Errors In Exposure Assessment Research Articles

A risk-based approach to measuring population micronutrient status from blood biomarker concentrations.

BackgroundNutrient biomarkers and their definitive cut-offs are used to classify individuals as nutrient-deficient or sufficient. This determinism does not consider any uncertainty, and a probability approach, using biomarker distributions, is then preferable to define the risk of nutrition deficiency when in populations.MethodHealthy 1–19-year-old children and adolescents were selected from the Comprehensive National Nutrition Survey (CNNS), to obtain probability distributions of their retinol, zinc and vitamin B12, along with erythrocyte folate. Model-based estimates of location, scale and shape parameters of these distributions were obtained across ages. Subsequently, in the entire sample of 1–19 year old children of CNNS, the population risk of deficiency (PRD) which is average risk of deficiency in individuals in the population was computed, which is “of concern” when >50%. When individual risk of deficiency is >97.5% it is called “severe risk of deficiency” (SRD).ResultsIn the entire CNNS sample, the PRD of concern was low for serum retinol (3.6–8.2%), zinc (0–5.5%), and SRD of vitamin B12 and erythrocyte folate were 2.3–7.2% and 4.2–9.7%, respectively, across age and sex groups.ConclusionThis proposed method assesses the adequacy of nutrient exposures without relying on pre-defined deterministic biomarker cut-offs to define micronutrient deficiency and avoids errors in exposure assessment.

Spatiotemporally resolved daily relative humidity predictions across Germany during 2000-2021: a Random Forest approach

BACKGROUND AND AIM The lack of high-resolution relative humidity (RH) datasets and the limited ability of available weather stations to fully capture the spatiotemporal RH variability might lead to errors in exposure assessment for epidemiological studies and biased health effects estimates. We aimed to predict German-wide 1×1 km daily mean RH during 2000-2021 by using a machine learning approach. METHODS We used data from multiple sources, including RH observations, modelled precipitation and wind speed as well as remote sensing elevation, vegetation and the visible red, green and blue light bands. Our main predictor for estimating RH was modelled daily mean air temperature that we previously estimated in 1×1 km across Germany through a multi-stage regression-based approach incorporating two linear mixed models. Additionally, we included date in our RH model, capturing the day-to-day variation of the response-explanatory variables relationship. All the aforementioned predictors were included in a Random Forest (RF) model, which was separately developed for each year. RESULTS Ten-fold cross validation showed that the RF model achieved high accuracy (R²=0.81) and low errors (Root Mean Square Error (RMSE)=5.26%). We also compared our output with an independent dataset from a dense monitoring network in Augsburg metropolitan area (R²=0.81, RMSE=5.42%). Our models displayed high RH overall (21y-average RH=78.8%) and high spatial variability within a year across the country, exceeding 12.1% on average. RH distribution followed spatial patterns including urbanization, mountains, rivers and coastlines. For instance, the Alps and the North Sea coast were areas with elevated RH, while extended urban cores or individual cities were less humid than the surrounding rural settings. CONCLUSIONS Our findings indicate the proposed RF model as suitable for countrywide RH modeling at high spatiotemporal resolution, providing a reliable dataset for subsequent epidemiological analyses and other research purposes. KEYWORDS: relative humidity, spatiotemporal modeling, random forest, exposure assessment

Predicting spatiotemporally-resolved mean air temperature over Sweden from satellite data using an ensemble model

Mapping of air temperature (Ta) at high spatiotemporal resolution is critical to reducing exposure assessment errors in epidemiological studies on the health effects of air temperature. In this study, we applied a three-stage ensemble model to estimate daily mean Ta from satellite-based land surface temperature (Ts) over Sweden during 2001–2019 at a high spatial resolution of 1 × 1 km2. The ensemble model incorporated four base models, including a generalized additive model (GAM), a generalized additive mixed model (GAMM), and two machine learning models (random forest [RF] and extreme gradient boosting [XGBoost]), and allowed the weights for each model to vary over space, with the best-performing model for each grid cell assigned the highest weight. Various spatial predictors were included as adjustment variables in all the base models, including land cover type, normalized difference vegetation index (NDVI), and elevation. The ensemble model showed high performance with an overall R2 of 0.98 and a root mean square error of 1.38 °C in the ten-fold cross-validation, and outperformed each of the four base models. Although each base model performed well, the two machine learning models (RF [R2 = 0.97], XGBoost [R2 = 0.98]) had better performance than the two regression models (GAM [R2 = 0.95], GAMM [R2 = 0.96]). In the machine learning models, Ts was the dominant predictor of Ta, followed by day of year, NDVI, latitude, elevation, and longitude. The highly spatiotemporally-resolved Ta can improve temperature exposure assessment in future epidemiological studies.

