Community empowerment and involvement of female sex workers in targeted sexual and reproductive health interventions in Africa: a systematic review

Sex workers are highly vulnerable to health and social problems, including sexually-transmitted infections (STIs), unintended pregnancy, violence, exploitation, discrimination and substance abuse [1]. For example, in countries with generalised HIV epidemics, the odds of a female sex worker (FSWs) to be living with HIV is 13.5 times that of other women [2]. Social marginalisation and a typically criminalised working environment limit the ability of FSWs to mitigate the impact of their occupational hazards, such as multiple sexual partners, difficulties in negotiating condom use, poor access to appropriate lubricants and high STI prevalence [3, 4]. Up to 4.3% of women and girls engage in sex work in sub-Saharan Africa [5], yet FSW interventions in the region have historically operated in isolation with limited national or international support, in strong contrast to larger coordinated sex work projects in other settings [6]. Despite being a priority population for HIV interventions globally, coverage of prevention, treatment and support programmes among FSWs in the region remains low [7, 8].

International discourse on the health of vulnerable populations has increasingly favoured ‘structural interventions’ – in other words, interventions addressing ‘macro-level’ structural forces that “shape distributions of power within and across societies”[9] – over those targeting bio-behavioural risk at the individual level [9‐12]. In India, targeted FSW interventions have for long adopted a community-centred approach, with a particular focus on transforming politico-legal or socio-cultural structures [13‐15]. In the early 1990s, the Sonagachi Project in West Bengal, India implemented one of the first developing country models of sex worker services that incorporated community and societal-level theories of change into HIV/STI prevention interventions [13]. As a measure of its success, the community-based organisation Durbar Mahila Samanwaya Committee (DMSC), which formed in 1995 and has a current membership of 65 000 sex workers, has been running the Sonagachi Project since 1999. Other sex worker-led organisations such as Ashodaya Samithi in Mysore (hereafter ‘Ashodaya’) have integrated the Sonagachi approach into broader community-led interventions to reduce violence and the structural underpinnings of HIV vulnerability. This model has since been rolled out across six states as part of the Avahan India AIDS Initiative [14]. Such projects have arguably achieved greater impact, in terms of clinical and social outcomes, coverage and sustainability, than those which seek to build individual responsibility for effective behaviour change [15, 16].

Lessons drawn from several large evaluations of Indian community empowerment models [13, 15, 17, 18] suggest that these models lend themselves to adaptation in the African context. However, much remains unknown about whether FSW community empowerment initiatives have been implemented in Africa or, indeed, whether they have the potential to make as impressive an impact as they have done in India. It is also unclear how interventions using this broad approach to empower sex worker communities in African settings have been structured. To date, health services across the world have primarily engaged with FSWs through provision of sexual and reproductive health (SRH) services, often in the form of targeted interventions designed to interrupt transmission of HIV and STIs between FSWs, their clients and the wider population [19]. This article presents findings from a systematic review of SRH interventions targeting FSWs in Africa, focusing on projects that include elements designed to empower communities. Firstly the review aims to describe the nature and structure of targeted SRH interventions encompassing empowerment processes. Thereafter, the review evaluates the progression of ‘community empowerment’ interventions in Africa from the perspective of SRH service provision – in other words, the extent to which SRH interventions have provided an enabling environment for community mobilisation and structural change. For this assessment, we use a four-stage framework of community empowerment devised by Ashodaya and DMSC in India (examples are primarily drawn from Ashodaya). This is a sub-study of a larger systematic review of targeted SRH service delivery models across Africa and India, undertaken for the DIFFER project (Diagonal Interventions to Fast-Forward Enhanced Reproductive health).

The most recent WHO guidelines on HIV and STI prevention and treatment for sex workers (2012) recommend that all health services, including primary health care, are made ‘available, accessible and acceptable to sex workers based on the principles of avoidance of stigma, non-discrimination and the right to health’[19]. Critically, these guidelines consider a package of interventions to enhance community empowerment among sex workers “absolutely necessary” for improving living and working conditions and redressing human rights violations. In the context of formulating a set of recommendations for such interventions, the WHO defines community empowerment as “a collective process through which the structural constraints to health, human rights and well-being are addressed by sex workers to create social and behavioural changes, and access to health services to reduce the risk of acquiring HIV” [19]. We use this definition throughout the article, since it approximates most closely the way ‘community empowerment’ is operationalised by Ashodaya (which, to our knowledge, has not explicitly developed a definition of its own).

