The RACs are frequently wrong in their assertions about what a physician, confronted with a patient in need of medical treatment, should have done months or years earlier. Indeed, hospitals report that when they pursue appeals through the administrative appeals process—an expensive and burdensome exercise—they are successful in overturning RAC denials 75 percent of the time. Despite this alarming error rate, when a RAC determines that a provider was paid for inpatient hospital services but that the patient in question should have been treated as an outpatient, [Centers for Medicare & Medicaid Services] takes back the entire Part A payment. Moreover, CMS takes the position that once an inpatient claim that was paid under Medicare Part A is later—usually years later—denied, the hospital cannot receive Medicare Part B payment except for a few ancillary services. As a result, when a RAC concludes that a hospital should have provided items and services on an outpatient, rather than an inpatient, basis, the hospital ends up receiving little if any reimbursement for reasonable and medically necessary items and services provided. The RACs fare significantly better. They keep a contingency percentage—9 percent to 12.5 percent—of the entire Part A payment.

RACs were made permanent through the Tax Relief and Health Care Act of 2006 [text] and were controversial at the time of enactment. RACs have a three-year lookback period and are the only Medicare audit option for hospitals in all 50 states. They were implemented in 2008 [press release], with the CMS describing their goal as "to identify improper payments made on claims of health care services provided to Medicare beneficiaries." In the CMS's 2010 report to Congress [pdf] it determined that RACs had "identified and corrected $92.3 million in combined overpayments and underpayments," with 82 percent being overpayments that were returned to the Medicare Trust Fund [Medicare backgrounder, PDF].