HIV testing and sero-prevalence among methamphetamine users seeking substance abuse treatment in Cape Town

Abstract

Introduction and Aims

Methamphetamine use is highly prevalent in parts of South Africa, and there is concern this will contribute to the country’s substantial HIV epidemic. We examined the feasibility of implementing routine HIV testing at a community-based substance abuse treatment centre in Cape Town, and determined the HIV sero-prevalence among methamphetamine users seeking treatment at this site.

Design and Methods

In this cross-sectional study, 293 participants completed measures of demographics, substance use and HIV treatment. HIV sero-prevalence was determined by a rapid finger-prick HIV test, and prior HIV diagnosis was confirmed via clinic records.

Results

The majority of participants were male and self-identified as “Coloured,” with a mean age of 28. The HIV sero-prevalence was 3.8%. Of the 11 participants who tested HIV-positive, four were newly diagnosed. HIV-positive and HIV-negative participants were comparable on demographic and substance use factors. Uptake of HIV testing among all clients at the drug treatment centre increased from <5% prior to study initiation to 89% after study completion. Measures implemented to ensure high rates of HIV testing were regarded as sustainable.

Discussion and Conclusions

Our study suggests that integrating routine HIV testing into substance abuse treatment is feasible in a community-based health centre. The low HIV prevalence among this sample of treatment-seeking methamphetamine users highlights the potential benefits of supporting expanded efforts to optimise HIV prevention with this young adult population.

Introduction

The co-occurrence of HIV infection and methamphetamine use is a significant public health problem and potential driver of HIV transmission through its link with drug injection and heightened sexuality [1]. South Africa, which has an estimated 6.4 million adults living with HIV/AIDS [2], has experienced a dramatic increase in methamphetamine use, particularly in the densely populated township communities of Cape Town [3]. Methamphetamine use in this setting has been associated with sexual risk behaviours, such as multiple partners, unprotected intercourse, and se trading for drugs/money, which may facilitate HIV transmission [4–6]. The proportion of patients entering drug treatment for primary methamphetamine use in Cape Town increased from 0.8% in 2002 to 52% in 2011 [7]. Drug treatment programmes provide an opportunity to reach high-risk users for HIV testing in order to deliver appropriate HIV prevention and HIV care services.

To date, methamphetamine use in South Africa is most common among groups with relatively low HIV prevalence – youth, males and individuals who identify as “Coloured” (a uniquely South African term representing people of mixed-race ancestry) [8]. National HIV prevalence for the Coloured population is 3.1%, and HIV prevalence is lower in the Western Cape province compared to the rest of South Africa (7.8% vs. 18.8%) [2]. However, overall HIV prevalence in the Western Cape tripled from 2005 to 2012 [2], and there is concern that methamphetamine use may be a contributing factor [9]. Efforts to address both methamphetamine use and HIV risk are needed, and a combined approach is likely to be most effective. However, HIV testing has thus far not been routine in drug treatment programs in South Africa.

The aims of this preliminary investigation were to establish whether routine provider-initiated HIV testing in a drug treatment facility is feasible, to report the HIV sero-prevalence of methamphetamine users seeking drug treatment, and to determine whether demographic and substance use characteristics are associated with HIV status.

Methods

Setting

The study was conducted in a peri-urban community at a certified drug treatment facility co-located in a government health clinic that serves a predominantly Coloured, low-income population. Treatment is free, with no referral required. It follows the Matrix Model^®, a standardised 16-week intensive outpatient program for stimulant abuse [10].

Procedures

After providing written informed consent, participants completed a brief survey. Clinic staff then accompanied participants to the HIV testing service situated in the same health centre as the Matrix site. A HIV counsellor completed pre-test counselling followed by a rapid finger-prick HIV test. Participants were provided with test results during post-test counselling. Since HIV testing was conducted as part of routine care in the clinic, the HIV test result was reported to clinic staff and made part of the client’s clinic record. The result was also shared with research staff, based on participants’ informed consent during study enrolment. Individuals with a positive result were referred to the clinic’s HIV care team, if not already engaged in HIV care.

