Three Essays on Health Outcomes

Citation:

Three Essays on Health Outcomes

We find that while all ages experienced an increase in the prevalence of those overweight and/or obese, the prevalence among young adults has grown at a faster rate than that of older age groups. We find that the increases in body mass index are primarily due to period effects, not cohort or age effects. From the ordered logistical regression analyses, we find that protective influence of factors such as education, income, and age on an individual’s body mass index have, decreased over time. The analyses suggest that the increase in the prevalence of those overweight or obese is a phenomenon experienced by all demographic groups in the U.S.

Chapter 2 uses data from a health survey, the Behavior Risk Factor Surveillance System, for the years 1994 through 2000 to analyze the effect of employment increases due to welfare reform (Temporary Assistance for Need Families) on body weight and stress in single mothers. A difference-in-difference approach is used to analyze the survey, supplemented by state-level welfare and unemployment information. Additionally, as concern exists regarding the presence of incorrect standard errors when using such a model, I attempt to adjust said errors by correcting the asymptotics. I find that even by using a health survey, welfare reform does result in significant employment increases for single mothers, especially minorities. I find no relationship between welfare reform and either body weight or stress. The analysis suggests that employment increases due to welfare reform did not contribute to the obesity epidemic in the adult U.S. women.

I continue my study of health outcomes in Chapter 3. Utilizing Office of Statewide Health Planning and Development data for 2000 through 2005, I uses the 1999 California nurse-topatient ratio legislation to analyze the effect of increases in nurse staffing on adverse patient health outcomes for general medical/surgical hospital units. This law provides a natural experiment where the treatment effect varies across hospital units depending on the difference between the unit’s pre-legislated ratio (in 2000) and the mandated ratio (in 2005). I use this variation as the basis for an instrumental variables methodology. I find that nurse-to-patient ratios increase significantly for hospital units that were most in need of raising their ratios to become compliant with the 2005 mandated ratios. However, I find no evidence that said increases had the desired effect of reducing adverse health outcomes in the population.

Additionally, I explore the systematic differences of various hospital sub-sets, and attempt to isolate which sub-sets are actually affected by the legislation. The two sub-sets that I use are by region and by type of hospital financial control. My results depict no evidence that adverse patient health outcomes were either positively or negatively affected for various sub-sets of the sample. Finally, as hospital management claim considerable financial concerns with implementation of this law, I analyze the effect of AB 394 on three economic indicators. My results suggest that the legislation had no significant economic consequences for those hospitals most affected by the law. However, I consider my economic results inconclusive, as I am unable to determine whether hospitals were forced to redistribute resources in order to comply with the law. If such financial restructuring did occur, there is the potential for unintended consequences that I have been unable to pick-up.

This dissertation explores two topics related to health outcomes in the adult U.S. population.Over the past fifteen years, the United States has seen a decrease in overall health. Two potentialreasons behind this deterioration in physical condition are the rise in overweight and obesity, andincreases in adverse health outcomes for individuals requiring hospital stays. My researchattempts to comprehend the extent to which these two causes are contributors in said health declines.

My dissertation focuses on three aspects of health outcomes: (1) identifying who among the adult U.S. population has been affected by the obesity epidemic; (2) whether employment increases played a role in the rise of obesity in single women; and (3) do increases in nurse staffing reduce adverse health outcomes for the inpatient hospital population.

Chapter 1 uses data from the National Health Interview Survey for years spanning 1976 and 2001 to present an age-period-cohort analysis of weight gain throughout the life course. This is a joint project with Beth Osborne Daponte, and has been published in the spring 2008 edition of Population Research and Policy Review

• View Full Text (.pdf)

Adobe .pdf files require the Adobe Acrobat Reader.
Download and install the Adobe Acrobat Reader.