Community Intervention with UV disinfection to estimate acute gastrointestinal illness from drinking untreated groundwater

Authors: Borchardt, Mark; KIEKE, BURNEYA - Marshfield Clinic Research; Spencer, Susan; LAMBERTINI, ELISABETTA - Rti International, Usa; LOGE, FRANK - University Of California, Davis

Submitted to: Symposium Proceedings
Publication Type: Abstract Only
Publication Acceptance Date: 6/8/2019
Publication Date: N/A
Citation: N/A

Interpretive Summary:

Technical Abstract: Approximately 85,000 public water systems in the United States are supplied by untreated groundwater. Illness outbreaks in the USA have been linked to drinking untreated groundwater. However, its role in endemic illness is not fully understood. Our team made previous reports on endemic acute gastrointestinal illness (AGI) in 14 untreated groundwater-supplied communities as related to: 1) contamination events within distribution systems; and 2) concentrations of enteric viruses at household taps. Here we estimate the fraction of endemic AGI attributable to their groundwater supplies. We conducted a community-wide UV disinfection intervention trial. UV disinfection (50 mJ/cm2) was installed on all supply wells of intervention communities, whereas in control communities participants continued to drink untreated water. AGI incidence was estimated from households (621, with 1,659 study subjects) completing weekly health diaries during four 12-week surveillance periods. Between the second and third surveillance periods UV disinfection units were switched from intervention to control communities as a means of accounting for intrinsic differences in AGI exposures among the communities (i.e., crossover design). Well water samples were collected monthly and analyzed for viruses by qPCR. Calculating the difference in AGI incidence between control and intervention periods within the same community (the crossover design) there was no evidence for groundwater-borne AGI. However, the types and concentrations of viruses in the study wells varied greatly over time, suggesting AGI risk from untreated groundwater was not constant. Directly comparing intervention and control communities by surveillance period (ignoring crossover), when norovirus genogroup I was present in the wells the fraction of AGI from untreated groundwater was 21% (95% CI 1-36%) for children < 5 years old and 18% (95% CI -9-38%) for adults. Enteric virus contamination in groundwater is temporally variable. However, when virus contamination does occur, untreated groundwater consumption can be responsible for a measureable fraction of AGI.