The study was approved by the ethics committees at the University of Cape Town and the Duke University Health System. Participants were given a small gift worth approximately US$5.

Participants

Clinic staff attempted to recruit all new enrolees at the Matrix site from November 2011 to June 2013 who met eligibility criteria (≥18 years old, methamphetamine as primary substance of abuse, methamphetamine use in the past 30 days, and no impaired mental status or current intoxication). Of 301 eligible clients who were approached, 300 enrolled and 293 completed HIV testing. Seven participants agreed to HIV testing but services were not immediately available.

Measures

The Alcohol, Smoking, and Substance Involvement Screening Test was administered as part of standard care [11,12]. A Substance Severity Index score was created, using standard cut-offs, to categorise individuals into low (0–3), medium (4–26) and high (>27) risk for methamphetamine-related health problems. The demographics questionnaire assessed gender, age, race, marital status, education, employment and drug treatment history. A self-report questionnaire assessed history of HIV testing and treatment. For HIV-positive participants, prior diagnosis and treatment information were also extracted from clinic records.

Data analysis

Descriptive statistics were calculated for demographics, substance abuse and HIV sero-prevalence. We compared the HIV-positive and HIV-negative participants on these variables using t-tests (continuous variables), chi-square tests (categorical variables) and Fisher’s exact tests (categorical variables with expected frequencies ≤5). To assess feasibility of implementing routine HIV testing, we determined the proportion of participants who refused an HIV test. We also compared HIV testing rates before study initiation versus after its closure using clinic-level data provided by the municipal health department.

Results

The sample included 116 women and 177 men ranging in age from 18 to 54 years [mean = 28, SD = 6.27]. Most were Coloured (99%), not married (72%), currently unemployed (86%) and seeking drug treatment for the first time (71%). In addition to methamphetamine, many reported concurrent use of alcohol (48%), marijuana (64%) and Mandrax (methaqualone) (42%).

HIV sero-prevalence was 3.8%. Of the 11 participants who tested positive, four (36%) were newly diagnosed. Of those with a new HIV diagnosis, 25% had no prior history of HIV testing. HIV was diagnosed on average 5.5 years (SD= 5.3) following initial methamphetamine use; only one participant reported receiving an HIV diagnosis before initiating methamphetamine use. The mean recent CD4 cell count was 479 (SD= 291.9, range= 118–1107, n= 10). Among the seven participants with a previous diagnosis, four (57%) were on antiretroviral therapy. The HIV-positive and HIV-negative groups were comparable in all respects (see Table 1).

Table 1.

Sample characteristics by HIV status (N=293)

	HIV-positive N= 11	HIV-negative N= 282	Statistic	P value
Demographics
Male gender, n (%)	6 (55)	171 (61)	χ2(1) = 0.16	0.685
Age, M (SD)	30.20 (8.60)	28.15 (6.17)	t (291) = 1.08	0.282
Race: Coloured, n (%)	11 (100)	279 (99)	FET	0.891
Married, n (%)	1 (9)	80 (28)	χ2(1) = 1.97	0.161
Education: completed grade 10, n (%)	3 (27)	152 (54)	χ2(1) = 3.01	0.083
Employed (casual or full-time), n (%)	1 (9)	41 (15)	χ2(1) = 0.26	0.613
First time seeking drug treatment, n (%)	8 (73)	199 (71)	χ2(1) = 0.02	0.877
History of injection drug use, n (%)	1 (9)	3 (1)	FET	0.143
Methamphetamine use
Substance Severity Index, M (SD)	29.64 (8.64)	29.68 (8.12)	t (291) = 0.02	0.986
High-risk use in past 3 months, n (%)	9 (82)	209 (74)	χ2(1) = 0.33	0.566
Daily use in past 3 months, n (%)	4 (36)	127 (45)	χ2(1) = 0.32	0.570
Other substance use in past 3 months
Alcohol, n (%)	8 (73)	133 (48)	χ(1)2 = 2.77	0.096
Marijuana, n (%)	5 (46)	162 (57)	χ(1)2 = 0.62	0.431
Methaqualone (“Mandrax”), n (%)	4 (36)	120 (43)	χ(1)2 = 0.17	0.684
Heroin, n (%)	0 (0)	35 (12)	χ(1)2 = 1.55	0.213

FET, Fisher’s exact test.

