Case 1:12-cv-00299-WMS Document 32 Filed 02/05/16 Page 1 of 58 UNITED STATES DISTRICT COURT FOR THE WESTERN DISTRICT OF NEW YORK UNITED STATES OF AMERICA EX REL. [UNDER SEAL] PLAINTIFF, CASE NO. 12-CV-0299S AMENDED COMPLAINT FOR VIOLATIONS OF THE FEDERAL FALSE CLAIMS ACT v. [UNDER SEAL], JURY TRIAL DEMANDED DEFENDANTS. FILED IN CAMERA & UNDER SEAL (AS REQUIRED BY 31 U.S.C. § 3730(b)(2)) FILED IN CAMERA AND UNDER SEAL DO NOT ENTER IN PACER DO NOT PLACE IN PRESSBOX {00043348; 1} 343281.2 Case 1:12-cv-00299-WMS Document 32 Filed 02/05/16 Page 2 of 58 UNITED STATES DISTRICT COURT FOR THE WESTERN DISTRICT OF NEW YORK UNITED STATES OF AMERICA ex rel. [UNDER SEAL RELATOR A], PLAINTIFF, CASE NO. 12-CV-0299S AMENDED COMPLAINT FOR VIOLATIONS OF THE FEDERAL FALSE CLAIMS ACT v. [UNDER SEAL DEFENDANT 1]; [UNDER SEAL DEFENDANT 2]; [UNDER SEAL DEFENDANT 3]; [UNDER SEAL DEFENDANT 4]; [UNDER SEAL DEFENDANT 5], and [UNDER SEAL DEFENDANT 6] DEFENDANTS. JURY TRIAL DEMANDED FILED IN CAMERA & UNDER SEAL (AS REQUIRED BY 31 U.S.C. § 3730(b)(2)) FILED IN CAMERA AND UNDER SEAL DO NOT ENTER IN PACER DO NOT PLACE IN PRESSBOX {00043348; 1} 343281.2 Case 1:12-cv-00299-WMS Document 32 Filed 02/05/16 Page 3 of 58 AMENDED COMPLAINT For their complaint, the United States of America ex rel. Under Seal Relator A (the “United States”) alleges as follows: I. INTRODUCTION 1. This is an action to recover damages and civil penalties on behalf of the United States, the real party in interest, under the Federal False Claims Act, 31 U.S.C. §§ 3729–33 (the “FCA”) against Under Seal Defendant 1, Under Seal Defendant 2, Under Seal Defendant 3, Under Seal Defendant 4, Under Seal Defendant 5, and Under Seal Defendant 6 (“Defendants”). 2. Defendants are engaged in a scheme to knowingly submit, cause to be submitted, and conspire to submit false claims for payment to the United States by submitting false “risk adjustment” information to the Centers for Medicare & Medicaid Services (“CMS”) to improperly increase the amounts CMS pays them or their clients through the Medicare Advantage program. Likewise, Defendants have knowingly retained overpayments received from CMS as a result of their false risk adjustment submissions. 3. The Medicare Advantage (“MA”) program is designed to apply to Medicare a form of the “managed care” model commonly used by private health insurance companies. Under the managed care model, an employer or other organization seeking health care for its members—here the United States through the Medicare {00043348; 1} 1 343281.2 Case 1:12-cv-00299-WMS Document 32 Filed 02/05/16 Page 4 of 58 Program—pays a managed care organization a fixed fee to provide health services to its members. The payment is typically a per-member-per-month (“PMPM”) rate, also known as a capitation rate. The managed care organization receiving capitation payments (often a hospital, physician group, or other health insurance company) is responsible for paying hospitals, physicians and all other medical providers for health care services provided to the plan’s members. This differs from traditional fee-forservice (“FFS”) models, where the organization pays individual physicians, hospitals, and other providers for each service they provide to the organization’s members. 4. Through the MA program, Medicare allows private health insurers to set up managed care plans to cover Medicare beneficiaries. Medicare pays a monthly capitation rate for each beneficiary enrolled as an MA plan member. MA plans must then use that money to pay hospitals, physicians, and other health care providers for the services plan members receive, and to cover the plans’ administrative expenses. Certain MA plans are also given money to pay for plan members’ prescription drugs. Under both types of plans, CMS adjusts the capitation rate for each beneficiary to reflect that beneficiary’s individual demographics (e.g., age and gender), geographic location, and health status. 5. The adjustment for each member’s health status is one of the most significant components of the capitation rate. Individuals with multiple and/or serious health conditions account for more health care costs than healthy members. Accordingly, {00043348; 1} 2 343281.2 Case 1:12-cv-00299-WMS Document 32 Filed 02/05/16 Page 5 of 58 CMS pays a substantially higher capitation rate for members who have been recently treated for one or more serious, expensive diseases or conditions. These increased payments are known as “risk adjustment” payments. On average, CMS pays an MA plan close to $3,000 per year for each condition a member has that requires a risk adjustment payment. 