INADEQUATE TOILET use directly contributes to high rates of morbidity and mortality within India. Approximately, one fourth of the global population without access to adequate sanitation resides in India; and over 600 million people in India defecate in the open UNICEF, 2012). The combined lack of access and use decreases India’s GDP by at least 6.4 per cent annually (WSP2011; Chambers and Von Medeazza 2013). Despite a strong research focus on sanitation interventions, understanding of the determinants of inadequate sanitation remains poor. This paper uses spatial analysis techniques to explain variation in toilet use at the district scale across India. A test of global spatial auto-correlation confirms that sanitation use is strongly clustered at the district level. Tests of local spatial auto correlation also reveal that poor sanitation is clustered in districts that have high under-five mortality rates. A spatial lag model is employed to demonstrate the persistence of spatial clustering even after accounting for standard socioeconomic and institutional correlates of toilet use.The location of these clusters and the explanatory power at district level appear to reflect the mechanisms that are not captured by state-level analysis.