This project will develop statistical methods relevant to two common forms of environmental epidemiologic studies. The primary goal is to provide methods that extract a concise assessment of health risks associated with environmental exposures, supplemented by appropriate statistical inference. The first topic will evaluate the association between exposure and the risk of a health outcome using diagnosis data on a cohort of individuals supplemented with screening information on undiagnosed participants. The methods will be applied to data from the Seveso Women's Health Study which addresses health risks in women exposed to high levels of dioxin. It is intended that the statistical methods will apply generally to similar studies that include a combination of diagnostic and screening data. The second project concerns statistical techniques to investigate the effects of multiple environmental exposures on health and developmental outcomes. The ideas will be applied to data from the CHAMACOS study of Latino women and their children in California, where information has been collected on environmental (largely pesticide) exposures, in utero and in childhood, for a cohort of women and their infants. Statistical issues involve estimation and ranking-in importance-of suitable causal effects of each exposure, supplemented by a rigorous assessment of which of these represent real effects rather than spurious associations, allowing appropriately for multiple comparisons. Both studies involve the study of vulnerable populations exposed to above average environmental exposures with the potential for elevated risk for poor health outcomes. Statistical and computational algorithms will be developed and provided in an open source user-friendly format allowing their rapid dissemination and use by other investigators. The relevance to public health is two-fold: first, the research will allow environmental epidemiologists to accurately describe the effects of (i) acute dioxin exposure on the reproductive health of women, in particularly on the onset of fibroids, and of (ii) pesticide exposures on birth outcomes and subsequent neurodevelopment of children born to Latino women, in a farmworking community. Second, the proposed research will provide appropriate statistical tools and software to allow other investigators to apply these complex methods to similar studies of the effects of environmental exposures in a wide variety of settings.

Agency
National Institute of Health (NIH)
Institute
National Institute of Environmental Health Sciences (NIEHS)
Type
Research Project (R01)
Project #
5R01ES015493-04
Application #
7799361
Study Section
Biostatistical Methods and Research Design Study Section (BMRD)
Program Officer
Dilworth, Caroline H
Project Start
2007-06-15
Project End
2013-03-31
Budget Start
2010-04-01
Budget End
2013-03-31
Support Year
4
Fiscal Year
2010
Total Cost
$275,430
Indirect Cost
Name
University of California Berkeley
Department
Biostatistics & Other Math Sci
Type
Schools of Public Health
DUNS #
124726725
City
Berkeley
State
CA
Country
United States
Zip Code
94704
Heggeseth, Brianna C; Jewell, Nicholas P (2013) The impact of covariance misspecification in multivariate Gaussian mixtures on estimation and inference: an application to longitudinal modeling. Stat Med 32:2790-803
Stitelman, Ori M; De Gruttola, Victor; van der Laan, Mark J (2012) A general implementation of TMLE for longitudinal data applied to causal inference in survival analysis. Int J Biostat 8:
Wester, C William; Stitelman, Ori M; deGruttola, Victor et al. (2012) Effect modification by sex and baseline CD4+ cell count among adults receiving combination antiretroviral therapy in Botswana: results from a clinical trial. AIDS Res Hum Retroviruses 28:981-8
Stitelman, Ori M; Wester, C William; De Gruttola, Victor et al. (2011) Targeted maximum likelihood estimation of effect modification parameters in survival analysis. Int J Biostat 7:19
Stitelman, Ori M; van der Laan, Mark J (2010) Collaborative targeted maximum likelihood for time to event data. Int J Biostat 6:Article 21
McKeown, Karen; Jewell, Nicholas P (2010) Misclassification of current status data. Lifetime Data Anal 16:215-30
Park, June-Soo; Holden, Arthur; Chu, Vivian et al. (2009) Time-trends and congener profiles of PBDEs and PCBs in California peregrine falcons (Falco peregrinus). Environ Sci Technol 43:8744-51
Young, Jessica G; Jewell, Nicholas P; Samuels, Steven J (2008) Regression analysis of a disease onset distribution using diagnosis data. Biometrics 64:20-8
Hubbard, Alan E; Laan, Mark J VAN DER (2008) Population intervention models in causal inference. Biometrika 95:35-47
VAN DER Laan, Mark J; Jewell, Nicholas P (2003) CURRENT STATUS AND RIGHT-CENSORED DATA STRUCTURES WHEN OBSERVING A MARKER AT THE CENSORING TIME. Ann Stat 31:512-535