Predicting spatiotemporally-resolved air temperature over Sweden from satellite data using an ensemble model

BACKGROUND AND AIM: Mapping of air temperature (Ta) at high spatiotemporal resolution is critical to reduce exposure assessment errors in epidemiological studies on the health effects of air temperature. We aimed to predict daily Ta at a spatial resolution of 1×1 km2 from 2001 to 2019 using an ensemble model based on the satellite-based daily land surface temperature. The results will provide insights in modeling Ta with an ensemble model to reduce exposure measurement error and serve as a foundation for epidemiological studies on the short-term and long-terms effects of air temperature on public health at a large geographical scale. METHODS: The generalized additive model based ensemble model incorporated four base models, including a generalized additive model, a generalized additive mixed model, and two machine learning models (random forest and extreme gradient boosting), and allowed the weights for each model to vary over space, with the best-performing model for each grid cell assigned the highest weight. RESULTS:The ensemble model showed high performance with an overall R2 of 0.98 and a root mean square error of 1.38 °C in the ten-fold cross-validation, and outperformed each of the four base models. Among base models, the two machine learning models outperformed the two regression models. In the machine learning models, Ts was the dominant predictor of Ta, followed by day of year, Normalized Difference Vegetation Index, latitude, elevation, and longitude. CONCLUSIONS:The GAM-based ensemble model exploited the predictive ability of base models by allowing the weights for each model to vary over space, and provided an improved estimation of air temperature over Sweden that outperformed estimations from each base model. Among the base models, two machine learning methods (RF and XGBoost) exhibited higher predictive power than two linear regression models. The estimations from ensemble models can be applied in epidemiological studies to minimize bias caused by exposure misclassification. KEYWORDS: random forest, extreme gradient boosting, generalized additive model, ensemble model, air temperature, health exposure

Fine Particle Exposure and Clinical Aggravation in Neurodegenerative Diseases in New York State.

Background:Adult-onset neurodegenerative diseases affect millions and negatively impact health care systems worldwide. Evidence suggests that air pollution may contribute to aggravation of neurodegeneration, but studies have been limited.Objective:We examined the potential association between long-term exposure to particulate matter in aerodynamic diameter [fine particulate matter ()] and disease aggravation in Alzheimer’s (AD) and Parkinson’s (PD) diseases and amyotrophic lateral sclerosis (ALS), using first hospitalization as a surrogate of clinical aggravation.Methods:We used data from the New York Department of Health Statewide Planning and Research Cooperative System (SPARCS 2000–2014) to construct annual county counts of first hospitalizations with a diagnosis of AD, PD, or ALS (total, urbanicity-, sex-, and age-stratified). We used annual concentrations estimated by a prediction model at a resolution, which we aggregated to population-weighted county averages to assign exposure to cases based on county of residence. We used outcome-specific mixed quasi-Poisson models with county-specific random intercepts to estimate rate ratios (RRs) for a 1-y exposure. We allowed for nonlinear exposure–outcome relationships using penalized splines and accounted for potential confounders.Results:We found a positive nonlinear association that plateaued above (, 95% CI: 1.04, 1.14 for a increase from 8.1 to ). We also found a linear positive association (, 95% CI: 1.01, 1.09 per increase), and suggestive evidence of an association with AD. We found effect modification by age for PD and ALS with a stronger positive association in patients of age but found insufficient evidence of effect modification by sex or urbanization level for any of the outcomes.Conclusion:Our findings suggest that annual increase in county-level concentrations may contribute to clinical aggravation of PD and ALS. Importantly, the average annual concentration in our study was , below the current American national standards, suggesting the standards may not adequately protect the aging population. https://doi.org/10.1289/EHP7425

Short-Duration Characterization of Source Emissions for Use in Predictive Software Models to Assess Worker Exposure: A Note of Caution

