Building on preliminary work, we propose to develop a new statistical methodology for epidemiologic studies, called Threshold Regression (TR), that will facilitate and enrich the assessment of health risks associated with occupational exposures. Threshold regression is based on the plausible hypothesis that the latent health status of a subject is a random process that is affected by disease and ultimately results in death or another adverse endpoint when the subject's latent health status falls below a critical threshold value. The methodology captures effects of occupational exposure in a disease progression scale that acts as a modified time scale for the health status process. The advantages of our proposed methodology over existing analytic methods used to assess disease risk associated with occupational exposures are (a) it can be applied to both quantitative and semi-quantitative exposure data; (b) unlike traditional survival methods that depend heavily on the assumption of proportional hazards, the proposed model does not depend on this assumption; (c) it can take account of multiple competing health risks in a natural manner; (d) it provides new types of graphs that display survival trends from new perspectives. This application addresses Risk Assessment Methods listed in the """"""""Research Tools and Approache's category of NIOSH/NORA Priority Research Areas. This project is aimed at contributing to applied quantitative methods and statistical research that can model the stochastic relationship between work-related environmental exposure and cancer risk, The model can be used for elucidation of susceptibility factors associated with cancer risk in individuals and population subgroups.
Specific aims are: ? 1. Develop threshold regression into a sound practical tool for occupational risk assessment. ? 2. Quantify hazards and the dose-response relationship. ? 3. Apply the methodology to two large occupational studies. ? 4. Develop requisite computational software tools. ? ? ?
|Erich, Roger; Pennell, Michael L (2015) Ornstein-Uhlenbeck threshold regression for time-to-event data with and without a cure fraction. Lifetime Data Anal 21:1-19|
|Whitmore, G A; Zhang, Guangyu; Lee, Mei-Ling Ting (2012) Constructing normalcy and discrepancy indexes for birth weight and gestational age using a threshold regression mixture model. Biometrics 68:297-306|
|Lee, Mei-Ling Ting; Whitmore, G A; Rosner, Bernard A (2010) Threshold regression for survival data with time-varying covariates. Stat Med 29:896-905|
|Weisskopf, M G; Knekt, P; O'Reilly, E J et al. (2010) Persistent organochlorine pesticides in serum and risk of Parkinson disease. Neurology 74:1055-61|
|Pennell, Michael L; Whitmore, G A; Ting Lee, Mei-Ling (2010) Bayesian random-effects threshold regression with application to survival data with nonproportional hazards. Biostatistics 11:111-26|
|Lee, Mei-Ling Ting; Whitmore, G A (2010) Proportional hazards and threshold regression: their theoretical and practical connections. Lifetime Data Anal 16:196-214|
|Yu, Zhangsheng; Tu, Wanzhu; Lee, Mei-Ling Ting (2009) A semi-parametric threshold regression analysis of sexually transmitted infections in adolescent women. Stat Med 28:3029-42|
|Lee, Mei-Ling Ting; Whitmore, G A; Laden, Francine et al. (2009) A case-control study relating railroad worker mortality to diesel exhaust exposure using a threshold regression model. J Stat Plan Inference 139:1633-1642|
|He, Xin; Tong, Xingwei; Sun, Jianguo (2009) Semiparametric analysis of panel count data with correlated observation and follow-up times. Lifetime Data Anal 15:177-96|
|Lee, Mei-Ling Ting; Chang, Mark; Whitmore, G A (2008) A threshold regression mixture model for assessing treatment efficacy in a multiple myeloma clinical trial. J Biopharm Stat 18:1136-49|