Drug and Alcohol Dependence 144 (2014) 78–86
Contents lists available at ScienceDirect
Drug and Alcohol Dependence journal homepage: www.elsevier.com/locate/drugalcdep
Correlates of selling sex among male injection drug users in New York City Kathleen H. Reilly a,∗ , Alan Neaigus a , Travis Wendel b , David M. Marshall IV c , Holly Hagan d a
HIV Epidemiology Program, New York City Department of Health and Mental Hygiene, 42-09 28th Street, Long Island City, NY 11101, USA St. Ann’s Corner of Harm Reduction, 310 Walton Ave., Bronx, NY 10451, USA c Department of Anthropology, John Jay College of Criminal Justice, 899 10th Ave., New York, NY 10019, USA d College of Nursing, New York University, 726 Broadway, New York, NY 10003, USA b
a r t i c l e
i n f o
Article history: Received 4 June 2014 Received in revised form 15 August 2014 Accepted 15 August 2014 Available online 24 August 2014 Keywords: Men Injection drug use Drug use Exchange sex Sexual behavior
a b s t r a c t Background: Compared to female IDUs, the correlates of receiving money, drugs, or other things in exchange for sex (“selling sex”) among male IDUs are not well understood. Methods: In 2012, IDUs were sampled in New York City for the National HIV Behavioral Surveillance crosssectional study using respondent driven sampling. Analyses were limited to male participants. Logistic regression was used to calculate crude and adjusted odds ratios (aOR) and 95% confidence intervals (95% CI) to determine the correlates of selling sex to (1) men and (2) women in the past 12 months. Results: Of 394 males, 35 (8.9%) sold sex to men and 66 (16.8%) sold sex to women. Correlates of selling sex to men included bisexual/gay identity (aOR: 31.0; 95% CI: 8.1, 119.1), Bronx residence (vs. Manhattan) (aOR: 38.1; 95% CI: 6.2, 235.5), and in the past 12 months, being homeless (aOR: 9.9; 95% CI: 2.0, 49.6), ≥3 sex partners (aOR: 26.2; 95% CI: 4.7, 147.6), non-injection cocaine use (aOR: 5.4; 95% CI: 1.6, 18.2), and injecting methamphetamine (aOR: 36.9; 95% CI: 5.7, 240.0). Correlates of selling sex to women included, in the past 12 months, ≥3 sex partners (aOR: 14.6; 95% CI: 6.6, 31.9), binge drinking at least once a week (aOR: 3.1; 95% CI: 1.6, 6.1), non-injection crack use (aOR: 3.3; 95% CI: 1.6, 6.7), most frequently injected “speedball” (vs. heroin) (aOR: 2.1; 95% CI: 1.1, 4.2), and receptively shared syringes (aOR: 2.4; 95%CI: 1.2, 4.8). Conclusions: Among male IDUs, those who sold sex had more sex partners, which may facilitate the sexual spread of HIV among IDUs and to non-IDU male and female sex partners. HIV prevention interventions aimed at male IDUs who sell sex should consider both their sexual and parenteral risks and the greater risk of engaging in exchange sex associated with the use of injection and non-injection stimulant drugs. © 2014 Elsevier Ireland Ltd. All rights reserved.
1. Introduction The availability of syringe exchange programs and the legalization of selling syringes at pharmacies without a prescription in New York City has correlated with a reduction in the number of new HIV infections among injection drug users (IDUs) (Des Jarlais et al., 2005). The risk of HIV infection from contaminated syringes is thought to have declined over time, however the risk from sexual transmission has increased among IDUs and many IDUs engage in both injection equipment sharing and sexual risk behaviors (Des Jarlais et al., 2009; Neaigus et al., 2013). The results of a 2011 study of HIV, hepatitis C, and herpes simplex virus type 2 among IDUs in
∗ Corresponding author. Tel.: +1 347 396 7755; fax: +1 347 396 7793. E-mail address: [email protected] (K.H. Reilly). http://dx.doi.org/10.1016/j.drugalcdep.2014.08.010 0376-8716/© 2014 Elsevier Ireland Ltd. All rights reserved.
NYC suggest that most recent HIV infections among IDUs occurred through sexual transmission rather than from injection drug use (Des Jarlais et al., 2011). IDUs may also serve to bridge the HIV epidemic to the general population. In a NYC study, HIV infection among heterosexuals with no history of injecting drugs was associated with having sexual partnerships with IDUs (Jenness et al., 2010). Those who exchange sex for things including drugs or money (“sell sex”) have been shown to be at greater risk for HIV and STIs (Dunkle et al., 2004; Rolfs et al., 1990; Rudolph et al., 2013). Exchange sex has previously been found to be associated with HIV infection among high-risk heterosexual men in NYC (Jenness et al., 2011). Those who sell sex may be incentivized to not use condoms (Jie et al., 2012) may have greater numbers of sex partners (Edwards et al., 2006), and may have more risky sex partners (Weber et al., 2002). Exchange sex has been found to be common among female
K.H. Reilly et al. / Drug and Alcohol Dependence 144 (2014) 78–86
IDUs (Astemborski et al., 1994; Davey-Rothwell and Latkin, 2008) and the majority of research concerning exchange sex has focused on females selling sex. By contrast, there is little literature available on the prevalence of exchange sex among male IDUs, and studies of exchange sex among drug-using men have mostly focused on men who have sex with men (MSM) (Newman et al., 2004; Semple et al., 2010). Male IDUs that sell sex are considered to be at increased risk for HIV infection compared to male IDUs that do not sell sex (Kuyper et al., 2004). Potential correlates for selling sex among male IDUs include age, polydrug and alcohol use, high risk sexual behavior (high number of sex partners and sex without a condom), homelessness, and HIV positive status (Edwards et al., 2006; Kuyper et al., 2004; Latkin et al., 2003; Salazar et al., 2007; Wood et al., 2007). The current study examines the prevalence and correlates of selling sex among a sample of male IDUs in NYC. 2. Methods 2.1. Study population Participants were recruited to participate in the Center for Disease Control and Prevention (CDC)-sponsored National HIV Behavioral Surveillance (NHBS) study of injection drug users (IDU) in NYC from August to November of 2012. NHBS is an ongoing national, cross-sectional study sponsored by the CDC that monitors HIV risk behaviors, testing history, exposure to and use of HIV prevention services, and HIV prevalence among men who have sex with men (MSM), IDU, and high-risk heterosexuals (HET) in three-year cycles (Gallagher et al., 2007; Lansky et al., 2007). NHBS is conducted in collaboration with the CDC by local public health departments, universities, and other organizations. Respondent-driven sampling (RDS) was used for participant recruitment. RDS has been shown to be effective at reaching hidden populations, for which no sampling frame exists, where individuals are connected by strong social networks (Heckathorn, 2007). Ethnographers selected 12 initial recruits (“seeds”) through community outreach. Eligibility criteria for the study were having injected drugs without a prescription in the past 12 months (participants must have had physical signs of recent injection [fresh track marks, needle-sized scabs, or abscesses] or sufficient knowledge of drug preparation, injection, and syringes), at least 18 years of age, NYC residence, and English or Spanish comprehension. Additionally, seeds were required to identify as either male or female and not transgender. Once the seeds completed the interview and testing, they were asked to recruit up to three peers. Non-seed participants had to be referred into the study by other participants. Potential participants who were referred to the study were screened for eligibility. Those who were screened as eligible and provided their informed consent were given a structured survey interview administered in private by trained interviewers and were offered voluntary HIV, hepatitis B, and hepatitis C tests, and were provided with up to three coupons for peers they could refer to the study. Successive waves were recruited until the desired sample size was reached. Subjects were compensated $20 for completing the survey, $10 for testing for HIV, $10 for testing for hepatitis B and C, and $10 for each eligible participant recruited. All study procedures involving human subjects were approved by the New York City Department of Health and Mental Hygiene (DOHMH) and John Jay College of Criminal Justice Institutional Review Boards. 2.2. Measures The survey instrument was developed by the CDC in collaboration with local NHBS project sites. Interview data were collected on demographics; sexual behavior with main and casual partners;
79
alcohol and drug use history; HIV testing experiences; medical history; and exposure to HIV prevention activities. Questions pertaining to sexual behavior were framed in terms of behaviors in the past 12 months and at last sex; questions pertaining to drug use were framed in terms of the past 12 months and last syringesharing event. Participants were asked if they had ever received “things like money or drugs” in exchange for sex in the past 12 months from any of their main or casual, male or female, sex partners. For the purpose of characterizing the direction of the exchange, this manuscript refers to receiving things in exchange for sex as “selling sex” even though this exchange does not necessarily indicate a monetary transaction. For those participants who reported multiple male or female, main or casual, partners in the past 12 months, participants were asked for each “how many gave you things like money or drugs in exchange for sex?”; for those who reported one male or female, main or casual, partner in the past 12 months, participants were asked did this partner “give you things like money or drugs in exchange for sex?” (yes/no). Participants were considered to have sold sex if they reported that at least one partner gave them things like money or drugs in exchange for sex. Separate variables were created for selling sex to female partners and male partners, respectively. Phlebotomists collected blood specimens using venipuncture. Specimens were screened for HIV antibody on HIV1/2 enzyme-linked immunosorbent assay (ELISA) and confirmed using HIV1 Western blot platforms (BioRad Laboratories, Hercules, CA, USA). Hepatitis C infection was determined using chemiluminescence immunoassay (CIA) (VITROS Anti-HCV assay, Ortho-Clinical Diagnostics, Raritan, NJ, USA). Participants were asked to return in 2 weeks for their test results.