This article reports on use of advanced Near-Field—Far-Field software for assessing short- versus long-duration data obtained minute-by-minute at two distances from a small source of an evaporating solvent located in an isolated subsurface structure (a type of confined space) accessed through a manhole containing one or two opening(s). The software uses this data to predict worker exposure to airborne chemical substances. Initial flash-off of volatile components was readily visible in graphs prepared from some tests and especially so in initial output from the calibration utility contained in the modelling software. The calibration utility orients the mathematics of the software to measured data. The calibration utility indicated constant magnitude from longer-duration emissions consistent with constant composition. Source characterization of emissions from solvents containing multiple ingredients and constant initial mass deserves careful consideration because initial emissions may not represent overall behavior. This situation indicates the potential to bias predictions of worker and other types of exposure utilizing the same mathematics. This is especially the case during source characterization using measurements of short duration. This study advocates for further investigation to develop guidelines for source characterization during use of modelling software that minimize the potential for error in exposure assessment.

가습기살균제 사용에 따른 피해 신청자들의 특성 및 노출평가 오류 영향요인 분석

Objectives: The lung injuries by exposure to the humidifier disinfectants (HDs) were reported in 2011, Korea. For the HD victims, environmental exposure level and clinical diagnosis were conducted to determine the levels of damage by HDs. Methods: The exposure assessment to the HDs from 1st to 4th q...

Smoking and prostate cancer: a life course analysis

BackgroundInconsistent associations between smoking status and prostate cancer (PC) could be due to exposure assessment error. Reconstructing smoking behaviors over the life course could reduce exposure assessment error.MethodsAs part of a case-control study, we identified 402 incident and histologically confirmed PC cases that were matched by age (±5 years) to 805 population controls. Through direct interview, we obtained information about: age at smoking onset, intensity and frequency of cigarette smoking at different life stages, and smoking cessation age. Smoking status at interview and average smoking index over the lifetime (packs/year) were estimated. Life course smoking patterns were obtained applying the k-means+ method for longitudinal data to the smoking index (pack/year) for each life stage.ResultsTwo life-course smoking patterns were identified among ever smokers: “pattern A” characterized by males who reported low and constant smoking intensity (87.8%), and “pattern B” (12.2%) males with an initial period of low intensity, followed by an increase during the second period. Compared to never smokers, pattern B was associated with higher poorly differentiated PC, (OR 2.30; 95% CI 1.21–4.38). No association was observed with average smoking index.ConclusionLife course smoking patterns seem to capture the smoking variability during life course and reduce the likelihood of reverse causation. Using this assessment strategy our findings support the potential role of tobacco smoking in PC, particularly poorly differentiated PC. Prospective studies with comprehensive smoking history during the lifetime are needed to confirm these findings.

Using spatio-temporal modeling for exposure assessment in an investigation of fine particulate air pollution and cardiovascular mortality

BackgroundU.S. urban air quality has improved dramatically over the past decades. We evaluated acute effects of fine particulate matter (PM2.5) on cardiovascular (CVD) mortality among residents of Allegheny County in SW Pennsylvania (1999–2011) using spatio-temporal modeling of air pollutants (AP) to reduce misclassification error in exposure assessment. MethodsSpatio-temporal kriging of daily PM2.5 and ozone (O3) was used to produce daily exposure estimates at the residence ZIP code. Time-stratified case-crossover analysis was conducted to examine short-term effects of PM2.5 on CVD mortality, adjusting for O3 and apparent mean temperature. We studied both single and distributed lags for days 0–5. All CVD mortality and subcategories of ischemic heart disease (IHD), acute myocardial infarction, cerebrovascular disease, peripheral vascular disease (PVD), heart failure and cardiac arrhythmia were examined. ResultsA total of 62,135 deaths were identified. We found significant associations of PM2.5 with IHD and PVD mortality at lag day 5: (2.1% (95% CI, 0.2–4.1%) and (7.6%, 95% CI, 0.05–15.7%) per 10µg/m3 increase of PM2.5 in single lag models and for IHD in distributed lag models. There were no statistically significant associations with PM2.5 for any of the other outcomes. ConclusionsThe application of finer scale geographically resolved AP exposures made it possible to study acute effects of PM2.5 on CVD mortality in a large metropolitan area. Our study results demonstrated the continued presence of a dose response relationship of increased risk of CVD mortality within this lower range of PM2.5 exposure.

