Association of Ambient Temperature With the Prevalence of Intimate Partner Violence Among Partnered Women in Low- and Middle-Income South Asian Countries

Abstract

Importance: Intimate partner violence (IPV), including physical, sexual, and emotional violence, constitutes a critical public health problem, particularly in low- and middle-income countries. While climate change could escalate violent events, data quantifying its possible association with IPV are scant.

Objective: To evaluate the association of ambient temperature with the prevalence of IPV among partnered women in low- and middle-income countries in South Asia, and to estimate the association of future climate warming with IPV.

Design, setting, and participants: This cross-sectional study used data from the Demographic and Health Survey and included 194 871 ever-partnered women aged 15 to 49 years from 3 South Asian countries (India, Nepal, and Pakistan). The study applied the mixed-effect multivariable logistic regression model to investigate the association of ambient temperature with IPV prevalence. The study further modeled the change in IPV prevalence under various future climate change scenarios. The data included in the analyses were collected from October 1, 2010, to April 30, 2018, and the current analyses were performed from January 2, 2022, to July 11, 2022.

Exposure: Annual ambient temperature exposure for each woman, estimated based on an atmospheric reanalysis model of the global climate.

Main outcomes and measures: The prevalence of IPV and its types (physical, sexual, and emotional violence) were assessed based on self-reported questionnaires from October 1, 2010, to April 30, 2018, and the changes in the prevalence with climate changes were estimated through the 2090s.

Results: The study included 194 871 ever-partnered women aged 15 to 49 years (mean [SD] age, 35.4 [7.6] years; overall IPV prevalence, 27.0%) from 3 South Asian countries. The prevalence of physical violence was highest (23.0%), followed by emotional (12.5%), and sexual violence (9.5%). The annual temperature ranges were mostly between 20 °C and 30 °C. A significant association was found between high ambient temperature and the prevalence of IPV against women, with each 1 °C increase in the annual mean temperature associated with a mean increase in IPV prevalence of 4.49% (95% CI, 4.20%-4.78%). According to the study's projections under the unlimited emissions scenarios (SSPs [shared socioeconomic pathways], as defined by the Intergovernmental Panel on Climate Change] 5-8.5), IPV prevalence would increase by 21.0% by the end of the 21st century, while it would only moderately increase under increasingly stricter scenarios (SSP2-4.5 [9.8%] and SSP1-2.6 [5.8%]). In addition, the projected increases in the prevalence of physical (28.3%) and sexual (26.1%) violence were greater than that of emotional violence (8.9%). In the 2090s, India was estimated to experience the highest IPV prevalence increase (23.5%) among the 3 countries, compared with Nepal (14.8%) and Pakistan (5.9%).

Conclusions and relevance: This cross-sectional, multicountry study provides ample epidemiological evidence to support that high ambient temperature may be associated with the risk of IPV against women. These findings highlight the vulnerabilities and inequalities of women experiencing IPV in low- and middle-income countries in the context of global climate warming.

Figure 1.. Annual Mean Temperatures for the Studied Clusters in South Asia, 2010-2018

Clusters refer to groups of adjacent households, which serve as the primary sampling unit for field work efficiency in a 2-stage cluster sampling procedure in Demographic and Health Surveys.

Figure 2.. Percentage Change in Intimate Partner Violence (IPV) Prevalence per 1 °C Increase in Annual Mean Temperature in 3 South Asian Countries

Error bars represent 95% CIs.

Figure 3.. Effect Estimates for the Association Between Annual Mean Temperatures and Intimate Partner Violence (IPV) Prevalence in Various Subgroups

Low wealth level refers to the wealth level of the poorer or poorest individuals; high wealth level refers to individuals with middle wealth level or higher; not finished refers to natural, rudimentary, or other materials; not married refers to individuals living with a partner, widowed, divorced, or separated; young or old refers to women or their partner under and over the mean age; low educational level refers to individuals with primary education or lower; and high educational level refers to secondary education or higher. Error bars represent 95% CIs.

Figure 4.. Spatial Distributions of City-Level Percentage Changes in Intimate Partner Violence (IPV) Prevalence Associated With Climate Warming From the Baseline Period (1985-2014) in South Asia Under Different Climate Scenarios

The SSPs (shared socioeconomic pathways), as defined by the Intergovernmental Panel on Climate Change, correspond to the increasing trajectories of atmospheric greenhouse gas (GHG) concentrations and characterize a range of warming in global climate from mild to extreme; for this study, 3 SSPs were used, categorized as strict (SSP1-2.6; lowest projected change), medium (SSP2-4.5), and unrestricted (SSP5-8.5; highest projected change) GHG emissions.

Figure 5.. Estimated Percentage Changes in Prevalence of Intimate Partner Violence (IPV) and its Types (Physical, Sexual, and Emotional) Associated With Climate Warming From the Baseline Period (1985-2014) in 3 South Asian Countries Under Different Climate Scenarios

The SSPs (shared socioeconomic pathways), as defined by the Intergovernmental Panel on Climate Change, correspond to the increasing trajectories of atmospheric greenhouse gas (GHG) concentrations and characterize a range of warming in global climate from mild to extreme; for this study, 3 SSPs were used, categorized as strict (SSP1-2.6; lowest projected change), medium (SSP2-4.5), and unrestricted (SSP5-8.5; highest projected change) GHG emissions. Whiskers represent 95% empirical CIs.