Searching over 5,500,000 cases.

Buy This Entire Record For $7.95

Download the entire decision to receive the complete text, official citation,
docket number, dissents and concurrences, and footnotes for this case.

Learn more about what you receive with purchase of this case.

Livinrite, Inc. v. Azar

United States District Court, E.D. Virginia, Alexandria Division

June 17, 2019

LIVINRITE, INC., Plaintiff,
ALEX M. AZAR, II, Secretary of the United States Department of Health and Human Services, Defendant.



         Plaintiff, a Medicare-certified home health services provider, brings this action against defendant, the Secretary of the United States Department of Health and Human Services ("the Secretary"), seeking reversal of a decision by the Medicare Appeals Council ("MAC") that plaintiff had been overpaid approximately $1 million for Medicare claims submitted from 2008 to 2010. Specifically, plaintiff challenges the MAC's determinations

(i) that ten claims submitted by plaintiff were not covered by Medicare,
(ii) that a valid statistical sampling methodology was used to derive through extrapolation plaintiffs total overpayment amount, and
(iii) that plaintiff was not entitled to a waiver of liability for the overpayment amount. In response, defendant argues that the MAC's decision passes muster under the deferential standard of review prescribed by the Administrative Procedures Act ("APA")[1] because each of the MAC's conclusions applied the correct legal standards and is amply supported in the record.

         At issue in this matter are the parties' cross-motions for summary judgment, which have been fully briefed and argued and are thus ripe for decision.


         The APA confines judicial review of agency decisions to the administrative record of proceedings before the agency. See 5 U.S.C. § 706; see also Camp v. Pitts, 411 U.S. 138, 142 (1973). Put another way, "when a party seeks review of agency action under the APA, the district judge sits as an appellate tribunal." Am. Bioscience, Inc. v. Thompson, 269 F.3d 1077, 1083 (D.C. Cir. 2001). Given the district court's limited role in reviewing the administrative record, the ordinary summary judgment standard does not apply. The key difference in an APA case is that "the presence or absence of a genuine dispute of material fact is not in issue, as the facts are all set forth in the administrative record." Hyatt v. U.S. Patent & Trademark Office, 146 F.Supp.3d 771, 780 (E.D. Va. 2015). Therefore, in a review of agency action under the APA, "[t]he 'entire case' on review is a question of law." Am. Bioscience, Inc., 269 F.3d at 1083.

         The administrative record pertaining to plaintiffs administrative appeal proceeding before the MAC reflects the following relevant facts.[2]

• On March 11, 2010, AdvanceMed, a Centers for Medicare and Medicaid Services ("CMS") contractor, opened an investigation based on a complaint that plaintiff was (i) admitting patients who did not qualify for home health services and (ii) continuing to provide physical therapy to patients even after those patients reached their maximum level of potential. In addition, AdvanceMed conducted preliminary data analysis that showed plaintiff ranked above average in peer comparison of Medicare billing. After determining that a full audit of all claims paid to plaintiff would not be feasible, AdvanceMed conducted a statistical sampling of the Medicare claims paid to plaintiff. In this respect, AdvanceMed reviewed the services provided by plaintiff to 30 randomly selected beneficiaries and then extrapolated the overpayment determinations to estimate the total amount plaintiff was overpaid by Medicare.
• AdvanceMed selected January 1, 2008 to June 30, 2010 as the period to be reviewed and used simple random sampling as the sample design. AdvanceMed defined the universe as all fully and partially paid claims submitted by the provider for the period covered. It defined the sampling unit as individual beneficiaries, with each unit identified by a health insurance claim ("HIC") number. AdvanceMed created the sampling frame by identifying those sampling units from the universe where at least one line of service on the claim was paid greater than $0 to the provider, and then sorted the frame by HIC number. The frame included 1, 717 units.
• AdvanceMed then used a random number-generator software to draw a simple random sample of 30 beneficiaries.
• Fifteen of the 1, 717 HIC numbers used in the sampling frame to identify the beneficiaries did not match the beneficiaries' actual HIC numbers.
• On October 14, 2010, AdvanceMed requested documentation from plaintiff to support 60 Medicare claims plaintiff had submitted and received reimbursement for on behalf of those 30 beneficiaries. AdvanceMed used the documentation to conduct a medical review of those claims and determined that plaintiff had been overpaid for 33 claims not covered by Medicare. Using the lower limit of the 90% two-sided confidence interval, [3] AdvanceMed then extrapolated the sampled overpayment to conclude that plaintiff had received a total overpayment of $2, 775, 432.
• On June 6, 2012, AdvanceMed provided Plaintiff with documentation supporting its overpayment extrapolation, including, inter alia, documentation of the sample design, the universe of claims, the sampling frame, the random numbers used, the random sample generated, and the statistical results.
• After receiving notice of the overpayment assessment, plaintiff sought redetermination of AdvanceMed's determination that the 33 claims were not covered. The resulting decision was partially favorable, as AdvanceMed reversed 2 of the claim denials. Plaintiff then requested reconsideration by a separate CMS contractor, which affirmed all 31 of the claim denials.
• Next, plaintiff requested and received a hearing before an Administrative Law Judge ("ALJ"). At the hearing, plaintiff both challenged AdvanceMed's individual claim denials and argued that AdvanceMed's statistical sampling could not be replicated based on the documentation provided by AdvanceMed. The ALJ rejected plaintiffs challenge to AdvanceMed's statistical sampling, but reversed certain claim denials.
• Finally, plaintiff requested review of the ALJ's decision by the MAC, which constituted the final stage of administrative review. On March 29, 2018, the MAC issued a decision that reversed 3 claim denials by the ALJ but affirmed the ALJ's conclusions with respect to the validity of AdvanceMed's statistical sampling and the remaining 15 claim denials.


