Abstract

Clinical decisions made in the delivery setting are often made under high pressure, great uncertainty, and have serious consequences for mother and baby. Theories of decision-making suggest that individuals in such settings may resort to using “heuristics”, or simplified decision-rules, to aid complex decision-making. This study investigates whether physicians’ delivery-mode decisions (i.e., when to perform a vaginal vs. a cesarean) are influenced by such a heuristic. Electronic health record data spanning 86,000 deliveries suggests that, if the prior patient had complications in one delivery-mode, the physician will be more likely to switch to the other -- and likely inappropriate -- delivery-mode on the subsequent patient, regardless of patient indication. There is evidence that this heuristic has small, suboptimal effects on patient health.

Abstract

A growing literature has documented racial disparities in health care. We argue that racial disparities may be magnified when hospitals operate at capacity, when behavioral and structural conditions associated with poor patient outcomes – e.g., limited provider cognitive bandwidth or reliance on biased care algorithms – are aggravated. Using detailed, time-stamped electronic health record data from two large hospitals, we document that in-hospital mortality increased more for Black patients than for White patients when hospitals approached capacity. We estimate that 8.5% of Black patient deaths were capacity-driven and thus avoidable. We then investigate the extent to which differential care inputs explain our findings. While strain exacerbated wait times similarly for Black and White patients, Black patients both waited the longest at high strain and faced greater mortality consequences from prolonged wait times. Finally, the largest racial disparities in mortality were among women and uninsured patients, highlighting biases in provider behavior and hospital processes as key mechanisms driving our results.

Abstract

Feedback interventions such as report cards are often used in healthcare to encourage physicians to improve performance. However, providing feedback is tricky because it often conveys negative information to individuals about their performance, which can induce behavioral responses (e.g., dejection, confusion, attributional errors, etc) that interfere with learning and improvement. In this paper, we use novel data from a surgeon report card (linked to 320,000 surgeries) to examine surgeon response to positive vs. negative feedback, and to identify which surgeons improve most from such feedback. Exploiting two sources of plausibly exogenous variation in report card information, we highlight two results: i) consistent with behavioral responses to feedback, patient outcomes improve for surgeons receiving positive feedback and deteriorate for those receiving negative feedback, and ii) the largest returns to the feedback accrue to those already skilled at learning from experience ex-ante (i.e., the inframarginal physician) because they respond more effectively to both positive and negative feedback. These results are replicated in the lab, where we explore mechanisms and test interventions to attenuate suboptimal behavioral responses, such as providing help interpreting feedback and adding improvement-based incentives. Overall, report cards should be carefully designed to avoid triggering counterproductive behavioral responses to negative feedback.

Abstract

Power, defined as the asymmetric control of valued resources, affects most human interactions. Yet there is little observational evidence on how power affects real-world behavior and resource allocation. We examine this question using the power differential in the doctor-patient encounter: while it favors the physician in the clinical setting, powerful patients may be able to reduce this asymmetry and influence physician behavior. We exploit the quasi-exogenous assignment of 1.5 million patients to physicians in US military emergency departments, using the difference in their military ranks to measure their power differential. We find that power confers nontrivial advantage to its possessor: “high-power” patients (those who outrank their physician) receive greater physician effort and have better outcomes than equivalently-ranked “low-power” patients. Furthermore, within-physician effort is higher for patients recently promoted than those about to be promoted. We document negative spillovers from a physician's high-power patients to their concurrently seen low-power patients, as well as predictable interactions of such power dynamics with doctor-patient concordance on race and sex. While power-driven variation in behavior is often undesirable, it is especially concerning in healthcare where it can harm society's most vulnerable patients.

Abstract

Limited ICU capacity has been said to exacerbate several consequences of COVID-19, such as mortality, misallocation of resources by physicians, and harmful spillovers on nonCOVID patients. However, whether greater ICU capacity would have avoided these pitfalls is not clear. The goal of this paper is to shed light on how ICU capacity affects physician decision-making and patient welfare when the healthcare system is strained. We use two sources of variation in ICU capacity to estimate its causal effect on physician thresholds for admitting patients to the ICU: i) ICU “expansions”, resulting from increases in hospital number of ICU beds, and ii) ICU “strain”, resulting from random fluctuations in ICU bed availability. Our analysis uses 100% inpatient EHR data from two hospitals (150K encounters. 2015-18) both of which expanded their ICUs separately. Importantly, we use lab test results to assign patients an objective and validated measure of ICU-need, called the eSOFA score. We motivate our empirical analyses using a model of physicians’ dynamic admission decisions when there is uncertainty about a patient’s ICU-need. Results show that increases in capacity cause physicians to lower their thresholds for ICU admission, with mixed effects on patient welfare. Patients are 1.8 pp more likely to be admitted to the ICU post-expansion (even when ICU strain is at pre-expansion levels), but there is significant heterogeneity in likelihood of admission and in-hospital mortality by ICU strain and ICU-need. Expanding ICU capacity does not always help the patients who need it most, and at times harms patients who need it least. Finally, the greatest benefits of ICU expansions accrue to patients in the general wards by allowing physicians to better allocate non-ICU resources, especially when the hospital is operating at capacity.