Abstract
Background
Transactional sex is a risk factor for HIV infection. Alcohol use may increase the risk of transactional sex. No nationally-representative studies have examined the relationship between multiple dimensions of alcohol use and transactional sex in women in South Africa. The aim of the study was to examine the relationship between alcohol dependence, binge drinking and frequency of drinking in the past month and transactional sex in adult women in South Africa.
Methods
A cross-sectional study using multi-stage, cluster sampling collected data from a nationally representative sample of 5,969 women aged 16–55 years in 2012. The analysis conducted for this paper was restricted to women reporting sexual activity in the past 12 months (n = 3,594). Transactional sex was defined as having received money/gifts in exchange for sex with any sex partner in the past year. Alcohol use measures included: alcohol dependence (≥2 positive responses to the CAGE questionnaire); binge drinking (≥4 drinks for women on one occasion); and drinking frequency in the previous month. Logistic regression models were built to test the hypotheses that each dimension of alcohol use was associated with transactional sex.
Results
About 6.3% (n = 225) of sexually active women reported transactional sex. Almost a third (30.6%) of sexually active women had ever drunk alcohol, and 19.2% were current (past month) drinkers. Among lifetime drinkers, 28.0% were alcohol dependent and 56.6% were binge drinkers. Alcohol dependent women were twice as likely to report transactional sex (AOR 2.0, 95% CI 1.1–4.3, p<0.05) than those not alcohol dependent. Binge drinkers were 3.1 times more likely to have had transactional sex (95% CI 1.5–6.6, p<0.01) than non-binge drinkers. There was no significant relationship between frequency of drinking in the past month and transactional sex.
Conclusion
Alcohol dependency and binge drinking are significantly associated with transactional sex in South African women. HIV prevention programmes need to target these women, and address both their alcohol use, as well as the HIV risks associated with transactional sex.
Introduction
With approximately 23.5 million HIV positive people living in sub-Saharan Africa, the region bears the brunt of the HIV epidemic [1]. South Africa has a very high HIV prevalence with 12.2% of South Africans of all ages living with HIV [2]. Women remain disproportionately affected by HIV. The latest national HIV prevalence survey found that 5.6% of young women aged 15–19 years were infected which was eight times higher than in men (0.7%) in the same age group. In the 20–24 year age range, prevalence was three times higher among women than men: 17.4% compared to 5.1% [2].
One of the factors which increase women’s vulnerability to HIV is engaging in transactional sex [3–5]. For example, Dunkle and colleagues (2004) found, while controlling for the lifetime number of male sex partners and the length of time a woman had been sexually active, transactional sex was associated with HIV seropositivity among women attending antenatal clinics in Soweto, South Africa [3].
Transactional sex is generally defined as a sexual relationship which is primarily motivated by financial or material exchange [6, 7]. However, there is ongoing confusion and disagreement as to what is meant by transactional sex and this concept has been measured and operationalised differently across studies [8]. In transactional sex, cash is most commonly exchanged, but a range of goods or services such as accommodation or transport, may be involved [3, 9, 10]. Although this transaction has both a financial and sexual component, it is differentiated from sex work in that the person who engages in transactional sex does not self-identify as a sex worker, nor is he or she viewed as such by their communities [11]. Research from sub-Saharan Africa has demonstrated that exchange of sex for material or financial resources is commonly practiced and the majority of women who engage in such transactions do not self-identify as sex workers [12–18]. It is worth noting that some studies have conflated transactional sex and sex work [8].
Young women have been reported to engage in transactional sex for a number of reasons including survival, consumption and, in some cases, to increase a sense of agency [9, 18–20]. The transactional nature of these relationships reflects a power imbalance, which for women, often means being less able to influence the timing and nature of sex [14, 17]. Women with little negotiating power to insist on use of condoms experience a higher risk of contracting sexually transmitted infections (STIs), including HIV [19]. Transactional sex is also commonly associated with other risk factors for HIV including intergenerational sex, concurrent sexual partners, unprotected sex [20] and the risk of male-perpetrated intimate partner violence [21].
Another driver of HIV transmission in the sub-Saharan African region is alcohol use, especially more frequent or binge drinking [22–26]. There is a large body of evidence which suggests that alcohol use is strongly associated with HIV incidence [25, 27–30]. Alcohol has been recognised as a key determinant of risky sexual behaviour and as a result, an indirect contributor to the transmission of HIV [31]. A systematic review of the association between HIV infection and alcohol use found that even when all other factors were taken into account, alcohol users had a 57% greater likelihood of being HIV positive than non-drinkers [32]. In recent years, the causal pathway between alcohol intoxication, unsafe sex and HIV acquisition has been more clearly delineated and broadly accepted [33].
Drinking alcohol is associated with a number of risky sexual behaviours including transactional sex [26, 34] and multiple sexual partners [25, 26, 34–36]. Alcohol use, and especially binge (heavy episodic) drinking, has been shown to be associated with unprotected sex and inconsistent condom use [25, 34, 35, 37, 38] with both regular and casual sexual partners [35, 39]. Alcohol use has also been found to be associated with gender-based violence [40], sexual violence [37, 39] and intimate partner violence [41].
Studies examining the relationship between alcohol and risky sexual behaviour have predominately been undertaken in high risk groups such as sex workers [37, 42], bar patrons [42, 43], bar/hotel workers [26] or high risk drinkers [37]. In addition, many of these studies have been venue based. However, as Chersich et al (2007) point out, the relationship between drinking patterns and unsafe sex varies between drinking contexts, population groups and other interacting factors [37]. Findings among high risk groups are likely to differ from the general population [37]. Morojele and colleagues (2006) suggest that a study to look at the extent of risky sexual behaviour among adult and risky drinkers in the general population of South Africa is needed [40].
The quantities of alcohol which people consume and their drinking patterns have an important bearing on their HIV risk. Across a number of studies, a strong dose response relationship between alcohol use and risky sexual behaviour is evident [34]. Frequency of drinking is also of importance. For example, Fritz (2002) found that frequency of drinking in the past week was correlated with the number of episodes of unprotected sex among men [31]. A meta-analysis of African studies found that binge drinking was associated with greater sexual risks than lighter or non-binge drinking [32]. Results from another meta-analysis show that internationally, those who binge drink have double the risk of acquiring HIV compared with non-binge drinkers [27].