         The standards of review that govern the district court's review of the MAC's final decision are set forth in the Medicare statute and the APA. First, the Medicare statute provides that the MAC's factual findings must be upheld "if supported by substantial evidence." 42 U.S.C. §§ 405(g), 1395ff(b)(1)(A). The scope of review of the MAC's factual findings under this standard is quite limited. As the Supreme Court has explained, substantial evidence "does not mean a large or considerable amount of evidence, but rather 'such relevant evidence as a reasonable mind might accept as adequate to support a conclusion.'" Pierce v. Underwood, 487 U.S. 552, 565 (1988) (quoting Consolidated Edison Co. v. NLRB, 305 U.S. 197, 229 (1938)). And importantly, the district court may not "reweigh conflicting evidence, make credibility determinations, or substitute [its] judgment for that of the [MAC]." Johnson v. Barnhart, 434 F.3d 650, 653 (4th Cir. 2005); see Jarvis v. Berryhill, 697 Fed.Appx. 251, 252 (4th Cir. 2017) ("The duty to resolve conflicts in the evidence rests with the [agency], not with a reviewing court.").

         Second, pursuant to the APA, the MAC's decision may be set aside only if it is "arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law." 5 U.S.C. § 706(2)(A). The Fourth Circuit has made clear that "[r]eview under this standard is highly deferential, with a presumption in favor of finding the agency action valid." Ohio Vail. Envt'l Coalition v. Aracoma Coal Co., 556 F.3d 177, 192 (4th Cir. 2009). In general, an agency decision will not be considered arbitrary and capricious as long as "the agency has examined the relevant data and provided an explanation of its decision that includes 'a rational connection between the facts found and the choice made.'" Id. at 192-93 (quoting Motor Veh. Mfrs. Ass'n v. State Farm Mut. Ins. Co., 463 U.S. 29, 43 (1983)).


         Before applying these standards of review to the MAC's decision, it is important to describe briefly the statutory and regulatory framework that governs the MAC's determination whether a healthcare provider was overpaid by Medicare.

         Medicare is a federal health insurance program for the elderly and disabled that is administered by the Secretary through CMS. Among other things, the Medicare program reimburses providers of certain medical and health services for the cost of services that are covered by the Medicare Act. See 42 U.S.C. § 1395 et seq. To promote the integrity of the Medicare program, the Secretary is authorized to enter into contracts with private entities to review claims for reimbursement submitted by providers, to determine whether Medicare payments should not be, or should not have been, made, and to recoup payments that should not have been made. Id. § 1395ddd; 42 C.F.R. § 405.371(a)(3). In light of the substantial volume of Medicare claims submitted by providers, Medicare contractors are permitted to use statistical sampling and extrapolation to determine the extent to which a provider was overpaid by Medicare. See CMS Ruling 86-1 at 11 (Feb. 20, 1986).

