# The effect of audits on physicians and patients: Evidence from Medicare

> **NIH AHRQ R36** · COLUMBIA UNIV NEW YORK MORNINGSIDE · 2020 · $34,415

## Abstract

Project Summary/Abstract:
 Each year, Medicare spends an estimated $13 billion on medically unnecessary care
(CMS 2011). Reducing payments for unnecessary services can ​improve the quality,
affordability, and safety of healthcare delivery.​ The proposed dissertation project uses a
data-driven, quasi-experimental approach to study the Recovery Audit Contractor (RAC)
program, a large Medicare audit program aimed at detecting medically unnecessary care.
Little research has been done evaluating the audit program, despite its large impact on
providers. It is also theoretically ambiguous whether auditing decisions made under
uncertainty, like the decisions that physicians make, can reduce unnecessary medical care
without negative consequences. ​The research objective is to provide causal evidence on
whether audits can reduce medically unnecessary care without compromising patient
outcomes or imposing undue burden on providers. ​The proposed project focuses on the
context of ​unnecessary hospital admissions​ and considers the effects of audits at both the
patient-level and hospital-level. ​Aim 1​ and ​Aim 2​ seek to estimate the effect of audits on
patient-level hospital admissions ​and​ health outcomes​. The project leverages exogenous
variation in audit rates across patient populations to causally estimate this effect. Preliminary
results demonstrate that a​ udits​ ​reduce inpatient admissions​. Providers responded to audits by
reducing hospital admissions for relatively healthy patients​, in line with a model where providers
reduce medically unnecessary hospital admissions in response to audits. Preliminary results
also demonstrate that the increased audit rate has ​no detectable effect on two patient outcomes​:
in-hospital mortality​ and r​ evisits​. ​Aim 3​ seeks to ​quantify the financial burden of audits on
providers ​by comparing balance sheet data (i.e., administrative costs, hiring, investments) of
hospitals that were differentially affected by RAC audits. The overarching objectives of the
proposed dissertation project are to be one of the first studies to evaluate the efficacy of the
RAC program, highlight the policy’s potential unintended consequences, and inform
policymakers on how to leverage audits as a policy tool to reduce unnecessary care.

## Key facts

- **NIH application ID:** 10049604
- **Project number:** 1R36HS027715-01
- **Recipient organization:** COLUMBIA UNIV NEW YORK MORNINGSIDE
- **Principal Investigator:** Mengdi Shi
- **Activity code:** R36 (R01, R21, SBIR, etc.)
- **Funding institute:** AHRQ
- **Fiscal year:** 2020
- **Award amount:** $34,415
- **Award type:** 1
- **Project period:** 2020-09-30 → 2022-09-29

## Primary source

NIH RePORTER: https://reporter.nih.gov/project-details/10049604

## Citation

> US National Institutes of Health, RePORTER application 10049604, The effect of audits on physicians and patients: Evidence from Medicare (1R36HS027715-01). Retrieved via AI Analytics 2026-05-23 from https://api.ai-analytics.org/grant/nih/10049604. Licensed CC0.

---

*[NIH grants dataset](/datasets/nih-grants) · CC0 1.0*