South Africa’s National Strategic Plan on HIV, STIs and TB promotes HIV prevention programming for vulnerable populations such as young women who engage in transactional sex and those who abuse alcohol [44]. Understanding the relationship between alcohol use and transactional sex in South African women will assist policymakers and planners to design and implement future interventions to address this important risk factor.
The overall aim of this study was to examine the relationship between alcohol dependence, binge drinking and frequency of drinking in the past month and transactional sex in adult women in South Africa in 2012 among the general population.
Materials and Methods
We used data collected for the Third National HIV Communication Survey (NCS) [45]. This survey was a nationally representative cross-sectional study of men and women aged 16–55 years conducted between February and May 2012. The analysis conducted for this paper was restricted to female respondents reporting sexual activity in the past 12 months (n = 3,594). Ethical approval for this study was granted by the University of the Witwatersrand Human Research Ethics Committee. Written informed consent was obtained from all participants. Parental assent was also obtained for participants aged 16 and 17 years.
Multi-stage, cluster sampling was used to collect data from a random sample of 10,034 respondents (n = 4,065 men and n = 5,969 women). The sample size was calculated based on the statistical rules applicable for multi-stage probability-based cluster methodologies: the population percentage, the standard error, the desired level of significance and the design effect [45]. Sample weights were introduced to correct for selection bias at the sub-place, household and individual levels. The sample was weighted back to be representative of the population in South Africa in respect of sex, age, race, settlement type and province. Sample weights were benchmarked using the 2007 Community Survey undertaken by Statistics South Africa [45].
Data were collected using a structured, interviewer-administered questionnaire. Fieldworkers used computer assisted personal interviewing to administer the questionnaire, which was translated into all eleven official South African languages. At each selected household, face-to-face interviews were conducted with respondents in their home language by trained interviewers matched as far as possible to the socio-demographics of the respondent.
Respondents were defined as sexually active if they reported having had sex within the past 12 months. Sex was defined in the questionnaire as “when the penis is in the vagina or anus”.
In this study, transactional sex was defined as having received money/gifts in exchange for sex. Receiving money/gifts in exchange for sex was measured by a positive response to a question in the sexual calendar: “In the past year did you receive gifts or money from this person in order to have sex with him/her?”. Transactional sex was measured across all types of sexual partnerships reported in a sexual calendar.
Lifetime drinking was measured by asking respondents “Have you ever had an alcoholic drink?”. Current drinking was measured by an item that asked “During the past month, how many times did you have an alcoholic drink?”. Those who responded that they had had one or more drinks were categorised as “Drank any alcohol in the past month”, and those who reported that they had not had any alcoholic drinking the past month were coded as “Did not drink any alcohol in the past month”. This is consistent with the way current drinking has been measured elsewhere [46].
Alcohol dependence was assessed using the four-item CAGE questionnaire [47]. The specific items, with yes/no responses, were: “Have you ever felt you should cut down on your drinking?”; “Have people annoyed you by criticising your drinking”; “Have you ever felt bad or guilty about your drinking?”; and “Have you ever had a drink first thing in the morning to steady your nerves or get rid of a hangover (eye-opener)?”. Claasen (1999) reports that the CAGE questionnaire showed a sensitivity of 100% and a specificity of 78% for alcohol dependence in rural South Africa [48]. As recommended by Dhalla and Kopec (2007)[49] and in line with other studies in South Africa, respondents were classified as alcohol dependent if they responded “yes” to at least two items [46, 47].There was acceptable internal consistency with this measure (Chronbach alpha = 0.7).
Binge drinking was measured using an item that asked: “How often do you have four or more drinks on one occasion?” [50]. Women who responded that they had done so “Less than once a month”; “A few times a month”; “Almost every week; and “Almost every day” were categorised as “Binge drinkers”. “Non-binge drinkers” were those who answered “Never” or “Hardly ever”.
Frequency of drinking was measured using the question: “During the past month, how many times did you have an alcoholic drink?”. Response options included: “Almost every day”; “Several times per week”; “At least once a week”‘ “At least once a month”; and “Never”.
Ideation refers to new ways of thinking that diffuse within a culture by means of social interaction [51]. It encompasses cognitive, emotional, and social determinants of behaviour such as knowledge, beliefs, perceived risk, self-efficacy, personal advocacy and social norms [52]. The ideational factor of relevance to this paper is HIV prevention knowledge. HIV prevention knowledge was measured using a scale of ten items which measured knowledge of HIV prevention methods.
HIV status was measured through self-report. Individuals who reported ever testing for HIV were asked if they knew their HIV status. Those who answered in the affirmative, and who were comfortable sharing their status, were asked what their status was.
Violence was measured as having been in a physical fight in the past 12 months.
Statistical Analysis
Data analysis was undertaken in Stata (version 12.0, STATA Corp,. College Station, Texas, USA). All analysis took into account the multistage, cluster sample design of the study.
To test the hypothesis that multiple dimensions of alcohol use increased transactional sex among women, logistic regression models were built that modelled alcohol dependence, binge drinking and frequency of drinking in the past month and transactional sex. Alcohol variables were only added into the models where they were significant (p<0.1) in the univariate analysis.
Analyses controlled for other covariates which were selected on a basis of theoretical relevance and being independently associated with the outcomes in bivariate analysis (p-value <0.10). Covariates included ideational factors, exposure to HIV communication programmes and behaviour. e.g. violence.
These models were developed using a backwards stepwise logistic approach. All significant variables were entered into the model, and at each step the least significant variable was removed until all the remaining variables had a statistically significant contribution to the model or were theoretically important (based on the literature). Variables were eliminated based on p-values > 0.1. Age and cohabitation status were controlled for in each model.
The results of the final logistic regression models were reported using Adjusted Odds Ratios (AORs) and 95% confidence intervals (CIs).
