Individual-level factors associated with self-reported gender-based violence among Kenya Medical Training College students

Abstract

Gender-based violence and lack of institutional protection contribute to the potential for abuse in higher education, which is a global public-health problem and highly underreported. This study was guided by the Ecological Model of Violence and the theory of revictimization and examined individual determinants of self-reported gender-based violence among students at the Kenya Medical Training College where the clinical hierarchy and professional pressures amplify students' vulnerabilities. An analytical cross sectional research design with a predictive modelling component was used for the study with the target population being pre-service health-training students age 18 years old and above. A multistage sampling technique was adopted in which campus was selected purposively, years of study and programme type were stratified, and systematic random sampling of the respondents was done, and data was collected from 478 respondents. The main data collection tool was an anonymous, structured online questionnaire that included demographic, behaviourally specific violence experiences, past victimisation experiences, and mental-health indicators. The quantitative data were analysed by descriptive statistics, chi-square tests and multivariate logistic regression. The findings showed that 15.1% of the students surveyed suffered gender-based violence in the form of any of the behaviourally specific types during enrolment and 10.0% of them suffered gender-based violence in the past 12 months. Descriptive analysis revealed that the most common types of abuse experienced by recent victims were unwanted sexual comments (56.3%), emotional/verbal abuse (54.2%) and non-consensual sexual touching or stalking (47.9%). Academic progression (Y3 vs Y1: aOR=5.31, p<0.001), childhood sexual abuse (aOR=5.65, p<0.001) and severe mental-health symptoms (aOR=5.50, p=0.004) were significant independent predictors of violence occurrence. Measurement analysis indicated that 0.0% of the students endorsed a general violence question while 15.1% endorsed behaviourally specific items, indicating a significant conceptual gap and lack of practical recognition of abuse. The study finds that vulnerabilities to gender-based violence in these settings are cumulative and structurally embedded, and are related to past trauma experiences, seniority in school and psychosocial distress. It shows that there is a need for trauma-informed, institution-specific strategies to support the welfare of students, and that a combination of behaviourally specific literacy and confidential mental-health screening and peer-mentorship referral pathways are needed for sustainable student resilience.