studies. (February 5, 2009). The NSDUH report: Trends in nonmedical use of prescription pain relievers: 2002 to 2007. Rockville, MD: Substance Abuse and Mental Health Services Administration, Office of Applied Studies. Ward, J., Darke, S., & Hall, W. (1990). The HIV risk-taking behavior scale (HRBS) manual. Sydney, Australia: National Drug and Alcohol Research Centre. Wechsberg, W. M., Wu, L., Zule, W. A., Parry, C. D., Browne, F. A., Luseno, W. K., . . . Gentry, A. (2009). Substance abuse, treatment needs and access among female sex workers and non-sex workers in Pretoria, South Africa. Substance Abuse Treatment, Prevention, and Policy, 4, 1–11. doi:10.1186/1747-597X-4-11 ASSIST Working Group. (2002). The Alcohol, Smoking and Substance Involvement Screening Test (ASSIST): Development, reliability and feasibility. Addiction, 97, 1183–1194. doi:10.1046/j.1360-0443.2002 .00185.x
Received April 23, 2013 Revision received August 6, 2013 Accepted September 16, 2013 䡲
© 2014 American Psychological Association 0893-164X/14/$12.00 DOI: 10.1037/a0035417
BRIEF REPORT
Prevalence and Correlates of Transactional Sex Among an Urban Emergency Department Sample: Exploring Substance Use and HIV Risk Rikki Patton
Frederic C. Blow, Amy S. B. Bohnert, Erin E. Bonar, Kristen L. Barry, and Maureen A. Walton
This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
University of Michigan
University of Michigan and Department of Veteran’s Affairs, Health Services Research and Development, Ann Arbor, Michigan Men and women involved in transactional sex (TS) report increased rates of HIV risk behaviors and substance use problems as compared with the general population. When people engaged in TS seek health care, they may be more likely to utilize the emergency department (ED) rather than primary care services. Our goal was to examine the prevalence and correlates of TS involvement among an ED sample of men and women. Adults ages 18 – 60 were recruited from an urban ED, as part of a larger randomized control trial. Participants (n ⫽ 4,575; 3,045 women, 1,530 men) self-administered a screening survey that assessed past 3-month substance use (including alcohol, marijuana, illicit drugs, and prescription drugs) and HIV risk behaviors, including TS (i.e., being paid in exchange of a sexual behavior), inconsistent condom use, multiple partners, and anal sex. Of the sample, 13.3% (n ⫽ 610) reported TS within the past 3 months (64.4% were female). Bivariate analysis showed TS was significantly positively associated with alcohol use severity, marijuana use, and both illicit and prescription drug use, and multiple HIV risk behaviors. These variables (except marijuana) remained significantly positively associated with TS in a binary logistic regression analysis. The prevalence of recent TS involvement among both male and female ED patients is substantial. These individuals were more likely to report higher levels of alcohol/drug use and HIV risk behaviors. The ED may be a prime location to engage both men and women who are involved in TS in behavioral interventions for substance use and sexual risk reduction. Keywords: transactional sex, emergency department, HIV, substance use, gender
as a consequence to being involved in TS for others (Belcher & Herr, 2005; Potterat et al., 1998). Further, substantial proportions of substance users report TS involvement, with 25% of male crack users and over 40% female crack-cocaine users in various samples reporting TS within the past 30 days (Edwards, Halpern, & Wechsberg, 2006; Leukefeld, 1999; Logan & Leukefeld, 2000). Regardless of the motivation for use, these prevalence rates suggest individuals who engage in TS are a population vulnerable to substance abuse problems and engaging TS-involved individuals into substance abuse prevention and intervention programs may provide a crucial avenue for addressing both their substance use and their risky sexual behaviors. TS-involved individuals are considered a difficult-to-reach population (Benoit, Jansson, Millar, & Phillips, 2005). Prior research suggests that substance-abusing women involved in TS tend to seek health care services through the emergency department (ED) compared with other substance-abusing women (Burnette, Lucas, Ilgen, Frayne, Mayo, Weitlauf, 2008). For instance, 39%–56% of women who self-identified as sex workers reported visiting the ED recently or as a result of their TS involvement (Raymond, Hughes, & Gomez, 2001; Shannon, Bright, Duddy, & Tyndall, 2005). These findings suggest that the ED may be a prime location for engaging individuals