In the academic literature it is generally considered more insightful to define empowerment as a process rather than an outcome [20, 21], or as an interaction between both process and outcome [22, 23]. Laverack’s conceptualisation of community empowerment as a ‘dynamic continuum’ from individual action to collective social and political change [20] is most consistent with Ashodaya’s approach. In this formulation, components of an empowerment intervention should be designed to enable people to maximise their potential to progress along this continuum [20]. Fundamentally, Ashodaya aims to respond to the needs of the community by encouraging community mobilisation and critical consciousness-raising among sex workers, to enable transformation of the power relations that shape their worlds. In this respect, the approach draws on the tradition of ‘liberatory pedagogy’ and notions of collective empowerment as advanced by Paolo Friere and others [24].

For example, after identifying the ‘governance’ of sex work and its policing as a key community concern, Ashodaya’s advocacy efforts to reduce vulnerability in this area included holding ‘sensitisation’ meetings with local police officials and sex workers. These meetings challenged popularly held stigmatised notions of sex workers and highlighted how police could change their attitudes towards sex workers and assist in HIV prevention. Collectively, sex workers were empowered to confront police action they considered improper, through provision of legal literacy, support from a 24-hour crisis intervention outreach team and nurturing a culture of rights. These efforts resulted in a marked reduction in violence from police, clients, local thugs and shopkeepers, and cultivated an image of sex workers as active agents for social change [15]. In short, the community’s response to violence and their progressive empowerment featured mobilising the sex worker community, collectively designing an intervention and ensuring community ownership of it, thereby shifting structural vulnerabilities in the long term.

Ashodaya applies a four-stage framework to conceptualise the progression of empowerment in the context of their efforts to reduce violence and HIV risk: (1) engagement with the sex worker community, (2) involvement of the community in targeted activities, (3) ownership of the project by the community (4) sustainability of action beyond the community (Figure 1) [15]. Ashodaya’s framework is useful insofar as it provides a means of analysing the complex relationship between empowerment and the challenge of confronting stigma, discrimination and gender-based violence [25].

To assess FSW community empowerment in African settings, this paper considers the extent to which projects undertake activities that facilitate the processes defining each stage of this framework. Focus is not placed on individual examples, since these are likely to be context-specific. Our aim is not to compare African and Indian interventions per se, much less to suggest that the Indian trajectory is necessarily the ‘gold standard’ that other regions should uncritically emulate. Rather, the framework is useful merely as a heuristic device to illuminate the progress made by African service delivery models from an empowerment perspective.

On 22 November 2012, we searched Medline (Pubmed interface) and Web of Science, without date restrictions, for studies of SRH service provision to FSWs in Africa. Search terms used in Medline were: “prostit*” or “sex work” or “sex worker” or “sex workers”, and all low- and middle-income countries [26] (MeSH term or any field). Articles were located in Web of Science using the terms “sex work” or “prostitution”, and then filtered to include only African countries. The full search strategy is available on request. Some experts in the field were contacted for additional service models, but we did not aim to comprehensively identify projects not reported in the peer-reviewed literature. Authors of articles were not contacted for additional information.

Titles and abstracts were screened using EPPI-Reviewer (United Kingdom, version 4) [27]. Duplicate references were removed, and abstracts and titles then screened using pre-specified inclusion and exclusion criteria. Full-text articles were reviewed where inclusion could not be established on review of title and abstract alone. Review methods follow the PRISMA guidelines [28], however, because the review examined programme design and not intervention outcomes, some modification was required, such as assessment of risk of bias. The EPPI-Reviewer 4 software and Endnote were used to index all documents located and later referenced.

Of the 5413 articles screened, 129 were included, describing 42 different projects incorporating elements of FSWs community empowerment in 26 different African countries (Figure 2 and Additional file 1: Table 1).

Results are presented in two sections: the first describes the range of service delivery models and the second assesses the overall progression of community empowerment in African FSW interventions, using the conceptual framework detailed above.

