Our findings have several implications for the design and implementation of behavioral interventions to reduce AOD use and AOD-related sex risks for HIV among vulnerable women in South Africa. First, findings highlight the need for AOD use interventions among vulnerable women, irrespective of ethnicity. However, findings also indicate that Black African and Coloured women have different profiles of AOD use. Specifically, we found that Coloured women had significantly greater odds of testing positive for methamphetamine and self-reporting heavy episodic alcohol use and significantly reduced odds of testing positive for cannabis than Black African women. These findings suggest that while Black African women are more likely to use cannabis than Coloured women, they are less likely to have problematic AOD use (as indicated by the pervasiveness of methamphetamine use and heavy episodic drinking among Coloured women in this sample). This is supported by the finding that Coloured participants in our study were also significantly more likely to report that they believed they had a drug problem compared to their Black African counterparts. These findings are broadly in keeping with those from earlier studies that reported less AOD use (with the exception of cannabis) among women from Black African communities relative to women from Coloured communities in the Western Cape [22, 23].
Continuing ethnic disparities in access to income provide a possible explanation for these different profiles of AOD use, with Coloured persons being marginally better off and therefore potentially more able to afford AODs than Black African citizens [19–21]. Yet differences in income do not account for our findings. Even after controlling for the potentially confounding effects of age income, education and employment status, ethnic differences in AOD use outcomes remain. It is possible that contextual (such as drug marketing practices ) and cultural differences between Black African and Coloured communities may contribute to these unique profiles of AOD use, however this requires further investigation. Regardless of the reasons for these differences, our findings imply that AOD use interventions should be tailored to the distinct profiles of AOD use among Black African and Coloured women. Our findings suggest that AOD interventions for vulnerable Coloured women should focus on reducing heavy episodic drinking and providing treatment for methamphetamine use. While there are several drug treatment programs that address methamphetamine-related problems in the Western Cape, vulnerable women face more affordability and geographic access barriers to utilizing these services than men . The high prevalence of methamphetamine use and self-identified drug problems among Coloured women in our sample indicate an urgent need to address these barriers to treatment through introducing (where absent) and scaling up (where available) community-based AOD treatment services that are affordable and easily accessible for vulnerable Coloured women.
In contrast, our findings suggest that AOD interventions for vulnerable Black African women should consist mainly of low-intensity, brief interventions that focus on reducing cannabis use and educating women about the risks associated with methamphetamine and other drug use. As several studies have shown that frequent and regular use of cannabis is associated with increased risk of transition to other illicit drug use [30, 31], providing Black African women with brief interventions focused on reducing cannabis use may not only help them reduce the negative consequences associated with their cannabis use, but may also prevent progression from cannabis to methamphetamine use.
However, this does not mean that AOD interventions focused on reducing methamphetamine use are not needed in Black African communities. Ensuring that Black African women have access to interventions for methamphetamine use is essential given that more than a third of Black African participants tested positive for the recent use of methamphetamine. This unexpected finding is in sharp contrast to earlier work which has consistently pointed to low rates of methamphetamine use among Black African populations in Cape Town [22, 32, 33]. This sudden increase in the use of methamphetamine among Black African women points to the urgent need for AOD prevention interventions that address the underlying risk factors that predispose Black African women to initiate methamphetamine use as well as treatment interventions that address risk factors for the continued use of methamphetamine. These interventions, combined with other interventions that address environmental and structural risk factors for drug use , may help curtail the further escalation of methamphetamine use within Black African communities.
Failure to provide and, where available, scale up AOD interventions for vulnerable Black African and Coloured women not only represents a missed opportunity to intervene with their AOD use but may have unintended negative consequences for efforts to curb the spread of HIV in the region. This study found that AOD-impaired sex is widespread among vulnerable AOD-using women, with close to 40% of our sample reporting that their last sexual encounter was not only AOD-impaired but also unprotected. Compared to Black African participants, Coloured participants had four-fold greater odds of reporting that their last sexual encounter was AOD-impaired and unprotected, even after adjusting for age, education, income and employment status. One explanation for this finding is that AOD use before or during sex reduces personal perceptions of risk for HIV thus leading to less condom use . Personal perceptions of risk for HIV might already be lower among Coloured women than among Black African women because of the lower prevalence of HIV in Coloured communities [2, 17]. This explanation is supported by prior studies conducted in this region that identified a negative relationship between HIV risk perception and engagement in sex risk behaviors for women . Regardless of the reason for these findings, they clearly point to the urgent need for behavioral interventions that focus on reducing AOD use before or during sex and reducing other barriers to condom use among vulnerable women in Cape Town. While findings suggest that vulnerable women from both ethnic groups would benefit from AOD-related sex risk reduction interventions, the need seems particularly great among women from Coloured communities where AOD-impaired sex is prevalent and condom use low.
Third, our findings suggest that for AOD-related sex risk reduction interventions to be impactful, such interventions should target AOD use among vulnerable women as well as their main sexual partners. Close to two-thirds of our sample reported that their main sexual partner had either been drunk or had used drugs during the month preceding the study and almost one in two participants reported that their main sexual partner had been AOD-impaired during their last sexual encounter. This is cause for concern as evidence from earlier studies suggests that women with partners who use AODs have a poorer response to AOD interventions than women with partners who do not use AODs . In addition, the likelihood of implementing sex risk reduction strategies (such as consistent condom use) within relationships is diminished if one or both partners are AOD-impaired during sex [38, 39]. Finally, as we found ethnic differences in partner-related AOD-impaired sex (with Coloured women having three-fold greater odds of reporting that their main partner was AOD-impaired at last sex relative to Black African women), the need for these proposed interventions seems most pressing among Coloured women and their main sexual partners.
Despite the important implications that these findings have for HIV prevention programming, these results should be considered in the light of some methodological limitations. First, this paper was based on the analysis of baseline data from a larger HIV and AOD risk reduction project. The strict inclusion and exclusion criteria of this study may have adversely impacted the ability to recruit a representative sample of Black African and Coloured South African women. Second, as the sample was recruited from poor communities within one region of the Cape Town metropole, the extent to which findings can be extrapolated to other parts of the province or country are unclear. Third, in this study women provided self-report data on their main sexual partner’s AOD use. These perceptions are open to several reporting biases and it is quite possible that social desirability and other processes may have affected the accuracy of the information provided. This last limitation highlights the need for future work among vulnerable women and their main sexual partners so that AOD use and related sexual risk behaviors that occur within the context of relationships can be further explored.