2.2 billion people lack access to safely managed drinking water, with microbiological water quality the major barrier to the achievement of SDG target 6.1. Contamination from the catchment is a well-established pathway for water quality degradation. We focus on a different pathway: microbiological contamination of drinking water from local environmental hygiene, where contamination enters the water due to human and animal interaction with the water at the point of collection or the household. Through analysis of two secondary data types we demonstrate significant contribution of local hygiene to drinking water quality. Firstly, we compared paired water quality samples from 5,080 water systems: a sample before cleaning the tap/spout, showing the ‘point of collection’ (PoC) water quality; and a sample after cleaning, showing the ‘point of delivery’ (PoD) water quality, i.e., the quality of water within the system. In 50% of the cases with E. coli-positive PoC results, the paired PoD results were negative; therefore, E. coli at the PoC is attributed to contamination from local hygiene-related pathways. Secondly, using 38 Multiple Indicator Cluster Survey (MICS) datasets, we analysed cases of households collecting PoU water direct from an E. coli-negative PoC to assess the impact of hygiene within the household. Household hygiene increased the proportion of households facing contamination by 5% to 70%. Better reporting of PoC and PoD methods is required to recognise this contribution from poor hygiene and ensure data are interpreted appropriately, thereby enabling more efficient targeting and prioritisation of interventions. Further research is needed to inform sampling protocols and management approaches to better understand seasonal variability, the role of biofilm growth, and interactions with chlorination practices particularly in piped water systems. There is immediate benefit to be gained from expanding communication and hygiene education to encourage cleaning of taps for drinking water by service providers and users.