Results
This study found that 86.3% (n = 4,978) of women had ever had sex and 74.7% (n = 3,594) of these reported having sex in the 12 months prior to the survey. The socio-demographic characteristics of women reporting sexual activity in the past 12 months (n = 3,594) who participated in the study are presented in Table 1. The median age of women reporting sexual activity in the past 12 months was 29 years and the interquartile range was 14 years. Some 21.5% of respondents were aged 20–24 years and one in five (20.2%) were 25–29 years of age. The majority of respondents (78.1%) were Black African. Over a third had completed high school while 39.8% had attended some high school. Nearly six out of ten (59.5%) women were unemployed, 8.5% were students, while two-thirds (64.1%) lived in urban areas. Some 13.1% of women reporting sexual activity in the past 12 months and who had ever tested for HIV and knew their status reported being HIV positive.
Table 1. Socio-demographic characteristics of women 16–55 years reporting sexual activity in the past 12 months.
| Characteristics | Number | Unweighted percentage (%) | Weighted percentage (%) |
|---|---|---|---|
| Age (years) n = 3,594 | |||
| 16–19 | 247 | 6.9 | 6.4 |
| 20–24 | 784 | 21.8 | 21.5 |
| 25–29 | 768 | 21.4 | 20.2 |
| 30–34 | 603 | 16.8 | 16.5 |
| 35–39 | 431 | 12.0 | 12.9 |
| 40–44 | 308 | 8.6 | 9.3 |
| 45–49 | 229 | 6.4 | 6.3 |
| 50–55 | 224 | 6.2 | 7.0 |
| Median (IQR) | 29 (14) | ||
| Race n = 3,593 | |||
| Black | 2,975 | 82.8 | 78.1 |
| Coloured | 525 | 14.6 | 8.5 |
| Indian | 28 | 0.8 | 2.5 |
| White | 65 | 1.8 | 10.9 |
| Education n = 3,588 | |||
| No schooling | 37 | 1.0 | 1.3 |
| Up to primary school | 324 | 9.0 | 9.0 |
| Up to grade 11 | 1,545 | 43.1 | 39.8 |
| Matric | 1,270 | 35.4 | 34.9 |
| Tertiary | 412 | 11.5 | 15.1 |
| Employment n = 3,543 | |||
| Employed | 1,055 | 29.8 | 32.1 |
| Unemployed | 2,198 | 62.0 | 59.5 |
| Student | 290 | 8.2 | 8.5 |
| Settlement type n = 3,594 | |||
| Urban | 2,468 | 68.7 | 64.1 |
| Rural | 1,126 | 31.3 | 35.9 |
| Self-reported HIV status n = 2,471 | |||
| HIV positive | 316 | 12.8 | 13.1 |
| HIV negative | 2,115 | 87.2 | 87.0 |
Transactional sex
About 6.3% (n = 225) of women reporting sexual activity in the past 12 months reported receiving money/gifts in exchange for sex with any of their sex partners in the past 12 months. Using weighted data, this translates into around 560,000 women in South Africa reporting transactional sex in the previous year.
Table 2 shows that knowing one’s HIV status was significantly associated with receiving money/gifts in exchange for sex among women. Women who knew their HIV status were significantly less likely (p = 0.02) to receive money/gifts in exchange for sex (86.0% vs 93.8%). There was a borderline significant relationship between self-reported HIV positive status and engaging in transactional sex (p = 0.06). Women who were in a physical fight in the past 12 months were significantly more likely (p<0.001) to have engaged in transactional sex in exchange for money/gifts than those who did not (14.6% vs 6.9%).
Table 2. Associations between socio-demographic characteristics, ideational factors, behaviour, and transactional sex among women 16–55 years reporting sexual activity in the past 12 months n (%).
| Received money/gifts in exchange for sex with any sex partner in the last 12 months n = 3,574 | |||
|---|---|---|---|
| Yes n (%) | No n (%) | P value | |
| Socio-demographic characteristics | |||
| Age | |||
| 16–24 | 77 (29.7) | 949 (27.9) | 0.1 |
| 25–34 | 97 (43.9) | 1,263 (36.1) | |
| 35–55 | 51 (26.4) | 1,137 (36.1) | |
| Marital status | |||
| Single/ Divorced/Widowed | 102 (41.2) | 995 (29.6) | 0.1 |
| Not married or living together but in a steady relationship | 44 (19.4) | 874 (23.8) | |
| Not married, but living with sexual partner | 31 (15.7) | 478 (14.8) | |
| Married, living together | 43 (21.7) | 876 (27.3) | |
| Married not living together | 5 (2.1) | 126 (4.5) | |
| Education | |||
| No/ primary school | 18 (7.0) | 342 (10.5) | 0.5 |
| Up to grade 11 | 98 (37.6) | 1,438 (39.8) | |
| Matric | 83 (38.1) | 1,178 (34.7) | |
| Tertiary | 25 (17.3) | 386 (15.0) | |
| Employment | |||
| Unemployed | 145 (61.5) | 2,042 (59.3) | 0.7 |
| Employed | 59 (31.6) | 988 (32.1) | |
| Student | 17 (6.9) | 273 (8.6) | |
| Food security | |||
| Food insecure | 60 (22.5) | 547(16.9) | 0.1 |
| Settlement type | |||
| Urban | 117 (60.4) | 2,338 (64.4) | 0.5 |
| Rural | 108 (39.6) | 1,011 (35.6) | |
| Ideational factors | |||
| Knowledge of HIV prevention methods | |||
| No/low knowledge | 42 (19.0) | 750 (22.3) | 0.1 |
| Medium knowledge | 73 (32.2) | 1,414 (40.6) | |
| High knowledge | 110 (48.8) | 1,185 (37.1) | |
| HIV status & exposure to HIV communication | |||
| Knows HIV status | |||
| Yes | 156 (86.0) | 2,641 (93.8) | 0.0 |
| Self-reported HIV status | |||
| HIV positive | 27 (19.1) | 289 (13.2) | 0.1 |
| HIV communication programmes | |||
| Exposed to at least one HCP | 202 (89.5) | 2,927 (84.4) | 0.1 |
| Behaviour | |||
| Violence | |||
| Been in a physical fight in the past year | 36 (14.6) | 239 (6.9) | <0.001 |
| Alcohol use | |||
| Binge drinking | |||
| Yes | 52 (78.5) | 20 (55.0) | <0.001 |
| Frequency of drinking | |||
| Never | 17 (21.0) | 361 (35.8) | 0.0 |
| At least once a month | 22 (33.5) | 299 (30.8) | |
| At least once a week | 19 (23.1) | 187 (23.1) | |
| Several times per week | 7 (14.5) | 89 (8.8) | |
| Almost every day | 8 (7.9) | 18 (1.6) | |
| Alcohol dependence | |||
| Yes | 37 (42.0) | 307 (26.9) | 0.0 |
Alcohol use
Nearly a third (30.6%) of women reporting sexual activity in the past 12 months (n = 1,048) reported ever having had an alcoholic drink and 19.2% (n = 650) were current (past month) drinkers. Rates of lifetime drinking differed substantially across the country. Only 15.9% of women who were sexually active in the past 12 months in Limpopo Province reported lifetime drinking compared with 84.2% in Northern Cape Province.
