Non-medical use of prescription drugs and HIV risk behaviour in transgender women in the Mid-Atlantic region of the United States

Abstract

Male-to-female transgender women (TGW) experience high rates of substance use and HIV. A recent substance use trend is the use of prescription medication without a doctor’s consent. No research to date has examined the associations between this non-medical use of prescription drugs and HIV risk behaviour in TGW. In the present study, TGW recruited from community venues (N = 104) in the Mid-Atlantic region of the United States completed surveys assessing demographic information, non-medical use of prescription drugs, other substance use, injection practices and sexual risk behaviour. Twenty-four per cent of the sample reported lifetime non-medical use of prescription drugs across the following categories: analgesics (21.2%), anxiolytics (14.4%), stimulants (12.5%) and sedatives (8.7%). Participants reporting non-medical use of prescription drugs were more likely to report other substance use, needle use to inject drugs, injecting silicone and sharing needles. In multivariable analyses, non-medical use of prescription drugs was associated with unprotected sex, sex after engaging in substance use, and commercial sex work, after controlling for demographic factors. Self-esteem and social support from family served as protective factors for non-medical use of prescription drugs. HIV-prevention programmes focused on TGW in the United States may wish to expand their assessment of substance use to include the use of prescription medications without a physician’s consent.

Keywords

Transgender prescription drug misuse HIV sexual behaviour commercial sex work injection practices self-esteem social support

Transgender women (TGW) are individuals classified as male at birth but identifying as female later in life.¹ Probability studies examining HIV prevalence in TGW have not been conducted, despite evidence that TGW are at very high risk for HIV. A recent meta-analysis using data from convenience samples in 15 countries estimated the HIV prevalence in TGW at 19.1% (95% confidence interval: 17.4%–20.7%).² A comparable meta-analysis examining HIV prevalence in TGW in the United States (US) found that, in studies conducting HIV testing, 27.7% (95% confidence interval: 24.8–30.6%) tested positive.³ This review also reported that HIV rates were consistently highest in African-American TGW, relative to other racial groups in the US. Research with TGW shows high rates of sexual risk behaviour, including unprotected sex and sex with multiple partners.^4–6 Prior work has also identified commercial sex work (CSW) as a crucial risk factor for HIV infection in TGW.⁷ Additional research has documented injection risk behaviours in TGW.⁸ In most groups, injection risk is due to the use of illicit drugs. Among TGW, some individuals also inject hormones or inject silicone directly into the body in an effort to change body shape and features.⁹ Sharing needles during this process may be an additional HIV risk behaviour for some TGW.

Substance use may represent a method to cope with the discrimination faced by TGW in the United States.¹⁰ High rates of substance use have been found in this group.¹¹ Excessive substance use is well documented to be associated with sexual risk behaviour.¹² A recent substance use trend in the US is the non-medical use of prescription drugs (NMUPD), the use of medications for managing pain and treating anxiety, attention-deficit disorder and insomnia, without a physician’s prescription. NMUPD has increased in the US and the misuse of prescription drugs has been linked to negative outcomes including dependence and overdose.¹³ Although NMUPD appears to have shown the largest increases in North America,¹⁴ some studies in Europe and Australia also show concerning rates of this behaviour.^15,16

A few studies have examined relationships between NMUPD and HIV risk behaviour in populations at high risk for HIV. Benotsch et al. found associations between NMUPD and sexual risk behaviour in men who have sex with men (MSM) in Denver, CO (US).¹⁷ Kelly and Parsons found associations between NMUPD and unprotected sex in HIV-negative MSM in New York, NY (US).¹⁸ In a study of opioid abusers, Surratt et al. found that 23.8% reported injecting prescription analgesics and some participants reported sharing needles.¹⁹

Some prior studies have examined resiliency factors as protective factors that reduce the likelihood of engaging in NMUPD.^20,21 Given that TGW face considerable stigma, examining resiliency factors may be especially important in this group.

Prior work has not examined associations between NMUPD and HIV risk behaviour in TGW. The present analysis is from a larger study of substance use in transgender adults.²² Here, we describe analyses specific to HIV risk behaviours in the TGW subset of the larger study. We hypothesised that NMUPD would be associated with sexual and injection HIV transmission risk behaviours. We also hypothesised that resiliency factors (self-esteem, social support) would be protective against engaging in NMUPD.

