Adverse Selection and Switching Costs in Health Insurance Markets: When Nudging Hurts, by Benjamin R. Handel (NBER)
This paper investigates consumer switching costs in the context of health insurance markets, where adverse selection is a potential concern. Though previous work has studied these phenomena in isolation, they interact in a way that directly impacts market outcomes and consumer welfare. Our identification strategy leverages a unique natural experiment that occurred at a large firm where we also observe individual-level panel data on health insurance choices and medical claims. We present descriptive results to show that (i) switching costs are large and (ii) adverse selection is present. To formalize this analysis we develop and estimate a choice model that jointly quantifies switching costs, risk preferences, and ex ante health risk. We use these estimates to study the welfare impact of an information provision policy that nudges consumers toward better decisions by reducing switching costs. This policy increases welfare in a naive setting where insurance plan prices are held fixed. However, when insurance prices change endogenously to reflect updated enrollee risk pools, the same policy substantially exacerbates adverse selection and reduces consumer welfare, doubling the existing welfare loss from adverse selection.
Morbidity and mortality of radical prostatectomy differs by insurance status, by Quoc-Dien Trinh et al. (Cancer)
BACKGROUND: Private insurance status may favorably affect various health outcomes including those associated with radical prostatectomy (RP). We explored the effect of insurance status on 5 short-term RP outcomes.
METHODS: Within the Health Care Utilization Project Nationwide Inpatient Sample (NIS) we focused on RPs performed within the 5 most contemporary years (2003-2007). We tested the rates of blood transfusions, extended length of stay, intraoperative and postoperative complications, as well as in-hospital mortality, stratified according to insurance status. Multivariable logistic regression analyses, fitted with general estimation equations for clustering among hospitals, adjusted for confounding factors.
RESULTS: Overall, 61,167 RPs were identified. Of those, private insurance accounted for the majority of cases (n = 41,312, 67.5%), followed by Medicare (n = 18,759, 30.7%) and Medicaid (n = 1096, 1.8%). Insurance status other than private was associated with higher rates of blood transfusions (P < .001), higher overall postoperative complication rates (P < .001), higher rates of hospital stay above the median (P < .001), as well as higher in-hospital mortality (P = .01). In multivariable analyses, compared with patients with private insurance, Medicaid patients had higher rates of blood transfusion (odds ratio [OR] = 1.45, P < .001), length of stay beyond the median (OR = 1.61, P < .001) postoperative complications (OR= 1.24, P = .02), and in-hospital mortality (OR = 4.91, = .01). Similarly, Medicare patients had higher rates of blood transfusions (OR = 1.21, P < .001), overall postoperative complications (OR = 1.17, P×< .001) and length of stay beyond the median (OR = 1.25, P < .001).
CONCLUSIONS: Even after adjusting for confounding factors, patients with private insurance have better outcomes than their counterparts with nonprivate insurance.