Global epidemiology of HIV among female sex workers: influence of structural determinants

Abstract

Female sex workers (FSWs) bear a disproportionately large burden of HIV infection worldwide. Despite decades of research and programme activity, the epidemiology of HIV and the role that structural determinants have in mitigating or potentiating HIV epidemics and access to care for FSWs is poorly understood. We reviewed available published data for HIV prevalence and incidence, condom use, and structural determinants among this group. Only 87 (43%) of 204 unique studies reviewed explicitly examined structural determinants of HIV. Most studies were from Asia, with few from areas with a heavy burden of HIV such as sub-Saharan Africa, Russia, and eastern Europe. To further explore the potential effect of structural determinants on the course of epidemics, we used a deterministic transmission model to simulate potential HIV infections averted through structural changes in regions with concentrated and generalised epidemics, and high HIV prevalence among FSWs. This modelling suggested that elimination of sexual violence alone could avert 17% of HIV infections in Kenya (95% uncertainty interval [UI] 1-31) and 20% in Canada (95% UI 3-39) through its immediate and sustained effect on non-condom use) among FSWs and their clients in the next decade. In Kenya, scaling up of access to antiretroviral therapy among FSWs and their clients to meet WHO eligibility of a CD4 cell count of less than 500 cells per μL could avert 34% (95% UI 25-42) of infections and even modest coverage of sex worker-led outreach could avert 20% (95% UI 8-36) of infections in the next decade. Decriminalisation of sex work would have the greatest effect on the course of HIV epidemics across all settings, averting 33-46% of HIV infections in the next decade. Multipronged structural and community-led interventions are crucial to increase access to prevention and treatment and to promote human rights for FSWs worldwide.

Figure 1. Structural HIV determinants framework for sex work

Adapted with permission from Shannon and colleagues. ART=antiretroviral therapy. PWID=people who inject drugs. MSM=men who have sex with men.

Figure 2. Model of dynamic pathways between macrostructural (eg, law reforms) and social, policy, and physical features of the work environment (eg, access to safer work venues, violence, and policing) on HIV acquisition in female sex workers in Vancouver, Canada

The flowchart represents the dynamic and iterative pathways between various structural conditions (eg, recent or non-recent exposure to police harassment or client physical violence, or a lifetime exposure to sexual violence), which vary by work environment (represented by the coloured boxes). Female sex workers (FSWs; PWID or non-PWID) are assumed to work in a given work environment for their lifetime. FSWs on the far left have not experienced any violence yet. FSWs can be repeatedly exposed to different violence events at a frequency specific to their work environment; orange arrows show the flow of FSWs between recent and non-recent exposure to violence. IRR >1 shows the increase in the rate of exposure to recent police harassment if recently exposed to client physical violence or vice versa (bold arrows). RRc >1 denotes the increased risk of inconsistent condom use after a recent or non-recent exposure to the given form of violence or police harassment for the duration that they have been in that situation. PWID=people who inject drugs. IRR=incident rate ratio. RRc=inconsistent condom use risk ratio. SW=sex worker.

Figure 3. HIV prevalence and percentage of infections potentially averted in FSWs and their clients in the next 10 years through structural changes in Vancouver, Canada

(A) Predicted HIV prevalence in FSWs between 1985 and 2040. Squares show the empirical estimates (and 95% CIs) from the data; bold line shows the median (shaded area shows 95% CI) of the model predictions from the multivariate parameter fits. (B) Predicted fractions of new HIV infections that could be averted in FSWs and their clients from structural changes in 2014–24; vertical bars show 95% uncertainty intervals. Elimination of exposure means no future exposure to this form of violence. Elimination of ICU means an immediate and lasting reduction of ICU associated with past exposure to the form of violence. The values used to construct the graphs and data for HIV prevalence and infections averted assuming no structural changes are made can be found in the appendix. FSW=female sex worker. ICU=inconsistent condom use.

Figure 4. HIV prevalence and percentage of infections potentially averted in FSWs and their clients in the next 10 years through structural changes in Mombasa, Kenya

(A) Predicted HIV prevalence in FSWs between 1985 and 2040. Squares show the empirical estimates (and 95% CIs) from the data; bold line shows the median (shaded area shows 95% CI) of the model predictions from the multivariate parameter fits. (B) Predicted fractions of new HIV infections that could be averted in FSWs and their clients from structural changes in 2014–24; vertical bars show 95% uncertainty intervals. Elimination of exposure means no future exposure to this form of violence. Elimination of ICU means an immediate and lasting reduction of ICU associated with past exposure to the form of violence. FSW=female sex worker. ICU=inconsistent condom use. ART=antiretroviral therapy. *Increase in CD4 eligibility according to WHO guidelines to less than 500 cells per μL. The values used to construct the graphs and data for HIV prevalence and infections averted assuming no structural changes are made can be found in the appendix.

Figure 5. HIV prevalence and percentage of infections potentially averted in female sex workers (FSWs) and their clients in the next 10 years through structural changes in Bellary, India

(A) Predicted HIV prevalence in FSWs between 1985 and 2040. Squares show the empirical estimates (and 95% CIs) from the data; bold line shows the median (shaded area shows 95% CI) of the model predictions from the multivariate parameter fits. (B) Predicted fractions of new HIV infections that could be averted in FSWs and their clients from structural changes in 2014– 24; vertical bars show 95% uncertainty intervals. Elimination of exposure means no future exposure to this form of violence. Elimination of ICU means an immediate and lasting reduction of ICU associated with past exposure to the form of violence. *Increase in CD4 eligibility according to WHO guidelines to less than 500 cells per μL. The values used to construct the graphs and data for HIV prevalence and infections averted assuming no structural changes are made can be found in the appendix. FSW=female sex worker. ICU=inconsistent condom use. ART=antiretroviral therapy.