작업환경측정 시료 채취기의 부착 위치에 따른 유기화합물의 농도 분포에 관한 연구

Objectives: The purpose of this study is to investigate differences in concentration according to the position at the left or right shoulder within a 30 cm of radius of workers' respirators and provide basic data for the establishment of an industrial health policy. Methods: Personal samples were collected from a total of 65 workers from 27 manufacturing firms in South Gyeongsang-do Province from November 5, 2011 to December 30, 2012 after classifying the laborers into left- and right-side groups. The organic compound samples were collected and analyzed in accordance with the NIOSH Manual of Analytical Methods (NMAM) 1501. Results: In terms of the concentration of organic compounds collected from both left and right shoulders at the position of workers' respirators, isobutyl acetate was the highest with 145 ppm at the left shoulder, followed by ethyl acetate (133.5 ppm) and toluene (38.13 ppm). At the right shoulder, on the contrary, ethyl acetate (149.3 ppm) was the highest, followed by toluene (46.26 ppm), xylene (29.63ppm) and isopropyl alcohol (28.06 ppm). Overall, the right shoulder was higher than the left shoulder in terms of concentrations. Conclusions: For the measurement of the working environment, workers' personal samples should be collected at the place closest to the respirator. In terms of the reduction of error, the attachment of two sample media is expected to reduce errors in exposure assessment.

Effects of Covered Solid Sorbent Tube Sample Holders on Organic Vapor Measurements

A study was conducted to examine whether there are significant differences between organic vapor concentrations measured using charcoal tubes with three different configurations: uncovered sample holder (open tube), SKC, and Buck brand covered sample holders. A fractional factorial experimental design was used with the following factors and levels: vapor (n-hexane vs. m-xylene), pump type (pulsating vs. continuous), exposure profile (variable vs. constant), flow rate (30 mL/min vs. 200 mL/min), duration (30 min vs. 80 min), and sample placement (mannequin vs. free hanging). Two of each sampler configuration (six total) were placed in an exposure chamber, and a dynamic test-atmosphere generation system was used to prepare atmospheres containing approximately 12–15 ppm n-hexane or m-xylene with exposure profiles and sampling conducted according to a run sheet generated for the experimental design. A total of 24 runs were completed with six samplers per run, yielding 144 samples that were analyzed by gas chromatography/flame ionization detector. Concentration results for each pair of SKC and Buck covered sample holders were averaged and normalized by dividing by the average result for the open tube sampler from the same run to eliminate the effect of daily variation in chamber concentrations. The resulting ratio of covered sample tube holder and open tube concentrations was used as the response variable. Results of analysis of variance using the general linear model (MINITAB 16) identified statistically significant main effects and/or interactions for pump type, exposure profile, flow rate, and sample holder. However, the magnitude of the effects was generally less than 10%, and overall mean concentration ratios were 0.989 and 1.02 for the Buck and SKC sample holders, respectively. These results show good agreement between covered sample holder results and open tube measurements and demonstrate that exposure assessment errors resulting from the use of covered sorbent tube sample holders for organic vapor monitoring are relatively small (<10%) and not likely to be of practical importance.

Chronic Disease Prevention Research Methods and Their Reliability, With Illustrations From the Women’s Health Initiative

This article reviews the status of statistical methods for chronic disease prevention research, with emphasis on the reliability of findings and on future methodological needs and opportunities. Observational studies, especially cohort studies, play a major role in disease prevention research, but depend on adequate confounding control methods for a useful interpretation. Stratification and regression methods that are commonly used to control confounding are described, and comparative findings from the Women’s Health Initiative (WHI) randomized controlled trial and companion cohort study of the benefits and risk of postmenopausal hormone therapy are used to illustrate the success of these methods. Measurement error in exposure assessment may be a potentially dominating source of bias in such important prevention research areas as nutrition and physical activity epidemiology. Statistical methods to correct for measurement error are briefly reviewed, and the need for methods to accommodate systematic bias in exposure assessment is described. Recent analysis using nutrient exposure biomarkers in WHI cohorts is used to illustrate the impact of such methods. Randomized, controlled intervention trials have the potential to obviate these biases and to reliably assess intervention benefits and risks. However, trials in healthy persons with chronic disease outcomes are typically large, long-term, and expensive. Statistical methods for randomized, controlled prevention trials are briefly reviewed, and the roles of trials in the overall chronic disease prevention research enterprise are examined. Given the logistical and cost challenges, a full-scale disease prevention trial needs to be preceded by careful hypothesis development and initial testing. The potential role of biomarkers, especially high-dimensional biomarkers, in disease prevention hypothesis development, is described and illustrated. The presentation concludes with comments on the methodological research and research infrastructure developments needed to invigorate the chronic disease prevention research agenda, with emphasis on the important role for statisticians in enhancing prevention research methods and applications.