Transactional sex (TS) involvement, defined as the exchange of a sexual behavior for money, drugs, or other needs, is associated with increased likelihood of substance use issues (Clarke, Clarke, RoeSepowitz, & Fey, 2012; Wechsberg et al., 2009), and greater risk of contracting sexually transmitted infections (STIs), including HIV (Bobashev, Zule, Osilla, Kline, & Wechsberg, 2009; Shannon et al., 2008). The relationship between TS involvement and substance abuse is complex, with research suggesting that substance abuse may act as the antecedent behavior to TS involvement for some individuals, but
Rikki Patton, The Substance Abuse Research Center, Department of Psychiatry, and School of Social Work, University of Michigan; Frederic C. Blow, Amy S. B. Bohnert, Erin E. Bonar, Kristen L. Barry, and Maureen A. Walton, Department of Psychiatry, University of Michigan. Rikki Patton is now at the University of Akron. This investigation was supported by the National Institutes of Health under Ruth L. Kirschstein National Research Service Award T32 DA007267 and NIDA #026029. Its contents are solely the responsibility of the authors and do not necessarily represent the official views of the NIH. Correspondence concerning this article should be addressed to Rikki Patton, Department of Counseling, 27 South Forge Street, Akron, OH 44325. E-mail: [email protected] 625
PATTON, BLOW, BOHNERT, BONAR, BARRY, AND WALTON
This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
626
involved in TS. Less is known about the degree to which men engaged in TS utilize health care services, although prior reports have suggested that, among a sample of men receiving substance abuse treatment, TS-involved men who were involved in TS reported greater likelihood of use of inpatient mental health services compared with other patients, but not emergency services (Burnette et al., 2008). One study examining the utility of a brief intervention with drug-positive men and women recruited through the ED who reported either cocaine or heroin use stated that approximately 12% of patients reported TS involvement in the past 30 days (Bernstein et al., 2012). Although this study highlights the prevalence of TS involvement within an ED setting among a high-risk substance-using group, findings are limited in their generalizability and applicability to the entire ED patient population.
Present Study The current study aims to address the gaps in the literature by, first, evaluating the prevalence of TS involvement among broad a sample of both men and women seeking care in an ED. In addition, given the prior research regarding increased HIV risk and substance abuse associated with TS involvement (Bobashev et al., 2009; Shannon et al., 2008), we evaluated whether substance abuse and HIV risk indicators differed among individuals involved in TS compared with other ED patients. We hypothesized that those ED patients who reported TS involvement would also report higher rates of substance use and HIV-related risk behaviors than those who did not report TS.
Method Study Design and Setting The current study used data collected as part of a screening survey aimed at identifying patients eligible for a randomized control trial of adult patients (ages 18 – 60) presenting to an ED located in an Academic Level 1 Trauma Center a Midwestern city with similar rates of crime and poverty as other large cities. The study was approved and conducted in compliance with Institutional Review Board (IRB) requirements. A Certificate of Confidentiality was obtained for this study. Data were collected from February, 2011 to March, 2013. Research staff approached participants and described the study; those who were interested in participating provided written informed consent and selfadministered a 15-min computerized screening survey. Participants were compensated for screening with a gift valuing $1 (e.g., playing cards or hand lotion). Patients who presented with acute psychosis, acute sexual assault, medically unstable, or who were in police custody were excluded from screening.
non-TS group and all other respondents were recoded into the group labeled as TS group. Demographic variables. Gender, race, age, household income, marital status, current employment, sexual orientation, and education level were queried using items from validated surveys (e.g., National Survey of Drug Use and Health, Office of Applied Studies, 2009; Psychiatric Outcomes Module: Substance Abuse Outcomes Module, Smith et al., 1996; Global Appraisal of Individual Needs, version 5.4.0., 2006).
