Can a national government implement a violence prevention and response strategy for key populations in a criminalized setting? A case study from Kenya

Abstract

IntroductionKey population (KP) members frequently experience violence that violates their human rights, increases their risk of HIV, and acts as a barrier to access and uptake of HIV services. To be effective, HIV programmes for members of KPs need to prevent and respond to violence against them. We describe a violence prevention and response strategy led by the national KP programme in Kenya and examine trends in reports of and responses to violence (provision of support to an individual who reports violence within 24 hours of receiving the report).MethodsQuarterly programme monitoring data on the number of reports of violence and the number of responses to violence from 81 implementing partners between October 2013 and September 2017 were aggregated annually and analysed using simple trend analysis. Reports of violence relative to KP members reached, expressed as a percentage, and the percentage of reports of violence that received a response were also examined.Results and DiscussionBetween 2013 and 2017, annual reports of violence increased from 4171 to 13,496 cases among female sex workers (FSWs), 910 to 1122 cases among men who have sex with men (MSM) and 121 to 873 cases among people who inject drugs (PWID). Reports of violence relative to KP members reached increased among FSWs (6.2% to 9.7%; p < 0.001) and PWID (2.1% to 6.0%; p < 0.001) and decreased among MSM (10.0% to 4.2%; p < 0.001). During the same period, timely responses to reports of violence increased from 53% to 84% (p < 0.001) among FSWs, 44% to 80% (p < 0.001) among MSM and 37% to 97% (p < 0.001) among PWID.ConclusionsOver the past four years in Kenya, there has been an increase in violence reporting among FSWs and PWID and an increase in violence response among all KPs. This case study demonstrates that violence against KP members can be effectively addressed under the leadership of the national government, even in an environment where KP members’ behaviours are criminalized. Creating an enabling environment to promote wellbeing and safety for KP members is a critical enabler for HIV prevention programmes to achieve 95‐95‐95 goals.

Building the evidence base to optimize the impact of key population programming across the HIV cascade

“…Bhattacharjee et al . describe successful efforts to integrate violence prevention and response services into the national key population programme in Kenya . Drawing on programme data over a four‐year period, this paper contributes important evidence that it is possible to address violence against key populations under the leadership of the national government, even in an environment where sex work, same‐sex sexual practices and drug use are criminalized .…”

Sexual health among Kenyan male sex workers in a time of COVID-19

“…Those requiring crisis response for severe mental health problems face many challenges. HOYMAS' rescue centre, staffed by trained paralegal workers (Bhattacharjee et al, 2018a), is no longer accessible due to COVID-19. Crisis response mechanisms have been crippled by mandatory curfews; lacking the status of an essential health worker, community paralegals can no longer move about at night to respond to violence.…”

Assessing an inclusive model to increase access to comprehensive gender-based violence response services and improve HIV outcomes in Puerto Plata, Dominican Republic

“…Increases in violence disclosure can suggest increased confidence in a functional and confidential GBV response system (Bhattacharjee et al 2018). Beyond this, it is difficult to interpret incidence of violence or compare populations' experiences using service statistics, especially because only seven peer educators were employed.…”

“…Making data-driven course corrections between phasessuch as expanding the geographic area of implementation for expanded impact and changing the data collection form for more feasible implementationlimits the inferences that can be made. Using only routinely collected data instead of populationbased surveyssuch as the polling booth surveys used in Kenya (Bhattacharjee et al 2018)also means that it is not possible to know whether the model affected the incidence of violence. Furthermore, because there was no baseline measure of quality of GBV response services, nor was the number of reports of violence to CEPROSH prior to the intervention known, providers and service users' observations cannot be compared with pre-intervention measures.…”