How do Hospitals Respond to Payment Incentives?
A literature has found that medical providers inflate bills and report more conditions given financial incentives. We evaluate whether Medicare reimbursement incentives are driven more by bill inflation or coding costs. Medicare reformed its payment mechanism for inpatient hospitalizations in 2007, increasing coding costs. We first examine whether increased extra reimbursements from reporting more diagnoses lead hospitals to report more high bill codes. We find that increases in reimbursements within narrow patient groups led to more high bill codes before 2007 but not after. Using the payment reform, we then test for costly coding by comparing hospitals that adopted electronic medical records (EMRs) to others. Adopters reported relatively more top bill codes from secondary diagnoses after the reform, exclusively for medical patients, with a negative effect for surgical patients. This is consistent with EMRs lowering coding costs for medical discharges but increasing them for surgical ones. We further use a 2008 policy where Medicare implemented financial penalties for certain hospital-acquired conditions. EMR hospitals coded relatively more of these conditions following the penalization, lowering revenues. Together, this evidence is contrary to bill inflation but consistent with costly coding. Reducing coding costs may increase inpatient Medicare costs by $1.04 billion annually.
We have received helpful comments from Sebastian Fleitas, Mike Geruso, Josh Gottlieb, Ben Handel, Kelli Marquardt, Jeff McCullough, Jessamyn Schaller, Bob Town, and seminar participants at several institutions. This paper includes material from and supersedes "Does Hospital Electronic Medical Record Adoption Lead to Better Information or Worse Incentives?" and "Does Health IT Adoption Lead to Better Information or Worse Incentives?'" The views expressed herein are those of the authors and do not necessarily reflect the views of the National Bureau of Economic Research.
• I am the co-PI of a research grant from the Agency for Healthcare Quality and Research (1R01HS024850-01) that evaluates narrow network plans, including in the small group health insurance market.
• I am the PI of a research grant from the National Science Foundation (SES-1425063) that evaluates bargaining models in the context of insurer-hospital price negotiation.
• I will serve on the U.S. Congressional Budget Office’s Technical Review Panel for evaluating health insurance simulation models.
• I am a Senior Advisor to Cornerstone Research and as such, I provide paid consulting services to healthcare providers.
• The insurance company which we consider in this study provided us with the data for this study and has graciously responded to many questions from us regarding the data and their policies in this market. They have not provided us with any financial support.
• I declare that I have no other relevant material or financial interests that relate to the research described in this paper.