2.3. Statistical analysis Weighted and unweighted frequencies were compared and differences were considered significant if their 95% confidence intervals did not overlap (Niccolai et al., 2010). Data were weighted to reduce recruitment biases common in chain-referral methods (preferential in-group recruitment [homophily] and large network sizes; Heckathorn, 2007). Estimates of participants’ network sizes were determined through self-report by asking separately how many males and females they know who inject and whom they have seen in the past 30 days; the total of male and female IDUs known by participants was considered their IDU social network size. Weights were generated using RDS Analysis Tool (RDSAT) 7.1 (Ithaca, NY, USA). Analyses were restricted to non-seed males. Medians and interquartile ranges (IQR) for non-normal continuous data, and frequencies and unweighted and weighted percentages and 95% confidence intervals (CI) for each level of categorical variables were calculated. RDS is a relatively new sampling method and analysis techniques are still being developed (Heckathorn, 1997). RDS weighted estimates may not be generalizable to the target population if assumptions are not met and large sample sizes may be needed to obtain precise weighted estimates (Salganik, 2006; Wejnert et al., 2012; White et al., 2012). Methods incorporating weights in regression models are still under development (Johnston et al., 2008; Winship and Radbill, 1994). Associations between selected variables and (1) selling sex to men in the past 12 months and (2) selling sex to women in the past 12 months were examined through the estimation of crude and adjusted odds ratios (respectively, OR and aOR) and 95% confidence intervals (95% CI) using unweighted logistic regression models. Variables significant (p < 0.1) in bivariate analyses were considered for inclusion in the multivariate logistic regression models. Variables were eliminated from multivariate models using stepwise selection with p < 0.1 for entry and p < 0.05
80
K.H. Reilly et al. / Drug and Alcohol Dependence 144 (2014) 78–86
Table 1 Unweighted and weighted percentages of participant characteristics among male injection drug users in New York City (n = 394). Variable Demographics Age 18–29 30–39 40–49 50+ Race/ethnicity Latino Black White Other Language preference English Spanish Marital status Married or cohabiting Never married, separated, divorced, widowed Place of birth United States Puerto Rico Other country Graduated high school Yes No Sexual identity Straight Gay/bisexual Borough Manhattan Brooklyn Bronx Queens Staten Island NYC Residence $10,000 Unemployed Yes No Arresteda Yes No Homelessa Yes No Sexual behavior ≥3 sex partnersa Yes No Sex without a condom with ≥3 partnersa Yes No Substance use Bingeb drinking at least once a weeka Yes No Non-injection cocaine usea Yes No Non-injection crack usea Yes No Non-injection methamphetamine usea Yes No Injection methamphetamine usea Yes No Most frequently injected drug Heroin Cocaine “Speedball” (heroin + cocaine)
n
Unweighted % (95% CI)
Weighted % (95% CI)
29 106 137 122
7.4% (5.0–10.4%) 26.9% (22.6–31.6%) 34.8% (30.1–39.7%) 31.0% (26.4–35.8%)
14.2% (3.6–24.7%) 20.2% (14.5–25.9%) 31.0% (23.8–38.2%) 34.6% (26.9–42.4%)
261 79 50 4
66.2% (61.3–70.9%) 20.1% (16.2–24.4%) 12.7% (9.6–16.4%) 1.0% (0.3–2.6%)
59.4% (51.0–67.9%) 27.7% (20.1–35.3%) 12.4% (7.1–17.6%) 0.5% (0.0–1.2%)
294 100
74.6% (70.0–78.8%) 25.4% (21.2–30.0%)
82.6% (77.4–87.7%) 17.4% (12.3–22.6%)
56 338
14.2% (10.9–18.1%) 85.8% (81.9–89.1%)
11.7% (7.2–16.3%) 88.3% (83.7–92.8%)
233 154 7
59.1% (54.1–64.0%) 39.1% (34.2–44.1%) 1.8% (0.1–3.6%)
70.5% (63.9–77.2%) 28.5% (21.9–35.1%) 0.9% (0.1–1.7%)
201 193
51.0% (46.0–56.0%) 49.0% (44.0–54.0%)
52.3% (43.5–61.2%) 47.7% (38.8–56.5%)
356 37
90.6% (87.3–93.3%) 9.4% (6.7–12.7%)
93.2% (90.1–96.4%) 6.8% (3.6–9.9%)
65 152 139 37 1
16.5% (13.0–20.5%) 38.6% (33.8–43.6%) 35.3% (30.6–40.2%) 9.4% (6.7–12.7%) 0.3% (0.0–1.4%)
15.0% (7.9–22.2%) 41.2% (33.1–49.2%) 31.5% (24.0–38.9%) 12.2% (4.6–19.8%) 0.1% (0.0–0.4%)
35 359
8.9% (6.3–12.1%) 91.1% (87.9–93.7%)
6.8% (3.6–10.0%) 93.2% (90.0–96.4%)
291 100
74.4% (69.8–78.7%) 25.6% (21.3–30.2%)
75.7% (69.1–82.3%) 24.3% (17.7–30.9%)
352 42
89.3% (85.9–92.2%) 10.7% (7.8–14.1%)
91.3% (86.6–96.0%) 8.7% (4.0–13.4%)
151 243
38.3% (33.5–43.3%) 61.7% (56.7–66.5%)
41.8% (32.8–50.7%) 58.2% (49.3–67.2%)
201 193
51.0% (46.0–56.0%) 49.0% (44.0–54.0%)
43.8% (35.0–52.6%) 56.2% (47.4–65.0%)
156 238
39.6% (34.7–44.6%) 60.4% (55.4–65.3%)
28.4% (21.8–34.9%) 71.6% (65.1–78.2%)
112 282
28.4% (24.0–33.2%) 71.6% (66.8–76.0%)
20.9% (15.2–26.5%) 79.1% (73.5–84.8%)
93 301
23.6% (19.5–28.1%) 76.4% (71.9–80.5%)
24.7% (16.4–33.1%) 75.3% (66.9–83.6%)
100 294
25.4% (21.2–30.0%) 74.6% (70.0–78.8%)
29.0% (19.4–38.7%) 71.0% (61.3–80.6%)
87 307
22.1% (18.1–26.5%) 77.9% (73.5–81.9%)
19.4% (13.3–25.5%) 80.6% (74.5–86.7%)
9 385
2.3% (1.0–4.3%) 97.7% (95.7–99.0%)
3.4% (0.0–7.4%) 96.6% (92.6–100.0%)
17 377
4.3% (2.5–6.8%) 95.7% (93.2–97.5%)
2.9% (0.9–4.9%) 97.1% (95.1–99.1%)
249 31 114
63.2% (58.2–68.0%) 7.9% (5.4–11.0%) 28.9% (24.5–33.7%)
65.7% (56.1–75.2%) 10.6% (2.6–18.6%) 23.7% (15.5–31.9%)
K.H. Reilly et al. / Drug and Alcohol Dependence 144 (2014) 78–86
81
Table 1 (Continued) Variable
n
Unweighted % (95% CI)
Weighted % (95% CI)
103 291
26.1% (21.9–30.8%) 73.9% (69.2–78.1%)
18.4% (13.4–23.4%) 81.6% (76.6–86.6%)
194 200
49.2% (44.2–54.3%) 50.8% (45.7–55.8%)
43.2% (34.5–52.0%) 56.8% (48.0–65.5%)
253 141
64.2% (59.3–68.9%) 35.8% (31.1–40.7%)
56.2% (47.4–65.0%) 43.8% (35.0–52.6%)
28 366
7.1% (4.8–10.1%) 92.9% (89.9–95.2%)
7.9% (4.0–11.8%) 92.1% (88.2–96.0%)
43 337
11.3% (8.3–14.9%) 88.7% (85.1–91.7%)
17.6% (8.5–26.6%) 82.4% (73.4–91.5%)
259 107
70.8% (65.8–75.4%) 29.2% (24.6–34.2%)
66.7% (57.7–75.7%) 33.3% (24.3–42.3%)
a
Receptively shared syringes Yes No Receptively shared any injection equipment (syringes, cookers, cotton, water)a Yes No Injects at least dailya Yes No Infection status Self-reported HIV status Positive Negative or unknown HIV study test result Positive Negative Hepatitis C study test result Positive Negative a b
Past 12 months. Had at least 5 drinks in one sitting.
for retention. Analyses were conducted using SAS 9.2 (Cary, NC, USA). 3. Results Twelve seeds recruited a total of 525 participants. Of these, 394 (75.1%) were male. The median age of male participants was 45 (IQR: 36, 51). There were no significant differences between unweighted and weighted percentages (Table 1) and unweighted results are presented in the following text. Participants identified as Latino (66.2%), black (20.1%), white (12.7%), and other race/ethnicity (1.0%). Most of the participants were more comfortable using English with family and friends (74.6%), were neither married nor cohabiting (85.8%), identified their sexual orientation as straight (90.6%), and had lived in NYC for more than 12 months (91.1%). The majority were unemployed (89.3%), had an annual household income ≤$10,000 (74.4%), and reported being homeless at some time in the past 12 months (51.0%). Approximately half had graduated from high school (51.0%); the majority were born in the United States (59.1%). Participants reported their borough of residence as Brooklyn (38.6%), the Bronx (35.3%), Manhattan (16.5%), Queens (9.4%), and Staten Island (0.3%). More than onethird (38.3%) reported having been arrested in the past 12 months. In the past 12 months, 39.6% reported having ≥3 sex partners and 28.4% reported having sex without a condom with ≥3 sex partners. One-quarter (23.6%) reported binge drinking (at least 5 drinks in one sitting) at least once a week, 25.4% reported noninjection cocaine use, 22.1% reported non-injection crack use, 2.3% reported non-injection methamphetamine use, and 4.3% reported injection methamphetamine use in the past 12 months. The most frequently injected drug reported was heroin by itself (63.2%), followed by “speedball” (heroin and cocaine) (28.9%), and cocaine by itself (7.9%). More than half (64.2%) reported injecting at least daily, 26.1% reported receptively sharing syringes, and 49.2% reported receptively sharing other injection equipment (cookers, cotton, or water) in the past 12 months. Fewer than one-tenth (7.1%) selfreported HIV positive status, whereas 11.3% had positive HIV study test results. The majority of male study participants (70.8%) had positive hepatitis C study test results. Approximately one-fifth of male participants (n = 85, 21.6%) reported selling sex in the past 12 months to either main or casual partners. Fewer than ten percent (n = 35, 8.9%) of male participants reported selling sex to male partners. The large majority of those who reported selling sex to men also reported having sex with
women in the past 12 months (n = 30, 85.7%). Selling sex to women was almost twice as common, with 66 (16.8%) male participants reporting selling sex to female partners. Of those who reported selling sex to women, 20 (30.3%) also reported having sex with men. Sixteen (4.1%) male participants reported selling sex to both men and women in the past 12 months. Table 2 shows variables associated with male participants selling sex to men in the past 12 months. In the bivariate analysis, variables significantly associated (p < 0.1) with selling sex to men in the past 12 months included being age 18–29 years and 30–39 years vs. ≥50 years, Spanish language preference, being foreignborn and Puerto Rican-born (vs. born in the United States), gay or bisexual identity, living in the Bronx vs. Manhattan, living in NYC less than 12 months, having been homeless in the past 12 months, and past 12 month behavior of having ≥3 sex partners, having sex without a condom with ≥3 sex partners, binge drinking at least once a week, non-injection cocaine use, non-injection crack use, noninjection methamphetamine use, injection methamphetamine use, injecting cocaine and “speedball” most frequently (vs. injecting heroin most frequently), and receptively sharing syringes. Variables independently associated (p < 0.05) with selling sex to men in the multivariate model included gay or bisexual sexual identity (vs. straight) (adjusted odds ratio (aOR): 31.0; 95% CI: 8.1, 119.1), Bronx residence (vs. Manhattan (aOR: 38.1; 95% CI: 6.2, 235.5), having been homeless in the past 12 months (aOR: 9.9; 95% CI: 2.0, 49.6), past 12 month behaviors of having ≥3 sex partners (aOR: 26.2; 95% CI: 4.7, 147.6), non-injection cocaine use (aOR: 5.4; 95% CI: 1.6, 18.2), and injection methamphetamine use (aOR: 36.9; 95% CI: 5.7, 240.0). Table 3 shows the variables associated with male participants selling sex to women in the past 12 months. In the bivariate analysis, variables associated (p < 0.1) with selling sex to women in the past 12 months included not being married or cohabiting, being foreign born (vs. born in the United States), gay or bisexual identity, residing in the boroughs of the Bronx and Queens (vs. Manhattan), and past 12 month behavior of ever having been homeless, having ≥3 sex partners, having sex without a condom with ≥3 sex partners, binge drinking at least once a week, non-injection cocaine use, non-injection crack use, injecting “speedball” most frequently (vs. heroin), receptively sharing syringes, and receptively sharing injection equipment (cookers, cotton, or water). Variables independently associated (p < 0.05) with selling sex to women in the multivariate model included past 12 month behavior of having ≥3 sex partners (aOR: 14.6; 95% CI: 6.6, 31.9), binge drinking at least
82
K.H. Reilly et al. / Drug and Alcohol Dependence 144 (2014) 78–86
Table 2 Factors associated with selling sex to males in the past 12 months among male injection drug users in New York City (n = 394). Variable Demographics Age 18–29 30–39 40–49 50+ Race/ethnicity Latino Black White Other Language preference English Spanish Marital status Married or cohabiting Never married, separated, divorced, widowed Place of birth United States Puerto Rico Other country Graduated high school Yes No Sexual identity Straight Gay/bisexual Borough Manhattan Brooklyn Bronx Queens Staten Island NYC residence $10,000 Unemployed Yes No Arresteda Yes No Homelessa Yes No Sexual behavior ≥3 sex partnersa Yes No Sex without a condom with ≥3 partnersa Yes No Substance use Bingeb drinking at least once a weeka Yes No Non-injection cocaine usea Yes No Non-injection crack usea Yes No Non-injection methamphetamine usea Yes No Injection methamphetamine usea Yes No Most frequently injected drug Heroin Cocaine “Speedball” (heroin + cocaine)
% Sold sex to males in the past 12 months
Odds ratio (95% CI)
Adjusted odds ratio (95% CI)
13.8% 14.2% 7.3% 4.9%
3.1 (0.8, 11.8)* 3.2 (1.2, 8.5)** 1.5 (0.5, 4.3) 1.0
10.7% 3.8% 6.0% 2.9%
1.0 0.3 (0.1, 1.1)* 0.5 (0.2, 1.8) 2.8 (0.3, 27.6)
6.8% 15.0%
1.0 2.4 (1.2, 4.9)**
5.4% 9.5%
1.0 1.8 (0.5, 6.3)
5.2% 13.0% 42.9%
1.0 2.7 (1.3, 5.8)** 13.8 (2.8, 68.8)**
10.0% 7.8%
1.3 (0.7, 2.6) 1.0
3.7% 59.5%
1.0 38.7 (16.4, 91.3)**
1.0 31.0 (8.1, 119.1)**
6.2% 2.0% 18.0% 8.1% 0.0%
1.0 0.3 (0.1, 1.4) 3.3 (1.1, 10.1)** 1.3 (0.3, 6.4) –
1.0 0.5 (0.05, 6.3) 38.1 (6.2, 235.5)** 9.4 (0.9, 98.5)* –
17.1% 8.1%
2.4 (0.9, 6.1)* 1.0
8.6% 10.0%
0.8 (0.4, 1.8) 1.0
9.1% 7.1%
1.3 (0.4, 4.4) 1.0
8.0% 9.5%
0.8 (0.4, 1.7) 1.0
14.9% 2.6%
6.6 (2.5, 17.4)** 1.0
9.9 (2.0, 49.6)** 1.0
19.2% 2.1%
11.1 (4.2, 29.3)** 1.0
26.2 (4.7, 147.6)** 1.0
19.6% 4.6%
5.1 (2.4, 10.5)** 1.0
14.0% 7.3%
2.1 (1.0, 4.3)* 1.0
19.0% 5.4%
4.1 (2.0, 8.3)** 1.0
20.7% 5.5%
4.5 (2.2, 9.1)** 1.0
66.7% 7.5%
24.5 (5.8, 103.2)** 1.0
47.1% 7.2%
11.5 (4.1, 32.3)** 1.0
6.0% 16.1% 13.2%
1.0 3.0 (1.0, 8.9)** 2.4 (1.1, 5.0)**
5.4 (1.6, 18.2)** 1.0
36.9 (5.7, 240.0)** 1.0
K.H. Reilly et al. / Drug and Alcohol Dependence 144 (2014) 78–86
83
Table 2 (Continued) Variable
% Sold sex to males in the past 12 months
Odds ratio (95% CI)
15.5% 6.5%
2.6 (1.3, 5.3)** 1.0
11.0% 7.0%
1.6 (0.8, 3.3) 1.0
11.5% 4.3%
2.9 (1.2, 7.2)** 1.0
10.7% 8.7%
1.3 (0.4, 4.4) 1.0
Adjusted odds ratio (95% CI)
a
Receptively shared syringes Yes No Receptively shared injection equipment (cookers, cotton, water)a Yes No Injects at least dailya Yes No Infection status Self-reported HIV status Positive Negative or unknown a b * **
Past 12 months. Had at least 5 drinks in one sitting. p < 0.1. p < 0.05.