Of those who reported ever drinking alcohol, 28.0% (345) were classified as alcohol dependent using the CAGE questionnaire. More than half, 56.6% of women (n = 535) who had ever drunk alcohol, were classified as binge drinkers. Using weighted data, this equates to an estimated 1.5 million women in South Africa.
The relationship between alcohol and transactional sex with any sex partner in the past 12 months
In the multivariate regression model, adjusting for confounding variables, women classified as binge drinkers were three times more likely to have engaged in transactional sex (AOR 3.1, 95% CI 1.5–6.6) than women who had ever had an alcoholic drink but who did not report binge drinking. This is shown in Table 3. Women who were alcohol dependent were 2.2 times more likely to report transactional sex (AOR 2.2, 95% CI 1.1–4.3). There was no significant relationship between frequency of drinking in the past month and transactional sex.
Table 3. Multivariate logistic regression results for transactional sex with any sex partner in the past 12 months and binge drinking, frequency of drinking in the past month and alcohol dependence among women * .
| Crude odds ratio | Adjusted odds ratio | 95% CI | P value | |
|---|---|---|---|---|
| Binge drinking | ||||
| No | Ref | Ref | ||
| Yes | 2.5 | 3.1 | 1.5–6.6 | <0.01 |
| Frequency of drinking in the past month | ||||
| Never | Ref | Ref | ||
| At least once a month | 1.6 | 1.5 | 0.6–3.3 | 0.4 |
| At least once a week | 2.2 | 1.4 | 0.6–3.5 | 0.5 |
| Several times per week | 1.7 | 2.2 | 0.7–6.9 | 0.2 |
| Almost every day | 9.4 | 3.7 | 0.6–22.6 | 0.2 |
| Alcohol dependence | ||||
| No | Ref | Ref | ||
| Yes | 2.0 | 2.2 | 1.1–4.3 | 0.0 |
*Controlled for age; cohabitation status; settlement type; food security; involvement in physical violence; HIV prevention knowledge; and HIV status.
Discussion
This study aimed to test the hypotheses that multiple dimensions of alcohol use were associated with transactional sex among women in the South African general population. Results support the hypotheses that binge drinking and alcohol dependence increases the likelihood of engaging in transactional sex. A conceptual model describing the relationship between alcohol use and transactional sex has been proposed (Fig 1). According to this model, socio-demographic characteristics have an independent effect on transactional sex. The model proposes that there is a bi-directional relationship between hazardous drinking and ideational factors.
Fig 1. The relationship between hazardous drinking and transactional sex based on the results of this study.
The relationship between alcohol dependence and transactional sex among women is worth further consideration. In a study undertaken with patrons of alcohol-serving establishments in Cape Town, Pitpitan and colleagues (2013) found that women who traded sex were more likely to be alcohol dependent [43]. In Pitpitan et al’s study (2013) the term problem drinking was used to characterise those who answered “yes” to at least two items of the CAGE questionnaire. This is the same measure used in this study, where the term alcohol dependence is used to describe individuals with a CAGE score of two or more. Viewed together, these findings suggest that the relationship between alcohol dependence and transactional sex may be bi-directional, which is what Townsend and colleagues propose in their model outlining how alcohol use and transactional sex are linked to risky sexual behaviour [53]. We have not found any evidence to support the idea that the relationship between binge drinking and transactional sex is bi-directional.
This study also found that 6.3% of women reporting sexual activity in the past 12 months reported that they had recently engaged in transactional sex. This is consistent with findings from 12 sub-Saharan African countries where the prevalence of receiving money or gifts among women aged 15 years and older ranged from 1.8% to 11% [6]. However, this rate is substantially lower than that found among women attending antenatal clinics in Soweto (21.1%) [3]. This discrepancy may be because the NCS was a national survey whereas Dunkle and colleagues’ study (2004) was only conducted in one province and in view of the different age bands-the national study included women up to the age of 55 years who are arguably less likely to engage in transactional sex.
Study results support the assertion that while levels of alcohol use in South Africa are relatively low, levels of hazardous and harmful drinking are high [46, 54].This study found that 28% of female drinkers were alcohol dependent, which is higher than results from the South African Demographic and Health Survey of 1998 (10%) and of 2003 (6.9%) [55, 56]. Of note is that over half of sexually active female drinkers, equating to roughly 1.5 million women, were classified as binge drinkers. This is consistent with other studies which found that although a relatively low percentage of women drink, those who do drink, do so at high risk levels [57].
These high levels of binge drinking among women may be partly explained by the growing body of research which shows that female drinking is often related to defiance of gender norms [58] and an increased sense of agency [59]. A number of studies have described the socially-scripted phenomenon which occurs in taverns. It is common place for men to buy women drinks and it is mutually understood that this is likely to end in sex [53, 59, 60]. Qualitative studies have revealed that alcohol is a common and desired form of currency in transactional relationships [41, 43]. Some women demonstrated increased agency by participating in the transactional sex dynamic [18, 59].
Another potential reason why hazardous alcohol use is high may be the lack of awareness around safe drinking relative to non-drinking. In neighbouring Namibia, LeBeau and Yoder (2009) found that the concept of moderate drinking was not one which was commonly understood, with few people understanding the difference between non-drinking and harmful drinking [60] This may well be the case in South Africa too. The high frequency of heavy drinking amongst women is likely to be further promoted by aggressive marketing of alcohol increasingly targeted at young women [61].