Risk Variables Reason for ED visit. Participants were asked a yes/no question regarding whether or not their visit to the ED was injury related, referring to cuts, bruises, broken bones, and so forth. Alcohol and drug use. Alcohol use severity over the past 3 months was assessed with the 10-item Alcohol Use Disorders Identification Test (AUDIT; Saunders, Aasland, Babor, De La Fluente, & Grant, 1993). Questions assessed frequency of alcohol use, number of drinks consumed on a typical drinking day, frequency of having five or more drinks, and negative consequences due to drinking. Participants’ responses on these items were summed to create a single total score reflecting alcohol use severity (Saunders et al., 1993). Cronbach’s alpha for this measure in the current sample was .90. The Alcohol, Smoking and Substance Involvement Screening Test (ASSIST; WHO ASSIST Working Group, 2002) was used to measure frequency of drug use. Respondents were asked if they had used the following drugs at least once in the past 3 months— cocaine, marijuana, methamphetamines, hallucinogens, inhalants, prescription stimulants, prescription sedatives, prescription opioids, street opioids, or any other drugs. Due to the distribution of the data, with low frequencies of use for all illicit drugs, excluding marijuana, and nonmedical use of prescription drugs, these variables were refined into the following groups representing using at least once in the past 3 months: (a) marijuana, (b) other illicit drugs, and (c) nonmedical use of prescription drugs. HIV risk behaviors. Sexual and drug use HIV risk behaviors during the past 3 months were assessed using items selected from the HIV Risk-Taking Behavior Scale (HRBS; Ward, Darke, & Hall, 1990). Specifically, injection drug use, number of sexual partners, inconsistent condom use, and anal sex were queried. Each variable was dichotomized to denote engaging in the risky behavior within the past 3 months. STIs were measured by asking respondents if a doctor or other medical professional ever told the patient that they had an STI, such as chlamydia, gonorrhea, herpes, or syphilis. Respondents answered yes or no to this question (see Substance Abuse & Mental Health Services Administration, 2009).
Data Analysis Plan Measures Transactional sex involvement. TS involvement was assessed using the participant’s responses to the following question from the HIV Risk-taking Behavior Scale (HRBS; Ward, Darke, & Hall, 1990)—“In the past three months, how often have you used condoms when you have been paid for sex?” Respondents who answered “no paid sex” were recoded into the group labeled
Data were analyzed using SAS version 9.3 (SAS Institute Inc., 2012). First, demographic characteristics of the sample were examined. Variable distributions were examined for normality and appropriate statistics were used. Bivariate analyses, including chisquare and independent samples t tests, were conducted to determine if there were significant differences between the TS group and other participants on demographic characteristics, substance
TRANSACTIONAL SEX INVOLVEMENT IN AN ED SAMPLE
use, and HIV risk behaviors. Finally, a hierarchical logistic regression was conducted to determine the associations of TS involvement with demographic and risk variables. A hierarchical model was used in order to control for demographic characteristics in later steps. Demographic factors were entered on Step 1, followed by substance use variables on Step 2, with sex risk behaviors entered on Step 3. All variables that were significant in bivariate analyses were retained in the regression model. Model fit statistics indicated no evidence of multicollinearity.
the local IRB, we were not able to collect any additional data about characteristics of patients who refused without written informed consent. See Table 1 for sample characteristics.
Bivariate Analysis See Table 1 for a full summary of findings from the bivariate analyses. Chi-square/t test analyses indicated there were several significant differences between the TS group and the non-TS group. Participants in the TS group were more likely to report an educational level of high school completion or below (2 ⫽ 22.9; df ⫽ 1), an annual income more than $20,000, and to be married (2 ⫽ 10.8; df ⫽ 1). The TS group had higher scores on all substance use and HIV risk measures as compared with the non-TS group. There were no significant differences regarding gender, race, age, or sexual orientation between the two groups.
