Investigating temporary relocation of Female Sex Workers: Implications for Health, Care, and Trials

What is the context for this research?
HIV is a leading contributor to the global burden of diseases. Ambitious targets to tackle HIV will only be met when the health needs of female sex workers (FSW) are addressed with effective interventions. FSW in low income countries face a high burden of HIV as well as social and institutional exclusion. Although there are promising advances in the use of antiretroviral therapy (ART) to reduce HIV transmission, proven interventions to reach and support FSW are not available. Many current or proposed interventions to increase ART for FSW rely on community activities and regular contact. However, many FSW frequently move for work from their main place of work. This temporary relocation (TR) of FSW is under researched, despite potentially having major health and healthcare implications and being informative for intervention strategy. Furthermore, TR can cover long distances and connect disparate places. Many intervention evaluation methodologies, including cluster randomized trials (CRT), assume that places are independent from one another, especially when far apart: understanding the implications of TR for this assumption will improve evaluations of public health interventions.

To what disease/conditions is this research relevant?
The investigation of TR of FSW will have particular relevance to sexual health research, including HIV. It will also contribute to the conditions associated with violence against women and mental health, and other conditions that are targeted with interventions delivered at the community or cluster level.

What is the research trying to achieve?
The overall aim is to improve the health and well-being of FSW by investigating the implications of TR for burden of disease, access to healthcare, and intervention design and evaluation. This will be achieved through three objectives: (1) quantifying TR and exploring contributory factors; (2) analysing the association between TR and health outcomes such as HIV status and access to interventions such as ART; (3) investigating how TR might introduce bias into evaluations of interventions.

Who is carrying out the research?
I, Calum Davey, will conduct this research as a PhD at the London School of Hygiene and Tropical Medicine with supervision from Dr James Hargreaves, who is an expert in HIV prevention research and the Director of the Centre for Evaluation. I will receive advisory support from Dr Frances Cowan, the director of the Centre for Sexual Heath and HIV/AIDS Research in Zimbabwe (CeSHHAR), and Professor Richard Hayes. The Karnataka Health Promotion Trust (KHPT) has collected data in India, as has CeSHHAR in Zimbabwe.

How is the research being conducted?
(1) KHPT and CeSHHAR will provide cross-sectional biological and behavioural data on 14,450 FSW in multiple sites in Karnataka and Zimbabwe. The dataset will include questions on TR in the last 12 months and the names of the places traveled. These place names will be located on a geographical information system (GIS) which will be used to estimate the extent and variation of TR.
(2) Statistical methods will be used to explore the causes of TR.
(3) Further analysis will explore how TR affects the probability of a woman being HIV positive, and the extent of access to services.
(4) Data generated from steps 1-3 will be incorporated into a mathematical model of a FSW population: i.e. a simulation that will be based on parameters such as time FSW spend in each location, associated service use, etc.
(5) The model FSW population will be used to explore how TR affects the risk of bias in CRT by simulating the data that would be collected during a CRT under different FSW population situations.
(6) The results of the research will be published in peer-reviewed journals, and disseminated through workshops, conference presentations, blogs, seminars, and press releases.

Background
In many low-income settings the burden of HIV falls disproportionately on female sex workers (FSW). Recent innovations in interventions for FSW use community-level approaches to tackle HIV and related issues. Interventions that improve treatment access and adherence among FSW may contribute to ART coverage targets, and may reduce incident infections. While migrant status and the movement of clients have been studied, the phenomenon of relocation for work temporarily (TR) has not been researched. TR may have implications for community-based interventions, social cohesion, adherence support, and evaluations of intervention effects.

Purpose
To investigate the causes and implications of TR among FSW living in Karnataka and Zimabwe between 2006 and 2015, and implications for intervention planning. To improve evaluations of interventions for FSW, and pragmatic cluster randomized trials (CRT), by investigating the implications of TR on CRT.

Methods
(1) KHPT collected three rounds of cross-sectional survey data from FSW between 2006 and 20011. CeSHHAR collected cross-sectional data from 2,800 FSW in 2013, and will collect data on approximately 5,000 more in 2015.
(2) The names of places worked other than the sampled location will be geo-located using GIS software ARC GIS, with assistance from local researchers in Karnataka and Zimbabwe.
(3) Potential causes of TR will be explored using multilevel modelling and pathway analysis. HIV and ART-use implications of TR will be analysed similarly.
(4) A model FSW population will be simulated in R. Parameters will be derived from data. CRT will be simulated with a range of parameters, including: number and distance between clusters, extent of TR, typology of FSW.

Scientific developments
Quantitative characterization of TR in FSW in variety of settings; implications of TR for intervention planning; risks associated with TR; estimation of risk of bias in impact evaluations.

The outputs of this research will benefit the following groups:
1. Those designing and implementing interventions for female sex workers (FSW)
This research will produce quantitative metrics of temporary relocation (TR) of FSW, and the variation over a number of settings and typologies. Furthermore, this research will explore how TR influences service access and vulnerability to HIV and other health outcomes. This new data will enable organizations implementing interventions for FSW to better tailor designs to FSWs' needs, and to mitigate against the effects of TR. In particular, this rigorous analysis of TR will provide quantitative characterization of a phenomenon that is well known among implementing organizations and often discussed in the context of community mobilization interventions, and initiatives to improve antiretroviral therapy (ART) access and adherence. KHPT and CeSHHAR are world leading in this field and deliver programmes for tens of thousands of FSW. Both are the primary service delivery organization for FSW in their respective settings. This research will have immediate relevance for their work and through them influence best practice in other parts of the world. Implementing organizations will also benefit from fairer assessment of their efforts though improved evaluation design as a consequence of this research.

2. Those who design evaluations of public health interventions
This research will appraise the appropriateness and the limitations of the cluster randomized trial (CRT) design for evaluating interventions for FSW, in particular whether or not there is vulnerability to bias under certain design conditions. This output will benefit researchers and evaluation practitioners who want to know 'what works' for tackling HIV and other health concerns of FSW, and who are making decisions about how to measure unbiased estimates of effect. This research will also benefit evaluation officers who are conducting pragmatic impact evaluations by providing a 'case study' and tools to tackle the measurement of TR and calculating study power.

3. Those who interpret results of trials of interventions for FSW
This research will contribute to the appraisal of the susceptibility to bias in trials and other evaluations of interventions for FSW, and other evaluations where there is a higher than optimal risk of contamination. This will have direct relevance for researchers reviewing and synthesising evidence, such as Cochrane Collaboration, the Campbell Collaboration, WHO, or independent researchers. These groups themselves influence policy and guidelines.

4. FSW and others who will benefit from better healthcare interventions
Through impact on better design, better evaluation, and better synthesis, this research has the potential to improve the lives of FSW, other minority populations, and the wider population in countries where sexual networks involving FSW contribute significantly to the continuation of the HIV epidemic. This research and associated communications will increase awareness of the nuances if the lives of FSW, and the broader challenge of engaging with FSW as a varied population of women with a particular kind of work with associated pressures and issues.

5. Policy makers, modellers, bodies who estimate cost effectiveness
As a consequence of the impacts previously outlined, this research will have important influence on policy makers. Furthermore, this research will contribute to mathematical models that estimate effectiveness of interventions when rolled out, and make recommendations for cost-effectiveness. This will come at an crucial moment for agenda setting as HIV funding as a proportion of global health spending is reduced.