once a week (aOR: 3.1; 95% CI: 1.6, 6.1), non-injection crack use (aOR: 3.3; 95% CI: 1.6, 6.7), injecting “speedball” most frequently (vs. heroin) (aOR: 2.1; 95% CI: 1.1, 4.2), and receptively sharing syringes (aOR: 2.4; 95% CI: 1.2, 4.8).
4. Discussion Approximately one-fifth of male study participants reported selling sex in the past 12 months, which is higher than the levels found among men in studies of heroin users in the UK and IDUs in Vancouver, Canada (Gossop et al., 1993; Kuyper et al., 2004). Almost twice as many participants in the current study reported selling sex to women than selling sex to men; this is unprecedented in the literature on exchange sex. To our knowledge, this is one of the only U.S. studies to examine factors associated with men selling sex to women. Although women may be more likely to sell sex than men (Jenness et al., 2011), the frequency of men selling sex to women found in this study challenges the conventional perception of the gendered direction of exchange sex. Other research has found that men who have sex with both men and women are more likely to sell sex, compared to those who only have sex with men (Dyer et al., 2013; Weber et al., 2001). Unsurprisingly, those who identified as gay or bisexual were more likely to report selling sex to men than those who did not. Almost all male IDUs who reported selling sex to men, however, also reported having sex with women; these male IDUs may not only be at potentially greater risk for acquiring HIV, but also be at increased risk for transmitting HIV to both their male and female sex partners. The ALIVE longitudinal study of IDUs in Baltimore found sex with a man to be associated with HIV seroconversion among male IDUs, whereas high-risk heterosexual sex was associated with HIV seroconversion among female IDUs (Strathdee et al., 2001). Bronx residence was also independently associated with selling sex to men. The Bronx is one of the most impoverished counties in the United States, with over 30% of all residents living below the poverty line (U.S. Census Bureau, 2012). Sex may be used as an alternative currency in high-poverty areas, where access to resources is limited. Many other studies have found that men who are homeless are more likely to sell sex than those who are stably housed (Bobashev et al., 2009; Newman et al., 2004). Although the current study was cross-sectional, previous research indicates that homelessness often precedes drug use and involvement in sex work among MSM (Clatts et al., 2005). Edwards et al. (2006) also found that those who sell sex have greater numbers of partners, which may increase their exposure risk for HIV and other STIs and position them as a potential core group for HIV transmission. Although all participants were IDUs, those who used stimulants (non-injection
cocaine and injection methamphetamine) were more likely to report selling sex to men. Previous research has also found associations between use of stimulants and engaging in exchange sex (Newman et al., 2004; Ober et al., 2009; Windle, 1997). Having greater numbers of sex partners in the past 12 months was independently associated with selling sex to women. Also similar to men who reported selling sex to men, stimulant use (noninjection crack and most often injected “speedball”) was associated with selling sex to women. Non-injection crack use has repeatedly been found to be associated with selling sex in other studies (Newman et al., 2004; Sherman et al., 2011). A study conducted among male and female drug users in 22 U.S. cities found that injectors who also smoked crack were more likely than injectors who did not smoke crack to sell sex for drugs or money (Booth et al., 2000). The current study found that binge drinking at least weekly was also independently associated with selling sex to women. Although there are limited data available on alcohol use among men who sell sex, binge drinking has previously been associated with high risk sexual behaviors including multiple sex partners and not using condoms (Caldeira et al., 2009; Vosburgh et al., 2012). A Baltimore study found that compared to moderate drinkers or alcohol abstainers, problematic alcohol users were more likely to inject drugs before or during sex, which may have implications for sexual risk behaviors (Scherer et al., 2013). The current study also found that those who sold sex to women were more likely to have receptively shared syringes in the past 12 months. Although no data could be found on syringe-sharing behaviors among men who sell sex to women, a previous study of IDUs in the South Bronx found that those who exchanged sex for drugs or other things were more likely to share syringes than those who did not (Fernando et al., 2003). This study is subject to several limitations. The cross-sectional nature of this study precludes the determination of temporal relationships. Although one of the advantages of RDS is that it has the potential to reduce bias by producing long recruitment chains that cross-bridge between relevant groups, it is possible that this study oversampled from well-connected social networks that differ from the target population, particularly with respect to income and employment. The results of this study may therefore not be generalizable to other areas or populations. The study questionnaire elicited sensitive information regarding sexual behavior and HIV status, and participants may not have felt comfortable disclosing this information to study staff interviewers. The limited sample size may have restricted the power to detect differences within the target population, and may have impeded the inclusion of other relevant variables in the final multivariate models, or may have led to spurious associations, which would be mitigated by a larger sample. Some point estimates had wide confidence intervals reducing the precision of correlate effects. Also, no information was collected
84
K.H. Reilly et al. / Drug and Alcohol Dependence 144 (2014) 78–86
Table 3 Factors associated with selling sex to females in the past 12 months among male injection drug users in New York City (n = 394). Variable Demographics Age 18–29 30–39 40–49 50+ Race/ethnicity Latino Black White Other Language preference English Spanish Marital status Married or Cohabiting Never married, separated, divorced, widowed Place of birth United States Puerto Rico Other country Graduated high school Yes No Sexual identity Straight Gay/bisexual Borough Manhattan Brooklyn Bronx Queens Staten Island NYC residence $10,000 Unemployed Yes No Arresteda Yes No Homelessa Yes No Sexual behavior ≥3 sex partnersa Yes No Sex without a condom with ≥3 partnersa Yes No Substance use Bingeb drinking at least once a weeka Yes No Non-injection cocaine usea Yes No Non-injection crack usea Yes No Non-injection methamphetamine usea Yes No Injection methamphetamine usea Yes No Most frequently injected drug Heroin Cocaine “Speedball” (heroin + cocaine)
% sold sex to females in the past 12 months
Odds ratio (95% CI)
13.8% 14.2% 18.2% 18.0%
0.7 (0.2, 2.3) 0.7 (0.4, 1.5) 1.0 (0.5, 1.9) 1.0
19.2% 20.2% 0.0% 0.0%
1.0 1.1 (0.6, 2.0) – –
15.0% 22.0%
1.0 1.6 (0.9, 2.8)
5.4% 18.6%
1.0 4.0 (1.2, 13.3)**
14.6% 17.5% 71.4%
1.0 1.2 (0.7, 2.2) 14.6 (2.7, 78.4)**
16.9% 16.6%
1.0 (0.6, 1.7) 1.0
45.9% 13.8%
1.0 5.3 (2.6, 10.9)**
9.2% 13.8% 21.6% 24.3% 0.0%
1.0 1.6 (0.6, 4.1) 2.7 (1.1, 6.9)** 3.2 (1.0, 9.8)** –
8.6% 17.5%
0.4 (0.1, 1.5) 1.0
16.2% 19.0%
0.8 (0.5, 1.5) 1.0
16.8% 16.7%
1.0 (0.4, 2.4) 1.0
19.2% 15.2%
1.3 (0.8, 2.3) 1.0
20.4% 13.0%
1.7 (1.0, 3.0)** 1.0
35.9% 4.2%
12.8 (6.3, 26.0)** 1.0
34.8% 9.6%
5.0 (2.9, 8.8)** 1.0
35.5% 11.0%
4.5 (2.6, 7.8)** 1.0
27.0% 13.3%
2.4 (1.4, 4.2)** 1.0
32.2% 12.4%
3.4 (1.9, 5.9)** 1.0
33.3% 16.4%
2.6 (0.6, 10.5) 1.0
17.6% 16.7%
1.1 (0.3, 3.8) 1.0
14.5% 3.2% 25.4%
1.0 0.2 (0.02, 1.5) 2.0 (1.2, 3.5)**
Adjusted odds ratio (95% CI)
14.6 (6.6, 31.9)** 1.0
3.1 (1.6, 6.1)** 1.0
3.3 (1.6, 6.7)** 1.0
1.0 0.1 (0.02, 1.0)* 2.1 (1.1, 4.2)*
K.H. Reilly et al. / Drug and Alcohol Dependence 144 (2014) 78–86
85
Table 3 (Continued) Variable
% sold sex to females in the past 12 months
Odds ratio (95% CI)
Adjusted odds ratio (95% CI)
28.2% 12.7%
2.7 (1.6, 4.7)** 1.0
2.4 (1.2, 4.8)** 1.0
23.8% 10.8%
2.6 (1.5, 4.5)** 1.0
16.6% 17.0%
1.0 (0.6, 1.7) 1.0
25.0% 16.1%
1.7 (0.7, 4.3) 1.0
a
Receptively shared syringes Yes No Receptively shared injection equipment (cookers, cotton, water)a Yes No Injects at least dailya Yes No Infection status Self-reported HIV status Positive Negative or unknown a b * **
Past 12 months. Had at least 5 drinks in one sitting. p < 0.1. p < 0.05.
on mental health or history of sexual abuse, which has previously been found to be associated with exchange sex among men (Haley et al., 2004; Weber et al., 2001). This study also did not differentiate between types of exchange sex (e.g., commercial sex, survival sex, and “sex-for-presents”). There may be differences in perceptions of what constitutes exchanging sex, since the definition in the question was broad and may be dependent on culturally-determined subjective interpretation. More data are needed to learn more about the contexts of selling sex including the characteristics of exchange partnerships, condom use, and what is being exchanged. A substantial proportion of male IDUs in NYC reported selling sex. Although this study did not find an association between selling sex and HIV infection, this behavior has the potential for increased risk for HIV acquisition and transmission, especially among those who sell sex to men, given that MSM are disproportionately affected by the HIV epidemic both in New York City and nationally. Increased risk may not only occur because of efficient transmission of HIV, but also because of greater exposure to high-risk partners through sexual networks. In the current study, almost all of those who reported selling sex to men also reported having female sex partners, which could lead to the introduction and spread of heterosexuallytransmitted HIV to women in partnerships with high-risk men who also have sex with men. HIV prevention interventions aimed at IDU men who sell sex should consider those factors which increase their vulnerability to engaging in exchange sex, sexual transmission risk, and other risk behaviors. Future studies should examine whether male IDUs who sell sex are more likely to become infected with or transmit HIV and STIs compared to those who do not sell sex. Interventions might not only be beneficial for preventing individual transmission; targeting those who sell sex could have populationlevel effects by reducing transmission risks in exchange networks.