Prior to study initiation, clients were simply informed that HIV testing was available at the health centre. Uptake of HIV testing on this basis was very low; staff estimated that it was <5%. Following study completion, from August through November 2013, the Matrix site screened 200 individuals, of whom 178 underwent HIV testing, yielding a testing rate of 89%.

Discussion

Despite strong evidence of HIV risk behaviours among methamphetamine users, HIV testing in drug treatment programmes in South Africa has thus far been neglected [13]. To our knowledge, this is the first study in Sub-Saharan Africa to examine the feasibility of integrating provider-initiated HIV testing into drug treatment, and to report sero-prevalence among methamphetamine users seeking treatment. The study documented the feasibility and acceptability of routine HIV testing, consistent with studies conducted in high-income countries [14]. We found a relatively low HIV sero-prevalence, and HIV status was unrelated to demographic and substance use variables.

There were several factors that facilitated the integration of HIV testing into services routinely offered at the drug treatment centre: clinic staff received HIV education; clients were escorted by Matrix staff to the HIV testing site and given preferential access to reduce wait times; and services were co-located in a single community health centre. These factors likely contributed to the substantial increase in uptake of HIV testing during the study period and may be integral to the maintenance of high HIV testing rates in this setting following study completion. By comparison, the average testing rate for all drug treatment sites in the City of Cape Town at the time of study completion was 49% [7]. When HIV services are not offered in the same facility, alternative ways to integrate HIV testing into drug treatment may include having a staff member certified in HIV testing.

The HIV prevalence rate of 3.8% in this sample is slightly higher than the national prevalence rate for the Coloured population (3.1%), but lower than expected given high rates of sexual risk behaviour among community-recruited methamphetamine users in South Africa [4, 5]. This likely reflects the low background prevalence of HIV in the Coloured population of the Western Cape, as well as the relatively young age of the sample and the lower proportion of women [2]. It is also possible that people who are aware of their HIV-positive status may not access drug treatment due to stigma and shame, or that individuals who seek out drug treatment have less HIV risk behaviours compared to those who do not seek treatment.

Among the HIV-positive methamphetamine users in this study, a third were unaware of their infection, underscoring the importance of routine HIV testing in this population [17]. Of participants with a prior HIV diagnosis who had been referred for antiretroviral therapy, less than half had initiated treatment. Drug treatment programs provide the opportunity to access this high-risk population for HIV testing and linkage to HIV care [18].

This single-site study has several limitations. While generally comparable to the broader population of drug treatment seekers in the Western Cape, our sample had a larger proportion of women and Coloured persons [7]. In addition, our study may not be representative of the general population of methamphetamine users, as the majority do not seek treatment [19]. The small number of HIV-positive participants and lack of comparison data from the community makes it difficult to draw concrete conclusions regarding risk factors for HIV infection among methamphetamine users. Finally, while HIV prevalence was determined via a highly accurate biological test, risk factors were based on self-report, which may be biased.

The results of this study point to a great need for integration of routine HIV testing into drug treatment as a way to identify new HIV-positive cases and link them to HIV care. Further research is needed to establish HIV prevalence in methamphetamine users who do not seek drug treatment, particularly in the Black African population with likely higher HIV levels. Additional studies examining barriers to accessing drug treatment services, and integration of HIV care with drug treatment in this resource-limited setting are needed.