         Pursuant to the Medicare Program Integrity Manual ("MPIM"), a Medicare contractor must follow six steps to conduct statistical sampling for overpayment calculation. MPIM Ch. 8 § (Pub. No. 100-08, Rev. 377) (2011). First, the contractor must select the provider to be reviewed. Id. Second, the contractor must select the period to be reviewed. Id. §§, Third, the contractor must define the universe, the sampling unit, and the sampling frame. Id. § The "universe" consists of all Medicare claims submitted by the provider during the period under review. Id. § The "sampling unit" is the element that will be reviewed (e.g., individual claims or beneficiaries). Id. § The "sampling frame" is the group of sampling units that remain after any limiting criteria are applied to the sampling universe. Id. § Fourth, the contractor must choose a sampling method and implement the method to select the sample. Id. § The sampling method that is used must be classified as "probability sampling," that is, (i) the method must be capable of selecting a set of enumerable, distinct samples from the sampling frame and (ii) each sampling unit must have a known probability of being selected that is greater than zero.[4] Fifth, the contractors must review each unit in the selected sample and determine if an overpayment has been made. Id. §§, Sixth, the contractor must estimate the total overpayment to the provider during the review period by extrapolating the results from the selected sample to the entire sampling frame. Id. §§, 8.4.5.

         A provider may challenge a Medicare contractor's calculation of overpayment through the administrative appeals process.[5] The use of statistical sampling by the contractor "creates a presumption of validity as to the amount of an overpayment." CMS Ruling 86-1 at 11. It is the provider's burden to overcome this presumption by demonstrating either (i) that the sample is not statistically valid or (ii) that the contractor's determinations of overpayment with respect to specific units in the selected sample are incorrect. Id.

         First, the provider may challenge the statistical validity of the sample selected by the contractor. A challenge to the validity of the sample "must be predicated on the actual statistical validity of the sample as drawn and conducted." MPIM § Accordingly, "[i]f a particular probability sample design is properly executed" in accordance with the six steps set forth above, "then assertions that the sample and its resulting estimates are 'not statistically valid' cannot legitimately be made." Id. § 8.4.2. Put simply, "a probability sample and its results are always 'valid.'" Id.

         Second, the provider may challenge the contractor's determination that certain sampling units in the selected sample are not covered by the Medicare Act and thus resulted in an overpayment to the provider. In this respect, home health services qualify for Medicare coverage if such services are "reasonable and necessary" and are provided to a beneficiary who is (i) confined to the home, (ii) under the care of a physician who establishes a plan of care in accordance with 42 C.F.R. § 409.43, and (iii) in need of "skilled services" as certified by a physician. 42 U.S.C. §§ 1395f(a)(2)(C), 1395y(a)(1)(A); 42 C.F.R. § 409.42. A skilled service is one that is "so inherently complex that it can be safely and effectively performed only by, or under the supervision of, professional or technical personnel." 42 C.F.R. § 409.32


         Plaintiff first argues that the MAC'S decision should be set aside because the MAC'S decision that AdvanceMed's sampling methodology could be accurately replicated is arbitrary and capricious, incorrectly applies the relevant legal standards, and is not supported by substantial evidence in the record. Notably, plaintiff does not challenge the MAC's determination that the sampling methodology applied by AdvanceMed was statistically valid.[6] Rather, plaintiffs argument, distilled to its essence, is that AdvanceMed's extrapolated overpayment determination must be invalidated because it is impossible to replicate the sample based on the materials in the record.

         In addition to providing contractors with instructions on the proper execution of statistical sampling for overpayment calculation, the MPIM also requires Medicare contractors to document the sampling methodology, the sampling universe and frame, and the random number selection process that were used to estimate overpayment. MPIM §§,,, The purpose of these documentation requirements is to ensure that the sampling frame and the sample can be replicated in the event that the methodology is challenged. Id. §§, As previous MAC decisions have concluded, failure to supply the provider with sufficient documentation to recreate the sampling frame and sample effectively deprives the provider of its right to challenge the statistical validity of the sample and thus may constitute a ground for invalidating the overpayment extrapolation. See William Vecchioni, D.C., M-13-3700 (H.H.S. Nov. 20, 2013); Global Home Care, Inc., M-11-116 (H.H.S. Jan. 11, 2011); Podiatric Medical Associates, M-10-230 (H.H.S. June 22, 2010).

         But here, as the MAC concluded, plaintiff was provided with ample documentation to enable plaintiff to replicate the sampling frame and the sample. The record confirms that the statistical sampling information CMS provided to plaintiff included, inter alia, an electronic spreadsheet of the frame used in the overpayment review; a memorandum explaining the universe, sampling frame, sampling unit, sample size, and sample design; the sample that was selected from the frame; and the exact random numbers that were generated and used to select the sample from the frame. And it is undisputed that applying the random numbers provided by CMS to the sampling frame provided by CMS would generate the same sample as the one selected and recorded by AdvanceMed. ...

Buy This Entire Record For $7.95

Download the entire decision to receive the complete text, official citation,
docket number, dissents and concurrences, and footnotes for this case.

Learn more about what you receive with purchase of this case.