Author disclosures None.
Role of funding source This research was funded by a cooperative agreement between the New York City Department of Health and Mental Hygiene (NYC DOHMH) and the Centers for Disease Control and Prevention (CDC) (grant# 5U1BPS003246-02). The CDC developed the study design and methodology, but had no further role in the collection, analysis and interpretation of data; in the writing of the report; nor in the decision to submit the paper for publication. The contents of this article are solely the responsibility of the authors and do not
necessarily represent the official views of The Centers for Disease Control and Prevention. Contributors Author KHR undertook the statistical analysis and wrote the first draft of the manuscript. AN and TW directed the research study from which the data were derived. DMM collected the data. All authors contributed to and have approved the final manuscript. Conflict of interest All authors declare that they have no conflicts of interest. Acknowledgements The authors acknowledge Gabriela Paz-Bailey, Dita Broz, and Michael “Trey” Spiller of CDC for their contributions to the National HIV Behavioral Surveillance (NHBS) study design; Sarah Braunstein, Kent Sepkowitz, John Rojas, and Jay Varma of the NYC DOHMH for reviewing previous drafts of this article; and the NYC NHBS field staff for all their efforts. References Astemborski, J., Vlahov, D., Warren, D., Solomon, L., Nelson, K., 1994. The trading of sex for drugs or money and HIV seropositivity among female intravenous drug users. Am. J. Public Health 84, 382–387. Bobashev, G.V., Zule, W.A., Osilla, K.C., Kline, T.L., Wechsberg, W.M., 2009. Transactional sex among men and women in the south at high risk for HIV and other STIs. J. Urban Health 86, 32–47. Booth, R.E., Kwiatkowski, C.F., Chitwood, D.D., 2000. Sex related HIV risk behaviors: differential risks among injection drug users, crack smokers, and injection drug users who smoke crack. Drug Alcohol Depend. 58, 219–226. Caldeira, K.M., Arria, A.M., O’Grady, K.E., Zarate, E.M., Vincent, K.B., Wish, E.D., 2009. Prospective associations between alcohol and drug consumption and risky sex among female college students. J. Alcohol Drug Educ. 53, nihpa115858. Clatts, M.C., Goldsamt, L., Yi, H., Viorst Gwadz, M., 2005. Homelessness and drug abuse among young men who have sex with men in New York City: a preliminary epidemiological trajectory. J. Adolesc. 28, 201–214. Davey-Rothwell, M., Latkin, C., 2008. An examination of perceived norms and exchanging sex for money or drugs among women injectors in Baltimore, MD, USA. Int. J. STD AIDS 19, 47–50. Des Jarlais, D.C., Arasteh, K., McKnight, C., Hagan, H., Perlman, D., Friedman, S.R., 2009. Using hepatitis C virus and herpes simplex virus-2 to track HIV among injecting drug users in New York City. Drug Alcohol Depend. 101, 88–91. Des Jarlais, D.C., Arasteh, K., McKnight, C., Hagan, H., Perlman, D.C., Semaan, S., 2011. Associations between herpes simplex virus type 2 and HCV with HIV among injecting drug users in New York City: the current importance of sexual transmission of HIV. Am. J. Public Health 101, 1277–1283. Des Jarlais, D.C., Perlis, T., Arasteh, K., Torian, L.V., Beatrice, S., Milliken, J., Mildvan, D., Yancovitz, S., Friedman, S.R., 2005. HIV incidence among injection drug users in New York City, 1990 to 2002: use of serologic test algorithm to assess expansion of HIV prevention services. Am. J. Public Health 95, 1439–1444.
86
K.H. Reilly et al. / Drug and Alcohol Dependence 144 (2014) 78–86
Dunkle, K.L., Jewkes, R.K., Brown, H.C., Gray, G.E., McIntryre, J.A., Harlow, S.D., 2004. Transactional sex among women in Soweto, South Africa: prevalence, risk factors and association with HIV infection. Soc. Sci. Med. 59, 1581–1592. Dyer, T.P., Regan, R., Wilton, L., Harawa, N.T., Ou, S.S., Wang, L., Shoptaw, S., 2013. Differences in substance use, psychosocial characteristics and HIV-related sexual risk behavior between black men who have sex with men only (BMSMO) and black men who have sex with men and women (BMSMW) in six US cities. J. Urban Health 90, 1181–1193. Edwards, J.M., Iritani, B.J., Hallfors, D.D., 2006. Prevalence and correlates of exchanging sex for drugs or money among adolescents in the United States. Sex. Transm. Infect. 82, 354–358. Fernando, D., Schilling, R.F., Fontdevila, J., El-Bassel, N., 2003. Predictors of sharing drugs among injection drug users in the South Bronx: implications for HIV transmission. J. Psychoact. Drugs 35, 227–236. Gallagher, K.M., Sullivan, P.S., Lansky, A., Onorato, I.M., 2007. Behavioral surveillance among people at risk for HIV infection in the US: the National HIV Behavioral Surveillance System. Public Health Rep. 122, 32–38. Gossop, M., Griffiths, P., Powis, B., Strang, J., 1993. Severity of heroin dependence and HIV risk. I. Sexual behaviour. AIDS Care 5, 149–157. Haley, N., Roy, E., Leclerc, P., Boudreau, J., Boivin, J., 2004. HIV risk profile of male street youth involved in survival sex. Sex. Transm. Infect. 80, 526–530. Heckathorn, D.D., 1997. Respondent-driven sampling: a new approach to the study of hidden populations. Soc. Prob. 44, 174–199. Heckathorn, D.D., 2007. Extensions of respondent driven sampling: analyzing continuous variables and controlling for differential recruitment. Sociol. Methodol. 37, 151–207. Jenness, S.M., Kobrak, P., Wendel, T., Neaigus, A., Murrill, C.S., Hagan, H., 2011. Patterns of exchange sex and HIV infection in high-risk heterosexual men and women. J. Urban Health 88, 329–341. Jenness, S.M., Neaigus, A., Hagan, H., Murrill, C.S., Wendel, T., 2010. Heterosexual HIV and sexual partnerships between injection drug users and noninjection drug users. AIDS Patient Care STDS 24, 175–181. Jie, W., Xiaolan, Z., Ciyong, L., Moyer, E., Hui, W., Lingyao, H., Xueqing, D., 2012. A Qualitative exploration of barriers to condom use among female sex workers in China. PLoS One 7, e46786. Johnston, L.G., Malekinejad, M., Kendall, C., Iuppa, I.M., Rutherford, G.W., 2008. Implementation challenges to using respondent-driven sampling methodology for HIV biological and behavioral surveillance: field experiences in international settings. AIDS Behav. 12, 131–141. Kuyper, L., Lampinen, T., Li, K., Spittal, P., Hogg, R., Schechter, M., Wood, E., 2004. Factors associated with sex trade involvement among male participants in a prospective study of injection drug users. Sex. Transm. Infect. 80, 531–535. Lansky, A., Sullivan, P.S., Gallagher, K.M., Fleming, P.L., 2007. HIV behavioral surveillance in the US: a conceptual framework. Public Health Rep. 122, 16–23. Latkin, C., Hua, W., Forman, V., 2003. The relationship between social network characteristics and exchanging sex for drugs or money among drug users in Baltimore, MD, USA. Int. J. STD AIDS 14, 770–775. Neaigus, A., Reilly, K.H., Jenness, S.M., Hagan, H., Wendel, T., Gelpi-Acosta, C., 2013. Dual HIV risk: receptive syringe sharing and unprotected sex among HIV-negative injection drug users in New York City. AIDS Behav. 17, 2501–2509. Newman, P.A., Rhodes, F., Weiss, R.E., 2004. Correlates of sex trading among drugusing men who have sex with men. Am. J. Public Health 94, 1998–2003. Niccolai, L.M., Toussova, O.V., Verevochkin, S.V., Barbour, R., Heimer, R., Kozlov, A.P., 2010. High HIV prevalence, suboptimal HIV testing, and low knowledge of
HIV-positive serostatus among injection drug users in St. Petersburg, Russia. AIDS Behav. 14, 932–941. Ober, A., Shoptaw, S., Wang, P.-C., Gorbach, P., Weiss, R.E., 2009. Factors associated with event-level stimulant use during sex in a sample of older, low-income men who have sex with men in Los Angeles. Drug Alcohol Depend. 102, 123–129. Rolfs, R.T., Goldberg, M., Sharrar, R.G., 1990. Risk factors for syphilis: cocaine use and prostitution. Am. J. Public Health 80, 853–857. Rudolph, A.E., Crawford, N.D., Latkin, C., Fowler, J.H., Fuller, C.M., 2013. Individual and neighborhood correlates of membership in drug using networks with a higher prevalence of HIV in New York City (2006–2009). Ann. Epidemiol. 23, 267–274. Salazar, L.F., Crosby, R.A., Holtgrave, D.R., Head, S., Hadsock, B., Todd, J., Shouse, R.L., 2007. Homelessness and HIV-associated risk behavior among African American men who inject drugs and reside in the urban south of the United States. AIDS Behav. 11, 70–77. Salganik, M.J., 2006. Variance estimation, design effects, and sample size calculations for respondent-driven sampling. J. Urban Health 83, 98–112. Scherer, M., Trenz, R., Harrell, P., Mauro, P., Latimer, W., 2013. The role of drinking severity on sex risk behavior and HIV exposure among illicit drug users. Am. J. Addict. 22, 239–245. Semple, S.J., Strathdee, S.A., Zians, J., Patterson, T.L., 2010. Social and behavioral characteristics of HIV-positive MSM who trade sex for methamphetamine. Am. J. Drug Alcohol Abuse 36, 325–331. Sherman, S.G., Reuben, J., Chapman, C.S., Lilleston, P., 2011. Risks associated with crack cocaine smoking among exotic dancers in Baltimore, MD. Drug Alcohol Depend. 114, 249–252. Strathdee, S.A., Galai, N., Safaiean, M., Celentano, D.D., Vlahov, D., Johnson, L., Nelson, K.E., 2001. Sex differences in risk factors for HIV seroconversion among injection drug users: a 10-year perspective. Arch. Intern. Med. 161, 1281–1288. U.S. Census Bureau, 2012. Small Area Income and Poverty Estimates [Data], 2011 American Community Survey. U.S. Census Bureau, http://www.census.gov/ did/www/saipe/data/statecounty/data/2011.html (accessed on December 11, 2013. Vosburgh, H.W., Mansergh, G., Sullivan, P.S., Purcell, D.W., 2012. A review of the literature on event-level substance use and sexual risk behavior among men who have sex with men. AIDS Behav. 16, 1394–1410. Weber, A.E., Boivin, J.-F., Blais, L., Haley, N., 2002. HIV risk profile and prostitution among female street youths. J. Urban Health 79, 525–535. Weber, A.E., Craib, K.J., Chan, K., Martindale, S., Miller, M.L., Schechter, M.T., Hogg, R.S., 2001. Sex trade involvement and rates of human immunodeficiency virus positivity among young gay and bisexual men. Int. J. Epidemiol. 30, 1449–1454. Wejnert, C., Pham, H., Krishna, N., Le, B., DiNenno, E., 2012. Estimating design effect and calculating sample size for respondent-driven sampling studies of injection drug users in the United States. AIDS Behav. 16, 797–806. White, R.G., Lansky, A., Goel, S., Wilson, D., Hladik, W., Hakim, A., Frost, S.D.W., 2012. Respondent driven sampling—where we are and where should we be going? Sex. Transm. Infect. 88, 397–399. Windle, M., 1997. The trading of sex for money or drugs, sexually transmitted diseases (STDs), and HIV-related risk behaviors among multisubstance using alcoholic inpatients. Drug Alcohol Depend. 49, 33–38. Winship, C., Radbill, L., 1994. Sampling weights and regression analysis. Sociol. Methods Res. 23, 230–257. Wood, E., Schachar, J., Li, K., Stoltz, J.-A., Shannon, K., Miller, C., Lloyd-Smith, E., Tyndall, M.W., Kerr, T., 2007. Sex trade involvement is associated with elevated HIV incidence among injection drug users in Vancouver. Addict. Res. Theory 15, 321–325.