Of the SRH interventions assessed in this review, very few appear to have focused on community-led service provision or were concerned with building the capacity of the FSW community to change their circumstances. Our results support findings of a recent review of sex worker community empowerment intervention evaluations in low and middle-income countries by Kerrigan et al.[154], which identified no projects in sub-Saharan Africa initiating community empowerment in the sense defined by Wallerstein, namely, as “a social action process that promotes the participation of people, organizations, and communities towards the goals of increased individual and community control, political efficacy, improved quality of community life, and social justice”[22].

Although the Ashodaya framework represents only one of many possible evaluation methodologies, it is telling that the vast majority of interventions reviewed had conducted activities to enable empowerment processes consistent with only stage one or two in this framework. Because achievement of each stage was conceptualised as being necessary but not sufficient to allow progression to the next, we did not consider whether individual processes in one stage facilitated achievement of others without following the defined stepwise continuum. An example of where this may have been relevant is the Zambian Corridors of Hope project, within which project staff initiated broader social engagement from the outset (a stage four component that within the Ashodaya framework would be undertaken by FSWs themselves). Similarly, we are aware of some younger African SRH projects that do not engage with FSWs from within the service delivery model (with its inherent patient-provider power relations), but nonetheless aim to impact health service provision by first supporting communities to analyse and act on health and human rights issues concerning them. One example (a study of which did not meet our review criteria) is a UNFPA-supported initiative in Namibia that in 2011 trained sex workers to conduct rapid assessments on HIV and sex work [155].

Notwithstanding apparent limitations of approaches taken by the majority of projects in this review, efforts to build close links with peer networks and involve FSWs in project design and implementation do improve acceptability and participation, a crucial first step in addressing conditions that heighten FSW vulnerability [16]. The significance of these efforts should therefore not be overlooked. Furthermore, the fact that isolated examples of empowerment ‘successes’ exist (i.e. interventions meeting stage two or three) demonstrates proof of concept for the progression of FSW community empowerment in African settings.

One explanation worth considering for why African projects have not (yet) progressed through Ashodaya’s four stages is that these projects have simply not had time to develop complex empowerment processes and the formal structures required to cultivate them. The level of collective organisation required to bring about significant social and political change is usually only achieved after several years of sustained community mobilisation [20]. Indeed the Sonagachi project evolved from a single STI clinic over more than a decade, in response to sex workers expressed concerns [13]. Yet public health and service engagement with African FSW communities has been happening since at least the mid-1990s and has not evolved in the same way as in India [52, 156‐158].

The fact that the success or failure of certain peer education projects in Africa could be attributed to the level of training and support offered, highlights not only the challenge of resource availability, but also of securing long-term commitment of programmers who facilitate empowerment by engaging and adapting to the needs and capacity of these communities. In most cases we were unable to determine project duration, however the fact that most projects were established as part of research studies, potentially with limited funding and timeframe, is worrisome, as this limits opportunities to develop effective and lasting community empowerment processes. Power differentials between researchers and study participants may also undermine such processes, with researchers primarily viewed as being scientific experts, rather than as being skilled to engage with stakeholders, such as police [20]. Further, although ongoing community engagement is clearly a crucial component of sustainability, our assessment does not consider financial sustainability, which may be of particular concern in Africa due to lack of government involvement and project isolation. In Ashodaya’s case however, successful community engagement itself contributed to financial sustainability, through establishment of an independent cooperative society to provide loans to sex workers and non-sex workers alike, the interest from which was reinvested into Ashodaya.

