A Trial of Tap Water Treatment in the Elderly

There is heated debate in the United States about the extent to which waterborne infectious diseases may be transmitted to human beings through drinking water that meets federal standards for pathogen removal. This debate is even more pronounced among elderly immunocompromised persons and their health care providers because of the prolonged illness syndromes (generally gastrointestinal) and even death that may occur in these persons after the ingestion of infectious agents. Concerns about drinking water have been heightened by the findings of Payment et. al. in Canada which suggested that approximately 25% of "highly credible" gastrointestinal illness in a community might be due to drinking water. Gastrointestinal illness and diarrhea are recognized as a significant cause of morbidity and mortality in the elderly. One study showed that 51% of deaths caused by diarrhea over a 9-year period occurred in individuals over the age of 74 years (Lew et al. 1991). In a review of sensitive populations, Gerba et al. (1996) reported case fatality rates for specific enteric pathogens 10 to 100 times higher in this group compared to the general population. Although many infectious diseases are more problematic in the elderly because of a decline in immune function and a higher incidence of pre-existing malnutrition and dehydration, it is still not known what the principal modes of transmission are and which infectious agents are responsible. Furthermore, studies on gastrointestinal illness in the elderly are important now because of an expanding elderly population and the heightened interest in the development of vaccine strategies. These developments necessitate a clearer understanding of the relative contribution of various pathogens to gastrointestinal illness in this population. Although many agents responsible for gastrointestinal illness have previously been difficult to identify reliably, state-of-the-art laboratory technology, including molecular methods, now make such study more feasible. We will use laboratory techniques to identify the specific causal agents of gastrointestinal illness in the intervention and placebo groups and to estimate the relative prevalence of these pathogens as a cause of gastrointestinal illness. This study is a randomized, triple-blinded, placebo-controlled, intervention trial in an elderly population (i.e. over 55 years of age). The intervention to be tested is household-level treatment of drinking water by joint use of ultraviolet light and filtration devices in the form of a point-of-use, countertop unit. The project site is the city of Sonoma, California, and adjacent regions of southern Sonoma County also served by the Sonoma County Water Agency, that receive their drinking water from ground water aquifers under the Russian River. The unit of randomization and intervention will be individual households. The unit of analysis will be the individual participants within households, adjusted for intra-household clustering (i.e. the within-household correlation). The participants, the study investigators and the data analysts will be blinded to the assignment of individual households to active or placebo device except for one study staff member who will keep the randomization codes. This un-blinded staff member will not have contact with any of the study participants, nor will he/she have access to any data on the study participants during the course of the study. The primary objective is to determine the incidence of gastrointestinal disease in a group of households using a home drinking water device (combined filtration/UV light) and in a group of households receiving a placebo device that is inactive but identical in appearance. This information will be used to estimate the burden of gastrointestinal illness in an elderly population due to the consumption, at home, of drinking water that meets regulatory standards but receives no additional treatment. A secondary objective is an estimate of the incidence of specific bacterial, viral, and protozoan agents in stool specimens collected from elderly individuals with gastrointestinal symptoms, and that might be related to water consumption. To determine the rates of asymptomatic infections, we will also test stool specimens from individuals who are not sick. Further questions about seroprevalence of these waterborne pathogens and the immune response they elicit will be answered by testing serum samples taken from all participants at critical junctures during the study.