Please note: All research in progress seminars are off-the-record. Any information about methodology and/or results are embargoed until publication.

Abstract:

For forty years, the Tuskegee Study of Untreated Syphilis in the Negro Male passively monitored hundreds of adult black males with syphilis despite the availability of effective treatment. The study's methods have become synonymous with exploitation and mistreatment by the medical community. We find that the historical disclosure of the study in 1972 is correlated with increases in medical mistrust and mortality and decreases in outpatient physician interactions for black men. Blacks possessing prior experience with the medical community, including veterans and women, appear to have been less affected by the disclosure. Our findings relate to a broader literature on how beliefs are formed and the importance of trust for economic exchanges involving asymmetric information.

Jointly with Marianne Wanamaker

Please note: All research in progress seminars are off-the-record. Any information about methodology and/or results are embargoed until publication.

Abstract

In the private market for Medicare supplemental insurance, also known as Medigap, policymakers have experimented with several regulatory solutions, including an initial open enrollment period, guaranteed renewal, bans on differential pricing, and bans on rejections. In this paper, I study how bans on differential pricing and rejections affect premiums and coverage levels, compared to a regime that combines an initial open enrollment period with guaranteed renewal. I document two important effects. First, bans on differential pricing and rejections lead to substantial cross-subsidization from young to old. Under a ban on differential pricing, the youngest buyers see premiums that are $240 (16 percent) higher; when this is combined with a ban on rejections, the youngest buyers see premiums that are $640 (36 percent) higher. Second, a ban on rejections undoes consumers’ incentives to buy early. A ban on differential pricing and rejections leads to a 12 percentage point (46 percent) reduction in early buying. I present evidence for the importance of this mechanism, which is often assumed in the theoretical literature but seldom documented empirically. This interpretation is corroborated by an event study of individuals who experience health shocks.

Please note: All research in progress seminars are off-the-record. Any information about methodology and/or results are embargoed until publication.

Abstract

Using modeling methodologies, at least three groups have suggested that the high expense of US healthcare is justified by the systematic increase in US life expectancy over the last 60 years. Papers describing these models are frequently cited in both the academic literature and in policy briefs. In this analysis, assumptions underlying the three models are linked to recent systematic literature reviews. Using estimates based on recent RCTs, the models are reconstructed and subjected to sensitivity analysis. The results suggest that the benefits of high-technology interventions have been overestimated, while the effects of social and behavioral factors, including cigarette smoking cessation, may have been underestimated. The analysis is highly sensitive to assumptions about the percentage of variance in outcomes attributable to medical technology.

Bio

Robert M. Kaplan, is currently a Fellow at the Center for Advanced Studies in the Behavioral Sciences at Stanford University, where he works with Stanford’s Clinical Excellence Research Center (CERC). He has served as Chief Science Officer at the US Agency for Health Care Research and Quality (AHRQ) and Associate Director of the National Institutes of Health, where he led the behavioral and social sciences programs.  He was formerly Distinguished Professor of Health Services and Medicine at UCLA, where he led the UCLA/RAND AHRQ health services training program and the UCLA/RAND CDC Prevention Research Center. He was Chair of the Department of Health Services from 2004 to 2009.  From 1997 to 2004 he was Professor and Chair of the Department of Family and Preventive Medicine, at the University of California, San Diego. He is a past President of several organizations, including the American Psychological Association Division of Health Psychology, Section J of the American Association for the Advancement of Science (Pacific), the International Society for Quality of Life Research, the Society for Behavioral Medicine, and the Academy of Behavioral Medicine Research. Kaplan is a former Editor-in-Chief of Health Psychology and of the Annals of Behavioral Medicine.  His 20 books and over 500 articles or chapters have been cited more than 28,000 times and the ISI includes him in the listing of the most cited authors in his field (defined as above the 99.5th percentile). Kaplan is an elected member of the National Academy of Medicine (formerly the Institute of Medicine).