The socio-political context in which empowerment interventions are implemented is also likely to be a key determinant of their success; differences in the characteristics and organisation of sex work and the overall socio-political context in Africa, compared to India, may therefore limit the effectiveness of activities aiming to mobilise communities, as well as the impact of community mobilisation itself. Cornish and Campbell [159] provide a useful conceptual framework of ‘interventions-in-context’, based on an ecological perspective of community psychology, to examine inconsistent outcomes in peer-education programmes between different settings [159]. Whilst FSWs worldwide have historically been marginalised and disenfranchised, it is arguable that the social fabric, infrastructural, economic and political context in many African settings is exceptionally disempowering [159]. In contrast to India’s relatively structured sex work industry, FSWs across Africa are a highly heterogeneous and mobile group of women with varying frequency, duration and location of sex work, many of whom do not even self-identify as sex workers [3, 160]. In addition to this, nascent democracies, criminalisation and weak welfare states leave many in this environment with few foundations from which to unite or assert their rights, indicating that even longer time frames and higher levels of support may be required to facilitate community collectivisation [155, 159]. Even for one of Africa’s ‘empowerment successes’, the Pumwani Majengo project in Nairobi, lack of political will and financial resources were initially major constraints to expanding programme coverage, despite strong evidence that this service delivery model could make an important contribution to HIV and STI prevention [68]. To some extent these constraints were, however, subsequently overcome and the project has now been replicated in several Kenyan cities. In terms of community engagement, Pumwani Majengo demonstrates that careful programming can mitigate adverse social contexts. Here community spirit was nurtured by using the STI clinic to promote awareness of the fact that individuals in an outwardly heterogeneous community still shared needs and interests [20, 130].

Examination of various SRH service delivery models targeting FSWs provides valuable insights into the way in which sex worker vulnerability is framed on the continent. Ashodaya’s approach encouraged and supported sex workers to re-conceptualise their vocation and demand services to counteract various occupational hazards, thus entrenching bottom-up community empowerment. A range of social interventions, including crisis response and political advocacy, as well as clinic-based SRH promotion were thus provided from the outset [15]. The rarity of structural interventions in African projects suggests a predominant bio-behavioural approach to effecting disease control, rather than a community development approach to service delivery. FSW vulnerability in African settings thus has, to date, been regarded not as an occupational health problem with structural antecedents, but a matter of individual risk behaviour. Arguably a consequence of this framing is that peer education has tended to be treated as a participatory end-point, rather than a means of achieving progressive involvement. By contrast, Sonagachi peer education provided the experiential context that promoted and led to wider experiences of effective community action [161]. ‘Exit programmes’ – which seemingly are rare in Africa – also exemplify activities which may be undertaken in good faith, but whose capacity to empower is limited when the emphasis is on ‘rescue’ from sex work, rather than promoting individual agency [162]. In this respect it is the agenda and purpose of the process, rather than merely the process itself, which determines it’s potential to be simply participatory or truly empowering [20]. Progression through Ashodaya’s framework is therefore not simply an inevitable outcome of certain participatory activities, but may in some cases mandate re-evaluation of a project’s philosophy and a strong commitment to pursuing the social and political changes required to meet the expressed needs of the community.

International public health discourse around SRH in the past decade has moved from a focus on adverse outcomes of sexual behaviour, such as teenage pregnancy and STI transmission, to one of broader SRH and rights [163]. A failure to recognise the socially transformative value of community empowerment processes – independently of their impact on behavioural and clinical outcomes – may lead to effective interventions being overlooked and underfunded [163]. This has implications for study design, since traditional clinical trials are seldom able to account for the complex associations caused by structural interventions [164]. Evaluation frameworks for health interventions are typically short-term in orientation, and therefore unable to accommodate, or even recognise, the longer timeframes needed for community empowerment processes to evolve.

Evidence suggests that no single approach can be used when designing SRH interventions; rather, a multifaceted approach that addresses individual behaviour as well as the social and political context is required [163]. The fact that processes rather than specific activities are prioritised within the Ashodaya framework means that its application is highly context specific and must be tailored to the specific needs of a community [165]. A range of approaches to alter the structural causes of FSW vulnerability has been trialled internationally, in order to identify locally appropriate and effective solutions. For example, Thailand’s 100% Condom Use Programme, which provides outreach to sex workers and mandates condom use in all sex establishments, has increased access to STI services and enabled sex workers to demand condom use. The programme was scaled up nationally in the early 1990s and has achieved marked public health benefits. Various 100% Condom Use Programmes have been successfully implemented in several Asian countries alongside multi-stakeholder mobilisation, to create enabling environments for outreach and increased access to services [165]. In the Dominican Republic, the Sonagachi ‘grassroots’ approach was most effective only in combination with top-down condom legislation akin to 100% Condom Use Programmes [166].