Methods

A brief (15-min) paper survey was administered to TGW recruited from community venues in the Mid-Atlantic region of the United States (see Table 1 for details). Participants were eligible if they were 18 or older, and classified as TGW (indicated that their birth sex was male and their current gender was female or transgender). Overall, 92.9% of individuals invited to participate agreed to do so. All study measures and procedures were approved by the relevant institutional review boards.

Table 1.

Recruitment sites and characteristics of transgender women in the study sample.

Recruitment site:
Transgender health clinics in
Richmond, VA	38.5%
Washington, DC	24.0%
Bars	21.2%
Transgender community event	13.4%
Commercial sex work areas	2.9%
Mean age = 34.9 years (SD = 12.2; range = 18–65)
Mean years of formal education: 12.5 years (SD = 2.2)
Race:
African-American	67.0%
White	25.2%
Asian-American	2.9%
Latina	1.0%
Native American	1.0%
Other	2.9%
Employment status:
Employed full-time	24.0%
Employed part-time	14.4%
Unemployed	40.4%
Student	2.9%
On disability	7.7%
Retired	1.9%
Other	8.7%
Personal Annual Income (U.S. Dollars):
<$16,000	62.5%
$16-30,000	23.1%
$31-45,000	7.7%
$46-60,000	2.9%
>$60,000	3.8%
HIV status:
HIV-negative	60.4%
HIV-positive	31.3%
HIV status unknown	8.3%
History of incarceration (lifetime)
Yes	56.4%
No	43.6%

Demographics

Participants were asked to report their age, birth gender, current gender, race/ethnicity, education, employment status and HIV status. Participants also indicated if they had been in jail (lifetime).

NMUPD

Participants were asked to respond with a Yes/No to the question: ‘In your lifetime, have you ever used a prescription medication WITHOUT a doctor’s prescription recreationally – for example, to get high, or because you were curious, or to try something new?’ Participants then indicated their lifetime NMUPD in four categories: analgesics (e.g. Vicodin), sedatives (e.g. Ambien), anxiolytics (e.g. Xanax) and stimulants (e.g. Adderall). Examples of medications in each category were listed on the survey. This measure has been used previously.¹⁷

Other substance use

Participants were asked about alcohol and illicit drug use in the previous three months. Participants also indicated if they had ever used a needle to inject drugs, hormones, or silicone and if they had ever shared needles. This measure is similar to measures employed previously.²³

Sexual practices

Participants were asked to report the number of times they had engaged in unprotected anal and vaginal intercourse in the past three months. Participants also recorded their number of sexual partners in the past three months. In addition, participants indicated if they had exchanged sex for money, drugs, or a place to stay in their lifetime and in the past three months. Participants were also asked the number of times they had had sex after having ‘too much’ to drink, and the number of times they had had sex after using drugs in the previous three months. In addition, participants indicated if they use condoms the first time they have sex with a new partner and if they find out about a partner’s HIV status before having sex. These items were rated on a 4-point scale from 1 (Never) to 4 (Always).

Resilience factors

Self-esteem was assessed with a shortened version of the Rosenberg scale. This measure had adequate internal consistency within this sample (α = .77) and has shown good test–retest reliability in a variety of populations, including those at high risk for HIV.²⁴

Social support was assessed with a shortened version of the multidimensional scale of perceived social support.²⁵ This instrument separately measures social support provided by members of an individual’s family of origin and support provided by friends. The measure had excellent internal consistency for both the family support scale (α = .94) and the friends support scale (α = .93).

Data quality assurances and statistical analyses

All surveys were examined for inconsistencies and invalid responses. Missing data were omitted from analyses, resulting in slightly different sample sizes for various statistical tests. Distributions of substance use and sexual behaviour were highly skewed so non-parametric analyses were used. Because many factors can be associated with substance use and sexual risk behaviour, multivariable analyses were used to examine associations between NMUPD and sexual risk, after controlling for the influence of demographic factors. Comparable multivariable analyses examined the role of resiliency factors in protecting against NMUPD.