6. To receive risk adjustment payments, MA plans submit claims to CMS each year for each member for each qualifying disease or condition. When the plan submits these claims, it must assert that the member received treatment for the diagnosed condition from a qualified health care provider in the twelve-month period before the payment year. MA organizations may only submit risk adjustment claims if the individual patient has been diagnosed with the condition in question, consistent with established coding standards, and there is documentation in the patient’s medical record that: (1) the diagnosis was treated or affected treatment; (2) in a face-to-face visit (except for pathology services performed by a pathologist); (3) by an appropriate provider; and (4) during the proper time period. 7. Under Seal Defendant 1, Under Seal Defendant 2, and Under Seal Defendant 6 are engaged in systematic fraud in which they routinely: (a) “Upcode” risk adjustment claims by submitting claims for diagnoses that the member does not have or for which the member was not treated in the relevant year, or by claiming that a member {00043348; 1} 3 343281.2 Case 1:12-cv-00299-WMS Document 32 Filed 02/05/16 Page 6 of 58 was treated for a more serious condition than the member actually has; and (b) refuse to correct (and refuse to reimburse Medicare for) previously submitted risk adjustment claims when defendants discover, or in the exercise of reasonable care should discover, that those previously submitted claims were false. 8. Under Seal Defendant 3, Under Seal Defendant 4, and Under Seal Defendant 5 are engaged in systematic fraud by assisting and causing MA organizations, including Under Seal Defendant 1, Under Seal Defendant 2, and Under Seal Defendant 6, to submit fraudulent risk adjustment claims, and failing to correct (and reimburse Medicare for) previously submitted false claims. Under Seal Defendant 4 and Under Seal Defendant 5 are top executives at Under Seal Defendant 3 and, in Relator’s understanding, are the driving force behind Under Seal Defendant 3’s fraudulent scheme. 9. Through this scheme, Defendants have defrauded the United States of millions of dollars. 10. Defendants’ conduct alleged herein violates the federal False Claims Act. The federal False Claims Act (the “FCA”) was originally enacted during the Civil War. Congress substantially amended the Act in 1986—and, again, in 2009 and 2010—to enhance the ability of the United States Government to recover losses sustained as a {00043348; 1} 4 343281.2 Case 1:12-cv-00299-WMS Document 32 Filed 02/05/16 Page 7 of 58 result of fraud against it. The Act was amended after Congress found that fraud in federal programs was pervasive, and that the Act, which Congress characterized as the primary tool for combating government fraud, was in need of modernization. Congress intended that the amendments would create incentives for individuals with knowledge of fraud against the Government to disclose the information without fear of reprisals or Government inaction, and to encourage the private bar to commit legal resources to prosecuting fraud on the Government's behalf. 11. The FCA prohibits, inter alia: (a) knowingly presenting (or causing to be presented) to the federal government a false or fraudulent claim for payment or approval; (b) knowingly making or using, or causing to be made or used, a false or fraudulent record or statement material to a false or fraudulent claim; (c) knowingly making, using, or causing to be made or used, a false record or statement material to an obligation to pay or transmit money or property to the Government, or knowingly concealing or knowingly and improperly avoiding or decreasing an obligation to pay or transmit money or property to the Government; and (d) conspiring to violate any of these three sections of the FCA. 31 U.S.C. §§ 3729(a)(1)(A)-(C), and (G). Any person who violates the FCA is liable for a civil penalty of up to $11,000 for each violation, plus three times the amount of the damages sustained by the United States. 31 U.S.C. § 3729(a)(1). 