Examining the influence of urban definition when assessing relative safety of drinking-water in Nigeria

Reducing inequalities is a priority from a human rights perspective and in water and public health initiatives. There are periodic calls for differential national and global standards for rural and urban areas, often justified by the suggestion that, for a given water source type, safety is worse in urban areas. For instance, initially proposed post-2015 water targets included classifying urban but not rural protected dug wells as unimproved. The objectives of this study were to: (i) examine the influence of urban extent definition on water safety in Nigeria, (ii) compare the frequency of thermotolerant coliform (TTC) contamination and prevalence of sanitary risks between rural and urban water sources of a given type and (iii) investigate differences in exposure to contaminated drinking-water in rural and urban areas. We use spatially referenced data from a Nigerian national randomized sample survey of five improved water source types to assess the extent of any disparities in urban–rural safety. We combined the survey data on TTC and sanitary risk with map layers depicting urban versus rural areas according to eight urban definitions. When examining water safety separately for each improved source type, we found no significant urban–rural differences in TTC contamination and sanitary risk for groundwater sources (boreholes and protected dug wells) and inconclusive findings for piped water and stored water. However, when improved and unimproved source types were combined, TTC contamination was 1.6 to 2.3 times more likely in rural compared to urban water sources depending on the urban definition. Our results suggest that different targets for urban and rural water safety are not justified and that rural dwellers are more exposed to unsafe water than urban dwellers. Additionally, urban–rural analyses should assess multiple definitions or indicators of urban to assess robustness of findings and to characterize a gradient that disaggregates the urban–rural dichotomy. - Highlights: • Urban improved sources of a given type are no more contaminated than rural ones. • Fecal contamination is 1.6–2.3 times more likely in rural areas, combining sources. • We find no evidence to justify different urban and rural water quality standards. • Comparisons of urban and rural areas are sensitive to definitions of urban extent