The results of this study should be interpreted in light of its limitations. Firstly, the NCS was cross sectional, limiting an assessment of causality conclusions between alcohol and risky sexual behaviour because measurement of exposure and effect occur simultaneously [62]. Secondly, there were potentially issues of measurement bias in relation to exposure. Alcohol measures are especially subject to recall bias and underestimation [37, 63]. The NCS questionnaire asked respondents who drank: “How often do you have (for men) five or more and (for women) four or more drinks on one occasion?”. This question did not include a timeframe [50], which is necessary to quantify frequency of binge drinking “to differentiate “binge drinking” from “alcoholism” or “alcohol dependence”“[64]. In addition, this study did not make use of AUDIT which has been recommended for alcohol measurement [65]. Not only does this make it difficult to compare to other studies in South Africa but literature suggests that structured instruments, like AUDIT, generally perform better than quantity-frequency questions [66].
Finally, there may have been issues in relation to measurement bias for the outcomes. Under-reporting of risky sexual behaviours is common [67] and this may have led to an underestimate of transactional sex. It is also likely that measurement of transactional sex in the NCS was incomplete. A more complete definition would have included alcohol and/or drugs as well as gifts and/or money as in Pitpitan and colleagues’ study (2014) where participants were asked: “Has someone given you money, alcohol, drugs or a place to stay in exchange for sex in the past 4 months?” [43].
Notwithstanding the above limitations, the findings of this study are important. This is the first national published study we are aware of which examines the relationship between multiple and more nuanced measures of alcohol use—alcohol dependence, binge drinking and frequency of drinking in the past month—and transactional sex in women. Although levels of lifetime drinking among women are quite low, the levels of binge drinking and alcohol dependency are scarily high. Findings demonstrate a clear, significant association between both binge drinking and alcohol dependence, and transactional sex amongst South African women in the general population. This confirms that alcohol misuse is an important driver of HIV risk among South African women, and tackling alcohol needs to be central to HIV prevention efforts especially for young women who are particularly vulnerable to transactional sex. Modelling data from Kenya suggests that with widespread uptake of interventions targeting unhealthy alcohol use, HIV infections could be reduced by five percent and some 18,000 AIDS-related deaths could be averted [68]. Moreover, studies have concluded in countries with severe HIV epidemics that designing and implementing interventions to promote safer drinking practices together with interventions to reduce risky sexual behaviour may have the potential to reduce HIV transmission. These data are relevant to South Africa where unsafe drinking and risky sex are both prevalent in the context of a generalised epidemic [68, 69].
The findings of this study have several implications for HIV prevention policy. Since young women and girls are particularly vulnerable to HIV, it is clear that programmes which address both alcohol and HIV prevention are needed for this group. This is especially important in light of the fact that alcohol consumption in sub-Saharan Africa is expected to increase in the next ten years due to improving economic conditions and aggressive alcohol marketing [70]. Programming needs to include interventions to prevent alcohol misuse particularly among young women as well as specifically targeting women who are already alcohol dependent and binge drinkers.
Further intervention research to design and evaluate the impact of such interventions on transactional sex and HIV incidence, within the South African context, would be helpful to inform programme implementation. While the need for a randomised controlled trial that evaluates the effectiveness of an alcohol focused intervention on HIV incidence has been identified, it does not focus specifically on women as a key population and/or include transactional sex as an outcome [71].
There are a number of other research implications arising from this study. It would be helpful to use other national level data, such as from the HIV prevalence surveys [2], to further analyse the relationship between different measures of alcohol use, transactional sex and HIV longitudinally. Longitudinal studies could be undertaken in the future to explore the hypothesised pathways between alcohol and transactional sex and HIV in relation to other potential paths linked to transactional sex such as experiences of intimate partner violence.
It is clear that national level programmes are needed as this study shows that both drinking and transactional sex is found in adult women throughout the country. Such programmes need to consider the multiple levels of intervention including increasing risk perception of hazardous drinking particularly for women; shifting drinking norms including those that facilitate transactional sex in taverns, creating safer drinking environments and, limiting/banning alcohol marketing towards women.
Acknowledgments
We acknowledge the organisations responsible for the Third National HIV Communication Survey, 2012 (NCS); Johns Hopkins Health and Education in South Africa, loveLife and Soul City, for supporting the use of this dataset.
Data Availability
We have uploaded the minimal data set to Dryad. The DOI number is: doi:10.5061/dryad.h7d87.
Funding Statement
The authors received no specific funding for this work. Anansi Health Consulting provided support in the form of salaries for authors SM and SJ, but did not have any additional role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript. The specific roles of these authors are articulated in the ‘author contributions’ section.
References
- 1.UNAIDS. UNAIDS report on the Global AIDS Epidemic, 2012. Geneva: 2012.
- 2.Shisana O, Rehle, T., Simbayi, LC., Zuma, K., Jooste, S., Zungu, N., et al. South African National HIV Prevalence, Incidence and Behaviour Survey, 2012. Cape Town: 2014. [DOI] [PubMed]
- 3. Dunkle KL, Jewkes RK, Brown HC, Gray GE, McIntryre JA, Harlow SD. Transactional sex among women in Soweto, South Africa: prevalence, risk factors and association with HIV infection. Social science & medicine (1982). 2004;59(8):1581–92. Epub 2004/07/29. 10.1016/j.socscimed.2004.02.003 . [DOI] [PubMed] [Google Scholar]
- 4. Jewkes R DK, Nduna M, Shai NJ. Transactional Sex and HIV Incidence in a Cohort of Young Women in the Stepping Stones Trial. J AIDS Clinic Res 2012;3:158 10.4172/2155-6113.1000158 [DOI] [Google Scholar]
- 5. Leclerc-Madlala S. Transactional sex, HIV and young African women: Are we there yet? Future Virology 2013;8(11):1041–3. [Google Scholar]
- 6. Chatterji M, Murray N, London D, Anglewicz P. The factors influencing transactional sex among young men and women in 12 sub-Saharan African countries. Social biology. 2005;52(1–2):56–72. Epub 2007/07/11. . [DOI] [PubMed] [Google Scholar]
- 7. Pettifor AE, Measham DM, Rees HV, Padian NS. Sexual power and HIV risk, South Africa. Emerging infectious diseases. 2004;10(11):1996–2004. Epub 2004/11/20. 10.3201/eid1011.040252 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Stoebenau K, Heise L, Wamoyi J, Bobrova N. Victim, Agent or Lover? Clarifying the Meaning of Transactional Sex for HIV Prevention in Sub-Saharan Africa London: STRIVE; 2015. [cited 2015 10 November 2015]. Available from: http://strive.lshtm.ac.uk/system/files/attachments/victim%20agent%20lover.pdf [Google Scholar]
- 9. Wamoyi J, Wight D, Plummer M, Mshana GH, Ross D. Transactional sex amongst young people in rural northern Tanzania: an ethnography of young women's motivations and negotiation. Reproductive health. 2010;7:2 Epub 2010/05/01. 10.1186/1742-4755-7-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Jewkes R, Morrell R, Sikweyiya Y, Dunkle K, Penn-Kekana L. Transactional relationships and sex with a woman in prostitution: prevalence and patterns in a representative sample of South African men. BMC public health. 2012;12:325 Epub 2012/05/04. 10.1186/1471-2458-12-325 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.World Health Organization. Preventing HIV in sex work in settings in sub-Saharan Africa. Geneva: 2011.