12. For purposes of the FCA, a person “knows” a claim is false if that person: “(i) has actual knowledge of [the falsity of] the information; (ii) acts in deliberate {00043348; 1} 5 343281.2 Case 1:12-cv-00299-WMS Document 32 Filed 02/05/16 Page 8 of 58 ignorance of the truth or falsity of the information; or (iii) acts in reckless disregard of the truth or falsity of the information.” 31 U.S.C. § 3729(b)(1). The FCA does not require proof that the defendants specifically intended to commit fraud. Id. Unless otherwise indicated, whenever the words “know,” “learn,” discover,” or similar words indicating knowledge are used in this Complaint, they mean knowledge as defined in the FCA. 13. Each claim for risk adjustment payments that defendants submitted or caused to be submitted to CMS, where the patient was not treated, by a qualified provider, for that condition in the year in question, and/or the treatment and condition are not properly documented in the medical record, is a false and/or fraudulent claim within the meaning of the FCA, so long as defendant knew the claim was false when it was submitted, or the defendant later discovered its falsity and refused to correct the claim. 14. The FCA allows any person having information about an FCA violation to bring an action on behalf of the United States, and to share in any recovery. The FCA requires that the Complaint be filed under seal for a minimum of 60 days (without service on the defendant during that time) to allow the government time to conduct its own investigation and to determine whether to join the suit. 15. Based on the foregoing laws, qui tam plaintiff / Relator Under Seal Relator A seeks, through this action, to recover damages and civil penalties arising from the false or fraudulent records, statements and/or claims that the Defendants made or {00043348; 1} 6 343281.2 Case 1:12-cv-00299-WMS Document 32 Filed 02/05/16 Page 9 of 58 caused to be made in connection with false and/or fraudulent claims for Medicare Advantage and Medicare Part D risk adjustment payments. II. PARTIES 16. Under Seal Relator A is Teresa Ross (“Relator”), a resident of Vancouver, Washington and a former employee of Under Seal Defendant 1. 17. Under Seal Defendant 1 is Group Health Cooperative (“GHC”), a Washington non-profit corporation with its principal place of business in Seattle, Washington. GHC was founded in 1947 and operates as a non-profit consumer-governed health care organization that provides managed care plans to members in twenty-two counties throughout Washington and Idaho. GHC operates Medicare Advantage plans in twenty Washington counties. In 2010, GHC received Medicare and Medicaid revenues of over $700 million. 18. Under Seal Relator A worked at GHC for over fourteen years. In her final position at GHC, she was the Director of Risk Adjustment Services. Prior to that she was the Director of Insurance and Health Data Analysis (“IHDA”). In that position, she implemented the standard risk adjustment claims verification procedures used by GHC and developed successful algorithms to identify and correct diagnosis coding issues and ensure accurate and complete risk adjustment claims submissions. Relator has extensive knowledge of the Medicare risk adjustment system developed both during her time running the GHC risk adjustment department and during her participation in the 2002 {00043348; 1} 7 343281.2 Case 1:12-cv-00299-WMS Document 32 Filed 02/05/16 Page 10 of 58 administrative process whereby CMS developed and implemented the risk adjustment system. 19. Under Seal Defendant 2 is Independent Health Corporation (“IHC”), a New York for-profit corporation with a principal place of business in Buffalo, New York. IHC is a subsidiary of Independent Health Association. 20. Under Seal Defendant 6 is Independent Health Association (“IHA”), a New York non-profit corporation with a principal place of business in Buffalo, New York. 21. For purposes of this complaint, IHA and IHC are referred to collectively as IH. Through contracts with CMS, IH offers Medicare Advantage plans to members across New York. 22. The United States, the real party in interest, has ongoing contracts with Defendants GHC and IH through CMS of the Department of Health and Human Services, in accordance with GHC and IH’s participation in the Medicare programs. 