Ozone and several polar volatile organic compounds (VOCs) including organic acids and carbonyls (aldehydes and ketones) were measured over an approximately 24 hour period in four residences during the winter of 1993 and in nine residences during the summer of 1993. All residences were in the greater Boston, Massachusetts area. The relation of the polar VOCs to the ozone concentration was examined. Indoor carbonyl concentrations were similar between the summer and winter, with the total mean winter concentration being 31.7 ppb and the total mean summer concentration being 36.6 ppb. However, the average air exchange rate was 0.9 hr 1 during the winter and 2.6 hr 1 during the summer. Therefore, the estimated carbonyl emission rates were significantly higher during the summer. Indoor organic acid concentrations were about twice as high during the summer as during the winter. For formic acid, the indoor winter mean was 9.8 ppb, and the summer indoor mean was 17.8 ppb. For acetic acid, the indoor winter mean was 15.5 ppb, and the summer indoor mean was 28.7 ppb. The concentrations of the polar VOCs were found to be significantly correlated with one another. Also, the emission rates of the polar VOCs were found to be correlated with both the environmental variables such as temperature and relative humidity and the ozone removal rate; however, it was difficult to apportion the relative effects of the environmental variables and the ozone removal.

Interest in examining both the uncertainty and variability in environmental health risk assessments has led to increased use of methods for propagating uncertainty. While a variety of approaches have been described, the advent of both powerful personal computers and commercially available simulation software have led to increased use of Monte Carlo simulation. Although most analysts and regulators are encouraged by these developments, some are concerned that Monte Carlo analysis is being applied uncritically. The validity of any analysis is contingent on the validity of the inputs to the analysis. In the propagation of uncertainty or variability, it is essential that the statistical distribution of input variables are properly specified. Furthermore, any dependencies among the input variables must be considered in the analysis. In light of the potential difficulty in specifying dependencies among input variables, it is useful to consider whether there exist rules of thumb as to when correlations can be safely ignored (i.e., when little overall precision is gained by an additional effort to improve upon an estimation of correlation). We make use of well-known error propagation formulas to develop expressions intended to aid the analyst in situations wherein normally and lognormally distributed variables are linearly correlated.

No abstract

Information on human time-activity patterns is often required to interpret environmental exposure data fully and to implement exposure assessment models. Data on short-term time-activity patterns for individuals, such as 1-day measurements, are relatively abundant. The reliability of such data for use in chronic exposure (e.g., 1 or more years) assessments performed for evaluation of health risks is not well understood. As part of the NHEXAS-Maryland investigation, daily time budget data for seven microenvironments were collected from 80 people during as many as six 1-week Cycles over a 12-month period. The data were summarized and analyzed statistically by sampling Cycle, day of week, and individual to characterize long-term average microenvironmental time budgets and to identify their determinants. Median times spent in transit, indoors at home, outside at home, indoors at work or school, outdoors at work or school, indoors at other locations, and outdoors at other locations were found to vary significantly, although not substantively in many cases, by time of year (i.e., Cycle), by day of week, and by individuals. Time budgets for most of the microenvironments also exhibited significant variability by gender, age group, education level, annual household income, and work status. The results indicate that short-term (e.g., 1-day) measures of microenvironmental time budgets for individuals are unlikely to be representative of their long-term patterns. Thus, health risk or epidemiological assessments performed for a population mean or specific quantile may be relatively insensitive to when time budget data were collected. However, the accuracy of such assessments performed for individuals is likely to be greatly improved by collection of time budget data from numerous points in time.

To examine the association between lead exposure and both individual and geographic area indicators of socioeconomic position, the authors measured tibia lead concentration, a biomarker of cumulative lead exposure, using K x-ray fluorescence in a cross-sectional survey of 538 white males aged 50-92 years who were healthy when enrolled in the Normative Aging Study (Boston, Massachusetts) in the 1960s. Data on individual risk factors, education, occupation, and income were collected by questionnaire. Using subjects' residential addresses at the time of the tibia lead measurements, the authors obtained geographic area-specific measures of education, social class, and poverty by linking records to 1990 US Census block group data. In multivariate linear regression analysis controlling for age and cumulative smoking, tibia lead concentrations were 10.39 microg/g (95% confidence interval (CI) 7.80-12.97) higher in men who did not graduate from high school than in men with > or =4 years of college. Among the former men (non-high school graduates), living in an undereducated area was associated with a 9.28 microg/g (95% CI 1.59-16.97) increase in tibia lead level compared with living in a non-undereducated area; among the latter men (college graduates), no difference existed by residential area education (beta = 0.72, 95% CI -5.35 to 6.78). The authors conclude that the influence of individual socioeconomic position on cumulative lead exposure is modified by geographic area conditions.