- 12. Leclerc-Madlala S. Transactional sex and the pursuit of modernity. Social Dynamics 2003;29:213–33. [Google Scholar]
- 13. Luke N. Age and economic asymmetries in the sexual relationships of adolescent girls in sub-Saharan Africa. Studies in family planning. 2003;34(2):67–86. Epub 2003/08/02. . [DOI] [PubMed] [Google Scholar]
- 14. MacPhail C, Campbell C. 'I think condoms are good but, aai, I hate those things': condom use among adolescents and young people in a Southern African township. Social science & medicine (1982). 2001;52(11):1613–27. Epub 2001/05/01. . [DOI] [PubMed] [Google Scholar]
- 15. Matasha E, Ntembelea T, Mayaud P, Saidi W, Todd J, Mujaya B, et al. Sexual and reproductive health among primary and secondary school pupils in Mwanza, Tanzania: need for intervention. AIDS care. 1998;10(5):571–82. Epub 1998/11/26. 10.1080/09540129848433 . [DOI] [PubMed] [Google Scholar]
- 16. Wojcicki JM, Malala J. Condom use, power and HIV/AIDS risk: sex-workers bargain for survival in Hillbrow/Joubert Park/Berea, Johannesburg. Social science & medicine (1982). 2001;53(1):99–121. Epub 2001/05/31. . [DOI] [PubMed] [Google Scholar]
- 17. Hunter M. The Materiality of Everyday Sex: Thinking beyond 'prostitution'. African Studies. 2002;61(1):99–120. 10.1080/00020180220140091 [DOI] [Google Scholar]
- 18. Maganja RK, Maman S, Groves A, Mbwambo JK. Skinning the goat and pulling the load: transactional sex among youth in Dar es Salaam, Tanzania. AIDS care. 2007;19(8):974–81. Epub 2007/09/14. 10.1080/09540120701294286 . [DOI] [PubMed] [Google Scholar]
- 19. Gregson S, Nyamukapa CA, Garnett GP, Mason PR, Zhuwau T, Carael M, et al. Sexual mixing patterns and sex-differentials in teenage exposure to HIV infection in rural Zimbabwe. Lancet (London, England). 2002;359(9321):1896–903. Epub 2002/06/12. 10.1016/s0140-6736(02)08780-9 . [DOI] [PubMed] [Google Scholar]
- 20. Zembe YZ, Townsend L, Thorson A, Ekstrom AM. "Money talks, bullshit walks" interrogating notions of consumption and survival sex among young women engaging in transactional sex in post-apartheid South Africa: a qualitative enquiry. Globalization and health. 2013;9:28 Epub 2013/07/20. 10.1186/1744-8603-9-28 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Dunkle KL, Jewkes R, Nduna M, Jama N, Levin J, Sikweyiya Y, et al. Transactional sex with casual and main partners among young South African men in the rural Eastern Cape: prevalence, predictors, and associations with gender-based violence. Social science & medicine (1982). 2007;65(6):1235–48. Epub 2007/06/15. 10.1016/j.socscimed.2007.04.029 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Kapiga SH, Lyamuya EF, Lwihula GK, Hunter DJ. The incidence of HIV infection among women using family planning methods in Dar es Salaam, Tanzania. AIDS (London, England). 1998;12(1):75–84. Epub 1998/02/10. . [DOI] [PubMed] [Google Scholar]
- 23. Mbulaiteye SM, Ruberantwari A, Nakiyingi JS, Carpenter LM, Kamali A, Whitworth JA. Alcohol and HIV: a study among sexually active adults in rural southwest Uganda. International journal of epidemiology. 2000;29(5):911–5. Epub 2000/10/18. . [DOI] [PubMed] [Google Scholar]
- 24. Hargreaves JR. Socioeconomic status and risk of HIV infection in an urban population in Kenya. Tropical medicine & international health: TM & IH. 2002;7(9):793–802. Epub 2002/09/13. . [DOI] [PubMed] [Google Scholar]
- 25. Zablotska IB, Gray RH, Serwadda D, Nalugoda F, Kigozi G, Sewankambo N, et al. Alcohol use before sex and HIV acquisition: a longitudinal study in Rakai, Uganda. AIDS (London, England). 2006;20(8):1191–6. Epub 2006/05/13. 10.1097/01.aids.0000226960.25589.72 . [DOI] [PubMed] [Google Scholar]
- 26. Fisher JC, Cook PA, Sam NE, Kapiga SH. Patterns of alcohol use, problem drinking, and HIV infection among high-risk African women. Sexually transmitted diseases. 2008;35(6):537–44. Epub 2008/04/18. 10.1097/OLQ.0b013e3181677547 . [DOI] [PubMed] [Google Scholar]
- 27. Baliunas D, Rehm J, Irving H, Shuper P. Alcohol consumption and risk of incident human immunodeficiency virus infection: a meta-analysis. International journal of public health. 2010;55(3):159–66. Epub 2009/12/02. 10.1007/s00038-009-0095-x . [DOI] [PubMed] [Google Scholar]
- 28. Kalichman SC, Simbayi LC, Vermaak R, Cain D, Jooste S, Peltzer K. HIV/AIDS risk reduction counseling for alcohol using sexually transmitted infections clinic patients in Cape Town, South Africa. Journal of acquired immune deficiency syndromes (1999). 2007;44(5):594–600. Epub 2007/02/28. 10.1097/QAI.0b013e3180415e07 . [DOI] [PubMed] [Google Scholar]