Results This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
627
Sample Characteristics As part of the larger randomized controlled trial (RCT), 6,161 individuals were approached for screening, of whom, 4,575 (74.3%) agreed to complete the screening survey. Respondents completing the screening survey were compared with those who were missed and who refused on gender and race. Males were more likely to be missed (2 ⫽ 94.1; p ⬍ .0001) and to refuse participation (2 ⫽ 30.95; p ⬍ .0001). There were no significant differences by race based on those missed (2 ⫽ 2.576; p ⫽ .11) or refused (2 ⫽ 3.526; p ⫽ .06). Per privacy-related policies of
Regression Analysis The hierarchical logistic regression analysis was conducted in order to examine the relationship between transactional sex involvement and Step 1: demographic characteristics (gender, race, educational level, income, marital status, and ED presentation);
Table 1 Frequencies and Bivariate Analysis for the Whole Sample and Subsamples Transactional sex involvement (n ⫽ 610; 13.3%)
Full sample (n ⫽ 4,575) Variable Demographics Female gender Caucasian race High school/GED education or lessⴱⴱ Age (mean/SD) Household income less than $20k/yrⴱ Nonheterosexual orientation† Married/living togetherⴱⴱ Correlates ED presentation for injuryⴱ Alcohol and/or drug use (past 3 months) Alcohol use severity (mean/SD)ⴱⴱ Any marijuana useⴱⴱ Any other illicit drugⴱⴱ (cocaine, heroin, methamphetamines, inhalants, hallucinogens) Cocaineⴱⴱ Heroinⴱⴱ Methamphetamineⴱⴱⴱ Inhalantsⴱⴱⴱ Hallucinogensⴱⴱⴱ Any prescription drug useⴱⴱ Stimulantsⴱ Sedativesⴱⴱ Opioidsⴱⴱ HIV risk behaviors Ever inject drugsⴱⴱ Diagnosed with STI in past 3 monthsⴱⴱ More than 1 sex partner in past 3 monthsⴱⴱ Inconsistent condom use in past 3 monthsⴱⴱ Engaged in anal sex past in three monthsⴱⴱ † ⴱ
No transactional sex involvement (n ⫽ 3,965)
Frequency
%
Frequency
%
Frequency
3045 2033 2645 34.3 2378 323 1458
66.6 44.4 57.8 12.1 52.0 8.1 31.9
393 249 407 33.9 291 58 230
64.4 40.8 66.7 11.5 47.7 10.0 37.7
2652 1784 2238 34.3 2087 265 1228
66.9 45.0 56.4 12.2 52.6 7.8 31.0
1330
29.1
198
32.5
1132
28.6
2.7 1124
5.2 24.6
3.9 185
7.0 30.3
2.5 939
4.9 23.7
184 149 51 11 4 17 206 24 103 139
4.0 3.3 1.1 0.2 0.1 0.4 4.5 0.5 2.2 3.0
62 47 22 3 2 4 52 7 35 37
10.2 7.7 3.6 0.5 0.3 0.7 8.5 1.1 5.7 6.1
122 102 29 8 2 13 154 17 68 102
140 243 684 2733 398
3.1 5.3 15.0 59.7 11.4
36 51 131 560 104
5.9 8.4 21.5 91.8 17.0
104 192 553 2173 294
n ⫽ 584 reported selves celibate, did not select any orientation. p ⬍ .05. ⴱⴱ p ⬍ 0.001. ⴱⴱⴱ Fisher’s exact test, due to small cell frequencies, nonsignificant finding.
