Climate change is projected to increase the frequency and intensity of droughts, posing a significant threat to public health, particularly in LMICs. Diarrheal diseases are a leading cause of death among children under five, and their incidence is exacerbated by poor WASH conditions. While previous studies have explored the relationship between rainfall and diarrhea, the association with more complex drought measures remains unclear. This study aims to quantify the association between drought, measured using SPEI, and diarrhea risk among children under five in LMICs, while considering the mediating and modifying roles of WASH practices and other factors. The study's importance lies in its potential to inform interventions aimed at mitigating the impact of climate change on child health in vulnerable populations.

Existing literature shows mixed results regarding the relationship between rainfall and diarrhea risk, with some studies reporting positive associations with low rainfall and others finding associations with high or moderate rainfall. The complexity arises from the multifaceted nature of drought, which involves not only low rainfall but also high temperatures and evapotranspiration. The lack of consensus on drought definition and sparse epidemiological evidence using complex drought measures hinders a comprehensive understanding of the relationship between drought and diarrheal diseases. Previous studies have highlighted the importance of WASH practices in reducing diarrhea risk. This study addresses the gap by using a more advanced drought measure (SPEI) and exploring the mediating and modifying effects of WASH practices.

The study utilized data from 141 nationally representative Demographic and Health Surveys (DHS) conducted in 51 LMICs between 1990 and 2019. Drought exposure was measured using the SPEI, a multiscalar drought index that accounts for both precipitation and evapotranspiration. The researchers used log-binomial generalized linear mixed models to examine the association between diarrhea risk (children with diarrhea in the two weeks preceding the interview) and drought exposure at different timescales (6, 12, 18, and 24 months). The models adjusted for various individual and household characteristics, including WASH practices (time to collect water, water and soap availability at handwashing sites), sex, age, mother's education, residence, wealth quintile, breastfeeding status, and climate zone. Mediation analysis was conducted to assess the extent to which WASH practices mediated the association between drought and diarrhea. Effect modification analysis explored how various factors might modify the relationship between drought and diarrhea risk. Sensitivity analyses were conducted to address potential biases due to missing data and model specifications. Multiple imputation techniques were used to handle missing data.

Exposure to 6-month mild or severe drought was associated with a 5% (95% CI: 3–7%) or 8% (95% CI: 5–11%) increased diarrhea risk, respectively. The association with mild drought was consistent across all timescales. The association was stronger among children in households with longer water collection times and lacking access to water or soap/detergent for handwashing. The association for 24-month drought was stronger in dry zones but weaker or null in tropical or temperate zones, while the 6-month drought association was only observed in tropical or temperate zones. Mediation analysis suggested that a low to moderate proportion (11.6% for 6-month mild drought) of the association between drought and diarrhea risk was mediated by WASH variables. Effect modification analysis revealed significant interactions between drought and climate zone, water collection time, and water/soap availability at handwashing sites. In dry zones, longer-term droughts (12 and 24 months) were strongly associated with diarrhea risk, suggesting limited capacity to adapt to persistent drought conditions. In contrast, tropical and temperate zones showed stronger associations with shorter-term droughts.

The findings highlight the significant association between long-term drought and increased diarrhea risk among children in LMICs. The use of SPEI as a drought indicator provided a more nuanced understanding compared to previous studies relying solely on rainfall data. The study’s findings underscore the vulnerability of children in households with inadequate WASH practices. The varying associations across climate zones suggest the need for context-specific interventions. The partial mediation effect of WASH practices indicates that improved WASH alone may not fully mitigate the impact of drought. Addressing drought's impact on water quantity and quality, food safety and security, and migration patterns are crucial for reducing diarrhea risk.

This study provides robust evidence of the association between long-term drought and increased diarrhea risk among children under five in LMICs, particularly in settings with poor WASH conditions. Improved WASH infrastructure and practices are critical, but broader adaptation strategies addressing water scarcity and food security are needed. Future research should explore the role of other factors like food safety, migration, and water supply interruption patterns in exacerbating diarrhea risk under drought conditions.

The study acknowledges limitations such as the potential for mismatching exposure data with survey clusters due to anonymization techniques. The reliance on maternal reports for diarrhea cases might introduce recall bias. The use of reduced sample sizes for mediation and effect modification analyses due to missing data introduces potential selection bias. Multiple comparison issues may affect effect modification results. Lack of laboratory-based diagnosis of diarrhea cases limits the ability to analyze associations by pathogen type. The inability to disentangle the complex interplay of drought with food security, migration, and other factors also limits the study's scope.