Contents lists available at ScienceDirect
Drug and Alcohol Dependence journal homepage: www.elsevier.com/locate/drugalcdep
Correlates of selling sex among male injection drug users in New York City Kathleen H. Reilly a,∗ , Alan Neaigus a , Travis Wendel b , David M. Marshall IV c , Holly Hagan d a
HIV Epidemiology Program, New York City Department of Health and Mental Hygiene, 42-09 28th Street, Long Island City, NY 11101, USA St. Ann’s Corner of Harm Reduction, 310 Walton Ave., Bronx, NY 10451, USA c Department of Anthropology, John Jay College of Criminal Justice, 899 10th Ave., New York, NY 10019, USA d College of Nursing, New York University, 726 Broadway, New York, NY 10003, USA b
a r t i c l e
i n f o
Article history: Received 4 June 2014 Received in revised form 15 August 2014 Accepted 15 August 2014 Available online 24 August 2014 Keywords: Men Injection drug use Drug use Exchange sex Sexual behavior
a b s t r a c t Background: Compared to female IDUs, the correlates of receiving money, drugs, or other things in exchange for sex (“selling sex”) among male IDUs are not well understood. Methods: In 2012, IDUs were sampled in New York City for the National HIV Behavioral Surveillance crosssectional study using respondent driven sampling. Analyses were limited to male participants. Logistic regression was used to calculate crude and adjusted odds ratios (aOR) and 95% confidence intervals (95% CI) to determine the correlates of selling sex to (1) men and (2) women in the past 12 months. Results: Of 394 males, 35 (8.9%) sold sex to men and 66 (16.8%) sold sex to women. Correlates of selling sex to men included bisexual/gay identity (aOR: 31.0; 95% CI: 8.1, 119.1), Bronx residence (vs. Manhattan) (aOR: 38.1; 95% CI: 6.2, 235.5), and in the past 12 months, being homeless (aOR: 9.9; 95% CI: 2.0, 49.6), ≥3 sex partners (aOR: 26.2; 95% CI: 4.7, 147.6), non-injection cocaine use (aOR: 5.4; 95% CI: 1.6, 18.2), and injecting methamphetamine (aOR: 36.9; 95% CI: 5.7, 240.0). Correlates of selling sex to women included, in the past 12 months, ≥3 sex partners (aOR: 14.6; 95% CI: 6.6, 31.9), binge drinking at least once a week (aOR: 3.1; 95% CI: 1.6, 6.1), non-injection crack use (aOR: 3.3; 95% CI: 1.6, 6.7), most frequently injected “speedball” (vs. heroin) (aOR: 2.1; 95% CI: 1.1, 4.2), and receptively shared syringes (aOR: 2.4; 95%CI: 1.2, 4.8). Conclusions: Among male IDUs, those who sold sex had more sex partners, which may facilitate the sexual spread of HIV among IDUs and to non-IDU male and female sex partners. HIV prevention interventions aimed at male IDUs who sell sex should consider both their sexual and parenteral risks and the greater risk of engaging in exchange sex associated with the use of injection and non-injection stimulant drugs. © 2014 Elsevier Ireland Ltd. All rights reserved.
1. Introduction The availability of syringe exchange programs and the legalization of selling syringes at pharmacies without a prescription in New York City has correlated with a reduction in the number of new HIV infections among injection drug users (IDUs) (Des Jarlais et al., 2005). The risk of HIV infection from contaminated syringes is thought to have declined over time, however the risk from sexual transmission has increased among IDUs and many IDUs engage in both injection equipment sharing and sexual risk behaviors (Des Jarlais et al., 2009; Neaigus et al., 2013). The results of a 2011 study of HIV, hepatitis C, and herpes simplex virus type 2 among IDUs in
∗ Corresponding author. Tel.: +1 347 396 7755; fax: +1 347 396 7793. E-mail address: [email protected] (K.H. Reilly). http://dx.doi.org/10.1016/j.drugalcdep.2014.08.010 0376-8716/© 2014 Elsevier Ireland Ltd. All rights reserved.
NYC suggest that most recent HIV infections among IDUs occurred through sexual transmission rather than from injection drug use (Des Jarlais et al., 2011). IDUs may also serve to bridge the HIV epidemic to the general population. In a NYC study, HIV infection among heterosexuals with no history of injecting drugs was associated with having sexual partnerships with IDUs (Jenness et al., 2010). Those who exchange sex for things including drugs or money (“sell sex”) have been shown to be at greater risk for HIV and STIs (Dunkle et al., 2004; Rolfs et al., 1990; Rudolph et al., 2013). Exchange sex has previously been found to be associated with HIV infection among high-risk heterosexual men in NYC (Jenness et al., 2011). Those who sell sex may be incentivized to not use condoms (Jie et al., 2012) may have greater numbers of sex partners (Edwards et al., 2006), and may have more risky sex partners (Weber et al., 2002). Exchange sex has been found to be common among female
K.H. Reilly et al. / Drug and Alcohol Dependence 144 (2014) 78–86
IDUs (Astemborski et al., 1994; Davey-Rothwell and Latkin, 2008) and the majority of research concerning exchange sex has focused on females selling sex. By contrast, there is little literature available on the prevalence of exchange sex among male IDUs, and studies of exchange sex among drug-using men have mostly focused on men who have sex with men (MSM) (Newman et al., 2004; Semple et al., 2010). Male IDUs that sell sex are considered to be at increased risk for HIV infection compared to male IDUs that do not sell sex (Kuyper et al., 2004). Potential correlates for selling sex among male IDUs include age, polydrug and alcohol use, high risk sexual behavior (high number of sex partners and sex without a condom), homelessness, and HIV positive status (Edwards et al., 2006; Kuyper et al., 2004; Latkin et al., 2003; Salazar et al., 2007; Wood et al., 2007). The current study examines the prevalence and correlates of selling sex among a sample of male IDUs in NYC. 2. Methods 2.1. Study population Participants were recruited to participate in the Center for Disease Control and Prevention (CDC)-sponsored National HIV Behavioral Surveillance (NHBS) study of injection drug users (IDU) in NYC from August to November of 2012. NHBS is an ongoing national, cross-sectional study sponsored by the CDC that monitors HIV risk behaviors, testing history, exposure to and use of HIV prevention services, and HIV prevalence among men who have sex with men (MSM), IDU, and high-risk heterosexuals (HET) in three-year cycles (Gallagher et al., 2007; Lansky et al., 2007). NHBS is conducted in collaboration with the CDC by local public health departments, universities, and other organizations. Respondent-driven sampling (RDS) was used for participant recruitment. RDS has been shown to be effective at reaching hidden populations, for which no sampling frame exists, where individuals are connected by strong social networks (Heckathorn, 2007). Ethnographers selected 12 initial recruits (“seeds”) through community outreach. Eligibility criteria for the study were having injected drugs without a prescription in the past 12 months (participants must have had physical signs of recent injection [fresh track marks, needle-sized scabs, or abscesses] or sufficient knowledge of drug preparation, injection, and syringes), at least 18 years of age, NYC residence, and English or Spanish comprehension. Additionally, seeds were required to identify as either male or female and not transgender. Once the seeds completed the interview and testing, they were asked to recruit up to three peers. Non-seed participants had to be referred into the study by other participants. Potential participants who were referred to the study were screened for eligibility. Those who were screened as eligible and provided their informed consent were given a structured survey interview administered in private by trained interviewers and were offered voluntary HIV, hepatitis B, and hepatitis C tests, and were provided with up to three coupons for peers they could refer to the study. Successive waves were recruited until the desired sample size was reached. Subjects were compensated $20 for completing the survey, $10 for testing for HIV, $10 for testing for hepatitis B and C, and $10 for each eligible participant recruited. All study procedures involving human subjects were approved by the New York City Department of Health and Mental Hygiene (DOHMH) and John Jay College of Criminal Justice Institutional Review Boards. 2.2. Measures The survey instrument was developed by the CDC in collaboration with local NHBS project sites. Interview data were collected on demographics; sexual behavior with main and casual partners;
79
alcohol and drug use history; HIV testing experiences; medical history; and exposure to HIV prevention activities. Questions pertaining to sexual behavior were framed in terms of behaviors in the past 12 months and at last sex; questions pertaining to drug use were framed in terms of the past 12 months and last syringesharing event. Participants were asked if they had ever received “things like money or drugs” in exchange for sex in the past 12 months from any of their main or casual, male or female, sex partners. For the purpose of characterizing the direction of the exchange, this manuscript refers to receiving things in exchange for sex as “selling sex” even though this exchange does not necessarily indicate a monetary transaction. For those participants who reported multiple male or female, main or casual, partners in the past 12 months, participants were asked for each “how many gave you things like money or drugs in exchange for sex?”; for those who reported one male or female, main or casual, partner in the past 12 months, participants were asked did this partner “give you things like money or drugs in exchange for sex?” (yes/no). Participants were considered to have sold sex if they reported that at least one partner gave them things like money or drugs in exchange for sex. Separate variables were created for selling sex to female partners and male partners, respectively. Phlebotomists collected blood specimens using venipuncture. Specimens were screened for HIV antibody on HIV1/2 enzyme-linked immunosorbent assay (ELISA) and confirmed using HIV1 Western blot platforms (BioRad Laboratories, Hercules, CA, USA). Hepatitis C infection was determined using chemiluminescence immunoassay (CIA) (VITROS Anti-HCV assay, Ortho-Clinical Diagnostics, Raritan, NJ, USA). Participants were asked to return in 2 weeks for their test results.