23. Under Seal Defendant 3 is DxID LLC, a New York limited liability company with a principal place of business in East Rochester, New York. DxID was founded in September 2011 as a subsidiary of IHC. DxID provides risk-adjustment review services to health care companies operating managed care plans under the Medicare Advantage program, such as GHC and IH, and Programs of All-Inclusive Care for the Elderly (“PACE”). It was founded to oversee and facilitate the submission of risk {00043348; 1} 8 343281.2 Case 1:12-cv-00299-WMS Document 32 Filed 02/05/16 Page 11 of 58 adjustment data from IH’s MA plans to CMS, including having its auditors perform retrospective chart reviews to identify additional chronic conditions to support new risk adjustment claims. Later, the company expanded to provide its risk-adjustment services to other health care companies that offer Medicare Advantage and PACE Plans. 24. On information and belief, DxID provides its risk-adjustment services to many of its MA and PACE plan clients on a contingency fee basis, i.e., in lieu of an hourly fee, DxID receives a percentage of payments received from CMS for additional risk adjustment claims DxID identifies for the plan. 25. CMS discourages the use of contingency fee arrangements (and considers them inherently suspect) because they create perverse incentives for vendors like DxID to find new risk adjustment claims, and no incentive to correct erroneous risk adjustment claims found during chart review. 26. Under Seal Defendant 4 is Dr. John Haughton, DxID’s Consulting Risk Adjustment Advisor. Dr. Haughton is responsible for the development of DxID’s risk adjustment claims review and submission methodology. He has extensive experience in risk adjustment and knowledge of Medicare’s coding rules and regulations. Notwithstanding that knowledge, he developed the DxID risk adjustment system that systematically violates those well-established rules and causes the submission of thousands of false risk adjustment claims. Mr. Haughton is believed to be 52 years old, currently residing in Severna Park, Maryland. {00043348; 1} 9 343281.2 Case 1:12-cv-00299-WMS Document 32 Filed 02/05/16 Page 12 of 58 27. Under Seal Defendant 5 is Betsy Gaffney, Co-Chief Executive Officer of DxID. She was the founder and Executive Vice President of Cognisight, LLC, prior to co-founding DxID. She was responsible for the design and development of Cognisight services, which were similar to DxID’s, and included retrospective chart reviews. At DxID, as at Cognisight, she has been directly and personally involved in developing and implementing risk adjustment claims review and submission practices. In pitching DxID’s fraudulent risk adjustment coding approach to GHC, Ms. Gaffney rationalized the scheme, stating: “[r]isk adjustment is a game, and you need to learn how to play it.” Ms. Gaffney is believed to be 59 years old and currently residing in Rochester, New York. 28. Defendants Haughton and Gaffney are named individually as defendants because of the direct, personal and substantial role they have played in the fraudulent conduct and scheme at issue in this complaint; however Defendants DxID, Haughton, and Gaffney are hereinafter referenced collectively as DxID. III. JURISDICTION AND VENUE 29. This Court has jurisdiction over the subject matter of this action pursuant to 28 U.S.C. § 1331 and 31 U.S.C. § 3732(a), the latter of which specifically confers jurisdiction on this Court for actions brought under 31 U.S.C. § 3730. 30. This Court has personal jurisdiction over the Defendants, pursuant to 31 U.S.C. § 3732(a), as one or more Defendants can be found in, reside in, transact business {00043348; 1} 10 343281.2 Case 1:12-cv-00299-WMS Document 32 Filed 02/05/16 Page 13 of 58 in, and have committed acts related to the allegations in this Complaint in the Western District of New York. Defendants DxID, IHA, and IHC are New York companies headquartered in the Western District of New York. 31. Venue is proper, pursuant to 31 U.S.C. § 3732(a), as the Defendants can be found in, reside in, and/or transact business in the Western District of New York, and because many of the violations of 31 U.S.C. § 3729 discussed herein occurred within this judicial district. IV. LEGAL PRINCIPLES 32. Medicare is a federally-funded health care program primarily serving people age 65 or older. Initially created in Title XVIII of the Social Security Act of 1965, Medicare now has four Parts, A through D. The two original components of Medicare are Part A, which covers inpatient hospital costs and related services, and Part B, which covers outpatient health care costs, such as physicians’ fees. 33. Traditionally, Medicare operates