Systemic caffeine clearance and urinary metabolite profiles were determined in 15 subjects with diverse exposure histories to cytochrome P-450 inducers (cigarette smoke) and inhibitors (oral contraceptive steroids). A correlation was observed between caffeine clearance and a urinary ratio based on the molar recovery of paraxanthine 7-demethylation products relative to a paraxanthine 8-hydroxylation product (r = 0.91; P less than 0.001). Analysis of urinary metabolites was undertaken in a larger population to assess the effects of gender, age, oral contraceptives, and smoking on the ratio. No gender differences were observed in either adults or children; children (n = 21) showed a higher (P less than 0.001) mean metabolite ratio than adults (n = 61), oral contraceptive users (n = 9) had lower (P less than 0.05) ratios than women not taking oral contraceptives (n = 30), and smokers (n = 26) had higher (P less than 0.001) ratios than nonsmokers (n = 61). The data indicate that a urinary metabolite ratio based on paraxanthine 7-demethylation/8-hydroxylation products reflects systemic caffeine clearance and likely monitors cytochrome P-450 activity inducible by polycyclic aromatic hydrocarbons.

This paper forms the second part of an introduction to a synoptic weather typing approach to assess differential and combined impacts of extreme temperatures and air pollution on human mortality, focusing on future estimates. A statistical downscaling approach was used to downscale daily five general circulation model (GCM) outputs (three Canadian and two US GCMs) and to derive six-hourly future climate information for the selected cities (Montreal, Ottawa, Toronto, and Windsor) in south-central Canada. Discriminant function analysis was then used to project the future weather types, based on historical analysis defined in a companion paper (Part I). Future air pollution concentrations were estimated using the within-weather-type historical simulation models applied to the downscaled future GCM climate data. Two independent approaches, based on (1) comparing future and historical frequencies of the weather groups and (2) applying within-weather-group elevated mortality prediction models, were used to assess climate change impacts on elevated mortality for two time windows (2040-2059 and 2070-2089). Averaging the five GCM scenarios, across the study area, heat-related mortality is projected to be more than double by the 2050s and triple by the 2080s from the current condition. Cold-related mortality could decrease by about 45-60% and 60-70% by the 2050s and the 2080s, respectively. Air pollution-related mortality could increase about 20-30% by the 2050s and 30-45% by the 2080s, due to increased air pollution levels projected with climate change. The increase in air pollution-related mortality would be largely driven by increases in ozone effects. The population acclimatization to increased heat was also assessed in this paper, which could reduce future heat-related mortality by 40%. It is most likely that the estimate of future extreme temperature-and air pollutionrelated mortality from this study could represent a bottomline figure since many of the factors (e.g., population growth, age structure changes, and adaptation measures) were not directly taken into account in the analyses.