Three essays on the impact of behavioral risk factors on health and mortality in the United States
This dissertation includes three chapters on the impact of behavioral risk factors on health and mortality in the United States. Chapter one analyzes interstate variation in cause-specific mortality rates to indirectly infer the role of risk factors in mortality patterns from 2000-2009. Specifically, we run a factor analysis on a death-rate-by-state matrix for ages 20-64 for men and women separately. We find three factors underlying interstate mortality variation that are pervasive in both men and women: smoking, substance abuse, and urban/rural residence. Chapter 2 estimates the impact of prescription drug abuse on mortality in the United States. Current estimates of mortality attributable to prescription drugs only consider directly attributable causes of death like overdoses. These are underestimates, as prescription drugs can influence many causes of death including liver failure and accidental injuries. I use prescription drug overdose deaths as an indicator of the damage done by prescription drug abuse, and then model the relationship between prescription drug overdose deaths and all other deaths to estimate the effect of prescription drug abuse on mortality for all causes. By modeling all cause mortality as a function of mortality from prescription drug overdoses, estimates of deaths attributable to prescription drug abuse increase by 73% for men and 87% for women in 2013 for non-Hispanic whites at ages 20-54. These deaths accounted for over 7% of all deaths at ages 20-54 in 2013 for non-Hispanic whites. The third chapter analyzes the relationship between alcohol consumption and body mass index (BMI). Obesity has emerged as one of the most important risk factors impacting health and mortality, yet there is uncertainty about the relationship between alcohol consumption and BMI. Using data from the 2003-2004 and 2005-2006 National Health and Nutrition Examination Survey, I find that alcohol consumption is associated with both higher calorie intake and lower BMI. I find that this paradox is partially explained by confounding variables including smoking and physical activity and largely explained, especially in men, by differences in serum insulin levels. This research has important implications for the effect of alcohol on weight, health, and mortality.^
Tencza, Christopher J, "Three essays on the impact of behavioral risk factors on health and mortality in the United States" (2015). Dissertations available from ProQuest. AAI3722013.