%
3.1 2.6 0.7 0.2 0.05 0.3 3.9 0.4 1.7 2.6 2.6 4.8 14.0 54.8 10.3
PATTON, BLOW, BOHNERT, BONAR, BARRY, AND WALTON
628
This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
Table 2 Hierarchical Binary Logistic Regression Assessing Correlates of Transactional Sex Involvement Variable
Model 1
Model 2
Model 3
Step 1 Male gender Caucasian race High school/GED education or less Household income less than $20k Married or living together ED presentation for injury Step 2 Alcohol use severity (past 3 months) Marijuana use (past 3 months) Other illicit drug use (past 3 months) Any prescription drug use (past 3 months) Step 3 Ever inject drugs? Diagnosed with STI (past 3 months) # sexual partners ⬎ 1 (past 3 months) Inconsistent condom use (past 3 months) Engaged in anal sex (past 3 months) Fit Statistics LR 2 (df) Generalized R2 Hosmer-Lemeshow goodness of fit 2
AOR [95%CI] 1.02 [0.85, 1.23] 0.80ⴱ [0.67, 0.95] 1.59ⴱⴱⴱ [1.32, 1.91] 0.83ⴱ [0.70, 0.98] 1.43ⴱⴱ [1.19, 1.72] 1.18 [0.97, 1.42]
AOR [95%CI] 0.88 [0.72, 1.06] 0.78ⴱⴱ [0.65, 0.93] 1.58ⴱⴱⴱ [1.31, 1.89] 0.79ⴱⴱ [0.67, 0.95] 1.53ⴱⴱⴱ [1.27, 1.84] 1.14 [0.94, 1.38]
AOR [95%CI] 0.94 [0.76, 1.15] 0.75ⴱⴱ [0.63, 0.91] 1.65ⴱⴱⴱ [1.36, 2.00] 0.68ⴱⴱⴱ [0.56, 0.83] 1.08 [0.89, 1.32] 1.15 [0.94, 1.41]
— — — —
1.03ⴱⴱⴱ [1.01, 1.04] 1.11 [0.90, 1.36] 2.62ⴱⴱⴱ [1.81, 3.81] 1.53ⴱ [1.06, 2.22]
1.03ⴱⴱⴱ [1.01, 1.05] 0.97 [0.78, 1.20] 2.19ⴱⴱ [1.44, 3.33] 1.60ⴱ [1.07, 2.41]
— — — — —
— — — — —
1.70ⴱ [1.05, 2.76] 0.58ⴱⴱⴱ [0.45, 0.73] 1.20 [0.94, 1.53] 9.75ⴱⴱⴱ [7.17, 13.24] 1.77ⴱⴱⴱ [1.36, 2.30]
51.01 (6) 0.02 6.67 (8)
127.79 (10) 0.05 10.41 (8)
523.48 (15) 0.20 11.85 (8)
ⴱ
p ⬍ .05.
ⴱⴱ
p ⬍ .01.
ⴱⴱⴱ
p ⬍ .001.
Step 2: substance use variables (alcohol use severity, any marijuana use, any other illicit drug use, and any prescription drug use); and Step 3: sexual risk behaviors (inject drugs, diagnosed with an STI, number of sexual partners, inconsistent condom use, and anal sex). Although gender and race were not significant in the bivariate analysis, these variables were included in the multivariate analysis due to the strong evidence in prior literature regarding the higher probability of women and racial minorities engaging in transactional sex behaviors. Model fit statistics indicated that each step improved the model (see Table 2). In Step 1, being Caucasian and income (making less than $20,000) were negatively associated with TS involvement (AOR ⫽ 0.80 and 0.83, respectively). Having a high school education or less (AOR ⫽ 1.59) and being married or living together as married (AOR ⫽ 1.43) were positively associated with TS involvement. No other demographic variables were significant. With the addition of substance abuse variables in Step 2 the following variables were positively associated with TS involvement: alcohol use severity (AOR ⫽ 1.03), illicit drug use (AOR ⫽ 2.62), prescription drug use (AOR ⫽ 1.53), educational status (AOR ⫽ 1.58), and marital status (AOR ⫽ 1.53). Race (being Caucasian) and income were negatively associated with TS involvement (AOR ⫽ 0.78 and 0.79, respectively). In the final model including HIV-risk variables, the following variables were positively associated with TS involvement: alcohol use severity (AOR ⫽ 1.03), illicit drug use (AOR ⫽ 2.19), prescription drug use (AOR ⫽ 1.60), ever injecting drugs (AOR ⫽ 1.70), inconsistent condom use (AOR ⫽ 9.75), and engaging in anal sex (AOR ⫽ 1.77). Individuals involved in TS were less likely to be Caucasian (AOR ⫽ 0.75), earn less than $20,000 annually (AOR ⫽ 0.68), and to ever be diagnosed with an STI (AOR ⫽ 0.58), and were more likely to report an education level of high school diploma or less (AOR ⫽ 1.65). Nonsignificant
variables in the final model included gender, marital status, reason for presenting to the ED, marijuana use, and having multiple sexual partners in the past 3 months.
