Background

HIV/AIDS is one of the world’s major public health challenges. Progress in the prevention and treatment of HIV/AIDS is not advancing at the same rate as the general population around the world, especially among the key populations with female sex workers (FSW) having a 30 times higher risk of acquiring HIV, and gay and other men who have sex with men (MSM), with 28 times greater risk than adults aged 15–49 years [1].

Some studies have documented that the elevated risk of acquiring HIV and other sexually transmitted infections (STI) is exacerbated by the use of illicit drugs and alcohol, particularly when multiple layers of stigma exist related to sex work, homosexuality, and violence as occurs in HIV key population [2,3,4]. Alcohol use can also co-occur with illicit drug use, and several studies reported a consistent and positive association of alcohol and drug use with sexual violence, and unprotected sexual intercourse among FSW and MSM [4, 5]. For these key populations, substance use often leads to poor decision-making and riskier behaviors, while also making them face more stigma and social exclusion, all of which contribute to the spread of HIV [6,7,8].

In sub-Saharan Africa, drinking is closely linked to unprotected sex, starting sex at a young age, and having multiple partners. Some research also points to higher risks when alcohol is used with other drugs [2]. The use of drugs and alcohol in FSW has been described as a coping mechanism to numb the challenges associated with the conditions of sex work and a facilitator for clients’ requests and wishes. Those data suggest that interventions to reduce drug and alcohol use may be important to prevent HIV and other sexually transmitted infections among high-risk. Thereby targeting these behavioral health factors, we can enhance the overall impact of public health initiatives designed to control and eliminate the spread of HIV [4, 6, 8, 9].

Furthermore, it is essential to consider the broader impact of substance use on mental health, social functioning, and overall well-being. The harmful use of substances not only increases the risk of HIV but also contributes to a range of mental health issues, disrupts social relationships, and deteriorates the overall quality of life of individuals. Addressing these broader implications is a critical component of public health initiatives aiming to control and eliminate the spread of HIV [10,11,12,13].

The WHO Global Information System on Alcohol and Health has developed The Alcohol Use Disorders Identification (AUDIT), an instrument to screen hazardous and harmful alcohol consumption, drinking behavior, and alcohol-related problems in healthcare settings and surveys [14]. The AUDIT tool can provide more accurate insights into the intricate links between alcohol consumption and risky sexual behavior [2].

Over the past few years, there has been increased advocacy to focus interventions among key populations, especially in low- and middle-income countries. The Global AIDS Strategy 2021–2026: End Inequalities emphasizes tackling inequalities, with equitable and equal access to HIV services to close the gaps preventing progress toward ending AIDS [15, 16]. Low engagement in services and subsequent low treatment outcomes among key population groups represent inequalities requiring specific attention.

Mozambique is one of the global countries most affected by the HIV epidemic, with an HIV prevalence of 12.5% in the general adult population, according to the recent Mozambique Population-based HIV Impact Assessments (PHIA) survey [17]. The 5th National Strategic Plan in Response to HIV/AIDS (2021–2025) prioritizes interventions for key populations, including FSW and MSM [18]. The contribution of alcohol and drug consumption to high-risk sexual behavior among FSWs and MSMs in Mozambique is not well documented.

In this context, recognizing the higher HIV prevalence among key populations, we conducted a secondary data analysis to describe patterns of substance use among FSW and MSM, who are particularly vulnerable to HIV infection and to assess the associated risk factors. This analysis was based on data from the second bio-behavioral surveys (BBS) conducted in Mozambique between 2019 and 2020. Understanding these patterns of hazardous alcohol and drug use is a crucial component in informing comprehensive HIV prevention efforts, especially in populations where the interplay of substance use and HIV risk is most pronounced.

Methodology

Survey design, participants, recruitment, and sampling

The second round of BBS among FSW and MSM in Mozambique was implemented between 2019 and 2020 in five major urban areas in the country (Maputo, Beira, Nampula, Zambezia, and Tete).