2.3. Statistical analysis Weighted and unweighted frequencies were compared and differences were considered significant if their 95% confidence intervals did not overlap (Niccolai et al., 2010). Data were weighted to reduce recruitment biases common in chain-referral methods (preferential in-group recruitment [homophily] and large network sizes; Heckathorn, 2007). Estimates of participants’ network sizes were determined through self-report by asking separately how many males and females they know who inject and whom they have seen in the past 30 days; the total of male and female IDUs known by participants was considered their IDU social network size. Weights were generated using RDS Analysis Tool (RDSAT) 7.1 (Ithaca, NY, USA). Analyses were restricted to non-seed males. Medians and interquartile ranges (IQR) for non-normal continuous data, and frequencies and unweighted and weighted percentages and 95% confidence intervals (CI) for each level of categorical variables were calculated. RDS is a relatively new sampling method and analysis techniques are still being developed (Heckathorn, 1997). RDS weighted estimates may not be generalizable to the target population if assumptions are not met and large sample sizes may be needed to obtain precise weighted estimates (Salganik, 2006; Wejnert et al., 2012; White et al., 2012). Methods incorporating weights in regression models are still under development (Johnston et al., 2008; Winship and Radbill, 1994). Associations between selected variables and (1) selling sex to men in the past 12 months and (2) selling sex to women in the past 12 months were examined through the estimation of crude and adjusted odds ratios (respectively, OR and aOR) and 95% confidence intervals (95% CI) using unweighted logistic regression models. Variables significant (p < 0.1) in bivariate analyses were considered for inclusion in the multivariate logistic regression models. Variables were eliminated from multivariate models using stepwise selection with p < 0.1 for entry and p < 0.05
80
K.H. Reilly et al. / Drug and Alcohol Dependence 144 (2014) 78–86
Table 1 Unweighted and weighted percentages of participant characteristics among male injection drug users in New York City (n = 394). Variable Demographics Age 18–29 30–39 40–49 50+ Race/ethnicity Latino Black White Other Language preference English Spanish Marital status Married or cohabiting Never married, separated, divorced, widowed Place of birth United States Puerto Rico Other country Graduated high school Yes No Sexual identity Straight Gay/bisexual Borough Manhattan Brooklyn Bronx Queens Staten Island NYC Residence $10,000 Unemployed Yes No Arresteda Yes No Homelessa Yes No Sexual behavior ≥3 sex partnersa Yes No Sex without a condom with ≥3 partnersa Yes No Substance use Bingeb drinking at least once a weeka Yes No Non-injection cocaine usea Yes No Non-injection crack usea Yes No Non-injection methamphetamine usea Yes No Injection methamphetamine usea Yes No Most frequently injected drug Heroin Cocaine “Speedball” (heroin + cocaine)
n
Unweighted % (95% CI)
Weighted % (95% CI)
29 106 137 122
7.4% (5.0–10.4%) 26.9% (22.6–31.6%) 34.8% (30.1–39.7%) 31.0% (26.4–35.8%)
14.2% (3.6–24.7%) 20.2% (14.5–25.9%) 31.0% (23.8–38.2%) 34.6% (26.9–42.4%)
261 79 50 4
66.2% (61.3–70.9%) 20.1% (16.2–24.4%) 12.7% (9.6–16.4%) 1.0% (0.3–2.6%)
59.4% (51.0–67.9%) 27.7% (20.1–35.3%) 12.4% (7.1–17.6%) 0.5% (0.0–1.2%)
294 100
74.6% (70.0–78.8%) 25.4% (21.2–30.0%)
82.6% (77.4–87.7%) 17.4% (12.3–22.6%)
56 338
14.2% (10.9–18.1%) 85.8% (81.9–89.1%)
11.7% (7.2–16.3%) 88.3% (83.7–92.8%)
233 154 7
59.1% (54.1–64.0%) 39.1% (34.2–44.1%) 1.8% (0.1–3.6%)
70.5% (63.9–77.2%) 28.5% (21.9–35.1%) 0.9% (0.1–1.7%)
201 193
51.0% (46.0–56.0%) 49.0% (44.0–54.0%)
52.3% (43.5–61.2%) 47.7% (38.8–56.5%)
356 37
90.6% (87.3–93.3%) 9.4% (6.7–12.7%)
93.2% (90.1–96.4%) 6.8% (3.6–9.9%)
65 152 139 37 1
16.5% (13.0–20.5%) 38.6% (33.8–43.6%) 35.3% (30.6–40.2%) 9.4% (6.7–12.7%) 0.3% (0.0–1.4%)
15.0% (7.9–22.2%) 41.2% (33.1–49.2%) 31.5% (24.0–38.9%) 12.2% (4.6–19.8%) 0.1% (0.0–0.4%)
35 359
8.9% (6.3–12.1%) 91.1% (87.9–93.7%)
6.8% (3.6–10.0%) 93.2% (90.0–96.4%)
291 100
74.4% (69.8–78.7%) 25.6% (21.3–30.2%)
75.7% (69.1–82.3%) 24.3% (17.7–30.9%)
352 42
89.3% (85.9–92.2%) 10.7% (7.8–14.1%)
91.3% (86.6–96.0%) 8.7% (4.0–13.4%)
151 243
38.3% (33.5–43.3%) 61.7% (56.7–66.5%)
41.8% (32.8–50.7%) 58.2% (49.3–67.2%)
201 193
51.0% (46.0–56.0%) 49.0% (44.0–54.0%)
43.8% (35.0–52.6%) 56.2% (47.4–65.0%)
156 238
39.6% (34.7–44.6%) 60.4% (55.4–65.3%)
28.4% (21.8–34.9%) 71.6% (65.1–78.2%)
112 282
28.4% (24.0–33.2%) 71.6% (66.8–76.0%)
20.9% (15.2–26.5%) 79.1% (73.5–84.8%)
93 301
23.6% (19.5–28.1%) 76.4% (71.9–80.5%)
24.7% (16.4–33.1%) 75.3% (66.9–83.6%)
100 294
25.4% (21.2–30.0%) 74.6% (70.0–78.8%)
29.0% (19.4–38.7%) 71.0% (61.3–80.6%)
87 307
22.1% (18.1–26.5%) 77.9% (73.5–81.9%)
19.4% (13.3–25.5%) 80.6% (74.5–86.7%)
9 385
2.3% (1.0–4.3%) 97.7% (95.7–99.0%)
3.4% (0.0–7.4%) 96.6% (92.6–100.0%)
17 377
4.3% (2.5–6.8%) 95.7% (93.2–97.5%)
2.9% (0.9–4.9%) 97.1% (95.1–99.1%)
249 31 114
63.2% (58.2–68.0%) 7.9% (5.4–11.0%) 28.9% (24.5–33.7%)
65.7% (56.1–75.2%) 10.6% (2.6–18.6%) 23.7% (15.5–31.9%)
K.H. Reilly et al. / Drug and Alcohol Dependence 144 (2014) 78–86
81
Table 1 (Continued) Variable
n
Unweighted % (95% CI)
Weighted % (95% CI)
103 291
26.1% (21.9–30.8%) 73.9% (69.2–78.1%)
18.4% (13.4–23.4%) 81.6% (76.6–86.6%)
194 200
49.2% (44.2–54.3%) 50.8% (45.7–55.8%)
43.2% (34.5–52.0%) 56.8% (48.0–65.5%)
253 141
64.2% (59.3–68.9%) 35.8% (31.1–40.7%)
56.2% (47.4–65.0%) 43.8% (35.0–52.6%)
28 366
7.1% (4.8–10.1%) 92.9% (89.9–95.2%)
7.9% (4.0–11.8%) 92.1% (88.2–96.0%)
43 337
11.3% (8.3–14.9%) 88.7% (85.1–91.7%)
17.6% (8.5–26.6%) 82.4% (73.4–91.5%)
259 107
70.8% (65.8–75.4%) 29.2% (24.6–34.2%)
66.7% (57.7–75.7%) 33.3% (24.3–42.3%)
a
Receptively shared syringes Yes No Receptively shared any injection equipment (syringes, cookers, cotton, water)a Yes No Injects at least dailya Yes No Infection status Self-reported HIV status Positive Negative or unknown HIV study test result Positive Negative Hepatitis C study test result Positive Negative a b
Past 12 months. Had at least 5 drinks in one sitting.
for retention. Analyses were conducted using SAS 9.2 (Cary, NC, USA). 3. Results Twelve seeds recruited a total of 525 participants. Of these, 394 (75.1%) were male. The median age of male participants was 45 (IQR: 36, 51). There were no significant differences between unweighted and weighted percentages (Table 1) and unweighted results are presented in the following text. Participants identified as Latino (66.2%), black (20.1%), white (12.7%), and other race/ethnicity (1.0%). Most of the participants were more comfortable using English with family and friends (74.6%), were neither married nor cohabiting (85.8%), identified their sexual orientation as straight (90.6%), and had lived in NYC for more than 12 months (91.1%). The majority were unemployed (89.3%), had an annual household income ≤$10,000 (74.4%), and reported being homeless at some time in the past 12 months (51.0%). Approximately half had graduated from high school (51.0%); the majority were born in the United States (59.1%). Participants reported their borough of residence as Brooklyn (38.6%), the Bronx (35.3%), Manhattan (16.5%), Queens (9.4%), and Staten Island (0.3%). More than onethird (38.3%) reported having been arrested in the past 12 months. In the past 12 months, 39.6% reported having ≥3 sex partners and 28.4% reported having sex without a condom with ≥3 sex partners. One-quarter (23.6%) reported binge drinking (at least 5 drinks in one sitting) at least once a week, 25.4% reported noninjection cocaine use, 22.1% reported non-injection crack use, 2.3% reported non-injection methamphetamine use, and 4.3% reported injection methamphetamine use in the past 12 months. The most frequently injected drug reported was heroin by itself (63.2%), followed by “speedball” (heroin and cocaine) (28.9%), and cocaine by itself (7.9%). More than half (64.2%) reported injecting at least daily, 26.1% reported receptively sharing syringes, and 49.2% reported receptively sharing other injection equipment (cookers, cotton, or water) in the past 12 months. Fewer than one-tenth (7.1%) selfreported HIV positive status, whereas 11.3% had positive HIV study test results. The majority of male study participants (70.8%) had positive hepatitis C study test results. Approximately one-fifth of male participants (n = 85, 21.6%) reported selling sex in the past 12 months to either main or casual partners. Fewer than ten percent (n = 35, 8.9%) of male participants reported selling sex to male partners. The large majority of those who reported selling sex to men also reported having sex with
women in the past 12 months (n = 30, 85.7%). Selling sex to women was almost twice as common, with 66 (16.8%) male participants reporting selling sex to female partners. Of those who reported selling sex to women, 20 (30.3%) also reported having sex with men. Sixteen (4.1%) male participants reported selling sex to both men and women in the past 12 months. Table 2 shows variables associated with male participants selling sex to men in the past 12 months. In the bivariate analysis, variables significantly associated (p < 0.1) with selling sex to men in the past 12 months included being age 18–29 years and 30–39 years vs. ≥50 years, Spanish language preference, being foreignborn and Puerto Rican-born (vs. born in the United States), gay or bisexual identity, living in the Bronx vs. Manhattan, living in NYC less than 12 months, having been homeless in the past 12 months, and past 12 month behavior of having ≥3 sex partners, having sex without a condom with ≥3 sex partners, binge drinking at least once a week, non-injection cocaine use, non-injection crack use, noninjection methamphetamine use, injection methamphetamine use, injecting cocaine and “speedball” most frequently (vs. injecting heroin most frequently), and receptively sharing syringes. Variables independently associated (p < 0.05) with selling sex to men in the multivariate model included gay or bisexual sexual identity (vs. straight) (adjusted odds ratio (aOR): 31.0; 95% CI: 8.1, 119.1), Bronx residence (vs. Manhattan (aOR: 38.1; 95% CI: 6.2, 235.5), having been homeless in the past 12 months (aOR: 9.9; 95% CI: 2.0, 49.6), past 12 month behaviors of having ≥3 sex partners (aOR: 26.2; 95% CI: 4.7, 147.6), non-injection cocaine use (aOR: 5.4; 95% CI: 1.6, 18.2), and injection methamphetamine use (aOR: 36.9; 95% CI: 5.7, 240.0). Table 3 shows the variables associated with male participants selling sex to women in the past 12 months. In the bivariate analysis, variables associated (p < 0.1) with selling sex to women in the past 12 months included not being married or cohabiting, being foreign born (vs. born in the United States), gay or bisexual identity, residing in the boroughs of the Bronx and Queens (vs. Manhattan), and past 12 month behavior of ever having been homeless, having ≥3 sex partners, having sex without a condom with ≥3 sex partners, binge drinking at least once a week, non-injection cocaine use, non-injection crack use, injecting “speedball” most frequently (vs. heroin), receptively sharing syringes, and receptively sharing injection equipment (cookers, cotton, or water). Variables independently associated (p < 0.05) with selling sex to women in the multivariate model included past 12 month behavior of having ≥3 sex partners (aOR: 14.6; 95% CI: 6.6, 31.9), binge drinking at least
82
K.H. Reilly et al. / Drug and Alcohol Dependence 144 (2014) 78–86
Table 2 Factors associated with selling sex to males in the past 12 months among male injection drug users in New York City (n = 394). Variable Demographics Age 18–29 30–39 40–49 50+ Race/ethnicity Latino Black White Other Language preference English Spanish Marital status Married or cohabiting Never married, separated, divorced, widowed Place of birth United States Puerto Rico Other country Graduated high school Yes No Sexual identity Straight Gay/bisexual Borough Manhattan Brooklyn Bronx Queens Staten Island NYC residence $10,000 Unemployed Yes No Arresteda Yes No Homelessa Yes No Sexual behavior ≥3 sex partnersa Yes No Sex without a condom with ≥3 partnersa Yes No Substance use Bingeb drinking at least once a weeka Yes No Non-injection cocaine usea Yes No Non-injection crack usea Yes No Non-injection methamphetamine usea Yes No Injection methamphetamine usea Yes No Most frequently injected drug Heroin Cocaine “Speedball” (heroin + cocaine)
% Sold sex to males in the past 12 months
Odds ratio (95% CI)
Adjusted odds ratio (95% CI)
13.8% 14.2% 7.3% 4.9%
3.1 (0.8, 11.8)* 3.2 (1.2, 8.5)** 1.5 (0.5, 4.3) 1.0
10.7% 3.8% 6.0% 2.9%
1.0 0.3 (0.1, 1.1)* 0.5 (0.2, 1.8) 2.8 (0.3, 27.6)
6.8% 15.0%
1.0 2.4 (1.2, 4.9)**
5.4% 9.5%
1.0 1.8 (0.5, 6.3)
5.2% 13.0% 42.9%
1.0 2.7 (1.3, 5.8)** 13.8 (2.8, 68.8)**
10.0% 7.8%
1.3 (0.7, 2.6) 1.0
3.7% 59.5%
1.0 38.7 (16.4, 91.3)**
1.0 31.0 (8.1, 119.1)**
6.2% 2.0% 18.0% 8.1% 0.0%
1.0 0.3 (0.1, 1.4) 3.3 (1.1, 10.1)** 1.3 (0.3, 6.4) –
1.0 0.5 (0.05, 6.3) 38.1 (6.2, 235.5)** 9.4 (0.9, 98.5)* –
17.1% 8.1%
2.4 (0.9, 6.1)* 1.0
8.6% 10.0%
0.8 (0.4, 1.8) 1.0
9.1% 7.1%
1.3 (0.4, 4.4) 1.0
8.0% 9.5%
0.8 (0.4, 1.7) 1.0
14.9% 2.6%
6.6 (2.5, 17.4)** 1.0
9.9 (2.0, 49.6)** 1.0
19.2% 2.1%
11.1 (4.2, 29.3)** 1.0
26.2 (4.7, 147.6)** 1.0
19.6% 4.6%
5.1 (2.4, 10.5)** 1.0
14.0% 7.3%
2.1 (1.0, 4.3)* 1.0
19.0% 5.4%
4.1 (2.0, 8.3)** 1.0
20.7% 5.5%
4.5 (2.2, 9.1)** 1.0
66.7% 7.5%
24.5 (5.8, 103.2)** 1.0
47.1% 7.2%
11.5 (4.1, 32.3)** 1.0
6.0% 16.1% 13.2%
1.0 3.0 (1.0, 8.9)** 2.4 (1.1, 5.0)**
5.4 (1.6, 18.2)** 1.0
36.9 (5.7, 240.0)** 1.0
K.H. Reilly et al. / Drug and Alcohol Dependence 144 (2014) 78–86
83
Table 2 (Continued) Variable
% Sold sex to males in the past 12 months
Odds ratio (95% CI)
15.5% 6.5%
2.6 (1.3, 5.3)** 1.0
11.0% 7.0%
1.6 (0.8, 3.3) 1.0
11.5% 4.3%
2.9 (1.2, 7.2)** 1.0
10.7% 8.7%
1.3 (0.4, 4.4) 1.0
Adjusted odds ratio (95% CI)
a
Receptively shared syringes Yes No Receptively shared injection equipment (cookers, cotton, water)a Yes No Injects at least dailya Yes No Infection status Self-reported HIV status Positive Negative or unknown a b * **
Past 12 months. Had at least 5 drinks in one sitting. p < 0.1. p < 0.05.
