Working Papers

Patient Peer Effects: Evidence from Nursing Home Room Assignments (with Martin Hackmann)

Media Mentions: NBER Bulletin on Health, McKnights; Gies College of Business

Awards: Best Paper in Aging and Long-Term Care at ASHEcon 2026

We provide causal evidence that patient peer effects generate mortality impacts comparable to provider quality differences. Drawing on administrative records covering 2.6 million stays (2000–2010) across 7,200 U.S. nursing homes, we exploit plausibly exogenous roommate assignments identified through unique room identifiers. We estimate that assignment to a roommate diagnosed with Alzheimer's disease (AD) or Alzheimer's disease related dementias (ADRD), relative to placement in a private room, increases 90-day mortality by 2.1 percentage points (14% of baseline)—equivalent to receiving care at a nursing home one full standard deviation worse in quality. Effects differ sharply by patient type: patients with AD/ADRD benefit substantially from cognitively healthy roommates but not from private rooms, suggesting important peer monitoring and support roles. In contrast, mortality of patients without AD/ADRD does not depend on roommate cognitive health but is reduced in private rooms. A simple assignment rule exploiting this heterogeneity could reduce overall mortality by 0.8 percentage points without additional resources.

Barriers to Choosing High-Quality Healthcare Providers: Evidence from the Nursing Home Market

This paper studies demand- and supply-side explanations for why consumers often choose low-quality providers. Using rich administrative data, I estimate wide quality dispersion across nursing homes in California based on risk-adjusted mortality. Structural estimates show very low average responsiveness to quality, with substantial heterogeneity, even after accounting for choice set constraints due to cream-skimming by nursing homes. After the star ratings were introduced, responsiveness rose heterogeneously, and nursing homes responded strategically, both in their quality investments and cream-skimming behavior. Counterfactual simulations show that increased responsiveness, quality-ranked shortlists from discharge planners, or stronger quality incentives may yield substantially larger benefits.

Racial Segregation and Disparities Across Nursing Homes: Mechanisms, Interactions, and Policy Tradeoffs

Racial minorities are disproportionately admitted to lower-quality nursing homes, but the forces generating segregation across facilities need not be the same as those generating racial gaps in quality. Using nationwide administrative data on nursing home residents in the U.S. from 2000–2010, I show that neighborhood sorting explains an important share of across-facility segregation, yet substantial disparities remain in the quality of nursing homes that different racial groups are admitted to, even conditional on fixed effects for zip code of prior residence. I then document three additional mechanisms that may drive across-facility segregation and disparities: (i) race-based admissions policies, where nursing homes become less likely to admit minority residents when capacity is strained; (ii) in-group preferences, whereby shocks to the minority share at nursing homes lead to persistent changes in the racial composition, and (iii) heterogeneity in responsiveness to quality, where information shocks lead to larger changes in where White residents are admitted. To quantify the contributions of these explanations, I estimate a two-sided matching model with latent choice set constraints in three large states, using distance and short-run within-facility occupancy shocks for identification. Counterfactual simulations show that neighborhood sorting and race-based admissions policies are complementary drivers of segregation: while the sum of their individual contributions accounts for just over one-half of segregation, eliminating both simultaneously reduces segregation by 85–90 percent. In contrast, the drivers of disparities do not always align with drivers of segregation and vary substantially across states as well as between Black-White and Hispanic-White comparisons within states, implying that policies which reduce segregation need not reduce—and may even worsen—disparities for some minority groups.

Estimation of Regression Discontinuity and Kink Designs with Multiple Running Variables [R Code]

In regression discontinuity designs with multiple running variables (MRD), units are assigned different treatments based on whether their values on several observed running variables exceed known thresholds. In such designs, applied work commonly analyzes each running variable separately, estimating a single-dimensional RD design in the first running variable after limiting the sample to the set of individuals qualifying on the second threshold, and vice versa. In this paper, I propose a new estimator for MRD designs using thin plate splines that improves upon the applied practice in two ways. First, the estimator can be used to estimate the conditional average treatment effect at every point on the boundary separating treated and untreated units, and second, it provides efficiency gains by using the entire sample. I also develop analogous estimators for multidimensional regression kink (MRK) and multidimensional regression discontinuity/kink designs (MRDK). I show consistency of these estimators and construct asymptotically valid confidence intervals (CIs), before presenting simulation results showing that they produce estimates and CIs that perform well in finite samples. Finally, I demonstrate the performance of my MRD estimator with two empirical applications: Londoño-Vélez, Rodríguez, and Sánchez (2020) on the effect of financial aid on college enrollment, and Keele and Titiunik (2015) on the effect of political ads on election turnout. R code for estimation and inference will soon be available.