- 29. Shuper PA, Neuman M, Kanteres F, Baliunas D, Joharchi N, Rehm J. Causal considerations on alcohol and HIV/AIDS—a systematic review. Alcohol and alcoholism (Oxford, Oxfordshire). 2010;45(2):159–66. Epub 2010/01/12. 10.1093/alcalc/agp091 . [DOI] [PubMed] [Google Scholar]
- 30. Cook RL, Clark DB. Is there an association between alcohol consumption and sexually transmitted diseases? A systematic review. Sexually transmitted diseases. 2005;32(3):156–64. Epub 2005/02/25. . [DOI] [PubMed] [Google Scholar]
- 31. Fritz K, Woelk G, Bassett M, McFarland W, Routh J, Tobaiwa O, et al. The Association Between Alcohol Use, Sexual Risk Behavior, and HIV Infection Among Men Attending Beerhalls in Harare, Zimbabwe. AIDS Behav. 2002;6(3):221–8. 10.1023/A:1019887707474 [DOI] [Google Scholar]
- 32. Fisher JC, Bang H, Kapiga SH. The association between HIV infection and alcohol use: a systematic review and meta-analysis of African studies. Sexually transmitted diseases. 2007;34(11):856–63. Epub 2007/12/01. 10.1097/OLQ.0b013e318067b4fd . [DOI] [PubMed] [Google Scholar]
- 33. Chersich MF, Rees HV. Causal links between binge drinking patterns, unsafe sex and HIV in South Africa: its time to intervene. International journal of STD & AIDS. 2010;21(1):2–7. Epub 2009/12/24. 10.1258/ijsa.2000.009432 . [DOI] [PubMed] [Google Scholar]
- 34. Weiser SD, Leiter K, Heisler M, McFarland W, Percy-de Korte F, DeMonner SM, et al. A population-based study on alcohol and high-risk sexual behaviors in Botswana. PLoS medicine. 2006;3(10):e392 Epub 2006/10/13. 10.1371/journal.pmed.0030392 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35. Lane T, Shade SB, McIntyre J, Morin SF. Alcohol and sexual risk behavior among men who have sex with men in South african township communities. AIDS Behav. 2008;12(4 Suppl):S78–85. Epub 2008/04/09. 10.1007/s10461-008-9389-x . [DOI] [PubMed] [Google Scholar]
- 36. Kalichman SC, Pitpitan E, Eaton L, Cain D, Carey KB, Carey MP, et al. Bringing it home: community survey of HIV risks to primary sex partners of men and women in alcohol-serving establishments in Cape Town, South Africa. Sexually transmitted infections. 2013;89(3):231–6. Epub 2012/12/18. 10.1136/sextrans-2012-050569 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37. Chersich MF, Luchters SM, Malonza IM, Mwarogo P, King'ola N, Temmerman M. Heavy episodic drinking among Kenyan female sex workers is associated with unsafe sex, sexual violence and sexually transmitted infections. International journal of STD & AIDS. 2007;18(11):764–9. Epub 2007/11/17. 10.1258/095646207782212342 . [DOI] [PubMed] [Google Scholar]
- 38. Coldiron ME, Stephenson R, Chomba E, Vwalika C, Karita E, Kayitenkore K, et al. The relationship between alcohol consumption and unprotected sex among known HIV-discordant couples in Rwanda and Zambia. AIDS Behav. 2008;12(4):594–603. Epub 2007/08/21. 10.1007/s10461-007-9304-x . [DOI] [PubMed] [Google Scholar]
- 39. Ghebremichael M, Paintsil E, Larsen U. Alcohol abuse, sexual risk behaviors, and sexually transmitted infections in women in Moshi urban district, northern Tanzania. Sexually transmitted diseases. 2009;36(2):102–7. Epub 2008/12/09. 10.1097/OLQ.0b013e31818b20e6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40. Morojele NK, Kachieng'a MA, Mokoko E, Nkoko MA, Parry CD, Nkowane AM, et al. Alcohol use and sexual behaviour among risky drinkers and bar and shebeen patrons in Gauteng province, South Africa. Social science & medicine (1982). 2006;62(1):217–27. Epub 2005/08/02. 10.1016/j.socscimed.2005.05.031 . [DOI] [PubMed] [Google Scholar]
- 41. Townsend L, Jewkes R, Mathews C, Johnston LG, Flisher AJ, Zembe Y, et al. HIV risk behaviours and their relationship to intimate partner violence (IPV) among men who have multiple female sexual partners in Cape Town, South Africa. AIDS Behav. 2011;15(1):132–41. Epub 2010/03/11. 10.1007/s10461-010-9680-5 . [DOI] [PubMed] [Google Scholar]
- 42. Lewis JJ, Garnett GP, Mhlanga S, Nyamukapa CA, Donnelly CA, Gregson S. Beer halls as a focus for HIV prevention activities in rural Zimbabwe. Sexually transmitted diseases. 2005;32(6):364–9. Epub 2005/05/25. . [DOI] [PubMed] [Google Scholar]
- 43. Pitpitan EV, Kalichman SC, Eaton LA, Watt MH, Sikkema KJ, Skinner D, et al. Men (and women) as "sellers" of sex in alcohol-serving venues in Cape Town, South Africa. Prevention science: the official journal of the Society for Prevention Research. 2014;15(3):296–308. Epub 2013/03/16. 10.1007/s11121-013-0381-y [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.SANAC. National Strategic Plan on HIV, STIs and TB, 2012–2016. Pretoria: 2011.
- 45.Johnson S, Kincaid D, Figueroa M, Delate R, Mahlasela L, Magni S. The Third National HIV Communication Survey, 2012. Pretoria: 2013.