once a week (aOR: 3.1; 95% CI: 1.6, 6.1), non-injection crack use (aOR: 3.3; 95% CI: 1.6, 6.7), injecting “speedball” most frequently (vs. heroin) (aOR: 2.1; 95% CI: 1.1, 4.2), and receptively sharing syringes (aOR: 2.4; 95% CI: 1.2, 4.8).
4. Discussion Approximately one-fifth of male study participants reported selling sex in the past 12 months, which is higher than the levels found among men in studies of heroin users in the UK and IDUs in Vancouver, Canada (Gossop et al., 1993; Kuyper et al., 2004). Almost twice as many participants in the current study reported selling sex to women than selling sex to men; this is unprecedented in the literature on exchange sex. To our knowledge, this is one of the only U.S. studies to examine factors associated with men selling sex to women. Although women may be more likely to sell sex than men (Jenness et al., 2011), the frequency of men selling sex to women found in this study challenges the conventional perception of the gendered direction of exchange sex. Other research has found that men who have sex with both men and women are more likely to sell sex, compared to those who only have sex with men (Dyer et al., 2013; Weber et al., 2001). Unsurprisingly, those who identified as gay or bisexual were more likely to report selling sex to men than those who did not. Almost all male IDUs who reported selling sex to men, however, also reported having sex with women; these male IDUs may not only be at potentially greater risk for acquiring HIV, but also be at increased risk for transmitting HIV to both their male and female sex partners. The ALIVE longitudinal study of IDUs in Baltimore found sex with a man to be associated with HIV seroconversion among male IDUs, whereas high-risk heterosexual sex was associated with HIV seroconversion among female IDUs (Strathdee et al., 2001). Bronx residence was also independently associated with selling sex to men. The Bronx is one of the most impoverished counties in the United States, with over 30% of all residents living below the poverty line (U.S. Census Bureau, 2012). Sex may be used as an alternative currency in high-poverty areas, where access to resources is limited. Many other studies have found that men who are homeless are more likely to sell sex than those who are stably housed (Bobashev et al., 2009; Newman et al., 2004). Although the current study was cross-sectional, previous research indicates that homelessness often precedes drug use and involvement in sex work among MSM (Clatts et al., 2005). Edwards et al. (2006) also found that those who sell sex have greater numbers of partners, which may increase their exposure risk for HIV and other STIs and position them as a potential core group for HIV transmission. Although all participants were IDUs, those who used stimulants (non-injection
cocaine and injection methamphetamine) were more likely to report selling sex to men. Previous research has also found associations between use of stimulants and engaging in exchange sex (Newman et al., 2004; Ober et al., 2009; Windle, 1997). Having greater numbers of sex partners in the past 12 months was independently associated with selling sex to women. Also similar to men who reported selling sex to men, stimulant use (noninjection crack and most often injected “speedball”) was associated with selling sex to women. Non-injection crack use has repeatedly been found to be associated with selling sex in other studies (Newman et al., 2004; Sherman et al., 2011). A study conducted among male and female drug users in 22 U.S. cities found that injectors who also smoked crack were more likely than injectors who did not smoke crack to sell sex for drugs or money (Booth et al., 2000). The current study found that binge drinking at least weekly was also independently associated with selling sex to women. Although there are limited data available on alcohol use among men who sell sex, binge drinking has previously been associated with high risk sexual behaviors including multiple sex partners and not using condoms (Caldeira et al., 2009; Vosburgh et al., 2012). A Baltimore study found that compared to moderate drinkers or alcohol abstainers, problematic alcohol users were more likely to inject drugs before or during sex, which may have implications for sexual risk behaviors (Scherer et al., 2013). The current study also found that those who sold sex to women were more likely to have receptively shared syringes in the past 12 months. Although no data could be found on syringe-sharing behaviors among men who sell sex to women, a previous study of IDUs in the South Bronx found that those who exchanged sex for drugs or other things were more likely to share syringes than those who did not (Fernando et al., 2003). This study is subject to several limitations. The cross-sectional nature of this study precludes the determination of temporal relationships. Although one of the advantages of RDS is that it has the potential to reduce bias by producing long recruitment chains that cross-bridge between relevant groups, it is possible that this study oversampled from well-connected social networks that differ from the target population, particularly with respect to income and employment. The results of this study may therefore not be generalizable to other areas or populations. The study questionnaire elicited sensitive information regarding sexual behavior and HIV status, and participants may not have felt comfortable disclosing this information to study staff interviewers. The limited sample size may have restricted the power to detect differences within the target population, and may have impeded the inclusion of other relevant variables in the final multivariate models, or may have led to spurious associations, which would be mitigated by a larger sample. Some point estimates had wide confidence intervals reducing the precision of correlate effects. Also, no information was collected
84
K.H. Reilly et al. / Drug and Alcohol Dependence 144 (2014) 78–86
Table 3 Factors associated with selling sex to females in the past 12 months among male injection drug users in New York City (n = 394). Variable Demographics Age 18–29 30–39 40–49 50+ Race/ethnicity Latino Black White Other Language preference English Spanish Marital status Married or Cohabiting Never married, separated, divorced, widowed Place of birth United States Puerto Rico Other country Graduated high school Yes No Sexual identity Straight Gay/bisexual Borough Manhattan Brooklyn Bronx Queens Staten Island NYC residence $10,000 Unemployed Yes No Arresteda Yes No Homelessa Yes No Sexual behavior ≥3 sex partnersa Yes No Sex without a condom with ≥3 partnersa Yes No Substance use Bingeb drinking at least once a weeka Yes No Non-injection cocaine usea Yes No Non-injection crack usea Yes No Non-injection methamphetamine usea Yes No Injection methamphetamine usea Yes No Most frequently injected drug Heroin Cocaine “Speedball” (heroin + cocaine)
% sold sex to females in the past 12 months
Odds ratio (95% CI)
13.8% 14.2% 18.2% 18.0%
0.7 (0.2, 2.3) 0.7 (0.4, 1.5) 1.0 (0.5, 1.9) 1.0
19.2% 20.2% 0.0% 0.0%
1.0 1.1 (0.6, 2.0) – –
15.0% 22.0%
1.0 1.6 (0.9, 2.8)
5.4% 18.6%
1.0 4.0 (1.2, 13.3)**
14.6% 17.5% 71.4%
1.0 1.2 (0.7, 2.2) 14.6 (2.7, 78.4)**
16.9% 16.6%
1.0 (0.6, 1.7) 1.0
45.9% 13.8%
1.0 5.3 (2.6, 10.9)**
9.2% 13.8% 21.6% 24.3% 0.0%
1.0 1.6 (0.6, 4.1) 2.7 (1.1, 6.9)** 3.2 (1.0, 9.8)** –
8.6% 17.5%
0.4 (0.1, 1.5) 1.0
16.2% 19.0%
0.8 (0.5, 1.5) 1.0
16.8% 16.7%
1.0 (0.4, 2.4) 1.0
19.2% 15.2%
1.3 (0.8, 2.3) 1.0
20.4% 13.0%
1.7 (1.0, 3.0)** 1.0
35.9% 4.2%
12.8 (6.3, 26.0)** 1.0
34.8% 9.6%
5.0 (2.9, 8.8)** 1.0
35.5% 11.0%
4.5 (2.6, 7.8)** 1.0
27.0% 13.3%
2.4 (1.4, 4.2)** 1.0
32.2% 12.4%
3.4 (1.9, 5.9)** 1.0
33.3% 16.4%
2.6 (0.6, 10.5) 1.0
17.6% 16.7%
1.1 (0.3, 3.8) 1.0
14.5% 3.2% 25.4%
1.0 0.2 (0.02, 1.5) 2.0 (1.2, 3.5)**
Adjusted odds ratio (95% CI)
14.6 (6.6, 31.9)** 1.0
3.1 (1.6, 6.1)** 1.0
3.3 (1.6, 6.7)** 1.0
1.0 0.1 (0.02, 1.0)* 2.1 (1.1, 4.2)*
K.H. Reilly et al. / Drug and Alcohol Dependence 144 (2014) 78–86
85
Table 3 (Continued) Variable
% sold sex to females in the past 12 months
Odds ratio (95% CI)
Adjusted odds ratio (95% CI)
28.2% 12.7%
2.7 (1.6, 4.7)** 1.0
2.4 (1.2, 4.8)** 1.0
23.8% 10.8%
2.6 (1.5, 4.5)** 1.0
16.6% 17.0%
1.0 (0.6, 1.7) 1.0
25.0% 16.1%
1.7 (0.7, 4.3) 1.0
a
Receptively shared syringes Yes No Receptively shared injection equipment (cookers, cotton, water)a Yes No Injects at least dailya Yes No Infection status Self-reported HIV status Positive Negative or unknown a b * **
Past 12 months. Had at least 5 drinks in one sitting. p < 0.1. p < 0.05.
on mental health or history of sexual abuse, which has previously been found to be associated with exchange sex among men (Haley et al., 2004; Weber et al., 2001). This study also did not differentiate between types of exchange sex (e.g., commercial sex, survival sex, and “sex-for-presents”). There may be differences in perceptions of what constitutes exchanging sex, since the definition in the question was broad and may be dependent on culturally-determined subjective interpretation. More data are needed to learn more about the contexts of selling sex including the characteristics of exchange partnerships, condom use, and what is being exchanged. A substantial proportion of male IDUs in NYC reported selling sex. Although this study did not find an association between selling sex and HIV infection, this behavior has the potential for increased risk for HIV acquisition and transmission, especially among those who sell sex to men, given that MSM are disproportionately affected by the HIV epidemic both in New York City and nationally. Increased risk may not only occur because of efficient transmission of HIV, but also because of greater exposure to high-risk partners through sexual networks. In the current study, almost all of those who reported selling sex to men also reported having female sex partners, which could lead to the introduction and spread of heterosexuallytransmitted HIV to women in partnerships with high-risk men who also have sex with men. HIV prevention interventions aimed at IDU men who sell sex should consider those factors which increase their vulnerability to engaging in exchange sex, sexual transmission risk, and other risk behaviors. Future studies should examine whether male IDUs who sell sex are more likely to become infected with or transmit HIV and STIs compared to those who do not sell sex. Interventions might not only be beneficial for preventing individual transmission; targeting those who sell sex could have populationlevel effects by reducing transmission risks in exchange networks.
