![[logo] US EPA](https://webarchive.library.unt.edu/eot2008/20081106024154im_/http://www.epa.gov/epafiles/images/logo_epaseal.gif){kind=logo}
Final Report: Statistics and Data Core
EPA Grant Number: R827355C009Subproject: this is subproject number 009 , established and managed by the Center Director under grant R827355
(EPA does not fund or establish subprojects; EPA awards and manages the overall grant for this center).
Center: Airborne PM - Northwest Research Center for Particulate Air Pollution and Health
Center Director: Koenig, Jane Q.
Title: Statistics and Data Core
Investigators: Sheppard, Lianne , Lumley, Thomas , Sampson, Paul , Wakefield, Jon
Institution: University of Washington
EPA Project Officer: Stacey Katz/Gail Robarge,
Project Period: June 1, 1999 through May 31, 2004 (Extended to May 31, 2005)
RFA: Airborne Particulate Matter (PM) Centers (1999)
Research Category: Particulate Matter
Description:
Objective:The overall objective of the statistics effort of this research project was to develop and clarify statistical methodology in the air pollution field. The specific objectives were to: (1) use source apportioned exposure data in health effects analyses, (2) interpret exposure effects in chronic effect studies, (3) review case-crossover methods used in analyzing air pollution exposure data, (4) develop a strategy for spatio-temporal modeling of PM, (5) review methods in air pollution panel studies, and (6) integrate statistical methods into biomarker research (methodological issues associated with below the limit of detection data).
The objectives of the data core are to: (1) compile and manage data to support PM Center research; (2) utilize existing data for analyses of health effects; (3) ensure quality statistical design and analysis for PM Center research; and (4) identify statistical methodology research needs for PM Center research and seek resources to perform such research.
Summary/Accomplishments (Outputs/Outcomes):Statistics
- Source Apportioned Exposure Data and Estimation of Health Effects
This research investigated the effects of using positive matrix factorization (PMF) imputed source contributions as exposure variables in health effects models. We reviewed the use of PMF for the identification of sources of PM and studied the properties of health effects estimated using source exposures imputed by PMF. We used a simulation study to demonstrate that directly substituting PMF results into a health effects model produces biased health effects estimates. Our results indicated that the uncertainty in PM source estimates imputed by PMF, although small enough to allow identification of sources and characterization of their contributions to overall particulate burden, has the potential to cause serious bias in estimated health effects.
We are preparing a paper to discuss the range of issues that affect the design of epidemiologic studies of the health effects of air pollution from PM sources, in particular using speciated data. This paper uses the framework of three models, a disease model, an exposure model, and a measurement model. These are integrated into a discussion of different study designs to clarify the opportunities and challenges associated with conducting health studies using speciated data with the intent of estimating the health effects of sources.
- Interpretation of Exposure Effects in Chronic Effect Studies
This research considered how one interprets the exposure effect parameter from a cohort study. The standard Cox model used in air pollution cohort study analyses assumes the exposure is fixed at baseline and the effect is propagated at the constant level throughout follow-up. We looked at sensitivity to this structure by establishing a framework for the effect of exposure on the disease outcome by considering the time resolution of the data, the exposure history, and the association history (i.e., how is risk at time t affected by exposure at an earlier time s?). Data were simulated from an individual-level model and then analyzed in the more typical aggregate framework where exposure is aggregated over time. We performed an “insensitivity” analysis to shed light on the underlying model parameter that would generate identical results from the standard analysis using mean exposure as the predictor. A paper describing this research is in preparation.
- Case-Crossover Methods in Air Pollution
In revising our case-crossover review paper, “Referent Selection Strategies in Case-Crossover Analyses of Air Pollution Exposure Data: Implications for Bias,” we conducted research in several new areas. First, we carried out an exhaustive review of air pollution case-crossover studies to date, and summarized their referent selection strategies. We also compared the statistical efficiency of the various referent selection strategies. We have proposed new terminology for case-crossover referent strategies, and have worked to clarify the issue of overlap bias, which is encountered when a referent strategy is improperly paired with an analysis method. We clarified the so-called “rare event” assumption in the case-crossover design, and have pointed out why this assumption is necessary. We also have made a point to distinguish between two types of models that have been used to describe case-crossover data, and have indicated which model is most appropriate for air pollution exposure data. Finally, we have researched the properties of various referent selection strategies for cases when individuals do not all share the same exposure series.