Oxidative Damage to DNA and Lipids as Biomarkers of Exposure to Air Pollution

BackgroundAir pollution is thought to exert health effects through oxidative stress, which causes damage to DNA and lipids.ObjectiveWe determined whether levels of oxidatively damaged DNA and lipid peroxidation products in cells or bodily fluids from humans are useful biomarkers of biologically effective dose in studies of the health effects of exposure to particulate matter (PM) from combustion processes.Data sourcesWe identified publications that reported estimated associations between environmental exposure to PM and oxidative damage to DNA and lipids in PubMed and EMBASE. We also identified publications from reference lists and articles cited in the Web of Science.Data extractionFor each study, we obtained information on the estimated effect size to calculate the standardized mean difference (unitless) and determined the potential for errors in exposure assessment and analysis of each of the biomarkers, for total and stratified formal meta-analyses.Data synthesisIn the meta-analysis, the standardized mean differences (95% confidence interval) between exposed and unexposed subjects for oxidized DNA and lipids were 0.53 (0.29–0.76) and 0.73 (0.18–1.28) in blood and 0.52 (0.22–0.82) and 0.49 (0.01–0.97) in urine, respectively. The standardized mean difference for oxidized lipids was 0.64 (0.07–1.21) in the airways. Restricting analyses to studies unlikely to have substantial biomarker or exposure measurement error, studies likely to have biomarker and/or exposure error, or studies likely to have both sources of error resulted in standardized mean differences of 0.55 (0.19–0.90), 0.66 (0.37–0.95), and 0.65 (0.34–0.96), respectively.ConclusionsExposure to combustion particles is consistenly associated with oxidatively damaged DNA and lipids in humans, suggesting that it is possible to use these measurements as biomarkers of biologically effective dose.

Potential exposure assessment errors associated with body‐worn RF dosimeters

Personal exposure meters for assessing exposure to RF electric or magnetic fields are subject to errors associated with perturbations of the fields by the presence of the human body. Although these alterations are complex they are not completely unpredictable. This article concludes that this error in a common worst-case scenario could reach up to 30 dB and therefore is of concern for exposure assessment. We present several guidelines to address this issue and a useful insight into the overall problem based on finite-difference time-domain simulations and experimental verification.

Chronic Exposure to Ambient Ozone and Lung Function in Young Adults

Tropospheric ozone (O3) is an oxidant, outdoor air pollutant. Chronic exposure has been associated with decreased lung function in children and adolescents. This study investigated the effects of long-term exposure to O3 on lung function in college freshmen. We recruited University of California, Berkeley students (n=255) who were lifelong residents of the Los Angeles and San Francisco Bay areas and who never smoked. Lifetime exposures to O3, small particulate matter (PM10), and nitrogen dioxide (NO2) were based on spatial interpolation of compliance monitor measurements to all residences at which students lived. Spirometry was performed between February and May, times when students would not have had recent exposure to increased levels of O3. Lifetime exposure to O3 was associated with decreased levels of measures of small airways (<2 mm) function (FEF75 and FEF25-75). There was an interaction with the FEF25-75/FVC ratio, a measure of intrinsic airway size. Subjects with a large ratio were less likely to have decreases in FEF75 and FEF25-75 for a given estimated lifetime exposure to O3. This association was not altered by history of chronic respiratory disease, allergy, second-hand exposure to environmental tobacco smoke, exposure to PM10 and NO2, or measurement errors in exposure assessment. A history of increased level of lifetime exposure to ambient O3 is associated with decreased function of airways in which O3 deposition in the lungs is the greatest. Adolescents with intrinsically smaller airways appear to be at greatest risk. Any environmental or genetic factors that lead to reduced airway size may lead to increased susceptibility to the adverse effects of ambient ozone.