on a fee-for-service basis, meaning that Medicare directly pays hospitals, physicians, and other health care providers for each service they provide to a Medicare beneficiary. Medicare beneficiaries are generally required to pay some portion of many of these services in the form of copayments, deductibles, coinsurance, or other set fees (collectively known as the members’ “out of pocket” expenses). {00043348; 1} 11 343281.2 Case 1:12-cv-00299-WMS Document 32 Filed 02/05/16 Page 14 of 58 34. In 1997, Congress created Medicare Part C, which provides similar benefits to Medicare members, but does so based on a managed care model rather than the traditional fee-for-service model. Under Part C, rather than pay providers directly, Medicare pays private managed care plans (later named “Medicare Advantage” or “MA” plans) a capitation rate (per member per month) and those plans are responsible for paying providers for the services they provide to members of that specific MA plan. 35. MA plans must provide Medicare beneficiaries benefits at least equivalent to those they would have received under the traditional Medicare Parts A and B. Depending on plan structure, MA plans may also provide additional benefits beyond what traditional Medicare would have covered, such as dental care, or cover some or all of their members’ out of pocket expenses associated with basic Medicare Parts A and B services. 36. In 2003, Congress passed the Medicare Prescription Drug, Improvement, and Modernization Act, creating Medicare Part D, which provides prescription drug coverage. Although a limited number of Medicare Part D plans are operated under a cost-reimbursement contract, the plans are generally financed under a managed care model. These managed care model plans are provided under both Part D prescription drug plans, which offer only prescription drug coverage, and Part C plans, which integrate the prescription drug coverage with the traditional Part C health care coverage. {00043348; 1} 12 343281.2 Case 1:12-cv-00299-WMS Document 32 Filed 02/05/16 Page 15 of 58 37. This Complaint refers, collectively, to Medicare Advantage plans with and without Part D coverage, and stand-alone managed care Medicare Part D Plans as “Medicare Advantage Plans” or “MA Plans.” A. Calculation of MA Plan Capitation Rates 38. Capitation rates Medicare pays to MA plans are determined based on a process involving consideration of past and expected future medical expenses, the location of the plan’s actual and expected members, the health status and demographics of those members, and whether the plan will include any additional benefits. That process is summarized in Medicare regulations as follows: In short, under the bidding methodology each plan’s bid for coverage of Part A and Part B benefits (i.e., its revenue requirements for offering original Medicare benefits) is compared to the plan benchmark (i.e., the upper limit of CMS’ payment, developed from the county capitation rates in the local plan’s service area or from the MA regional benchmarks for regional plans). The purpose of the bid-benchmark comparison is to determine whether the plan must offer supplemental benefits or must charge a basic beneficiary premium for A/B benefits. Medicare Managed Care Manual (“MMCM”), ch. 8, § 60. 39. In other words, it is a three-step process involving: (a) development of the MA plan’s bid rate; (b) review of the CMS benchmark rate; and (c) comparison of those {00043348; 1} 13 343281.2 Case 1:12-cv-00299-WMS Document 32 Filed 02/05/16 Page 16 of 58 two rates to develop the base capitation rate and determine whether any adjustments in the plan benefits or member premiums are required. 40. First, the MA plan develops a bid rate. This rate is the amount the MA plan expects it will be required to pay to provide Medicare Part A and B benefits to a hypothetical average plan member. This estimate must be based on either the MA plan’s prior experience covering Medicare members, or an actuarially validated data analysis of expected costs. To represent an “average” plan member, the bid rate must make adjustments to standardize the effect of expected geographic diversity (because some areas are more expensive than others) and the relative health status (i.e., the number and nature of chronic conditions) of the members whose claims experience provided the basis for the bid. The bid rate also includes an amount the MA plan expects to spend on administrative costs, and a profit margin. 