Discussion This study presents novel findings regarding the prevalence and correlates of TS involvement within an urban ED sample of adult men and women. Findings indicated that 13.3% of patients sampled had engaged in TS within the past 3 months and, of those, gender distribution was similar to the larger sample. Findings also showed that patients involved in TS were more likely to report substance use and HIV-risk behaviors compared with other patients. That more than one in 10 patients presenting to this urban ED reported recent TS involvement has several implications for engagement and treatment. Most research examining TS involvement includes substance-abusing samples only, with rates of TS involvement ranging from 12%– 44% (e.g., Bernstein et al., 2012; Burnette et al., 2008). The current sample included individuals with and without substance abuse problems and the prevalence was still within the range of substance-abusing only samples. Thus, the ED may be a useful venue for engaging individuals involved in transactional sex in brief interventions without limiting services to only those individuals who also present with substance abuse problems. Additionally, the current findings showed that TS involvement was not associated with gender, suggesting that both men and women involved in TS may be reached in one location—in the ED. The present findings for ED patients also support previous reports of increased alcohol and illicit drug use among individuals involved in TS in the general population (Clarke et al., 2012; Wechsberg et al., 2009). A unique finding from this study is that TS involvement was related to increased odds of misusing pre-
This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
TRANSACTIONAL SEX INVOLVEMENT IN AN ED SAMPLE
scription drugs, including opioids and sedative/hypnotics. Given these findings and the recent increases of prescription drug misuse throughout the United States (Gu, Dillon, & Burt, 2010; SAMHSA, 2009), additional research is needed to understand this association in order to appropriately tailor intervention strategies to address this emerging pattern of substance use. Prior research suggests that individuals who engage in TS may have higher levels of substance use for multiple reasons, including coping with the stresses related to TS involvement (Belcher & Herr, 2005; Burnette et al., 2008; Wechsberg et al., 2009). Alternatively, TS may be partially a result of greater involvement in substance use among impoverished populations as a means to maintain substance use (Clarke et al., 2012; Potterat et al., 1998). Although the causal nature of this relationship requires future study, our findings suggest that the ED may be an appropriate venue for interventions addressing substance use and TS, both of which are associated with risk for STIs and HIV infection, among this vulnerable population. Finally, the present findings also showed a significant association between TS and HIV-risk behaviors, including injecting drugs, STI diagnosis, inconsistent condom use, and anal sex among patients recruited from the ED, which are consistent with prior work with substance-using samples (Bobashev et al., 2009; Shannon et al., 2008). Interestingly, current findings indicate that men and women who engage in TS are less likely to have received a formal diagnosis of a STI, despite also being more likely to report inconsistent condom use. This may not necessarily reflect a true decreased risk of STI in this population, however, because it is possible that individuals from this vulnerable population may have more barriers to STI screening and treatment services (Kurtz, Surratt, Kiley, Marion, & Inciardi, 2005). Overall, current findings suggest that the ED may be a relevant site for intervention efforts focused on sexual risk reduction among men and women, particularly those who engage in TS given the high prevalence (91.8%) of inconsistent condom use, one of the most potent predictors of HIV/STI.
