Quantitative microbiological risk assessment of a rural intermittent water supply in Shandong

Background Intermittent water supply (IWS) is widely used in rural China, and related studies have shown that IWS increases microbial contamination of tap water and increases health risks for the population in coverage.

Objective This study aims to understand the current situation of microbial pollution of a typical IWS in rural China, and to quantitatively evaluate its health risks.

Methods A typical rural IWS in Zibo, Shandong was selected as a study subject, and a continuous water supply (CWS) with similar water source type and water treatment was selected as a control. Finished water samples (1 water sample from each water plant), tap water samples (20 sampling points for each water plant; for IWS, 1 tap water sample was collected at the moment of supplying water, 5 min after water supply restart, and 30 min after water supply restart, respectively; for CWS, 1 tap water sample was collected at the moment of supplying water and 5 min after water supply, respectively), and household stored water samples (only for IWS, 10 sampling pionts, 1 sample from each point) were collected and tested for microbiological parameters, and the results were evaluated according to the Standards for drinking water quality (GB 5749—2006) and compared among the three types of water samples and between the two water supply systems. Total coliform (TC) was used as an indicator of pathogenic microorganisms, and quantitative microbial risk assessment (QMRA) method was adopted to evaluate the annual infection probability from piped water for the IWS and CWS serviced populations. Monte-Carlo simulations were used for uncertainty analysis.

Results For the finished water samples from the two water supply systems, the testing results of TC and total bacteria (TB) met the national standard. The unqualified rates of TB and TC for IWS tap water samples were 48.3% (29/60) and 23.3% (14/60), the rates for household stored water samples were 80.0% (8/10) and 50.0% (5/10), and the rates for CWS tap water samples were 40.0% (16/40) and 20.0% (8/40), respectively. The TC count of IWS tap water samples at the moment of supplying water (logarithm median, 2.11 lg CFU·100 mL^-1) was higher than that of corresponding CWS tap water samples (logarithm median, 0.30 lg CFU·100 mL^-1) (P < 0.05); the TB count of IWS tap water samples at 30 min after water supply restart (logarithm mean, 2.04 lg CFU·mL^-1) was higher than that of CWS water samples at 5 min after water supply restart (logarithm mean, 1.62 lg CFU·mL^-1) (P < 0.05). The TB count of IWS household stored water samples (logarithm mean, 3.20 lg CFU·mL^-1) was higher than that of tap water samples from IWS and CWS (P < 0.05); the TC count of IWS household stored water samples (logarithm median, 1.52 lg CFU·100 mL^-1) was higher than that of tap water samples at 30 min after water supply restart (P < 0.05). The Monte-Carlo simulation results showed that the annual infection probability M (P₅-P₉₅) for IWS was 47.67×10^-4 (0-1392.46×10^-4), and that for CWS was 4.85×10^-4 (0-182.37×10^-4) (P < 0.05).

Conclusion Compared with CWS, IWS increases the risk of microbial exposure and infection for the populations in service in the selected area.