41. The mechanism for standardizing the bid for individuals’ demographic factors and health status is known as the “risk score.” It is an artificial score that CMS assigns to every beneficiary. CMS starts with a score of zero, then adds points for the beneficiary’s demographic condition (such as age and gender) and individual disease states (such as diabetes or congestive heart failure). The average risk score is one, with most Medicare beneficiaries having scores under three. The risk score model is designed so that a population with an average risk score of two would be expected to use twice as much health care (in dollars) as a population with a score of one. The bid rate the MA {00043348; 1} 14 343281.2 Case 1:12-cv-00299-WMS Document 32 Filed 02/05/16 Page 17 of 58 plans develop must reflect the amount they will require to provide services to a hypothetical population with a risk score of one. 42. Second, the MA plan must review the Medicare benchmark rate provided by CMS. This rate is the amount the Medicare program would spend to provide Part A and B benefits to an average member in the geographic area covered by the MA plan’s bid. The benchmark rate also includes several other adjustments, including until recently a bonus payment to incentivize health insurance companies to enter the MA market. 43. Third, the bid rate and the benchmark rate are compared to determine whether the MA plan must charge its members a premium, or, instead, must offer them enhanced benefits. If the bid rate is greater than the benchmark rate, Medicare will only pay the MA plan the benchmark rate per member per month. That benchmark rate becomes the base capitation rate that CMS pays the MA plan for a member with a 1.0 risk score (described below). To make up the shortfall between the bid rate and the base capitation rate, the MA plan must then charge beneficiaries who join its plan a monthly premium. See MMCM, ch. 8, § 60.1. 44. If, on the other hand, the bid rate is less than the benchmark rate, the bid rate becomes the base capitation rate. The difference between the benchmark rate and the bid rate is then split between the MA plan and the Medicare program. The first 25% of the difference is retained by the Medicare program as plan savings. The remaining 75% is returned to the MA plan, which must use the rebate to either provide enhanced benefits {00043348; 1} 15 343281.2 Case 1:12-cv-00299-WMS Document 32 Filed 02/05/16 Page 18 of 58 to its plan members or to cover the members’ out of pocket expenses. Ultimately, in such situations, the base capitation rate equals the bid rate, and the MA plan receives 75% of the difference between the bid rate and the benchmark rate. 45. Medicare does not, however, pay plans the base capitation rate. Instead, when payments are made, the base capitation rate is adjusted, for each member, to reflect his or her age, gender, location, and, most important, health status. 46. MA plans must rebid their rates every year. B. Calculation of Part D Plan Capitation Rates 47. The process of calculating the capitation rates for the Part D portion of MA plans is very similar to the process used for the base portion of the MA rate. Annually, the plan develops and submits a bid rate based on the plan’s estimate of the monthly revenue it will require to provide qualified prescription drug coverage for an average, eligible individual. 42 C.F.R. § 423.265(c). As for the base MA rate, a Medicare prescription drug coverage plan’s average monthly bid rate is adjusted to take into account the geographic differences in pricing and the relative health status of the members on whom the bid calculation was based. 48. Risk score calculations for the Medicare Part D portion of the plans, like the calculation for the basic MA rate, are determined by each beneficiary’s demographic information and health status. Each plan’s bid must reflect the revenue the plan will {00043348; 1} 16 343281.2 Case 1:12-cv-00299-WMS Document 32 Filed 02/05/16 Page 19 of 58 require to provide services to a population of “average” members, i.e., those with a risk score equal to one. C. Risk Adjustment Depends on Accurate, Substantiated Health Condition Codes 49. As described above, CMS pays MA plans at a capitation rate that reflects, among other things, each member’s health status. The process of adjusting the capitation rate to reflect a member’s disease