Two dimensions of measurement error: Classical and Berkson error in residential radon exposure assessment

Measurement error in exposure assessment is unavoidable. Statistical methods to correct for such errors rely upon a valid error model, particularly regarding the classification of classical and Berkson error, the structure and the size of the error. We provide a detailed list of sources of error in residential radon exposure assessment, stressing the importance of (a) the differentiation between classical and Berkson error and (b) the clear definitions of predictors and operationally defined predictors using the example of two German case-control studies on lung cancer and residential radon exposure. We give intuitive measures of error size and present evidence on both the error size and the multiplicative structure of the error from three data sets with repeated measurements of radon concentration. We conclude that modern exposure assessment should not only aim to be as accurate and precise as possible, but should also provide a model of the remaining measurement errors with clear differentiation of classical and Berkson components.

Epidemiology of environmental tobacco smoke exposure

The health hazards due to exposure to environmental tobacco smoke (ETS) are increasingly established. ETS contains thousands of chemicals including 43 known carcinogens. One of the most important known health effects of ETS exposure is lung cancer in non-smokers, based on epidemiologic evidence and knowledge of the uptake and metabolism of ETS. Epidemiologic studies need to carefully take into account confounding and potential errors in exposure assessment. More research is needed to understand the genetic factors that influence ETS-induced lung cancer. Studies of the patterns of ETS exposure suggest higher rates of exposure in people employed as blue collar workers, in service occupations, earning lower incomes, and among the less educated. Certain racial/ethnic groups (e.g. Blacks, American Indians) may be at higher risk of ETS exposure. Despite substantial progress in protecting individuals from ETS exposure, additional efforts are needed in improving and enforcing policies to reduce exposure.

Effects of systematic exposure assessment errors in partially ecologic case-control studies

In ecologic studies, group-level rather than individual-level exposure data are used. When using group-level exposure data, established by sufficiently large samples of individual exposure assessments, the bias of the effect estimate due to sampling errors or random assessment errors at the individual-level is generally negligible. In contrast, systematic assessment errors may produce more pronounced errors in the group-level exposure measures, leading to bias in ecologic analyses. We focus on effects of systematic exposure assessment errors in partially ecologic case-control studies. Individual-level information on disease status, group membership, and covariates is obtained from registries, whereas the exposure is a group-level measure obtained from an established exposure database. Effects on bias and coverage of 95% CI in various error situations are investigated under the linear risk model, using both simulated and empirical ecologic data on exposures that are binary at the individual level. Our simulations suggest that the bias produced by systematic exposure assessment errors under the linear risk model is generally approximately equal to the ratio of the slope bias and the intercept bias in ordinary linear regression with measurement errors in the independent variable. Consequently, bias in either direction can occur. Exposure assessment errors that systematically distort the group-level exposure measures have more pronounced effects on bias and coverage than errors producing random fluctuations of the group-level measures, which imply bias towards the null. The results indicate the need for careful consideration of potential effects of systematic distortions of the group-level exposure measures when constructing and applying group-level exposure databases, such as probabilistic job exposure matrices.

Residential magnetic fields predicted from wiring configurations: II. Relationships to childhood leukemia

Case-control data on childhood leukemia in Los Angeles County were reanalyzed with residential magnetic fields predicted from the wiring configurations of nearby transmission and distribution lines. As described in a companion paper, the 24-h means of the magnetic field's magnitude in subjects' homes were predicted by a physically based regression model that had been fitted to 24-h measurements and wiring data. In addition, magnetic field exposures were adjusted for the most likely form of exposure assessment errors: classic errors for the 24-h measurements and Berkson errors for the predictions from wire configurations. Although the measured fields had no association with childhood leukemia (P for trend=.88), the risks were significant for predicted magnetic fields above 1.25 mG (odds ratio=2.00, 95% confidence interval=1.03-3.89), and a significant dose-response was seen (P for trend=.02). When exposures were determined by a combination of predictions and measurements that corrects for errors, the odds ratio (odd ratio=2.19, 95% confidence interval=1.12-4.31) and the trend (p =.007) showed somewhat greater significance. These findings support the hypothesis that magnetic fields from electrical lines are causally related to childhood leukemia but that this association has been inconsistent among epidemiologic studies due to different types of exposure assessment error. In these data, the leukemia risks from a child's residential magnetic field exposure appears to be better assessed by wire configurations than by 24-h area measurements. However, the predicted fields only partially account for the effect of the Wertheimer-Leeper wire code in a multivariate analysis and do not completely explain why these wire codes have been so often associated with childhood leukemia. The most plausible explanation for our findings is that the causal factor is another magnetic field exposure metric correlated to both wire code and the field's time-averaged magnitude.

