Rainfall and water-related diseases, malnutrition and mortality in Low- and Middle- Income Countries: a systematic review of the epidemiological evidence

<h2>Abstract</h2><h3>Background</h3> Climate change is altering rainfall patterns. Rainfall has been linked to numerous health outcomes, through the impacts on water quality and quantity, but the coherence and strength of evidence across outcomes remain unclear. <h3>Objectives</h3> Understand and evaluate the strength of evidence on associations between rainfall (both low and heavy events) and health outcomes in Low- and Middle- Income Countries (LMICs). <h3>Methods</h3> A systematic review of peer-reviewed epidemiological studies quantifying associations between rainfall and human health outcomes in LMIC populations was conducted. Seven databases were searched including MEDLINE and EMBASE. Study quality was evaluated using 9 modified criteria that were previously used to assess environmental epidemiology studies. The strength of evidence for each health outcome was assessed across rainfall exposures. <h3>Results</h3> Of 23,579 papers identified, 177 met the inclusion criteria. Health outcomes included diarrheal diseases (n = 119); malnutrition (n = 35); mortality (n = 21); helminth infections (n = 6), and eye infections (n = 4). There was moderately strong evidence for positive associations between both heavy and low rainfall and all-cause diarrhea. Evidence for undernutrition was mixed, with moderate evidence of a positive association with low rainfall. Despite sharing causal pathways, diarrheal disease and nutrition studies found contrasting results for heavy rainfall, likely due to differing study designs. Studies were heterogenous in design, rainfall exposure definitions, and lag times. Studies also often lacked a clear hypothesis. <h3>Discussion</h3> There is substantial evidence that rainfall affects health in LMICs through multiple pathways. Limitations in the data (often from cross-sectional surveys) and study designs, limit the strength of evidence for several health outcomes. Specifically, studies frequently used inappropriate exposures or lags to reflect the causal pathways. In future studies, efforts should be directed to incorporating pathways in models and better defining low rainfall exposures. Conceptualising the interaction between rainfall and inter-related health outcomes can also strengthen future research.