Results

Participants

As shown in Table 1, participants were predominantly African-American, with modest educational attainment and income.

NMUPD

Almost a quarter of the sample (24.0%) reported lifetime NMUPD. The most commonly used type of medication was analgesics (21.2% of the sample), followed by anxiolytics (14.4%), stimulants (12.5%) and sedatives (8.7%). Participants who reported NMUPD were significantly more likely to report being HIV positive (52.0%) than individuals who did not report NMUPD (23.9%), χ² (1, N = 96) = 6.47, p < .05. Participants who reported NMUPD were also more likely to report having been in jail in their lifetime (79.2%) than participants not reporting NMUPD (49.4%), χ² (1, N = 101) = 7.04, p < .01. NMUPD was unrelated to other demographic factors assessed.

NMUPD, other substance use and injection practices

As seen in Table 2, participants who engaged in NMUPD were more likely to report the recent use of illicit drugs. Participants who reported NMUPD were also significantly more likely to report injecting drugs, injecting silicone, and sharing needles.

Table 2.

Non-medical use of prescription drugs, other substance use, and injection risk behaviours in a sample of transgender women in Richmond, VA and Washington, DC, USA.

	Individuals reporting non-medical use of prescription drugs (N = 25)	Individuals NOT reporting non-medical use of prescription drugs (N = 79)	χ²
Substance use, past three months	% reporting use of:	% reporting use of:
Alcohol	58.3, n = 14	65.4, n = 51	0.39
Marijuana	62.5, n = 15	29.1, n = 23	8.57*
Nitrites (‘Poppers’)	17.4, n = 4	5.2, n = 4	3.05
Ecstasy (‘X’)	26.1, n = 6	7.8, n = 6	4.84*
Cocaine	35.3, n = 6	9.2, n = 7	3.87*
Methamphetamine (‘crystal,’ ‘tina’)	21.1, n = 4	3.9, n = 3	4.12*
Heroin	17.4, n = 4	2.6, n = 2	5.59*
‘Other’ recreational drugs	21.7, n = 5	3.9, n = 3	6.32*
Injection practices, lifetime	% reporting:	% reporting:
Use of a needle to inject drugs	32.0, n = 8	2.5, n = 2	15.85***
Use of a needle to inject silicone	37.5, n = 9	11.4, n = 9	7.66**
Use of a needle to inject hormones	43.5, n = 10	23.3, n = 17	3.34
Needle sharing	12.0, n = 3	0.0, n = 0	8.84**

N = 104. *p < .05, ** p < .01, *** p < .001.

NMUPD and sexual risk behaviour

As seen in Table 3, NMUPD was associated with sexual risk behaviour. Individuals who reported NMUPD were more likely to report unprotected sex, sex after using substances and CSW. Participants who reported NMUPD also indicated they were less likely to use condoms with a new partner and were less likely to find out a partner’s HIV status before having sex.

Table 3.

Non-medical use of prescription drugs and sexual behaviour in a sample of transgender women in Richmond, VA and Washington, DC, USA.

	Individuals reporting non-medical use of prescription drugs (N = 25)	Individuals NOT reporting non-medical use of prescription drugs (N = 79)	χ²
Behaviour	% reporting:	% reporting:
Any unprotected (no condom used) anal or vaginal sex, past three months	61.0, n = 14	37.7, n = 29	3.87*
Multiple anal or vaginal sex partners, past three months	50.0, n = 12	36.4, n = 28	1.40
Sex after having ‘too much’ to drink, past three months	40.9, n = 9	19.7, n = 15	3.84*
Sex after using drugs, past three months	50.0, n = 11	13.2, n = 10	12.15***
Exchanged sex for money, drugs, or a place to stay, past three months	40.0, n = 8	17.1, n = 13	4.39*
Exchanged sex for money, drugs, or a place to stay, lifetime	80.0, n = 20	34.2, n = 27	16.72***
	Mean (SD)	Mean (SD)	Z ^a
Total unprotected anal and vaginal sex acts, past three months	2.91 (3.78)	1.55 (5.59)	2.62**
Total number of anal or vaginal partners, past three months	7.21 (17.61)	1.94 (3.58)	1.30
Sex after having ‘too much’ to drink, number of times, past three months	2.32 (3.66)	0.55 (1.54)	2.38*
Sex after using drugs, number of times, past three months	2.86 (5.01)	1.00 (3.98)	3.59***
Use condoms with new sexual partners^a	2.83 (0.96)	3.62 (0.85)	−4.11***
Find out partner’s HIV status before having sex^a	2.30 (1.14)	2.93 (1.18)	−2.24*