- Spatio-Temporal Modeling of PM
We developed a strategy for spatial modeling and estimation of PM concentrations that combines data from monitoring sites operating at different temporal scales, recording integrated average concentrations at either daily, 3-day, or 2-week intervals. The modeling builds on the spatial-deformation modeling strategy for nonstationary spatial covariance structure introduced by Drs. Sampson and Guttorp and further developed in a Bayesian framework in collaboration with one of their students. We have combined this modeling strategy with a new flexible, but parsimonious approach to modeling the spatio-temporal trend in such monitoring data. This approach represents seasonal structure that varies in space and from year to year. It is based on a modified (smoothed) singular value decomposition of the (typically incomplete) space x time matrix of observed concentrations. This approach (for a monitoring at a single time scale) has been explained and illustrated with an application to ozone monitoring data in a recently submitted chapter. Separate empirical analyses of PM, NOx, and NO2 data for southern California at daily and 2-week average scales have been carried out to confirm the applicability of the model structure prior to the implementation of the strategy for integration of monitoring data at different time scales.
- Review of Methods in Air Pollution Panel Studies
We currently are working on writing a paper that is aimed at clarifying the statistical issues that arise in the design and analysis of air pollution panel studies. We portray the appropriate types of exposures and outcomes, contrast the various modeling approaches, discuss methods for controlling for confounding, and describe appropriate exploratory data analysis techniques. We illustrate the concepts throughout using data from the Seattle Panel Study.
- Methodological Issues Associated With Below the Limit of Detection Data
We investigated the relative bias and efficiency of regression models for environmental data when some observations lie below the limit of detection and others are missing at random. We investigated the performance of linear regression models and parametric survival models for left censored data when only complete cases are included and after using multiple imputation to account for random missingness in the data. Our simulated data were based on the distribution of measurements of methoxyphenol compounds in ambient air collected at Panel Study outdoor and central sites.
Data Core
The following activities were completed by the Data Core:
- Support of the agricultural burning study data analyses.
- Support of management and analysis of project size-distributed data, including data management, data validation, data analyses, and development of new software and methods.
- Support of epidemiology studies activities, particularly with respect to the Women’s Health Initiative (WHI) study, Multi-Ethnic Study of Atherosclerosis (MESA) Air Pollution Study, cystic fibrosis analysis, Myocardial Infarction Triage and Intervention (MITI) study analysis, and bronchiolitis study.
- Support of the Panel Study Health Effects project, particularly health effects analyses (heart rate variability, bloods, lung).
- Support of biomarker research.
- Support of the Diesel Facility Project, including data forms development, data organization, database management, data validation, technical systems review (TSR) support, and statistical collaboration/support.
- General support of data analyses and data sharing, including preparation of a final PM Center data release.
- Implementation and support of statistical analysis plans (SAPs).
Journal Articles on this Report: 19 Displayed | Download in RIS Format
| Other subproject views: | All 43 publications | 29 publications in selected types | All 28 journal articles |
| Other center views: | All 191 publications | 97 publications in selected types | All 94 journal articles |
| Type | Citation | ||
|---|---|---|---|
|
|
Allen R, Wallace L, Larson T, Sheppard L, Liu L-JS. Estimated hourly personal exposures to ambient and nonambient particulate matter among sensitive populations in Seattle, Washington. Journal of the Air & Waste Management Association 2004;54(9):1197-1211. |
