Contemporary Issues in Population Health: Disability, Pain Management, and Opioid Overdose in the United States

Abstract

This dissertation addresses several contemporary health challenges affecting the US population, drawing on methods and perspectives from health services research, social demography, and medical sociology. Paper 1 examines the pathways from educational attainment to difficulty with activities of daily living (ADLs) and instrumental activities of daily living (IADLs), helping to elucidate the means by which education shapes health outcomes throughout life. Less educated individuals experience higher rates of disability than those with higher educational attainment, and this disparity appears to be growing at both middle and older ages. Yet the mechanisms underlying it are not well understood. In this study, I use nationally representative data from the Panel Study of Income Dynamics to estimate the contributions of three mediators—excess body mass index (BMI), cigarette smoking, and manual labor—to educational disparities in ADL/IADL disability incidence. Disparities are evident in both younger and older adults (33-64 years, 65-96 years) and larger in women. At younger ages, these factors account for an estimated 60-70% of disparities in disability incidence between the most and least educated. Among women ages 65 and over, they account for nearly 40% of that disparity. Estimates in older men are more variable, suggesting an explanatory power of 20 to 60%.

Papers 2 and 3 speak to the US opioid crisis, which resulted in nearly 47,000 deaths in 2018. Paper 2 centers on unequal treatment in pain management. Research has shown that black and Hispanic chronic pain patients are less likely than their white counterparts to receive opioid prescriptions, and there is evidence that provider bias contributes to this disparity. Using Optum healthcare claims data, I investigate whether the 2016 CDC Guideline for Prescribing Opioids for Chronic Pain influenced seven measures of opioid prescribing—and, if so, whether these changes varied by patient race/ethnicity, thereby influencing preexisting disparities. Across racial/ethnic groups, the guideline is associated with a gradual decline in the frequency of opioid prescribing, the average daily dose prescribed, and the frequency of high-dose prescribing. However, point estimates suggest these declines may have tended to be sharper in black and Hispanic patients, potentially amplifying black/white and Hispanic/white disparities in prescribing. While the guideline aimed to provide clinicians an evidence-based framework to inform opioid prescribing decisions, its emphasis on providers’ use of personal judgment to determine each patient’s prescribing needs may have increased reliance on racial stereotypes and bias amid clinical uncertainty.

Finally, Paper 3 asks how the scarce resources available to reduce opioid-related deaths can be allocated for maximum health benefit. Distribution of naloxone for reversing opioid-related overdose reduces mortality, but it is not known whether distribution remains cost-effective in the context of rising naloxone prices and fentanyl-related overdose rates. Moreover, it is unclear whether distribution to people likely to witness or experience overdose (“laypeople”), police and firefighters, or emergency medical services (EMS)—as well as combinations of these—are equally cost-effective strategies. I conduct a cost-effectiveness analysis examining these questions and find that high distribution to all three target groups minimized overdose deaths, averting 21% of fatalities compared to “bare minimum” distribution. This strategy is highly cost-effective from a health sector perspective, cost-saving from a societal perspective, and robust to hypothetical moral hazard. The results suggest that communities with insufficient resources for distributing to all three groups should prioritize distribution to laypeople and EMS.