The two surveys used the same respondent driven sampling (RDS) methodology, a peer-to-peer sampling methodology widely used to recruit high-risk and hidden populations. In this method, initial respondent, identified as “seeds,” are recruited based on their role in the community, and they are issued coupons used to invite additional participants from within their network to participate in the study. RDS sampling methodology uses responses about a participant’s personal network size to apply adjustments to the sample data that represent the larger target population in the specific geographic location. The weights represent the probability of selection. An estimator is used to produce weights for each participant that are used to generate population proportions of estimates and corresponding standard errors. Details about the survey methodology are described elsewhere [19].

FSW were eligible to participate in the survey if they were 15 years of age or older (FSW aged 15–17 are officially considered sexually exploited minors and for the purpose of the study are classified as emancipated minors and were therefore allowed to provide written informed consent to participate in the survey); reported receiving money or goods in exchange for sex from someone other than a steady partner in the six months preceding the survey; live, work or socialize in the survey area; and had a valid referral coupon. Following the formative assessment, in each city we selected three seeds, and each seed was offered five coupons. In Maputo, Beira, Tete, Quelimane and Nampula, 419, 521, 519, 514 and 519 FWS were recruited to participate to participate in the study, respectively. The sample size for each city was calculated following WHO 2017 bio-behavioral survey guidelines for populations at higher risk of HIV infection [20]. The BBS surveys carried out with FSW in Mozambique included young women aged 15–17. This decision is well documented and was made after careful consideration, drawing on evidence from previous research indicating that this data is crucial for devising effective policies and programs targeted at this high-risk HIV group [19, 21,22,23].

MSM were eligible if they were biologically male; 18 years of age or older; reported oral or anal sex with another male in the 12 months preceding the survey; live, work or socialize in the survey area; and had a valid coupon. In both surveys Exclusion criteria included inability to provide informed consent or being under influence of alcohol and drugs. For MSM, the sample size calculation followed a similar approach as that for FWS. In the survey, 530 MSM were recruited in Maputo city, 527 in Beira, 559 in Tete, 525 in Quelimane, and 528 in Nampula. Further details on sample size calculation can be found in the report available at www.ins.gov.mz.

An interview using a standardized questionnaire was followed by two sequential rapid HIV tests (Determine™ and Uni-Gold™), along with pre-and post-test counseling on site, in line with Mozambique Ministry of Health guidelines [24].

Outcome variables

Bio-behavior variables were assessed using the WHO Biobehavioral Survey Guidelines for populations at risk for HIV [25]. Participants were asked about alcohol use and illicit drug consumption. Drug use was measured by asking participants the following questions, “During the last 12 months have you consumed any illicit drug (drugs without medical permission)?” and “Which drug did you use?” Responses included injectable (cocaine and heroin) and non-injectable drugs.

We adopted the Alcohol Use Disorders Identification (AUDIT) instrument developed by the WHO Global Information System on Alcohol and Health to screen hazardous and harmful alcohol consumption, drinking behavior and alcohol-related problems in healthcare settings and surveys [14]. While the AUDIT provides an in-depth screening for hazardous, harmful, and potentially dependent drinking with a scoring range of 0 to 40, the AUDIT-C offers a more streamlined assessment of severe binge drinking (heavy drinking and/or active alcohol abuse or dependence), suitable for rapid screening in primary care settings, with a scoring system ranging from 0 to 12. This distinction makes AUDIT-C particularly useful in our study for identifying individuals engaging in risky drinking behaviors [2, 14]. For the purpose of this study, The 3 AUDIT questions analyzed were “How often did you have a drink containing alcohol in the past year?”, “How many drinks containing alcohol did you have on a typical day when you were drinking in the past year? ”, and “How often did you have six or more drinks on one occasion in the past year?” a composite score was calculated depending on the responses.

We utilized a cut-off score of 4 for men and 3 for women, as recommended in clinical guidelines. This threshold was chosen to identify individuals with risky drinking behaviors effectively. For the purpose of our analysis, the AUDIT-C score was treated as a binary variable: scores at or above the cut-off indicated hazardous alcohol use, while scores below suggested lower-risk consumption. This binary categorization allowed us to clearly delineate between participants with potentially harmful drinking patterns and those without, facilitating a more focused investigation into the correlates of hazardous alcohol use within the study population.

