Quantitative Bayesian predictions of source water concentration for QMRA from presence/absence data for E. coli O157:H7

摘要

A hierarchical Bayesian framework was applied for describing variability in pathogennconcentration (with associated uncertainty) from presence/absence observations for E. colinO157:H7. Laboratory spiking experiments (method performance) and environmental samplenassays were undertaken for a surface drinking water source in France. The concentrationnestimates were strongly dependent upon the assumed statistical model used (gamma, log-gammanor log-gamma constrained), highlighting the need for a solid theoretical basis for model choice.nBayesian methods facilitate the incorporation of additional data into the statistical analysis; thisnwas illustrated using faecal indicator results of E. coli (Colilertw) to reduce the posteriornparameter uncertainty and improve model stability. While conceptually simple, application ofnthese methods is still specialised, hence there is a need for the development of data analysisntools to make Bayesian simulation techniques more accessible for QMRA practitioners.