Waterborne microbial risk assessment : a population-based dose-response function for Giardia spp. (E.MI.R.A study)
1 INSERM-ERI 11, Nancy University Medical School – 9 av de la Forêt de Haye, BP 184 – 54505 Vandoeuvre-les-Nancy Cedex, France
2 Grenoble University Hospital, 38700 La Tronche, France
BMC Public Health 2006, 6:122 doi:10.1186/1471-2458-6-122Published: 3 May 2006
Dose-response parameters based on clinical challenges are frequently used to assess the health impact of protozoa in drinking water. We compare the risk estimates associated with Giardia in drinking water derived from the dose-response parameter published in the literature and the incidence of acute digestive conditions (ADC) measured in the framework of an epidemiological study in a general population.
The study combined a daily follow-up of digestive morbidity among a panel of 544 volunteers and a microbiological surveillance of tap water. The relationship between incidence of ADC and concentrations of Giardia cysts was modeled with Generalized Estimating Equations, adjusting on community, age, tap water intake, presence of bacterial indicators, and genetic markers of viruses. The quantitative estimate of Giardia dose was the product of the declared amount of drinking water intake (in L) by the logarithm of cysts concentrations.
The Odds Ratio for one unit of dose [OR = 1.76 (95% CI: 1.21, 2.55)] showed a very good consistency with the risk assessment estimate computed after the literature dose-response, provided application of a 20 % abatement factor to the cysts counts that were measured in the epidemiological study. Doing so, a daily water intake of 2 L and a Giardia concentration of 10 cysts/100 L, would yield an estimated relative excess risk of 12 % according to the Rendtorff model, against 11 % when multiplying the baseline rate of ADC by the corresponding OR. This abatement parameter encompasses uncertainties associated with germ viability, infectivity and virulence in natural settings.
The dose-response function for waterborne Giardia risk derived from clinical experiments is consistent with epidemiological data. However, much remains to be learned about key characteristics that may heavily influence quantitative risk assessment results.