This systematic review has several limitations. Most obviously, we examined only the structure of interventions, rather than uptake or effectiveness. This approach was adopted partly to focus our efforts within a wide subject area, and partly because of the conceptual decision to define empowerment in terms of processes rather than a simplified dichotomy of whether empowerment had been achieved or not. We do acknowledge, however, that alternative theoretical frameworks may be appropriate for further research, particularly for individual project evaluations. Some of our findings allude to potentially important empowerment ‘outcomes’ for assessment: for example, we note that in some circumstances, too wide a forum may have discouraged the most vulnerable from voicing or acting on their concerns, thereby limiting the impact of the intervention. Here, the outcome in question would be the extent to which a range of individuals was able to participate. Other outcomes such as coverage, rates of consistent condom use, or a measure of progressive involvement of FSWs are further examples of important clinical, process and empowerment ‘outcomes’ not examined in our approach.

A second limitation is that by undertaking a systematic review focusing largely on peer-reviewed literature, we have likely excluded a large number of projects in Africa led by non-governmental or community-based organisations, which potentially have closer working relations with FSW communities than researcher- or government-led services. For example, through other research in South Africa, the authors are familiar with the work of SWEAT, and therefore recognise that SWEAT’s activities as reported in the academic literature (and which are detailed in this article) do not do justice to the broad range of activities and complex democratic structures that characterise this organisation in practice [167]. Importantly, SWEAT’s partnership with the sex worker organisation Sisonke has facilitated a national network of sex workers, thus substantially increasing sex worker involvement in project activities. Similarly, the Sex Workers’ Outreach Programme (SWOP) in Kenya runs a nationwide network of drop-in centres, yet was not identified in our literature review [168]. Although not necessarily directly related to SRH service provision, numerous other sex worker associations, such as those affiliated with the African Sex Worker Alliance (ASWA), are also involved in important community development work [169, 170]. These activities are known to us because of our familiarity with these organisations, and not because they are described in the published literature.

Notwithstanding these limitations to our search methods, we regard a systematic review of the peer-reviewed literature on this subject to be a valid methodology. A database search of this literature provides a systematic way of surveying evidence on this subject, a difficult task with grey literature. Also, it highlights the need for more research in this field to be published (whether descriptive or evaluative), in order to facilitate informed and transparent project design in the future. The false dichotomy between academic research and community activism should not prevent information sharing that may promote more effective realisation of health and human rights of vulnerable populations.

A third limitation is that we assessed empowerment activities only within the parameters of SRH service provision. As previously stated, adopting wider parameters was considered beyond the scope of this research. However, we caution the reader to be aware that highly valuable efforts to shift social vulnerabilities outside the health sector are potentially excluded here. The fourth limitation concerns the focus on FSWs, with the review unable to assess services targeting the sizable population involved in transactional sex in Africa. Moreover, by focusing on FSW services in Africa as a whole, we have not accounted for the high level of heterogeneity in socio-political contexts between and within countries, thereby limiting the applicability of our conclusions. Finally, some methodological constrains warrant mention. Single data extraction was done, which incurs more errors than duplicate independent extractions. Also, though the review adhered to the principles underlying systematic reviews, it differs from most quantitative systematic reviews, for example in omitting an assessment of risk of bias.

Although there are clearly lessons to be learnt from previous experiences, there is currently insufficient evidence within the academic literature to support the scale up of any particular model of community empowerment within SRH interventions in African settings. Further rigorous research in this region is needed, using appropriate methods and follow-up periods to measure intervention outcomes longitudinally. Progression through Ashodaya’s framework may not be possible in all African settings, nor can we determine whether collectivisation alone is necessary or sufficient to achieve significant improvements in health and human rights - highlighting the particular need to investigate context-specific, innovative or combination approaches to reduce structural vulnerability of FSWs in Africa.

Our findings suggest that FSW SRH interventions in African settings that seek to engage with the community need to rethink their lack of attention to structural factors and reframe FSW vulnerability as an occupational health problem, to be addressed through collective action built on genuine involvement of FSWs themselves. This means moving away from a narrow disease-control focus and creating an environment which is both receptive and responsive to the expressed needs of the community, thereby promoting the transition from technical to transformative communication that facilitates social and political change [161]. Ultimately, higher levels of human and financial resources are needed to ensure sustainability of long-term interventions that are ‘owned’ by African FSW communities in much more meaningful ways than before.