Health and water supply: Implementation of quantitative microbial risk assessment

Safe water supplies are fundamental to public health protection. The assessment and control of pathogen risk is of primary concern, as a large disease burden is prevented. Due to the insensitivity of laboratory techniques and public health surveillance, quantitative microbial risk assessment (QMRA) is used to quantify public health risk associated with water supply. This thesis aims to establish new evidence to inform QMRA policy and practice. It does this from the perspective of a practitioner-researcher. This thesis comprises three analytical studies. Each answers a research question arising in the professional practice of the researcher. The first study uses the systematic review method to synthesise usage patterns of QMRA for drinking water supplies globally. The second uses interrupted timeseries analysis to examine an intervention into the sensitivity of protozoal enumeration results used in QMRA. The final study uses stochastic QMRA modelling to examine the fitness of a new method, ‘detailed verification’, for the routine quantification of public health risk for water recycling. The studies are contextualised by a narrative literature review, a narrative case study, and an overarching discussion and conclusion. The first study found that current QMRA approaches varied most for deriving dose, that dose was influential in determining risk, and that the complexity of selected QMRA approaches is generally incommensurate with the purpose served. The second study found that the intervention overall provided greater precision of public health risk estimation. The intervention was associated with an around two-thirds reduction in the estimated public health risk associated with Cryptosporidium and Giardia (rate ratio [RR] = 0.35, p < 0.05, and RR = 0.32, p < 0.001, respectively) in one system, and an increased frequency of non-zero protozoal counts in three scenarios (p < 0.05). The third study found that quantitative public health requirements were met. Sensitivity analysis found that assumptions relating pathogen to surrogate levels were influential on the probability of infection (Spearman’s rank correlation co-efficient, ρ, = 0.38 to 0.90). This thesis establishes new evidence for QMRA implementation for public water supplies. Its findings will support public health protection through the improved assessment and control of pathogen risk.