Exploratory variables

Explanatory variables were chosen based on the literature and programmatic importance including socio-demographic variables, HIV infection, self-reported STIs in the past 12 months, experience with violence (physical and sexual), comprehensive HIV knowledge (which entails a thorough understanding of how HIV is transmitted, prevented, and managed), own HIV risk perception, and health-seeking behaviors: access of health and health services in the last 12 months, self-reported STI (symptoms or diagnosis), and HIV test. High-risk social and sexual behaviors including hazardous drinking (defined by-Audit C), drug use, age of first sex work experience, number of sexual partners during the last month and condom use were also included.

Self-reported STIs were assessed by responding “yes” to at least one of the following questions: “During the last six months, have you had an abnormal discharge from your vagina, anus or penis?”, “During the last six months, have you had a sore or ulcer near your vagina, anus or penis?,” and “In the last six months, did someone inform you that you had or could have a sexually transmitted infection?”

Data analysis

Responses from the FSW and MSM, BBS surveys conducted during 2019–2020 were aggregated across survey-cities to produce pooled estimates for each population. Aggregate weighted estimates were computed using the RDS R package to analyze self-reported alcohol and drug use prevalence [26]. To produce aggregate weighted estimates, we considered the Gile’s successive sampling (RDS-SS) estimator, which accounts for the self-reported network sizes of participants, recruitment patterns and estimated size of the study population [27]. Details in the estimated size population for both FSW and MSM can be found in respective main reports. Unweighted bivariate analyses by Chi-square test were performed to identify risk factors associated with self-reported alcohol and drug use. The sample size of MSM reporting drug use was small (n = 36) and therefore were not included this sub-analysis.

All the analyses were done using the R statistical software 4.2.2 [28]. The code to run all the analyses is available upon request and approval.

Ethical considerations

The bio-behavioral survey among key populations adhered to stringent ethical measures that were rigorously observed. Participants provided written informed consent, and the study protocols received approval from the Institutional Ethics Committee (CIE-INS) and Mozambican National Bioethics Committee for Health (CNBS). To ensure confidentiality, personal identifiers were not collected, and survey responses were anonymized. Data were stored securely, accessible only to authorized personnel, and were encrypted to maintain privacy. The research team received comprehensive training in ethical research practices, with an emphasis on safeguarding participant confidentiality given the sensitivity of the data. Moreover, the study was subject to regular ethical audits to ensure adherence to established ethical standards and protection of participants’ rights and welfare. Minor FSWs [14,15,16], were connected with the Office for Women and Children of the Mozambican police, under the Ministry of the Interior. This office facilitates access to the legal system within a secure environment for victims of violence. Additionally, these women were directed to the Mozambican League of Human Rights, which offers legal advice at no cost.

Results

Socio-demographic characteristics

A total of 2,567 FSW and 2,669 MSM from survey cities participated in the study. The majority of FSW were < 25 years old (60.5%), single/ (61.1%) and had secondary/high education (72.8%). Among MSM participants the majority were < 25 years (78%), single (89.2%) and had secondary or high education (99.0%). The overall HIV prevalence among FSW participants was 26.5% and among MSM was (7.0%). Main socio-demographic and health characteristics are summarized in Table 1.

Table 1 General characteristics and HIV status of FSW and MSM, 2019–2020
Full size table

Hazardous alcohol and illicit drug among FSW and MSM

Among FSW in the five survey cities, nearly half (47,1%), reported hazardous drinking and 12.4% reported illicit drugs use. The prevalence of hazardous drinking among of MSM was 46.5% (Table 2).

Table 2 Prevalence of drug and alcohol use among FSW and MSM, 2019–2020
Full size table

Risk factors associated with hazardous drinking among FSW and MSM

Among the sub-sample of FSW who met AUDIT-C criteria for hazardous drinking, the median age was 24 years old (range:15–58). Hazardous drinking was more frequent in FSW participants from Maputo city (61.7% vs. 56.3% from Tete, 54.5% from Beira and 41.4% from Quelimane), > 25 years (54.3% vs. 42.5% for 15–24, p < 0.001) and those cohabitating or married (53.1% vs. 53.1% for widowed/divorced/separated and 43.5% for single, p < 0.001). More FSW participants with hazardous drinking behaviors reported having 2 or more clients in the last month (55.3% vs. 43.7% FSW with 1 client, p = 0.003), had higher STI self-reported (62,5% vs. 48,2%, p < 001) within the last month, had a higher own HIV risk perception (53.3% vs. 42.6% low), experience more physical (53.5% vs. 46.7%, p < 0.0001) and sexual violence (54.7% vs. 44.2%, p < 0.001) in the last 12 months preceding the survey, and were infected with HIV (55.2% vs. 44.2 p < 0.001) (Table 3).

