Particulate air pollution, in association with other common urban air pollutants, has been associated with various measured health endpoints, including the incidence and duration of respiratory symptoms, lung function, absence from work or school due to respiratory illness, hospitalization for respiratory disease, and cardiopulmonary disease mortality. In this study, the association between daily mortality and air pollution was assessed in Zurich, Basle, and Geneva (Switzerland) for the time period 1984 through 1989. Various regression modeling techniques were used to estimate the effect of air pollution on mortality, to control for time trends, seasonal factors, and weather variables, and to assess the sensitivity of the results. A positive, statistically significant association between daily mortality counts and measures of ambient air pollution in all three cities was observed. Mortality was associated with total suspended particulate pollution, sulfur dioxide, and nitrogen dioxide. The strongest association was with a 3-day moving average (including the concurrent day and the preceding 2 days) of these pollutants. The estimated mortality-air pollution effects were not highly sensitive to regression modeling techniques used to control for seasonality, long-term trends, and weather variables.