R827355 (2004) R827355 (Final) R827355C003 (2003) R827355C003 (2004) R827355C003 (Final) R827355C008 (Final) R827355C009 (Final) |
|
|
|
Haneuse S, Wakefield J, Sheppard L. The interpretation of exposure effect estimates in chronic air pollution studies. Statistics in Medicine 2007;26(16):3172-3187. |
R827355 (Final) R827355C009 (Final) |
|
|
|
Janes H, Sheppard L, Lumley T. Case-crossover analyses of air pollution exposure data: referent selection strategies and their implication for bias. Epidemiology 2005;16(6):717-726. |
R827355 (Final) R827355C009 (Final) |
|
|
|
Janes H, Sheppard L, Lumley T. Overlap bias in the case-crossover design, with application to air pollution exposures. Statistics in Medicine 2005;24(2):285-300. |
R827355 (2004) R827355 (Final) R827355C009 (2003) R827355C009 (Final) |
|
|
|
Koenig JQ, Mar TF, Allen RW, Jansen K, Lumley T, Sullivan JH, Trenga CA, Larson TV, Liu L-JS. Pulmonary effects of indoor-and outdoor-generated particles in children with asthma. Environmental Health Perspectives 2005;113(4):499-503. |
R827355 (2004) R827355 (Final) R827355C002 (2003) R827355C002 (2004) R827355C002 (Final) R827355C003 (2004) R827355C003 (Final) R827355C009 (Final) |
|
|
|
Larson TV, Covert DS, Kim E, Elleman R, Schreuder AB, Lumley T. Combining size distribution and chemical species measurements into a multivariate receptor model of PM2.5. Journal of Geophysical Research 2006;111:D10S09. |
R827355 (Final) R827355C004 (Final) R827355C008 (Final) R827355C009 (Final) |
|
|
|
Lumley T, Sheppard L. Assessing seasonal confounding and model selection bias in air pollution epidemiology using positive and negative control analyses. Environmetrics 2000;11(6):705-717. |
R827355 (2001) R827355 (Final) R827355C001 (2000) R827355C001 (2001) R827355C009 (Final) R825173 (1999) |
|
|
|
Lumley T, Levy D. Bias in the case-crossover design: implications for studies of air pollution. Environmetrics 2000;11(6):689-704. |
R827355 (2001) R827355 (Final) R827355C009 (Final) R825173 (1999) |
|
|
|
Moolgavkar SH, Hazelton W, Leubeck G, Levy D, Sheppard L. Air pollution, pollens, and admissions for chronic respiratory disease in King County, Washington. Inhalation Toxicology 2000;12(Suppl 1 to Issue 1):157-171(15). |
R827355 (2001) R827355 (Final) R827355C009 (Final) R825266 (Final) |
|
|
|
Sheppard L, Damian D. Estimating short-term PM effects accounting for surrogate exposure measurements from ambient monitors. Environmetrics 2000;11(6):675-687. |
R827355 (2001) R827355 (Final) R827355C009 (Final) |
|
|
|
Sheppard L, Kaufman J. Sorting out the role of air pollutants in asthma initiation. (Editorials) Epidemiology 2000;11(2):100-101. |
R827355 (2004) R827355 (Final) R827355C001 (1999) R827355C001 (2001) R827355C009 (Final) |
|
|
|
Sheppard L, Levy D, Checkoway H. Correcting for the effects of location and atmospheric conditions on air pollution exposures in a case-crossover study. Journal of Exposure Science and Environmental Epidemiology 2001;11(2):86-96. |
R827355 (2004) R827355 (Final) R827355C001 (2001) R827355C009 (Final) R825173 (1999) R825173 (2000) |
|
|
|
Sheppard L. Insights on bias and information in group-level studies. Biostatistics 2003;4(2):265-278. |
R827355 (2004) R827355 (Final) R827355C009 (Final) |
|
|
|
Sheppard L. Acute air pollution effects: consequences of exposure distribution and measurements. Journal of Toxicology and Environmental Health, Part A 2005;68(13-14):1127-1135. |
R827355 (2004) R827355 (Final) R827355C009 (2003) R827355C009 (Final) |
|
|
|
Sheppard L, Slaughter JC, Schildcrout J, Liu L-JS, Lumley T. Exposure and measurement contributions to estimates of acute air pollution effects. Journal of Exposure Science and Environmental Epidemiology 2005;15(4):366-376. |
R827355 (2004) R827355 (Final) R827355C009 (2003) R827355C009 (Final) |
|
|
|
Slaughter JC, Kim E, Sheppard L, Sullivan JH, Larson TV, Claiborn C. Association between particulate matter and emergency room visits, hospital admissions and mortality in Spokane, Washington. Journal of Exposure Science and Environmental Epidemiology 2005;15(2):153-159. |
R827355 (Final) R827355C008 (Final) R827355C009 (2002) R827355C009 (2003) R827355C009 (Final) |
|
|
|
