Title: Is intimate partner violence decreasing? Analysis of population-level trends in 21 countries

IntroductionAround the world, 193 countries have pledged to eliminate violence against women and girls by 2030, the most common form of which is intimate partner violence (IPV). To verify if progress is being made, we examined trends in prevalence of IPV against women and girls across 21 countries with three or more Demographic and Health Surveys. Additionally, we examine whether trends differ by type, frequency and severity of IPV; and whether changes are concentrated in specific sub-groups of women and girls. MethodsWe pooled individual-level data from within-country surveys conducted over time and harmonised outcomes to ensure comparability. We estimated weighted crude prevalence ratios and used log binomial regression models to calculate age adjusted risk ratios and predict annual age adjusted prevalence of 12-month physical and/or sexual IPV and severe IPV against women and girls aged 15-49, overall and by age group, education, wealth and place of residence. We used an ordered logit model to estimate the frequency of 12-month physical and/or sexual IPV. ResultsWe found no consistent cross-country trends over time in IPV prevalence: five countries showed declines, 15 demonstrated mixed directions or no changes, and one a consistent increase. In most countries, levels of frequent or severe IPV changed in the same way as overall prevalence, but four countries with an overall decline showed an increase in the proportion of severe IPV. Changes were not systematically concentrated in younger, wealthier, or more educated women and girls. ConclusionProgress towards reducing IPV against women and girls cannot be assumed. Overall, less than a quarter of examined countries showed consistent declines over approximately a decade prior to the most recent data collection. Further efforts leveraging evidence-based and locally adapted, scalable interventions and policies are urgently needed to accelerate prevention and to respond to IPV against women and girls. Key messagesO_ST_ABSWhat is already known on this topic?C_ST_ABSAround the world, 193 countries have pledged to eliminate violence against women and girls by 2030. A recent paper has concluded that prevalence of IPV is declining in low- and middle-income countries, with an average annual rate of change of -0.2% (95% CI -0.4 to - 0.03) (https://www.thelancet.com/journals/langlo/article/PIIS2214-109X(23)00417-5/fulltext). However, this recent paper pooled multiple survey datasets from 21 countries, across only two time points, despite more surveys being available, which limits the ability to identify overall trends. In addition, the study does not ensure consistent items are included across surveys. This leaves open the possibility that the declines observed are due to methodological differences between surveys and/or are not consistent within countries. What this study addsWe address many of the methodological concerns raised from the previous analysis, including the standardisation of outcomes, and the inclusion of countries which have at least three measurement points--thus more confidently identifying trends, typically over a decade or longer. This leads us to a different overall conclusion. We find that the global prevalence of physical and/or sexual IPV against women and girls is not systematically declining across countries. Overall, less than a quarter of examined countries showed consistent declines in physical and/or sexual IPV over time, namely in Colombia, Kenya, Philippines, Tanzania, Uganda. Further, we find that in Zambia, Kenya, Nigeria and Zimbabwe, as overall physical IPV prevalence reduces, the proportion of women and girls who experience severe physical IPV increases. How this study might affect research, practice or policyOur results suggest that further efforts leveraging evidence-based and scalable interventions and policies are urgently needed to accelerate prevention and response to IPV against women and girls. Second, careful monitoring is also needed to ensure gains are sustained equally across vulnerable groups. Finally, sustaining and strengthening national data collection architecture is critical to enable continued progress towards elimination of IPV against women and girls.