N = 104. *p < .05, ** p < .01, *** p < .001

Mann-Whitney test. ^aRated on a 1 (never) to 4 (always) scale.

Because multiple factors can influence sexual behaviour, a series of logistic regression analyses were conducted predicting the categorical sexual risk behaviours shown in Table 3 after accounting for the influence of demographic factors. Adjusted odds ratios and 95% confidence intervals are presented in Table 4. As can be seen in Table 4, NMUPD remained a significant predictor of sexual risk behaviour after accounting for the influence of demographic variables.

Table 4.

Non-medical use of prescription drugs and sexual risk behaviour in a sample of transgender women in Richmond, VA and Washington, DC, USA.

	Any unprotected anal or vaginal sex, past three months AOR (95% CI)	Multiple anal or vaginal sex partners, past three months AOR (95% CI)	Sex after having ‘too much’ to drink, past three months AOR (95% CI)	Sex after using drugs, past three months AOR (95% CI)	Exchanged sex for money, drugs, or a place to stay, past three months AOR (95% CI)	Exchanged sex for money, drugs, or a place to stay, lifetime AOR (95% CI)
NMUPD	3.31^* (1.07, 10.30)	1.50 (0.52, 4.34)	5.51* (1.22, 24.78)	15.66** (3.23, 75.96)	5.49* (1.28, 23.57)	12.89** (2.34, 71.07)

AOR: Adjusted Odds Ratio (values reported are adjusted for the influence of age, education, income, race, and HIV status); CI = confidence interval; NMUPD: non-medical use of prescription drugs.

p < .05, ** p < .01.

Resiliency factors

As shown in Table 5, the resiliency factors of self-esteem and family social support were protective against engaging in NMUPD. There was no significant protective effect of social support from friends and NMUPD.

Table 5.

Non-medical use of prescription drugs and resiliency factors in a sample of transgender women in Richmond, VA and Washington, DC, USA.

	NMUPD
Resiliency factor:	AOR (95% CI)
Self-esteem	0.74** (0.63, 0.89)
Social support from family of origin	0.90* (0.82, 0.99)
Social support from friends	0.94 (0.83, 1.05)

AOR: adjusted odds ratios (odds ratios are adjusted for the influence of age, education, income, race and HIV status); CI: confidence interval; NMUPD: non-medical use of prescription drugs.

p < .05, ** p < .01

Discussion

In this study, nearly one in four TGW reported lifetime NMUPD. These findings are consistent with increasing NMUPD prevalence in other groups in North America. As with other studies, NMUPD was associated with the use of illicit drugs.^17,19 It may be more appropriate to consider NMUPD in the context of a polysubstance use pattern. Using prescription medications with illicit drugs can have adverse effects, including death.¹³ TGW may be especially vulnerable to deleterious effects of the combined use of illicit drugs, prescription drugs used non-medically, as well as prescription drugs that are prescribed by a physician (e.g. hormone therapy, antiretrovirals).

NMUPD was also associated with both sexual and injection-related HIV transmission risk, including CSW, a crucial risk factor for HIV acquisition in TGW.²⁶ In prior work, TGW engaged in CSW reported the use of illicit substances to cope with the stress of that activity.²⁷ Previous research has documented stress-reduction motivations for NMUPD, and NMUPD appears higher in individuals who have experienced sexual trauma.²⁸ Use of these medications among TGW may reflect an attempt to cope with the stress of CSW. For some of these medications (e.g. anxiolytics), reducing stress-related symptoms is consistent with how they might be prescribed. TGW frequently report poor access to culturally sensitive medical care.²⁹ We specifically assessed recreational use; however, it may be that these medications were being used without a physician’s prescription because that is the only option available to them. Having high self-esteem and high social support from family were protective factors for NMUPD, perhaps by buffering individuals from the effects of stigma and other stressors.³⁰