Research Needs Relating to Health Effects of Exposure to Low Levels of Airborne Particulate Matter

Research Needs Relating to Health Effects of Exposure to Low Levels of Airborne Particulate Matter Robert F. Phalen andjefrey S . Lee University of California, Irvine, California 92697 University of Utah, Salt take City, Utah 841 12 oncern over the consistent and reproducible epidemio- logical associations between particulate air pollution and human mortality and morbidity stimulated a series of two international scientific colloquia. The first was held January 24-25, 1994, in Irvine, California, and the second was held May 1-3, 1996, in Park City, Utah. Each colloquium was attended by about 200 epidemiologists, toxicologists, chemists, clinicians, regulators, and other concerned specialists. At both colloquia, attendees were asked to submit written suggestions for future related research after they had listened to numerous formal papers and after participating in several open discus- sions. It is important to note that these research suggestions represent snapshots in time of the ideas of a large group of well-informed experts with vaned backgrounds; these recom- mendations do not carry the imprimatur of the colloquium sponsors, any regulatory body, or any other agency. This article summarizes the research suggestions from the second colloquium; a similar article was published relating to the earlier colloquium.(') The suggestions from attendees of the second colloquium, over 160 in total, were first sorted into broad categories based on the type of project suggested (epidemiology, toxicology, sampling, etc.), and then similar studies were combined under a single description. Some projects were suggested by five or more individuals, most by two or three individuals, and about 25 percent by only one person. In the summary lists that follow, the projects listed first within each category are those that were most frequently suggested. C Studies Related to Air Pollution Epidemiology and Epidemiologic Methods 1. Longitudinal panel studies of morbidity and mortality in healthy and compromised individuals (bronchitis, cardio- vascular-compromised, etc.), with personal exposure characterizations. 2. Effects of long-term exposures with extensive data collec- tion, including personal exposures and weather parame- ters. 3. Studies of the effects of exposure assessment errors on epidemiology study findmgs. 4. Meta-analyses in which studies are combined in order to identify roles of specific exposure agents. 5. Studies in cities (including those in Eastern and Western AyPL.OCCUP.ENVIRON.HYG. 13(6) JUNE 1998 Occupational Cohort Studies 1. Follow-up of particle-exposed occupational cohorts after retirement to look for chronic illnesses and possible induced sensitivity to environmental particulate air pollutants. 2. Studies of cohorts with current heavy particle exposures to look for excess cardiopulmonary-related morbidity and mortality. Research Needs by Category (Projects Are Not Necessarily Mutually Exclusive) Europe) which have high and low extreme levels of pollutants. Complete monitoring data are required. 6. Investigations of the effects (roles) of personal exposures on epidemiologic findings. 7. Collaborative (between economists and epidemiologists) research focusing on cost-benefit analyses and health val- uations. 8. Comparisons of mortality rates and incidence of diseases for the following: clean and highly polluted days; clean and highly polluted cities; and cities that have improving air quality. 9. Health effects of personal exposures, including studies of hospital deaths (with thorough monitoring of chemical species). 10. Incorporation of autopsy data into epidemiologic investi- gations. 11. Investigations directed at threshold phenomena. 12. Studies to explain the observed increases in asthma rates in light of improving outdoor air quality. 13. Investigation of population characteristics (as opposed to individual characteristics) in relation to morbidity and mortality. 14. Additional investigations of the roles of nonparticulate pollutants in epidemiologic investigations of particulate matter. Clinical Studies 1. Investigations of pollutant deposition and clearance in sub- jects with existing pulmonary diseases. 2. Investigations of pollutant effects in subpopulations that are expected to have enhanced sensitivity to particulate air pollutants. 3. Investigations of ventilatory and activity patterns for various types of people. 4. Investigations relating to the possibility of differentiating particulate-caused cardiovascular (and pulmonary) deaths from deaths produced by other causes. 1047-322X/98/1306-352$19.00/4 0 1998 AIH PI1 Sl047-322X(98)00004-1