Respiratory tract infections and diarrhoea are the two biggest killers of children in low income contexts. They are closely related to access to, and use of improved water, sanitation and hygiene (WASH). However, there is no high quality systematic review that quantifies the effect of WASH improvements on childhood mortality. Existing systematic reviews of WASH improvements measure effects on morbidity, under the (often implicit) assumption that morbidity is closely correlated with mortality. This is at least partly because the impact evaluations on which they are based are only designed to detect changes in morbidity with statistical precision, whereas mortality is a relatively rare outcome. The proposed review will address this evidence synthesis gap, using the greater statistical power of meta-analysis to pool findings across studies.