Stan Judek

Meta-Analysis of Time-Series Studies of Air Pollution and Mortality: Effects of Gases and Particles and the Influence of Cause of Death, Age, and Season

Abstract A comprehensive, systematic synthesis was conducted of daily time-series studies of air pollution and mortality from around the world. Estimates of effect sizes were extracted from 109 studies, from single- and multipollutant models, and by cause of death, age, and season. Random effects pooled estimates of excess all-cause mortality (single-pollutant models) associated with a change in pollutant concentration equal to the mean value among a representative group of cities were 2.0% (95% CI 1.5-2.4%) per 31.3 μg/m3 particulate matter (PM) of median diameter <10 μm (PM10); 1.7% (1.2-2.2%) per 1.1 ppm CO; 2.8% (2.1-3.5%) per 24.0 ppb NO2; 1.6% (1.1-2.0%) per 31.2 ppb O3; and 0.9% (0.7-1.2%) per 9.4 ppb SO2 (daily maximum concentration for O3, daily average for others). Effect sizes were generally reduced in multipollutant models, but remained significantly different from zero for PM10 and SO2. Larger effect sizes were observed for respiratory mortality for all pollutants except O3. Heterogeneity among studies was partially accounted for by differences in variability of pollutant concentrations, and results were robust to alternative approaches to selecting estimates from the pool of available candidates. This synthesis leaves little doubt that acute air pollution exposure is a significant contributor to mortality.

Ambient air pollution, birth weight and preterm birth: A systematic review and meta-analysis

Low birth weight and preterm birth have a substantial public health impact. Studies examining their association with outdoor air pollution were identified using searches of bibliographic databases and reference lists of relevant papers. Pooled estimates of effect were calculated, heterogeneity was quantified, meta-regression was conducted and publication bias was examined. Sixty-two studies met the inclusion criteria. The majority of studies reported reduced birth weight and increased odds of low birth weight in relation to exposure to carbon monoxide (CO), nitrogen dioxide (NO(2)) and particulate matter less than 10 and 2.5 microns (PM(10) and PM(2.5)). Effect estimates based on entire pregnancy exposure were generally largest. Pooled estimates of decrease in birth weight ranged from 11.4 g (95% confidence interval -6.9-29.7) per 1 ppm CO to 28.1g (11.5-44.8) per 20 ppb NO(2), and pooled odds ratios for low birth weight ranged from 1.05 (0.99-1.12) per 10 μg/m(3) PM(2.5) to 1.10 (1.05-1.15) per 20 μg/m(3) PM(10) based on entire pregnancy exposure. Fewer effect estimates were available for preterm birth and results were mixed. Pooled odds ratios based on 3rd trimester exposures were generally most precise, ranging from 1.04 (1.02-1.06) per 1 ppm CO to 1.06 (1.03-1.11) per 20 μg/m(3) PM(10). Results were less consistent for ozone and sulfur dioxide for all outcomes. Heterogeneity between studies varied widely between pollutants and outcomes, and meta-regression suggested that heterogeneity could be partially explained by methodological differences between studies. While there is a large evidence base which is indicative of associations between CO, NO(2), PM and pregnancy outcome, variation in effects by exposure period and sources of heterogeneity between studies should be further explored.

Meta-analysis of time-series studies of air pollution and mortality: update in relation to the use of generalized additive models.