Limitations and Future Directions Although the present study augments current literature by examining the prevalence and correlates of TS involvement among men and women in the ED, several limitations should be noted. The criterion we used to define TS was limited by the use of a single item assessment regarding condom use with partners who paid for sex. Although this question asked about recent involvement in exchanging sex for money, the depth of involvement in sex work for these participants is unknown. In addition, this question only inquired about being paid for sex, and may not have identified those who trade sex for drugs or other goods. Further, one of the exclusion criteria for the original study was presenting to the ED for acute sexual assault. Individuals involved in exchanging sex for money experience repeated sexual assaults (Dalla, Xia, & Kennedy, 2003; Karandikar & Prospero, 2010) and these exclusion criteria may have biased the sample, producing a conservative estimate of the prevalence of TS involvement among ED patients. Despite these limitations, the results showed that 13.3% of individuals attending the ED, including both men and women, report engaging in TS. Additionally, although our sample was racially diverse, these results may not be generalizable to other populations
629
or racial/ethnic groups (e.g., Hispanics, Asians), or to other geographic locales due to recruitment from a single site within an ED located in an economically depressed urban area. Further, the analysis was cross-sectional in design, thereby limiting our ability to understand causal relationships among TS and substance use or other factors. To address these limitations, future research should explore the relationship between TS involvement and use of the ED longitudinally in order to determine how these individuals use the ED to meet their needs and their connection to other services, such as substance abuse treatment. More in-depth assessment of TS involvement is warranted to allow further understanding of how interventions can be developed to effectively intervene with individuals in the ED who engage in high-risk sexual behaviors and substance use. Additionally, given that so many men receiving services in the ED also reported TS involvement, it may be helpful to explore gender differences in TS to inform prevention, behavioral intervention, and medical screening and treatment of STIs. Focusing on these individuals, defined as a difficult-to-reach population, for intervention services in the ED may prove to be a fruitful area for research and practice. Future research is needed to develop and implement combined HIV risk and substance use intervention programs tailored for patients in the ED engaged in both behaviors.
References Belcher, J. R., & Herr, S. (2005). Development of grounded theory: Moving towards a theory of the pathways into street prostitution among low-income women. Journal of Addictions Nursing, 16, 117–124. doi: 10.1080/10884600500196651 Benoit, C., Jansson, M., Millar, A., & Phillips, R. (2005). CommunityAcademic research on hard-to-reach populations: Benefits and challenges. Qualitative Health Research, 15, 263–282. doi:10.1177/ 1049732304267752 Bernstein, E., Ashong, D., Heeren, T., Winter, M., Bliss, C., Madico, G., & Bernstein, J. (2012). The impact of a brief motivational intervention on unprotected sex and sex while high among drug-positive emergency department patients who receive STI/HIV VC/T and drug treatment referral as standard of care. AIDS and Behavior, 16, 1203–1216. doi:10.1007/s10461012-0134-0 Bobashev, G. V., Zule, W. A., Osilla, K. C., Kline, T. C., & Wechsberg, W. M. (2009). Transactional sex among men and women in the South at high risk for HIV and other STIs. Journal of Urban Health, 86, 32– 47. doi:10.1007/s11524-009-9368-1 Burnette, M. L., Lucas, E., Ilgen, M., Frayne, S. M., Mayo, J., & Weitlauf, J. C. (2008). Prevalence and health correlates of prostitution among patients entering treatment for substance use disorders. Archives of General Psychiatry, 65, 337–344. doi:10.1001/archpsyc.65.3.337 Clarke, R. J., Clarke, E. A., Roe-Sepowitz, D., & Fey, R. (2012). Age at entry into prostitution: Relationship to drug use, race, suicide, education level, childhood abuse, and family experiences. Journal of Human Behavior in the Social Environment, 22, 270 –289. doi:10.1080/ 10911359.2012.655583 Dalla, R. L., Xia, Y., & Kennedy, H. (2003). “You just give them what they want and pray they don’t kill you”: Street-level sex workers’ reports of victimization. Violence Against Women, 9, 1367–1394. doi:10.1177/ 1077801203255679 Edwards, J. M., Halpern, C. T., & Wechsberg, W. M. (2006). Correlates of exchanging sex for drugs or money among women who use crack cocaine. AIDS Education and Prevention, 18, 420 – 429. doi:10.1521/ aeap.2006.18.5.420