Sullivan JH, Schreuder AB, Trenga CA, Liu SL, Larson TV, Koenig JQ, Kaufman JD. Association between short term exposure to fine particulate matter and heart rate variability in older subjects with and without heart disease. Thorax 2005;60(6):462-466. |
R827355 (Final) R827355C001 (Final) R827355C009 (2003) R827355C009 (Final) |
|
|
|
Sullivan J, Sheppard L, Schreuder A, Ishikawa N, Siscovick D, Kaufman J. Relation between short-term fine particulate matter exposure and onset of myocardial infarction. Epidemiology 2005;16(1):41-48. |
R827355 (Final) R827355C001 (2003) R827355C001 (Final) R827355C009 (2003) R827355C009 (Final) |
|
|
|
Trenga CA, Sullivan JH, Schildcrout JS, Shepherd KP, Shapiro GG, Liu L-JS, Kaufman JD, Koenig JQ. Effect of particulate air pollution on lung function in adult and pediatric subjects in a Seattle panel study. Chest 2006;129(6):1614-1622. |
R827355 (Final) R827355C002 (Final) R827355C009 (Final) |
|
ambient particles, fine particles, combustion, health, exposure, biostatistics, susceptibility, human susceptibility, sensitive populations, air toxics, genetic susceptibility, indoor air, indoor air quality, indoor environment, tropospheric ozone, California, CA, polyaromatic hydrocarbons, PAHs, hydrocarbons, acute cardiovascular effects, aerosols, air pollutants, air pollution, air quality, airborne pollutants, airway disease, airway inflammation, allergen, ambient aerosol, ambient aerosol particles, ambient air, ambient air quality, ambient particle health effects, animal model, assessment of exposure, asthma, atmospheric aerosols, atmospheric chemistry, biological markers, biological response, cardiopulmonary response, cardiovascular disease, children, children’s vulnerability, combustion, combustion contaminants, combustion emissions, epidemiology, exposure, exposure and effects, exposure assessment, harmful environmental agents, hazardous air pollutants, health effects, health risks, human exposure, human health effects, human health risk, incineration, inhalation, lead, morbidity, mortality, mortality studies, particle exposure, particle transport, particulates, particulate matter, risk assessment,
,
ENVIRONMENTAL MANAGEMENT, Air, Geographic Area, Scientific Discipline, Health, RFA, PHYSICAL ASPECTS, Susceptibility/Sensitive Population/Genetic Susceptibility, Risk Assessment, Risk Assessments, Air Pollutants, genetic susceptability, Northwest, Health Risk Assessment, Physical Processes, Epidemiology, Air Pollution Effects, air toxics, Atmospheric Sciences, Biochemistry, particulate matter, Environmental Chemistry, State, aerosols, exposure assessment, California (CA), exposure and effects, environmental hazard exposures, ambient air quality, cardiovascular disease, health effects, inhalation, mortality, epidemelogy, air quality, cardiopulmonary response, hazardous air pollutants, atmospheric aerosols, cardiopulmonary responses, human health risk, particle exposure, toxics, mortality studies, biomarker based exposure inference, acute cardiovascular effects, biostatistics, dose-response, human health effects, particulates, sensitive populations, toxicology, ambient particle health effects, air pollution, atmospheric chemistry, exposure, biomarker, human susceptibility, ambient aerosol, health risks, human exposure, Human Health Risk Assessment, morbidity, PM, particle transport
Relevant Websites:
http://depts.washington.edu/pmcenter/
Progress and Final Reports:
2002 Progress Report
2003 Progress Report
Original Abstract
Main Center Abstract and Reports:
R827355 Airborne PM - Northwest Research Center for Particulate Air Pollution and Health
Subprojects under this Center:
(EPA does not fund or establish subprojects; EPA awards and manages the overall grant for this center).
R827355C001 Epidemiologic Study of Particulate Matter and Cardiopulmonary
Mortality
R827355C002 Health Effects
R827355C003 Personal PM Exposure Assessment
R827355C004 Characterization of Fine Particulate Matter
R827355C005 Mechanisms of Toxicity of Particulate Matter Using Transgenic Mouse Strains
R827355C006 Toxicology Project -- Controlled Exposure Facility
R827355C007 Health Effects Research Core
R827355C008 Exposure Core
R827355C009 Statistics and Data Core
R827355C010 Biomarker Core
R827355C011 Oxidation Stress Makers