Lifetime reports of injecting drugs (10%), silicone (17%) and hormones (27%) were also high. Risks associated with injecting illicit drugs are well established. TGW face additional risks associated with injecting silicone, which is seen as a rapid method of feminising body parts.³¹ Consequences of silicone injection can include acute infection and pulmonary dysfunction leading to death.³² Injecting silicone and drugs was associated with NMUPD. To our knowledge, this is the first study that documents associations between NMUPD and injection risk behaviours in TGW.

Limitations and conclusion

These data were collected from a relatively small (N = 104) convenience sample of TGW in the Mid-Atlantic region of the US. The study population was predominantly African-American and had low incomes and high rates of unemployment; generalisation to other regions and populations may not be warranted. The use of a cross-sectional design limits drawing causal conclusions. We assessed the use of broad classes of medication (e.g. analgesics) but did not separately assess the use of specific medications (e.g. Vicodin) and only assessed lifetime NMUPD. Future work should assess the non-medical use of specific medications within each class as well as assessing NMUPD over more recent time frames. In addition, a more detailed assessment of sexual behaviour (e.g. separately assessing insertive/receptive anal sex) is warranted.

Despite these limitations, this study provides additional information about patterns of substance use in TGW and is the first to document associations between NMUPD and HIV risk behaviour in this group. Given that almost 1/3 of the sample reported being HIV positive, the associations between NMUPD and behaviour include increased risk for both the transmission and acquisition of the virus. This study is also one of the first quantitative studies to link NMUPD and injection risk behaviours. HIV-prevention programmes focused on TGW may wish to expand their assessment of substance use to include NMUPD. If validated through additional research, these results may indicate the need for specialised interventions to be made available that incorporate HIV prevention and substance use treatment. Given the high rates of unemployment, low incomes, and engagement in CSW, organisations working with TGW may also wish to include job training as part of a comprehensive risk reduction intervention. Comprehensive HIV-prevention programmes should also focus on HIV testing (8.3% of participants in this high-risk group reported not knowing their HIV status) and linkage to care.

Footnotes

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This project was funded by internal institutional grants to the first and second author.

References

Feinberg

. Transgender liberation: a movement whose time has come, New York: World View Forum, 1992.

Baral

Poteat

Strömdahl

. Worldwide burden of HIV in transgender women: a systematic review and meta-analysis. Lancet Infect Dis 2013; 13: 214–222.

Herbst

Jacobs

Finlayson

. Estimating HIV prevalence and risk behaviors of transgender persons in the United States: a systematic review. AIDS Behav 2008; 12: 1–17.

Bockting

Robinson

Forberg

. Evaluation of a sexual health approach to reducing HIV/STD risk in the transgender community. AIDS Care 2005; 17: 289–303.

Hill

Daniel

Benzie

. Sexual health of transgender sex workers attending an inner-city genitourinary clinic. Int J STD AIDS 2011; 22: 686–687.

Johnston

Valliant

Dolores

. HIV, hepatitis B/C and syphilis prevalence and risk behaviors among gay, transsexuals and men who have sex with men, Dominican Republic. Int J STD AIDS 2012; 24: 313–321.

Nuttbrock

Hwahng

Bockting

. Lifetime risk factors for HIV/sexually transmitted infections among male-to-female transgender persons. J Acquir Immune Defic Syndr 2009; 52: 417–421.

Crosby

Pitts

. Caught between different worlds: How transgendered women may be “forced” into risky sex. J Sex Res 2007; 44: 43–48.

Garofalo

Deleon

Osmer

. Overlooked, misunderstood, and at-risk: exploring the lives and HIV risk of ethnic minority male-to-female transgender youth. J Adol Health 2006; 38: 230–236.

10.

Testa

Sciacca

Hendricks

. Effects of violence on transgender people. Professional Psychol-Res Pract 2012; 43: 452–459.

11.