states is known as risk adjustment. Risk adjustment is intended to improve the accuracy of payments CMS makes to these plans. To this end, CMS pays a higher future premium for enrollees whom the MA plan represents have been treated for certain diseases and conditions in the current year, based on the expectation that those enrollees will require treatment and/or management for the conditions in the following year. See 2008 Risk Adjustment Training for Medicare Advantage Organizations Participant Guide (“Participant Guide”), at 6.4.1 (for purposes of this Complaint, “treatment” is defined as treatment and management within the meaning of the Participant Guide). 50. Conversely, CMS pays a lower premium for enrollees who, although they may have certain typically expensive conditions, did not require care, treatment, or management for those conditions in the current year. For these members, the risk adjustment methodology assumes that because their condition did not require treatment in the current year, it has improved or otherwise changed so that it is not expected to require treatment in the following year. {00043348; 1} 17 343281.2 Case 1:12-cv-00299-WMS Document 32 Filed 02/05/16 Page 20 of 58 51. As a practical matter, the CMS risk adjustment model evaluates enrollee health (and establishes risk adjustment payment rates) using diagnosis classifications set forth in the International Classification of Diseases, 9th Edition, Clinical Modification (“ICD-9-CM”) system. The ICD-9 system assigns each diagnosis a specific code “used to describe the clinical reason for a patient’s treatment.” Participant Guide at 6.2. Under the MA model, these individual diagnosis codes are then organized into groups, called Hierarchical Condition Categories (“HCCs”). MMCM, ch. 8, § 50. Every HCC consists of several ICD-9-CM diagnosis codes that are clinically related and are expected to require a similar level of resources to treat. Id. For example, there are five HCCs for members with diabetes: HCC 15 (diabetes with renal or vascular manifestation); HCC 16 (diabetes with neurologic or other specified manifestation); HCC 17 (diabetes with acute complications); HCC 18 (diabetes with ophthalmologic or unspecified manifestation); and HCC 19 (diabetes without complication). Generally speaking, members grouped in HCC 15 have the most serious diabetes-associated manifestations, and are expected to cost the most to treat. Members in HCC 19 have the least cost-intensive type of diabetes, and therefore the CMS risk adjustment system provides a smaller enhanced payment for these members. 52. CMS uses the same model for the Part D portion of risk adjustment. However, because certain diagnoses will be expected to increase liability for prescription drugs covered under Part D, but not hospital costs and physician fees covered under Part {00043348; 1} 18 343281.2 Case 1:12-cv-00299-WMS Document 32 Filed 02/05/16 Page 21 of 58 C, and vice versa, a distinct list of Hierarchical Condition Categories (“RxHCCs”) with corresponding diagnosis codes was created for Part D risk adjustment. See Participant Guide at 8.2.5.2. For example, RxHCC 75 represents attention deficit disorder, a condition predicted to increase drug spending. However, because attention deficit disorder is unlikely to result in hospitalization, RxHCC 75 has no corresponding HCC. On the other hand, HCC 77, respirator dependence/tracheostomy status, a condition category predictive of Part C medical costs, but not necessarily predictive of Part D drug expenses, has no RxHCC equivalent. 53. Although the HCC and RxHCC systems are not identical, they overlap significantly. Certain HCCs have equivalent RxHCCs, meaning that the condition categories consist of identical ICD-9-CM diagnosis codes. For example, HCC 5 (opportunistic infections) is equivalent to RxHCC 2 (opportunistic infections), and HCC 37 (bone/joint/muscle infections/necrosis) is equivalent to RxHCC 39 (bone/joint/muscle infections/necrosis). Even where they are not identical, most HCCs overlap with one or more RxHCCs. For example, of the thirty-seven diagnosis codes that fall within HCC 45 (disorders of immunity), twenty-seven fall within RxHCC 52 (disorders of immunity), seven fall within RxHCC 51 (severe hematological disorders), and three do not fall within any RxHCCs. Thus, the majority of ICD-9-CM diagnosis codes that capture an HCC will also capture an RxHCC. {00043348; 1} 19 343281.2 Case 1:12-cv-00299-WMS Document 32 Filed 02/05/16 Page 22 of 58 54. An individual ICD-9-CM code included in the HCC system for a particular member corresponds on average to nearly $3,000 in extra revenue for the plan over the course of the following year for that member. 