Mixed-method longitudinal investigation of sexual and gender-based violence following COVID-19 in South Africa.

Hartmann M, Stoner MCD, Storey S, Giovenco D, Zondi YZ, Qwabe N, Ekström AM, Pettifor AE, Bekker LG, Kågesten A

BMJ Public Health 3 (1) e001697 [2025-01-00; online 2025-04-16]

Throughout the COVID-19 pandemic, concerns were raised about unintended effects of measures taken to prevent its spread, on sexual and gender-based violence (SGBV). The United Nations called for understanding how national lockdowns put young people at risk of SGBV. This research is particularly needed in contexts such as South Africa, where pre-existing levels of SGBV are high and limited data has been released. This mixed-method longitudinal study characterised trajectories of household-level and partner-level SGBV exposure over 6 months, approximately 1 year after the initial COVID-19 lockdown. Utilising group-based trajectory modelling, survey data from 535 male and female participants, ages 13-24 and qualitative insights from 20 in-depth interviews were analysed. Two trajectory groups emerged for both household-level and partner-level SGBV: (1) groups of participants with consistently low SGBV levels (household: 77.5%; partner: 89.4%) and (2) groups with high baseline levels of SGBV, followed by decreases to moderate levels (household: 22.5%; partner: 10.8%). Characteristics significantly associated with the latter groups included being female, not employed or in school, food insecurity and symptoms of probable common mental disorders. Qualitative data supported these findings and revealed the mitigating role of positive household communication skills, along with potentially unmeasured levels of technology-facilitated partner violence, occurring over phones and social media during lockdown. Findings should inform the targeting of financial, food and mental health support to those at higher risk of ongoing violence during future times of crises. Further research on technology-facilitated violence should be conducted to better understand its prevalence.

PubMed 40260124

DOI 10.1136/bmjph-2024-001697

Crossref 10.1136/bmjph-2024-001697

pmc: PMC12010298
pii: bmjph-2024-001697