This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
630
PATTON, BLOW, BOHNERT, BONAR, BARRY, AND WALTON
Gu, Q., Dillon, C. F., & Burt, V. L. (2010). Prescription drug use continues to increase: U. S. prescription drug data for 2007–2008. NCHS Data Brief, 42. Karandikar, S., & Prospero, M. (2010). From client to pimp: Male violence against female sex workers. Journal of Interpersonal Violence, 25, 257–273. doi:10.1177/0886260509334393 Kurtz, S. P., Surratt, H. L., Kiley, M. C., & Inciardi, J. A. (2005). Barriers to health and social services for street-based sex workers. Journal of Health Care for the Poor and Underserved, 16, 345–361. doi:10.1353/ hpu.2005.0038 Leukefeld, C. (1999). HIV risk behavior and drug use among heterosexual male crack users by prostitution involvement. Population Research and Policy Review, 18, 23–38. doi:10.1023/A:1006192426092 Logan, T. K., & Leukefeld, C. (2000). Sexual and drug use behaviors among female crack users: A multi-site study. Drug and Alcohol Dependence, 58, 237–245. doi:10.1016/S0376-8716(99)00096-4 Office of Applied Studies. (2009). 2010 national survey on drug use and health: CAI specifications for programming English version. Rockville, MD: Substance Abuse and Mental Health Services Administration. Potterat, J. J., Rothenberg, R. B., Muth, S. Q., Darrow, W. M., & PhillipsPlummer, L. (1998). Pathways to prostitution: The chronology of sexual and drug abuse milestones. Journal of Sex Research, 35, 333–340. doi:10.1080/00224499809551951 Raymond, J. G., Hughes, D. M., & Gomez, C. J. (2001). Sex trafficking of women in the United States: International and domestic trends. Washington, DC: U.S. Department of Justice. doi:10.1037/e624522007-001 SAS Institute Inc. (2012). What’s new in SAS® 9.3. Cary, NC: SAS Institute Inc. Saunders, J. B., Aasland, O. G., Babor, T. F., De La Fuente, J. R., & Grant, M. (1993). Development of the Alcohol Use Disorders Identification Test (AUDIT): WHO Collaborative project on early detections of persons with harmful alcohol consumption-II. Addiction, 88, 791– 804. doi:10.1111/j.1360-0443.1993.tb02093.x
Shannon, K., Bright, V., Duddy, J., & Tyndall, M. W. (2005). Access and utilization of HIV treatment and services among women sex workers in Vancouver’s downtown eastside. Journal of Urban Health, 82, 488 – 497. doi:10.1093/jurban/jti076 Shannon, K., Kerr, T., Allinott, S., Chettiar, J., Shoveller, J., & Tyndall, M. W. (2008). Social and structural violence and power relationship in mitigating HIV risk of drug-using women in survival sex work. Social Science & Medicine, 66, 911–921. doi:10.1016/j.socscimed.2007.11.008 Smith, G. R., Kramer, T., Babor, T., Burnam, M. A., Mosley, C. L., & Rost, K., Burns, B. (1996). Substance and outcomes module–User’s manual. 1996 –2007. Little Rock, AR: University of Arkansas for Medical Sciences. Substance Abuse and Mental Health Services Administration, Office of Applied Studies. (February 5, 2009). The NSDUH report: Trends in nonmedical use of prescription pain relievers: 2002 to 2007. Rockville, MD: Substance Abuse and Mental Health Services Administration, Office of Applied Studies. Ward, J., Darke, S., & Hall, W. (1990). The HIV risk-taking behavior scale (HRBS) manual. Sydney, Australia: National Drug and Alcohol Research Centre. Wechsberg, W. M., Wu, L., Zule, W. A., Parry, C. D., Browne, F. A., Luseno, W. K., . . . Gentry, A. (2009). Substance abuse, treatment needs and access among female sex workers and non-sex workers in Pretoria, South Africa. Substance Abuse Treatment, Prevention, and Policy, 4, 1–11. doi:10.1186/1747-597X-4-11 ASSIST Working Group. (2002). The Alcohol, Smoking and Substance Involvement Screening Test (ASSIST): Development, reliability and feasibility. Addiction, 97, 1183–1194. doi:10.1046/j.1360-0443.2002 .00185.x
Received April 23, 2013 Revision received August 6, 2013 Accepted September 16, 2013 䡲
