Role of Insurance in Monitoring Physician Fraud and Other Healthcare Delivery Crimes
Presenter: Margarita Khvan, USC Discussant: Ian McCarthy, Emory University and NBER
ASHEcon, June 14, 2023
Waste in U.S. Health Care
Total waste is estimated at over $900 billion per year (nearly 35% of all health care expenditures)
Failures of care delivery: $102 to $154 billion
Failures of care coordination: $25 to $45 billion
Overtreatment: $158 to $226 billion
Administrative complexity: $107 to $389 billion
Pricing failures: $84 to $178 billion
Fraud and abuse: $82 to $272 billion
Potential for fraud in Medicare and Medicaid
GAO…designated Medicare and Medicaid as being at “high risk” for fraud, abuse, and improper payments. Both programs were designed to enroll “any willing provider” and to reimburse claims quickly for services provided.
Enormous volume: 4.5 million Medicare claims from 1.5 million providers daily, 30,000 enrollment applications from providers monthly
Emphasis on rapid payment
Eastern European crime syndicates stealing the identifies of beneficiaries and billing the programs for treatments that didn’t take place at clinics that don’t exist
What role do insurers play in combatting fraud?
What does this paper do?
Identify physicians that are eventually excluded due to fraud
Compare to physicians not excluded
Focus on inpatient stays in Florida from 2012 to 2017
Outcomes: discharges, length of stay, charges
Excluded physicians do more…but only in Medicare and Medicaid FFS
at least 150% more discharges, 180% longer patient stays, 200% higher charges
Excluded physicians do less in Medicare Advantage
around 50% less in all outcomes
Selection into insurance types
A heavily fraudulent provider should focus where fraud is easiest
Is there sufficient overlap in non-FFS insurance?
How much of the estimates are mechanical?
Fraud is identified by outlier behaviors in Medicare and Medicaid FFS claims. By construction, excluded physicians are outliers in Medicare and Medicaid FFS
Physicians only have so much capacity. 150% more in FFS should be offset by a reduction somewhere
Is this fraud?
“Doing more” can mean lots of things
Can we separate fraud from overtreatment here?
How to quantify the reduction in waste/fraud versus the added administrative complexity from other insurers?
What mechanisms are these other insurers using to pull down quantity? Prior approvals? Alternative payment models?
Show balance in payer mix among excluded and non-excluded physicians
Consider additional analysis on a per-patient basis? May avoid mechanical offsets due to capacity constraints
Leverage data post-exclusion to asses role of capacity constraints
Reframe in terms of waste? Or bound estimates using other estimates of fraud vs overtreatment?
Any opportunity to exploit bundled payments or CJR to identify mechanisms?
Can insurers successfully combat fraud? Great question with clear policy implications
Ian McCarthy, Emory University & NBER ianmcccarthyecon.com email@example.com