- 46. Peltzer K, Davids A, Njuho P. Alcohol use and problem drinking in South Africa: findings from a national population-based survey. African journal of psychiatry. 2011;14(1):30–7. Epub 2011/04/22. . [DOI] [PubMed] [Google Scholar]
- 47. Ewing JA. Detecting alcoholism. The CAGE questionnaire. Jama. 1984;252(14):1905–7. Epub 1984/10/12. . [DOI] [PubMed] [Google Scholar]
- 48. Claassen JN. The benefits of the CAGE as a screening tool for alcoholism in a closed rural South African community. South African medical journal = Suid-Afrikaanse tydskrif vir geneeskunde. 1999;89(9):976–9. Epub 1999/11/11. . [PubMed] [Google Scholar]
- 49. Dhalla S, Kopec JA. The CAGE questionnaire for alcohol misuse: a review of reliability and validity studies. Clinical and investigative medicine Medecine clinique et experimentale. 2007;30(1):33–41. Epub 2007/08/25. . [DOI] [PubMed] [Google Scholar]
- 50. Chavez PR, Nelson DE, Naimi TS, Brewer RD. Impact of a new gender-specific definition for binge drinking on prevalence estimates for women. American journal of preventive medicine. 2011;40(4):468–71. Epub 2011/03/17. 10.1016/j.amepre.2010.12.008 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 51. Cleland J WC. Demand Theories of the Fertility Transition: An Inconoclastic View. Population Studies. 1987;41(1):5–30. [Google Scholar]
- 52. Kincaid DL. Mass Media, Ideation, and Behavior. A Longitudinal Analysis of Contraceptive Change in the Philippines. Communication Research. 2000;27(6):723–63. [Google Scholar]
- 53. Townsend L, Ragnarsson A, Mathews C, Johnston LG, Ekstrom AM, Thorson A, et al. "Taking care of business": alcohol as currency in transactional sexual relationships among players in Cape Town, South Africa. Qualitative health research. 2011;21(1):41–50. Epub 2010/07/31. 10.1177/1049732310378296 . [DOI] [PubMed] [Google Scholar]
- 54. Parry CD. South Africa: alcohol today. Addiction (Abingdon, England). 2005;100(4):426–9. Epub 2005/03/24. 10.1111/j.1360-0443.2005.01015.x . [DOI] [PubMed] [Google Scholar]
- 55.Department of Health. South African Demographic and Health Survey 1998. Pretoria: 1998.
- 56.Department of Health. South African Demographic and Health Survey 2003. Pretoria: 2003.
- 57. Ojo OA, Louwagie G, Morojele N, Rendall-Mkosi K, London L, Olorunju S, et al. Factors associated with female high-risk drinking in a rural and an urban South African site. South African medical journal = Suid-Afrikaanse tydskrif vir geneeskunde. 2010;100(3):180–2. Epub 2010/05/13. . [DOI] [PubMed] [Google Scholar]
- 58. Wolff B, Busza J, Bufumbo L, Whitworth J. Women who fall by the roadside: gender, sexual risk and alcohol in rural Uganda. Addiction (Abingdon, England). 2006;101(9):1277–84. Epub 2006/08/17. 10.1111/j.1360-0443.2006.01516.x . [DOI] [PubMed] [Google Scholar]
- 59. Watt MH, Aunon FM, Skinner D, Sikkema KJ, Kalichman SC, Pieterse D. "Because he has bought for her, he wants to sleep with her": alcohol as a currency for sexual exchange in South African drinking venues. Social science & medicine (1982). 2012;74(7):1005–12. Epub 2012/02/14. 10.1016/j.socscimed.2011.12.022 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 60.LeBeau D, Yoder P. Alcohol Consumption, Sexual Partners, and HIV Transmission in Namibia. Calverton, Maryland 2009.
- 61.European Centre for Monitoring Alcohol Marketing. Women—The New Market Trends in Alcohol Marketing. Netherlands: 2008.
- 62. Bonita R, Beagleholde R, Kjellstrom T. Basic epidemiology 2nd ed. Geneva: World Health Organization; 2006. [Google Scholar]
- 63. Harrison L, Hughes A. Introduction-the validity of self-reported drug use: improving the accuracy of survey estimates. NIDA research monograph. 1997;167:1–16. Epub 1997/01/01. . [PubMed] [Google Scholar]
- 64. Courtney KE, Polich J. Binge drinking in young adults: Data, definitions, and determinants. Psychological bulletin. 2009;135(1):142–56. Epub 2009/02/13. 10.1037/a0014414 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 65. Fritz K, Kapiga S, Heise L, Francis J. Measuring alcohol-related HIV risk STRIVE: 2013. [Google Scholar]
- 66. Fiellin DA, Reid MC, O'Connor PG. Screening for alcohol problems in primary care: a systematic review. Archives of internal medicine. 2000;160(13):1977–89. Epub 2000/07/11. . [DOI] [PubMed] [Google Scholar]
- 67. Myer L, Karim S. Precision and validity in epidemiological studies: error, bias and confounding In: Joubert G, Ehrlich R, editors. Epidemiology: A Research Manual for South Africa. 2nd ed. Cape Town: Oxford University Press; 2007. p. 155–69. [Google Scholar]
- 68. Braithwaite RS, Nucifora KA, Kessler J, Toohey C, Mentor SM, Uhler LM, et al. Impact of interventions targeting unhealthy alcohol use in Kenya on HIV transmission and AIDS-related deaths. Alcoholism, clinical and experimental research. 2014;38(4):1059–67. Epub 2014/01/17. 10.1111/acer.12332 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 69.Fritz K. Prevention of alcohol-related HIV risk behaviour: Technical brief. Arlington, VA: 2009.
- 70. Jernigan DH, Babor TF. The concentration of the global alcohol industry and its penetration in the African region. Addiction (Abingdon, England). 2015;110(4):551–60. Epub 2015/03/17. 10.1111/add.12468 . [DOI] [PubMed] [Google Scholar]
- 71. Carrasco MA, Esser MB, Sparks A, Kaufman MR. HIV-Alcohol Risk Reduction Interventions in Sub-Saharan Africa: A Systematic Review of the Literature and Recommendations for a Way Forward. AIDS Behav. 2015. Epub 2015/10/30. 10.1007/s10461-015-1233-5 . [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
We have uploaded the minimal data set to Dryad. The DOI number is: doi:10.5061/dryad.h7d87.