MSM participants with hazardous drinking had a median age of 22 years old (range:18–61). Hazardous drinking was more frequent in MSM participants from Beira City (66.1% vs. 59.9% from Maputo, 59.9% from Tete, 26.3% from Nampula and 24.9 from Quelimane) younger MSM participants aged 19–24 years old (61.6% vs. 42.5% for 25 years and older, p < 0.001), and widowed/divorced/separated (65.9% vs. 59.3% cohabitating/married and 44.0% single). Most MSM participants with hazardous drinking had their first sexual experience with other men after 18 years of age (54.7% vs. 47.0 from 15 to 18, p < 0.001), had a higher HIV risk perception (53.4% vs. 45.8 low), higher self-reported STI (52.8% vs. 45.4%, p < 001. HIV infection was also more frequent among MSM participants reporting hazardous drinking (53.0% vs. 46.3 p < 0.001) (Table 3).

Table 3 Socio-demographic and behavior correlates of hazardous drinking (AUDIT-C) and risk factors among Female Sex Workers (FSW) and Men who have Sex with Men (MSM), Mozambique 2019–2020
Full size table

Risk factors associated with illicit drug use among FSW

Among the sub-sample of FSW participant who use illicit drug use in the last 12 months before the survey the median age was 22 years old (range 15–22). Illicit drug use in the past year was more frequent among FSW from Maputo City (25.5% vs. 13.7% from Beira, 10.5% Tete, 8.4% Quelimane and 6.3% from Nampula p < 0.001), and among FSW in a union (15.5% vs. 13.0% single and 10.7% widowed/divorced/separated, p = 0.003). More FSW that used illicit drugs had a high perception of their own HIV risk (14.2% vs. 9.7% low perception, p = 0.005) and have their first sexual experience when younger than 15 years old (15.4% vs. 12.6% for 15–17 years old and 5.3% for more than 24 years old, p < 0.001). More FSW who used drugs self-reported an STI in the last year (17.9% vs. 10.2% those did not, p < 0.001), and experienced physical (17.4% vs. 7.0% those that did not, p < 0.001) and sexual violence (18.6% vs. 8.9% those that did not, p < 0.001) in the last year before the survey (Table 4).

Table 4 Socio-demographic and behavioral correlates of drug use among Female Sex Workers (FSW) participants, Mozambique 2019–2020
Full size table

Discussion

Our findings indicate a high prevalence of hazardous alcohol consumption among FSW and MSM where close to half are considered at risk according to the AUDIT-C scale, which is consistent with the literature [5, 6, 29, 30]. For FSW, hazardous drinking use may serve as a coping mechanism to deal with the nature of their work (stress, violence, criminalization and stigmatization), while for MSM, alcohol consumption may serve as a mechanism to deal with stigma and related stress [4, 30]. These results emphasize the need for alcohol risk reduction programmes for key populations specifically focused on the adoption of safer drinking practices integrated into HIV prevention packages.

In our study, hazardous drinking was more frequent among MSM with younger age. This likely mirrors social behavior related to alcohol use in the general population to enhance pleasure, to be more social, and/or life histories of traumatic experiences such as sexual orientation-based discrimination and childhood sexual abuse [31]. Several studies demonstrate alcohol use in younger ages to be associated with a greater number of lifetime sexual partners, non-protected sexual intercourse, elevated levels of depressive symptoms, and alcohol abuse later in life [29, 31, 32]. Within younger MSM, where there is an increased risk of acquiring HIV compared to their older counterparts, alcohol use preceding sexual intercourse represents an increase risk for the acquisition of HIV and other STI [7, 31, 33].