In an earlier paper, we reported the results of a metaanalysis of the association between air pollution and mortality based on the results of time-series studies published since 1985.1 In the earlier analysis, estimates of effect size were extracted from 109 studies, from single and multipollutant models, and by cause of death, age, and season. PM10, CO, NO2, O3, and SO2 were all positively and significantly associated with allcause mortality. Effect sizes generally were reduced in multipollutant models but remained significantly different from zero for PM10 and SO2. Larger effect sizes were observed for respiratory mortality for all pollutants except O3. Heterogeneity among studies was partially accounted for by differences in variability of pollutant concentration, and results were robust to alternative approaches to selecting estimates from the pool of available candidates. Generalized additive models (GAMs) evolved as the preferred approach to time-series analysis in this area because, compared with fully parametric methods, they permitted greater flexibility in modeling nonlinear relationships, they were better able to deal with missing data, and they were perhaps less prone to investigator bias in selecting the optimum multivariate model. However, it has recently been observed that some estimates of the association between air pollution and acute health effects derived from time-series studies may be incorrect because of previously unrecognized problems with GAMs.2 The principal issues that have been identified are that the default convergence criterion for the estimation procedure was not sufficiently stringent and that results differed when fully parametric versus nonparametric smoothing functions were used to control for effects of time and weather. The problem appears to have a particularly important impact when adjustments for both temporal cycles and weather are made using nonparametric smoothing functions. An additional issue is that the standard error associated with regression parameters is underestimated when the default approximate method is used to estimate this value, although in the context of a random effects model employed in meta-analyses, standard errors associated with pooled estimates of effect size appear to be only slightly affected, because greater within-study variance is offset by reduced betweenstudy variance.3 In light of these recent revelations, we revisited our original analysis and classified estimates of effect size from primary studies according to whether they were GAM-based. Any estimate that was based on nonparametric smoothing functions of time or weather was considered to be potentially affected. Figure 1 shows the prevalence of use of GAMs based on the 272 effectsize estimates for all-cause mortality from single and multipollutant models, which served as the basis of pooled estimates in our original paper. The prevalence of GAM-based estimates has clearly increased dramatically since these methods were first applied in this area in 1996. However, the frequency varies significantly among the pollutants and geographic regions considered in the original analysis. The percentage of GAMbased primary estimates ranged from 42% for SO2 (28 of 67 estimates) to 85% for CO (29 of 34), and from 33% for Australia and New Zealand (4 of 12) to 100% for Canada (27 of 27). Table 1 presents pooled estimates of effect size according to whether primary estimates were GAM-based. Effect sizes were calculated based on the same changes in pollutant concentrations as reported in the original paper. Based on single-pollutant models, all pollutants were positively and significantly associated with mortality. For all pollutants, pooled estimates from single-pollutant models were greater than those from multipollutant models. As reported in the earlier paper, pooled estimates from multipollutant models remained significantly different from zero for PM10 and SO2, based on both GAMand non-GAM-based estimates. Based on single-pollutant models, the largest effect sizes were observed for CO (non-GAM-based) and NO2 (GAM-based), while for multipollutant models, PM10 NOTEBOOK PAPER ISSN 1047-3289 J. Air & Waste Manage. Assoc. 53:258–261

Air pollution, aeroallergens and cardiorespiratory emergency department visits in Saint John, Canada.

Existing studies of the association between air pollution, aeroallergens and emergency department (ED) visits have generally examined the effects of a few pollutants or aeroallergens on individual conditions such as asthma or chronic obstructive pulmonary disease. In this study, we considered a wide variety of respiratory and cardiac conditions and an extensive set of pollutants and aeroallergens, and utilized prospectively collected information on possible effect modifiers which would not normally be available from purely administrative data. The association between air pollution, aeroallergens and cardiorespiratory ED visits (n=19,821) was examined for the period 1992 to 1996 using generalized additive models. ED visit, air pollution and aeroallergen time series were prefiltered using LOESS smoothers to minimize temporal confounding, and a parsimonious model was constructed to control for confounding by weather and day of week. Multipollutant and multi-aeroallergen models were constructed using stepwise procedures and sensitivity analyses were conducted by season, diagnosis, and selected individual characteristics or effect modifiers. In single-pollutant models, positive effects of all pollutants but NO2 and COH were observed on asthma visits, and positive effects on all respiratory diagnosis groups were observed for O3, SO2, PM10, PM2.5, and SO42−. Among cardiac conditions, only dysrhythmia visits were positively associated with all measures of particulate matter. In the final year-round multipollutant models, a 20.9% increase in cardiac ED visits was attributed to the combination of O3 (16.0%, 95% CI 2.8–30.9) and SO2 (4.9%, 95%CI 1.7–8.2) at the mean concentration of each pollutant. In the final multipollutant model for respiratory visits, O3 accounted for 3.9% of visits (95% CI 0.8–7.2), and SO2 for 3.7% (95% CI 1.5–6.0), whereas a weak, negative association was observed with NO2. In multi-aeroallergen models of warm season asthma ED visits, Ascomycetes, Alternaria and small round fungal spores accounted for 4.5% (95% CI 1.8–7.4), 4.7% (95% CI 1.0–8.6) and 3.0% (95% CI 0.8–5.1), respectively, of visits at their mean concentrations, and these effects were not sensitive to adjustment for air pollution effects. In conclusion, we observed a significant influence of the air pollution mix on cardiac and respiratory ED visits. Although in single-pollutant models, positive associations were noted between ED visits and some measures of particulate matter, in multipollutant models, pollutant gases, particularly ozone, exhibited more consistent effects. Aeroallergens were also significantly associated with warm season asthma ED visits.

