CAUSAL INFERENCE WITH A CONTINUOUS TREATMENT AND OUTCOME: ALTERNATIVE ESTIMATORS FOR PARAMETRIC DOSE-RESPONSE FUNCTIONS WITH APPLICATIONS.

Causal inference with a continuous treatment is a relatively under-explored problem. In this dissertation, we adopt the potential outcomes framework. Potential outcomes are responses that would be seen for a unit under all possible treatments. In an observational study where the treatment is continuous, the potential outcomes are an uncountably infinite set indexed by treatment dose. We parameterize this unobservable set as a linear combination of a finite number of basis functions whose coefficients vary across units. This leads to new techniques for estimating the population average dose-response function (ADRF). Some techniques require a model for the treatment assignment given covariates, some require a model for predicting the potential outcomes from covariates, and some require both. We develop these techniques using a framework of estimating functions, compare them to existing methods for continuous treatments, and simulate their performance in a population where the ADRF is linear and the models for the treatment and/or outcomes may be misspecified. We also extend the comparisons to a data set of lottery winners in Massachusetts. Next, we describe the methods and functions in the R package causaldrf using data from the National Medical Expenditure Survey (NMES) and Infant Health and Development Program (IHDP) as examples. Additionally, we analyze the National Growth and Health Study (NGHS) data set and deal with the issue of missing data. Lastly, we discuss future research goals and possible extensions.

Effects of Exposure to the Sound from Seismic Airguns on Pallid Sturgeon and Paddlefish - Data Set

Dataset for publication in PLOS One

LUNG CANCER IN NEPAL: THE ROLE OF TRADITIONAL TOBACCO PRODUCTS AND COMBUSTION RELATED HOUSEHOLD AIR POLLUTION

The burden of chronic diseases such as cancer is increasing in low and middle income countries around the globe. Nepal, one of the world’s poorest countries, is no exception to this trend, with lung cancer as the leading causes of cancer deaths. Despite this, limited data is available on the environmental and behavioral risk factors that contribute to the lung cancer etiology in Nepal. The objectives of this dissertation are to: 1) investigate the ethnic differences in consumption of local tobacco products and their role in lung cancer risk in Nepal; 2) evaluate urinary metabolite of 1,3-butadiene as a biomarker of exposure to combustion related household air pollution (CRHAP); 3) investigate the association between CRHAP exposure and lung cancer risk using urinary metabolite of 1,3-butadiene as a biomarker of exposure; 4) investigate the association between CRHAP exposure and lung cancer risk using questionnaire based measure of exposure. Lung cancer cases (n=606) and frequency matched controls (N=606) were recruited from B.P. Koirala Memorial Cancer Hospital. We obtained biological samples and information on lifestyles including cooking habits and type of fuels used. We used liquid chromatograph tandem mass spectrometer (LC-MS/MS) to quantify urinary metabolites of 1,3-butadiene in urine samples. We employed a combination of logistic and linear regression models to detect any exposure-disease associations while controlling for known confounding variables. Overall, we found that ethnic groups in Nepal use different tobacco products that have different differing cancer potency -we observed the highest odds ratios for the traditional tobacco products. The biomarker analysis showed strong evidence that monohydroxybutyl mercapturic acid is associated with biomass fuel use among participants. However, we did not find significant association between urinary MHMBA and lung cancer risk. When we used questionnaire based measure of exposure to household air pollution, we observed significant, dose-response associations between CRHAP exposure and lung cancer risk, particularly among never-smokers. Our results show that important role of local tobacco products in lung cancer risk in Nepal. Furthermore, we demonstrate that CRHAP exposure is a risk factor for lung cancer risk, independent of tobacco smoking.