Nearly half (47.1%) of our FSW participants reported hazardous drinking, a figure that aligns closely with the 41% prevalence of hazardous/harmful/dependent alcohol use among FSWs reported in 32 low- and middle-income countries. This rate is substantially higher than the global prevalence of alcohol use disorders among women in the general population, which stands at 5.1% [5]. Unweighted pooled estimates conducted in our study demonstrate that FSW with hazardous alcohol consumption have an increased number of sexual partners, a higher occurrence of self-reported STI, higher perception of their own HIV risk, and have higher HIV prevalence. Other studies demonstrate that hazardous alcohol consumption likely account for the positive association between frequency of drinking and increased number of clients [2]. Previous studies have demonstrated long-term links between alcohol consumption and HIV infection in the general population [4, 15, 34]. Particularly among FSW Alcohol affects decision-making about negotiating for safer sex, which can increase risk of HIV and other STI transmission [8]. In addition hazardous heavy drinking and drug use have also been associated with poor adherence to antiretroviral therapy (ART) among HIV positive people and the interaction between all those substances usually leads to a higher susceptibility to co-morbidities and opportunistic infections [8, 35]. The high burden of alcohol use among FSWs also carries serious health and social implications, as excess alcohol consumption is linked with a range of adverse physical and mental health outcomes, underscoring the need for targeted interventions in this population [4, 8, 36].

Physical and sexual violence is recognized as a widespread public health problem and a violation of human rights. We found FSWs that use alcohol are enmeshed in a dynamic of physical and sexual violence victimization. A prior study conducted among FSW in the country found a high prevalence of physical and sexual violence, confirming the need for specific interventions to address this vulnerability [23]. Furthermore, it’s important to note that while sex work is not criminalized in Mozambique [37], it remains highly stigmatized, and the legal system does not include specific protections for sex workers, creating additional challenges for those affected by alcohol-related issues.

“While our study found a prevalence of illicit drug use among FSW participants at 13.3%, it is noteworthy that the systematic review by Iversen et al. identified a higher pooled prevalence of lifetime illicit drug use among sex workers globally at 35% [38]. It is important to note that the prevalence of drug use among FSWs is a complex issue that is influenced by various factors such as social, economic, and cultural determinants. Consequently, the prevalence of drug use among FSWs may vary across different regions and countries. However, the risk profile of FSWs who use illicit drugs in our study is consistent with what has been observed in other studies among key populations (KP) in Mozambique. These studies consistently show that drug use tends to occur more frequently in younger individuals with a higher number of sexual partners, and it is associated with higher HIV and STI prevalence [19, 39]. Several additional studies demonstrate that there is a strong relationship between substance use and unsafe sexual practices that increases the risk for HIV and STIs [7, 30, 36, 40,41,42]. Additionally, FSW may use substances as a coping mechanism to numb the challenges associated with their sexual practices, which can also contribute to hazardous drinking [36, 40,41,42].

Our study had some potential limitations. First, we measured the alcohol and drug use consumption by self-report thus, the data might be under-reported due to social desirability bias. Second, due to missing data we were not able to assess the drug consumption among MSM. Third, like other cross-sectional surveys, we could not assess cause-and-effect relationships, for example it was not possible to assess whether hazardous drinking behaviors and illicit drug use preceded sexual risk behaviors. Fourth, the FSW and MSM network assessed in the study may be missing important sub-groups (E.g. FSW and MSM with higher economic status) and therefore may not capture the specific drug and alcohol behaviors of these sub-populations. In addition, the analysis of pooled results from across the survey cities may affect the representativeness of sample to the general KP population in the survey cities. Finally, these findings need to be interpreted with caution and cannot be generalized to the entire MSM and FSW population in the survey cities nor at the national level. Although we were able to find significant association between the outcome variables and the exposure, we did not provide the strength of the association, considering that the analysis was more descriptive than inferential. Therefore, results should be interpreted with caution. Additionally, it should be noted that RDS is only valid within each area.

Conclusion

The data from this study indicate that hazardous drinking use are associated with behaviors that increase the risk of HIV among FSW and MSM, and illicit drug among FSW. As these two groups are among the most vulnerable populations for HIV, there is a need to integrate substance abuse screening, referral, and treatment into HIV prevention and treatment services. Additional studies are needed to further explore the relationship between alcohol consumption, illicit drug use and mental health issues among these populations to inform efforts that address the inequalities of the HIV epidemic.