Protocol for a discrete choice experiment: understanding preferences for seeking health services for survivors of sexual violence in Uganda

• Published in: BMJ open 2024-08, Vol.14 (8), p.e081663
• Main Authors: Stark, Lindsay, Mutumba, M, Ssewamala, Fred, Brathwaite, Rachel, Brown, Derek S, Atwebembere, Raymond, Mwebembezi, Abel
• Format: Article
• Language: English
• Subjects: Adult; Behavior; Birth control; Choice Behavior; Decision making; Experiments; Female; Females; Health Services; Health services utilization; Humans; Mental health; Patient Acceptance of Health Care - psychology; Patient Acceptance of Health Care - statistics & numerical data; Patient Preference; Preferences; Protocol; Qualitative Research; Research Design; Self help; Sex Offenses - psychology; Sexual Health; SEXUAL MEDICINE; Sexually transmitted diseases; STD; Survivor; Survivors - psychology; Uganda; Violence; Womens health
• ISSN: 2044-6055; 2044-6055
• DOI: 10.1136/bmjopen-2023-081663

IntroductionSexual violence is a significant public health concern with severe physical, social and psychological consequences, which can be mitigated by health service utilisation. However, in Uganda and much of sub-Saharan Africa, these services are significantly underused, with 9 out of 10 survivors not seeking care due to a range of psychological, cultural, economic and logistical factors. Thus, there is a strong need for research to improve health service utilisation for survivors of sexual violence.Methods and analysisThe proposed study seeks to address the underutilization of health services for female survivors of sexual violence using a discrete choice experiment (DCE). The study will be conducted in the greater Masaka region of southwestern Uganda and target adult female survivors of sexual violence. We will first undertake qualitative interviews with 56 survivors of sexual violence to identify the key attributes and levels of the DCE. In order to ensure a sufficiently powered sample, 312 women who meet inclusion criteria will be interviewed. Our primary analysis will employ a mixed (random parameters) logit model. We will also model the role of individual-specific characteristics through latent class models.Ethics and disseminationThe study protocol was reviewed and approved by the following ethics review boards in Uganda and the USA: the Uganda Virus Research Institute (UVRI), the Uganda National Council for Science and Technology (HS2364ES), Washington University in St Louis and the University of Michigan. Our methods conform to established guidelines for the protection of human subjects involved in research. Our dissemination plan targets a broad audience, ranging from policymakers and government agencies to healthcare providers, academic communities and survivors themselves.