Author disclosures None.
Role of funding source This research was funded by a cooperative agreement between the New York City Department of Health and Mental Hygiene (NYC DOHMH) and the Centers for Disease Control and Prevention (CDC) (grant# 5U1BPS003246-02). The CDC developed the study design and methodology, but had no further role in the collection, analysis and interpretation of data; in the writing of the report; nor in the decision to submit the paper for publication. The contents of this article are solely the responsibility of the authors and do not
necessarily represent the official views of The Centers for Disease Control and Prevention. Contributors Author KHR undertook the statistical analysis and wrote the first draft of the manuscript. AN and TW directed the research study from which the data were derived. DMM collected the data. All authors contributed to and have approved the final manuscript. Conflict of interest All authors declare that they have no conflicts of interest. Acknowledgements The authors acknowledge Gabriela Paz-Bailey, Dita Broz, and Michael “Trey” Spiller of CDC for their contributions to the National HIV Behavioral Surveillance (NHBS) study design; Sarah Braunstein, Kent Sepkowitz, John Rojas, and Jay Varma of the NYC DOHMH for reviewing previous drafts of this article; and the NYC NHBS field staff for all their efforts. References Astemborski, J., Vlahov, D., Warren, D., Solomon, L., Nelson, K., 1994. The trading of sex for drugs or money and HIV seropositivity among female intravenous drug users. Am. J. Public Health 84, 382–387. Bobashev, G.V., Zule, W.A., Osilla, K.C., Kline, T.L., Wechsberg, W.M., 2009. Transactional sex among men and women in the south at high risk for HIV and other STIs. J. Urban Health 86, 32–47. Booth, R.E., Kwiatkowski, C.F., Chitwood, D.D., 2000. Sex related HIV risk behaviors: differential risks among injection drug users, crack smokers, and injection drug users who smoke crack. Drug Alcohol Depend. 58, 219–226. Caldeira, K.M., Arria, A.M., O’Grady, K.E., Zarate, E.M., Vincent, K.B., Wish, E.D., 2009. Prospective associations between alcohol and drug consumption and risky sex among female college students. J. Alcohol Drug Educ. 53, nihpa115858. Clatts, M.C., Goldsamt, L., Yi, H., Viorst Gwadz, M., 2005. Homelessness and drug abuse among young men who have sex with men in New York City: a preliminary epidemiological trajectory. J. Adolesc. 28, 201–214. Davey-Rothwell, M., Latkin, C., 2008. An examination of perceived norms and exchanging sex for money or drugs among women injectors in Baltimore, MD, USA. Int. J. STD AIDS 19, 47–50. Des Jarlais, D.C., Arasteh, K., McKnight, C., Hagan, H., Perlman, D., Friedman, S.R., 2009. Using hepatitis C virus and herpes simplex virus-2 to track HIV among injecting drug users in New York City. Drug Alcohol Depend. 101, 88–91. Des Jarlais, D.C., Arasteh, K., McKnight, C., Hagan, H., Perlman, D.C., Semaan, S., 2011. Associations between herpes simplex virus type 2 and HCV with HIV among injecting drug users in New York City: the current importance of sexual transmission of HIV. Am. J. Public Health 101, 1277–1283. Des Jarlais, D.C., Perlis, T., Arasteh, K., Torian, L.V., Beatrice, S., Milliken, J., Mildvan, D., Yancovitz, S., Friedman, S.R., 2005. HIV incidence among injection drug users in New York City, 1990 to 2002: use of serologic test algorithm to assess expansion of HIV prevention services. Am. J. Public Health 95, 1439–1444.
86
K.H. Reilly et al. / Drug and Alcohol Dependence 144 (2014) 78–86
Dunkle, K.L., Jewkes, R.K., Brown, H.C., Gray, G.E., McIntryre, J.A., Harlow, S.D., 2004. Transactional sex among women in Soweto, South Africa: prevalence, risk factors and association with HIV infection. Soc. Sci. Med. 59, 1581–1592. Dyer, T.P., Regan, R., Wilton, L., Harawa, N.T., Ou, S.S., Wang, L., Shoptaw, S., 2013. Differences in substance use, psychosocial characteristics and HIV-related sexual risk behavior between black men who have sex with men only (BMSMO) and black men who have sex with men and women (BMSMW) in six US cities. J. Urban Health 90, 1181–1193. Edwards, J.M., Iritani, B.J., Hallfors, D.D., 2006. Prevalence and correlates of exchanging sex for drugs or money among adolescents in the United States. Sex. Transm. Infect. 82, 354–358. Fernando, D., Schilling, R.F., Fontdevila, J., El-Bassel, N., 2003. Predictors of sharing drugs among injection drug users in the South Bronx: implications for HIV transmission. J. Psychoact. Drugs 35, 227–236. Gallagher, K.M., Sullivan, P.S., Lansky, A., Onorato, I.M., 2007. Behavioral surveillance among people at risk for HIV infection in the US: the National HIV Behavioral Surveillance System. Public Health Rep. 122, 32–38. Gossop, M., Griffiths, P., Powis, B., Strang, J., 1993. Severity of heroin dependence and HIV risk. I. Sexual behaviour. AIDS Care 5, 149–157. Haley, N., Roy, E., Leclerc, P., Boudreau, J., Boivin, J., 2004. HIV risk profile of male street youth involved in survival sex. Sex. Transm. Infect. 80, 526–530. Heckathorn, D.D., 1997. Respondent-driven sampling: a new approach to the study of hidden populations. Soc. Prob. 44, 174–199. Heckathorn, D.D., 2007. Extensions of respondent driven sampling: analyzing continuous variables and controlling for differential recruitment. Sociol. Methodol. 37, 151–207. Jenness, S.M., Kobrak, P., Wendel, T., Neaigus, A., Murrill, C.S., Hagan, H., 2011. Patterns of exchange sex and HIV infection in high-risk heterosexual men and women. J. Urban Health 88, 329–341. Jenness, S.M., Neaigus, A., Hagan, H., Murrill, C.S., Wendel, T., 2010. Heterosexual HIV and sexual partnerships between injection drug users and noninjection drug users. AIDS Patient Care STDS 24, 175–181. Jie, W., Xiaolan, Z., Ciyong, L., Moyer, E., Hui, W., Lingyao, H., Xueqing, D., 2012. A Qualitative exploration of barriers to condom use among female sex workers in China. PLoS One 7, e46786. Johnston, L.G., Malekinejad, M., Kendall, C., Iuppa, I.M., Rutherford, G.W., 2008. Implementation challenges to using respondent-driven sampling methodology for HIV biological and behavioral surveillance: field experiences in international settings. AIDS Behav. 12, 131–141. Kuyper, L., Lampinen, T., Li, K., Spittal, P., Hogg, R., Schechter, M., Wood, E., 2004. Factors associated with sex trade involvement among male participants in a prospective study of injection drug users. Sex. Transm. Infect. 80, 531–535. Lansky, A., Sullivan, P.S., Gallagher, K.M., Fleming, P.L., 2007. HIV behavioral surveillance in the US: a conceptual framework. Public Health Rep. 122, 16–23. Latkin, C., Hua, W., Forman, V., 2003. The relationship between social network characteristics and exchanging sex for drugs or money among drug users in Baltimore, MD, USA. Int. J. STD AIDS 14, 770–775. Neaigus, A., Reilly, K.H., Jenness, S.M., Hagan, H., Wendel, T., Gelpi-Acosta, C., 2013. Dual HIV risk: receptive syringe sharing and unprotected sex among HIV-negative injection drug users in New York City. AIDS Behav. 17, 2501–2509. Newman, P.A., Rhodes, F., Weiss, R.E., 2004. Correlates of sex trading among drugusing men who have sex with men. Am. J. Public Health 94, 1998–2003. Niccolai, L.M., Toussova, O.V., Verevochkin, S.V., Barbour, R., Heimer, R., Kozlov, A.P., 2010. High HIV prevalence, suboptimal HIV testing, and low knowledge of
HIV-positive serostatus among injection drug users in St. Petersburg, Russia. AIDS Behav. 14, 932–941. Ober, A., Shoptaw, S., Wang, P.-C., Gorbach, P., Weiss, R.E., 2009. Factors associated with event-level stimulant use during sex in a sample of older, low-income men who have sex with men in Los Angeles. Drug Alcohol Depend. 102, 123–129. Rolfs, R.T., Goldberg, M., Sharrar, R.G., 1990. Risk factors for syphilis: cocaine use and prostitution. Am. J. Public Health 80, 853–857. Rudolph, A.E., Crawford, N.D., Latkin, C., Fowler, J.H., Fuller, C.M., 2013. Individual and neighborhood correlates of membership in drug using networks with a higher prevalence of HIV in New York City (2006–2009). Ann. Epidemiol. 23, 267–274. Salazar, L.F., Crosby, R.A., Holtgrave, D.R., Head, S., Hadsock, B., Todd, J., Shouse, R.L., 2007. Homelessness and HIV-associated risk behavior among African American men who inject drugs and reside in the urban south of the United States. AIDS Behav. 11, 70–77. Salganik, M.J., 2006. Variance estimation, design effects, and sample size calculations for respondent-driven sampling. J. Urban Health 83, 98–112. Scherer, M., Trenz, R., Harrell, P., Mauro, P., Latimer, W., 2013. The role of drinking severity on sex risk behavior and HIV exposure among illicit drug users. Am. J. Addict. 22, 239–245. Semple, S.J., Strathdee, S.A., Zians, J., Patterson, T.L., 2010. Social and behavioral characteristics of HIV-positive MSM who trade sex for methamphetamine. Am. J. Drug Alcohol Abuse 36, 325–331. Sherman, S.G., Reuben, J., Chapman, C.S., Lilleston, P., 2011. Risks associated with crack cocaine smoking among exotic dancers in Baltimore, MD. Drug Alcohol Depend. 114, 249–252. Strathdee, S.A., Galai, N., Safaiean, M., Celentano, D.D., Vlahov, D., Johnson, L., Nelson, K.E., 2001. Sex differences in risk factors for HIV seroconversion among injection drug users: a 10-year perspective. Arch. Intern. Med. 161, 1281–1288. U.S. Census Bureau, 2012. Small Area Income and Poverty Estimates [Data], 2011 American Community Survey. U.S. Census Bureau, http://www.census.gov/ did/www/saipe/data/statecounty/data/2011.html (accessed on December 11, 2013. Vosburgh, H.W., Mansergh, G., Sullivan, P.S., Purcell, D.W., 2012. A review of the literature on event-level substance use and sexual risk behavior among men who have sex with men. AIDS Behav. 16, 1394–1410. Weber, A.E., Boivin, J.-F., Blais, L., Haley, N., 2002. HIV risk profile and prostitution among female street youths. J. Urban Health 79, 525–535. Weber, A.E., Craib, K.J., Chan, K., Martindale, S., Miller, M.L., Schechter, M.T., Hogg, R.S., 2001. Sex trade involvement and rates of human immunodeficiency virus positivity among young gay and bisexual men. Int. J. Epidemiol. 30, 1449–1454. Wejnert, C., Pham, H., Krishna, N., Le, B., DiNenno, E., 2012. Estimating design effect and calculating sample size for respondent-driven sampling studies of injection drug users in the United States. AIDS Behav. 16, 797–806. White, R.G., Lansky, A., Goel, S., Wilson, D., Hladik, W., Hakim, A., Frost, S.D.W., 2012. Respondent driven sampling—where we are and where should we be going? Sex. Transm. Infect. 88, 397–399. Windle, M., 1997. The trading of sex for money or drugs, sexually transmitted diseases (STDs), and HIV-related risk behaviors among multisubstance using alcoholic inpatients. Drug Alcohol Depend. 49, 33–38. Winship, C., Radbill, L., 1994. Sampling weights and regression analysis. Sociol. Methods Res. 23, 230–257. Wood, E., Schachar, J., Li, K., Stoltz, J.-A., Shannon, K., Miller, C., Lloyd-Smith, E., Tyndall, M.W., Kerr, T., 2007. Sex trade involvement is associated with elevated HIV incidence among injection drug users in Vancouver. Addict. Res. Theory 15, 321–325.
