On February 12, 2014, the U.S. Court of Appeals for the Third Circuit held that Federal courts do not have jurisdiction to review HHS’s determination of a “sustained or high level of payment error,” one of two possible prerequisites for use of extrapolation to determine Medicare overpayments. John Balko & Associates, Inc. v. U.S. Dept. of Health & Human Services, Case No. 13-1568 (Feb. 12, 2014). The court’s decision also upheld the Medicare Appeal Council’s ruling that a Medicare contractor may use the same audit sample to determine whether a high level of payment error exists and to extrapolate the Medicare overpayment.
John Balko & Associates, Inc., d/b/a Senior Healthcare Associates (Balko), provides podiatry, audiology, and optometry services to Medicare patients in nursing homes. In 2008, SafeGuard Services (SafeGuard), a Medicare contractor, audited Balko’s claims for reimbursement. During the audit, Safeguard used statistical sampling, as outlined in the Medicare Program Intermediary Manual, to review Balko’s claims. SafeGuard first identified a universe of 5,445 Medicare beneficiaries associated with the claims at issue, and then selected a random sample of 81 beneficiaries with a total of 581 claims. After reviewing these claims, SafeGuard found that 99.85% of the claims were paid improperly. Considering this to be a high error rate, HHS directed SafeGuard to extrapolate an overpayment. SafeGuard used the same audit sample to calculate an extrapolated overpayment of $857,109.07.
Through several levels of administrative and judicial review, Balko challenged SafeGuard’s overpayment demand. Federal law (42 U.S.C. § 1395ddd(f)(3)) allows extrapolation to determine overpayments only if there is a high level of payment error or if there is evidence the provider was informed of the payment error and failed to correct it. Balko argued that SafeGuard’s extrapolation was inappropriate because it used the same sample to determine the high error rate and then to extrapolate the amount of the overpayment. Balko also argued that HHS’s determination of a sustained or high level of payment error was not supported by substantial evidence. After an ALJ invalidated SafeGuard’s use of statistical sampling based on a lack of documentation, the Medicare Appeals Council reviewed the case on its own motion. The Appeals Council reversed the ALJ’s ruling, finding that 42 U.S.C. § 1395ddd(f)(3) precludes administrative and judicial review of high claim error rate determinations. The Appeals Council also found that the Medicare statute did not require a high error rate finding before beginning audit, and that audits can use statistical sampling. Balko appealed to the district court, which upheld the Appeals Council’s determination and granted summary judgment in favor of the Secretary.
On appeal, the Third Circuit found that the Appeals Council was correct in interpreting 42 U.S.C. § 1395ddd(f)(3) to preclude administrative and judicial review, and that the district court was correct in upholding the Appeals Council’s decision. As such, the Third Circuit could not review the Secretary’s interpretation of 42 U.S.C. § 1395ddd(f)(3), which allows the use of a single sample to calculate both high error rates and extrapolate overpayment amounts. Balko attempted to avoid the jurisdictional bar by challenging the procedures SafeGuard used in reaching its determination, rather than the merits of the determination itself, but the court was unpersuaded, finding no such distinction in the statute. In addition, the court found unpersuasive Balko’s challenge that the Appeals Council’s decision was not supported by substantial evidence. The court noted that because SafeGuard complied with Medicare rulings and procedures, Balko did not meet its heavy burden to prove the sample was statistically invalid, and not merely that another statistician could select a different or more precise sample. The court’s decision may be found here.
Reporter, Paige Fillingame, Houston, +1 713 615 7632, pfillingame@kslaw.com.