Effect of airborne allergens on emergency visits by children for conjunctivitis and rhinitis

The effect of fungal spores and pollen grains on morbidity from childhood conjunctivitis and rhinitis is mostly unknown. We therefore studied the association between daily concentrations of these airborne allergens and daily emergency visits to a children hospital between 1993 and 1997. An increase of 551 basidiomycetes spores per m(3), or of 72 ragweed grains per m(3), was associated with an increase of about 10% in visits for conjunctivitis and rhinitis (p<0.01). Our results suggest that conjunctivitis and rhinitis could be caused by fungal spores and pollens in the air.

Airborne endotoxin is associated with respiratory illness in the first 2 years of life.

To determine the influence of endotoxin on the incidence of acute respiratory illness during the first 2 years of life, we carried out a longitudinal follow-up study, beginning at birth, of 332 children born in Prince Edward Island, Canada. We measured 5-day averaged air endotoxin in the homes of children, whose parents provided information by daily symptom diaries and twice-monthly telephone contact for up to 2 years. Endotoxin concentration was 0.49 ± 3.49 EU/m3 (geometric mean ± geometric SD), and number of annualized illness episodes was 6.83 ± 2.80 (mean ± SD). A doubling of the air endotoxin concentration was associated with an increase of 0.32 illness episodes per year (p = 0.0003), adjusted for age, year of study, breast-feeding, environmental tobacco smoke, child care attendance, indoor temperature, and income. Indoor mold surface area and fungal ergosterol were not significantly associated with endotoxin. Airborne endotoxin appears to be a risk factor for clinically symptomatic respiratory illnesses during the first 2 years of life independent of indoor fungus.

Tree Pollen and Hospitalization for Asthma in Urban Canada

Background: Allergy to tree pollen is common and aeroallergens have been associated with severe asthma exacerbations in the community setting. To determine the impact of different trees on asthma, we tested the association between daily hospitalizations for asthma and daily concentrations of different tree pollens in 10 large Canadian cities. Methods: Daily time-series analyses were employed to remove unwanted temporal trends. For each family or genus, results were adjusted for day of the week, temperature, barometric pressure and relative humidity. Results were expressed as the percentage increase in asthma hospitalizations related to an increase in tree pollen concentration equivalent in magnitude to its interquartile range. Results: For an interquartile increase in daily tree pollen concentration, percent increases in daily hospitalization for asthma were: 2.63% (95% CI 1.19–4.07) for Ulmus (elm), 2.45% (1.12–3.78) for the group containing Pinaceae (pine, fir, spruce), Tsuga (hemlock) and Larix (larch, tamarack); 2.32% (0.93–3.71) for the group containing Quercus (oak) and Castanea (chestnut), and 2.16% (0.70–3.62) for Acer (boxelder and maple). Statistically significant (p < 0.05) but small (<2%) effects were observed for Fraxinus (ash), Populus (aspen, poplar), Alnus (alder), Betula (birch) and Corylus (hazelnut). Conclusions: Several common tree pollens are an important cause of acute exacerbations of asthma severe enough to require hospitalization.

Do gender, education, and income modify the effect of air pollution gases on cardiac disease?

Objective: We sought to determine whether gender, education, and income influence the susceptibility to ambient air pollution. Methods: We determined the association between daily cardiac hospitalizations and daily concentrations of gaseous air pollutants in 10 large Canadian cities using time-series analyses adjusted for day-of-the week, temperature, barometric pressure, relative humidity. Results: Percentage increases in hospitalization associated with an increase in air pollution equivalent to its mean value were statistically significant for ozone, carbon monoxide and nitrogen dioxide individually (P < 0.05) and the combined pollutant effect was 8.5% (95% confidence interval: 1.8, 14.6). The air pollution-cardiac disease association was not significantly influenced by gender or community level of education or income. Conclusion: Short-term changes in air pollution may adversely affect cardiac disease but gender, and community education and income do not accurately identify those with increased susceptibility.

Economic evaluation of the benefits of reducing acute cardiorespiratory morbidity associated with air pollution

BackgroundFew assessments of the costs and benefits of reducing acute cardiorespiratory morbidity related to air pollution have employed a comprehensive, explicit approach to capturing the full societal value of reduced morbidity.MethodsWe used empirical data on the duration and severity of episodes of cardiorespiratory disease as inputs to complementary models of cost of treatment, lost productivity, and willingness to pay to avoid acute cardiorespiratory morbidity outcomes linked to air pollution in epidemiological studies. A Monte Carlo estimation procedure was utilized to propagate uncertainty in key inputs and model parameters.ResultsValuation estimates ranged from

Personal Exposure of Adults with Cardiorespiratory Disease to Particulate Acid and Sulfate in Saint John, New Brunswick, Canada

13 (1997, Canadian) (95% confidence interval,