Please note: All research in progress seminars are off-the-record. Any information about methodology and/or results are embargoed until publication.

"Modeling Disease for Effective Control: Tuberculosis in India"

Please note: All research in progress seminars are off-the-record. Any information about methodology and/or results are embargoed until publication.

Abstract:

Simulation and optimization frameworks that incorporate individual heterogeneity can be powerful tools to inform health policy decisions, particularly decisions about how to efficiently control infectious diseases in resource-constrained settings.  We apply such models to assess policies for control of tuberculosis (TB) in India, where more than two million people have TB.

We first use a microsimulation model to uncover the changing dynamics of drug-resistant (DR) TB. We find that nearly half of new DR TB cases in India are transmission-generated, as opposed to treatment-generated, and we project this proportion to continue to rise, implying that strategies that disrupt DR transmission may provide greater DR prevalence reductions over time.  We then incorporate healthcare costs into the simulation and find that both new diagnostics and institutional reform policies that refer patients in informal, private TB clinics to public clinics using approved treatment regimens would both be cost-effective ways of combatting TB in India. However, these institutional reforms should be prioritized if insufficient resources are available to implement both types of policies nationally. Building on the microsimulation results, we use dynamic programming methods to design patient-specific DR TB testing algorithms that can reduce over-testing, reduce costs, and quickly identify DR TB patients. We estimate that the optimal DR TB testing algorithm identified by our analysis will decrease healthcare costs by an average of $4000 per patient by averting downstream transmission.

"An Efficient Gaussian Approximation and Regression Metamodeling Approach to Value of Information Analysis"

Please note: All research in progress seminars are off-the-record. Any information about methodology and/or results are embargoed until publication.

Abstract

Value of information (VOI) analysis is based on statistical decision theory, and has recently gained increased recognition for its potential application in resource allocation, research prioritization, and future data collection designs.  However, VOI remains underutilized due to many conceptual, mathematical and computational challenges of implementing Bayesian decision theoretic approaches in models of sufficient complexity for real-world decision making.  In this study, I propose a practical approach for conducting VOI analysis using a combination of probabilistic sensitivity analysis, linear regression metamodeling, and Gaussian approximation as an efficient replacement for traditional Bayesian updating.  I leverage the Central Limit Theorem to simplify the Bayesian updating process.  I illustrate my approach using a previously published cost-effectiveness analysis of several treatment strategies of gouty arthritis.  I present several measures of VOI, including the expected value of sample information for various sample sizes and the optimal sample size that maximizes the benefit of research while addressing many of the challenges of traditional VOI analyses.
 

"Private provision of social insurance: drug-specific price elasticities and pricing in Medicare Part D"

Please note: All research in progress seminars are off-the-record. Any information about methodology and/or results is embargoed until publication.

Abstract:

Although optimal social insurance theory suggests that consumer cost-sharing should increase in a drug's elasticity of demand, publicly provided drug coverage typically involves uniform cost-sharing across drugs. Does the private market behave differently? We examine this question in the context of Medicare Part D. To do so, we first exploit the famous donut hole – at which insurance becomes discontinuously generous at the margin – together with detailed claim-level data for about 2 million beneficiaries to estimate price elasticities of demand for almost 200 different common drugs and 80 different common therapeutic classes. We document substantial heterogeneity in the price elasticity of demand across drugs and diseases, with an average elasticity estimate across drugs of -0.22 and a standard deviation of 0.5. Drawing on additional detail on the contract design of hundreds of Medicare Part D private plans, we document - in contrast to pricing in public prescription drug plans - substantial heterogeneity in the cost-sharing rate private plans charge for different drugs. We find that private plans vary cost-sharing in the direction of the social optimum: charging higher consumer coinsurance for drugs and classes with more elastic demand. Results from a highly stylized model are consistent with the empirical findings: profit maximizing private firms have incentives to vary cost-sharing across drugs in the socially efficient direction. Our findings suggest that benefit design may be more efficient in privately rather than publicly provided insurance.