RESUMO
Objective To assess if water from improved sources are microbiologically safe in Niassa province, Mozambique, by examining the presence of total coliforms in different types of water sources. Methods We conducted a cross-sectional household survey in two rural districts of Niassa province during the dry season, from 21 August to 4 October 2019. We observed water sources and conducted microbiological water quality tests and structured household interviews. Findings We included 1313 households, of which 812 (61.8%) used water from an improved source. There was no significant difference in presence of total coliforms between water sampled at improved and unimproved water sources, 62.7% (509 samples) and 65.7% (329 samples), respectively (P-value=0.267). Households using improved water sources spent significantly longer time collecting water (59.1 minutes; standard deviation, SD: 55.2) than households using unimproved sources (49.8 minutes; SD: 58.0; P-value <0.001). A smaller proportion of households using improved sources had access to water sources available 24 hours per day than that of households using unimproved sources, 71.7% (582 households) versus 94.2% (472 households; P-value <0.001). Of the 240 households treating water collected from improved sources, 204 (85.4%) had total coliforms in their water, while treated water from 77 of 107 (72.0%) households collecting water from an unimproved source were contaminated. Conclusion Current access to an improved water source does not ensure microbiological safety of water and thereby using access as the proxy indicator for safe drinking and cooking water is questionable. Poor quality of water calls for the need for integration of water quality assessment into regular monitoring programmes.