Henderson

. Substance use disorders in lesbian, gay, bisexual & transgender clients: assessment and treatment. Soc Work 2010; 55: 286–287.

12.

Wilson

Garofalo

Harris

Belzer

. Sexual risk taking among transgender male-to-female youths with different partner types. Am J Pub Health 2010; 100: 1500–1505.

13.

Substance Abuse and Mental Health Services Administration (SAMSHA). Prescription Medications: misuse, abuse, dependence and addiction. Substance Abuse Treatment Advisory, http://store.samhsa.gov/product/Prescription-Medications-Misuse-Abuse-Dependence-and-Addiction/SMA12-4175 (2006, accessed 24 June 2015).

14.

Fischer

Keates

Bühringer

. Non-medical use of prescription opioids and prescription opioid-related harms: why so markedly higher in North American compared to the rest of the world? Addiction 2014; 109: 177–181.

15.

Fotiou

Kanavoi

Richardson

. Trends in the association between prescribed and non-prescribed use of tranquillisers or sedatives among adolescents in 22 European countries. Nordic Stud Alc Drug 2014; 31: 371–387.

16.

Australian Institute of Health and Welfare. 2010 National Drug Strategy Household survey report, Canberra, Australia: Author, 2011.

17.

Benotsch

Martin

Koester

. Non-medical use of prescription drugs and HIV risk behavior in gay and bisexual men. Sex Transm Dis 2011; 38: 105–110.

18.

Kelly

Parsons

. Prescription drug misuse and sexual risk taking among HIV-negative MSM. AIDS Behav 2013; 17: 926–930.

19.

Surratt

Kurtz

Cicero

. Alternate routes of administration and risk for HIV among prescription opioid abusers. J Addict Dis 2011; 30: 334–341.

20.

Heck

Livingston

Flentje

. Reducing risk for illicit drug use and prescription drug misuse: high school gay-straight alliances and lesbian, gay, bisexual, and transgender youth. Addict Behav 2014; 39: 824–828.

21.

Jeffers

Benotsch

Koester

. Misuse of prescription stimulants for weight loss, psychosocial variables and eating disordered behaviors. Appetite 2013; 65: 8–13.

22.

Benotsch

Zimmerman

Cathers

. Non-medical use of prescription drugs, polysubstance use, and mental health in transgender adults. Drug Alcohol Depend 2013; 132: 391–394.

23.

Martin

Benotsch

Perschbacher

. Transmission risk behaviors in a subset of HIV-positive individuals: the role of narcissistic personality features. Pers Indiv Diff 2013; 54: 256–260.

24.

Bauermeister

Johns

Sandfort

TGM

. Relationship trajectories and psychological well-being among sexual minority youth. J Youth Adolesc 2010; 39: 1148–1163.

25.

Cheng

Chan

ACM

. The multidimensional scale of perceived social support: Dimensionality and age and gender differences in adolescents. Pers Indiv Diff 2004; 37: 1359–1369.

26.

Operario

Nemoto

. HIV in transgender communities: syndemic dynamics and a need for multicomponent interventions. J Acquir Immune Defic Syndr 2010; 55: S91–S93.

27.

Nemoto

Operario

Keatley

. Social context of HIV risk behaviours among male-to-female transgenders of colour. AIDS Care 2004; 16: 724–735.

28.

McCauley

Amstadter

Kmett Danielson

. Mental health and rape history in relation to non-medical use of prescription drugs in a national sample of women. Addict Behav 2009; 34: 641–648.

29.

Sanchez

Danoff

. Health care utilization, barriers to care, and hormone usage among male-to-female transgender persons in New York City. Am J Pub Health 2009; 99: 713–719.

30.

Zimmerman

Darnell

Rhew

. Resilience in community: A social ecological development model for young adult sexual minority women. Am J Commun Psychol 2015; 55: 179–190.

31.

Wallace

. Finding self: a qualitative study of transgender, transitioning, and adulterated silicone. Health Educ J 2010; 69: 439–446.

32.

Hariri

Gaissert

Brown

. Progressive granulomatous pneumonitis in response to cosmetic subcutaneous silicone injections in a patient with HIV-1 infection. Arch Pathol Lab Med 2012; 136: 204–207.