Selection on Unobservables in Discrete Choice Models

Selection on unobservables is an important concern for causal inference in observational studies, and accordingly, previous papers have developed methods for sensitivity analysis for OLS, binary choice models, instrumental variables, and movers designs. In this paper, I develop methods for sensitivity analysis for a setting that has not been previously studied — discrete choice models. In particular, I derive bounds for the omitted variables bias under an assumption about how much the consumer values the omitted variable(s) relative to the included control variables, and about the relationship between the omitted variable and the variable of interest. I provide theoretical results for my bounding procedure, and demonstrate its performance in simulations. Finally, I show in several empirical applications that my procedure produces economically meaningful bounds.

Research in Progress

Peer Effects in the Workplace: Evidence from the Illinois Workplace Wellness Study (with Damon Jones, David Molitor, and Julian Reif)

We study peer effects in health screening behavior using a large-scale field experiment in which individuals were randomly assigned to a workplace wellness program. Friendship networks were measured at baseline and at two annual follow-ups. We document substantial homophily in the baseline network: individuals are significantly more likely to be friends with others who share similar observable characteristics, underscoring the importance of randomization for identification of peer effects. We find that having friends assigned to the treatment group has a positive causal effect on program participation. Using treatment status as an instrument, we estimate that each additional friend induced to participate by treatment increases an individual’s probability of participating in health screenings by over 5 percentage points. Finally, we show that treatment assignment causally influenced both the structure of the network and the degree of homophily in the two years following randomization.

Economics of Threshold-Based Regulations: Evidence from Nursing Home Staffing Mandates (with Riley League)

Minimum staffing mandates are a prominent feature of the nursing home industry, where quality of care is an issue of great concern to policymakers. Yet, the effects of these regulations are often unclear, given strategic responses by nursing homes. In this paper, we develop a novel and simple method of estimating the effects of regulation on firm behavior and the regulated variable in a setting where firms are able to avoid a regulation by manipulating their size. Comparing the distribution of facility size in California, which has a staffing requirement that applies only to facilities with at least 100 beds, to those in other states without such a requirement, we document extensive bunching---facilities in California are 70 times more likely to be just below the threshold compared to just meeting the threshold, with no such pattern for facilities in control states. We find that the excess bunching is almost entirely driven by for-profit facilities, and that the size distortions induced by the regulation limits access, particularly for Black and Medicaid-insured patients. Moreover, given that size distortions are concentrated among facilities with low staffing, we estimate that the regulation has little effect on staffing.

Challenges in Measuring Mental Health Trends

Public awareness of mental health issues has grown in recent years, and there is a common perception that mental health in the population is worsening, concerns that are supported by descriptive evidence. However, changes in public attitudes towards mental health make the interpretation of these trends challenging: low response rates to mental health surveys make their results sensitive to changes in sample selection bias, and even diagnosis rates in comprehensive data sets such as the Medicare data are a function of individuals' willingness to seek professional help. In this paper, I address these measurement issues using a comprehensive data set on all nursing home residents, so it does not suffer from sample selection bias. Similar to Medicare data, it contains information on whether each resident was diagnosed with depression in the recent past, but in addition, it contains a rich set of psychosocial measures, which provides us with a detailed picture of the resident's underlying mental health. I find that while depression diagnoses at admission increased from 19 to 24 percent for residents admitted to a nursing home in California between 2000 and 2010, underlying mental health of these residents (based on observed psychosocial behavior as well as machine-learning predictions using hundreds of covariates) was roughly constant over the same period. These results illustrate the perils of inferring mental health trends from survey evidence or diagnosis trends alone. 

Past Work

Cigarette Consumption and Tax Salience
This paper studies how cigarette consumption responds over time to changes in tax rates. Using a panel of state data, I estimate that the cumulative effect of an excise tax rise on consumption is larger than the cumulative effect of an increase in sales tax, in line with a theory of tax salience. In addition, I find that consumption falls in advance of an excise tax hike, whereas it only falls in the year after a sales tax increase. The pattern of consumption response to sales taxes is also consistent with consumer learning over time.