55. Because submitting incorrect diagnosis codes increases risk adjustment payments, CMS requires MA plans to follow strict guidelines when submitting codes. Only services provided by an eligible provider type may be included. CMS expressly prohibits MA plans from submitting “risk adjustment diagnoses based on any diagnostic radiology services” or laboratory services. Participant Guide, at 3.2.2, 4-3. The reason CMS prohibits MA plans from submitting codes based on radiology charts, for example, is that “[d]iagnostic radiologists typically do not document confirmed diagnoses. Confirmed diagnoses come from referring physician or physician extenders.” Id. at 4-3 (emphasis added). Because radiologists generally list on their charts the diagnoses a doctor wants them to look for, not which diagnoses the member actually has, CMS excludes radiology services as a valid provider type (i.e., source of risk adjustment data). Also, except in the case of interventional radiology (which does qualify for risk adjustment), it is rare that the radiologist sees the patient face-to-face. 56. The treating provider must document the facts supporting the coded diagnosis in the member’s medical record and sign and date the record. At a minimum, the plan must record five elements for submission to CMS: (a) {00043348; 1} the member’s Health Insurance Claim (“HIC”) number; 20 343281.2 Case 1:12-cv-00299-WMS Document 32 Filed 02/05/16 Page 23 of 58 (b) the ICD-9-CM diagnosis code; (c) the “service from” date; (d) the “service through” date; and (e) the provider type (e.g., hospital inpatient, hospital outpatient, physician). 57. MA plans are responsible for the content of risk adjustment data submissions to CMS, regardless of whether they submit the data themselves or through an intermediary. Participant Guide, at 3-13. Before submitting data to CMS, MA plans are required to filter the data “to ensure that they submit data from only appropriate data sources.” Participant Guide, at 4-11. For example, filters should check that physician data comes from face-to-face encounters with members and ensure that data does not come from non-covered providers, such as diagnostic radiology services. 58. MA plans that filter risk adjustment claims by CPT codes must also filter the data to ensure that only diagnoses treated through approved procedure types are included. Id. at 4-11. MA organizations typically classify professional (e.g., physician) procedures using Current Procedural Terminology (“CPT”) codes, and institutional procedures using revenue codes. These codes show whether the type of service in question was a face-to-face procedure such as a physical examination, or a nonqualifying remote procedure, such as a laboratory test or radiology exam. {00043348; 1} 21 343281.2 Case 1:12-cv-00299-WMS Document 32 Filed 02/05/16 Page 24 of 58 59. MA plans are required to correct the risk adjustment data they submit to CMS. When the MA plan learns that information in a risk adjustment claim (i.e., HIC number, diagnosis code, service dates, and provider type) contains an error, it must submit a “delete record” to CMS for that claim. 60. CMS also requires that diagnosis codes used as the basis for a risk adjustment claim be substantiated through documentation in a medical record. Upon request by CMS, MA plans must provide documentation to support each diagnosis and substantiate that the provider followed proper coding guidelines. Id. at 6-5; 5-52. 61. In general, CMS sets risk scores based on risk adjustment data submitted for services provided during the year preceding the payment year. 42 C.F.R. §§ 422.310(g), 423.329(b)(3). The annual deadline for submitting risk adjustment data to CMS is in early September. Id. The data submitted by the September deadline determines members’ preliminary risk scores for the following year. 62. Despite the September deadline, CMS accepts submissions of risk adjustment data for a period after the end of service year and, through a reconciliation process, adjusts its payments to the MA plan retroactively to account for codes submitted after the September deadline. MA plans are allowed to submit risk adjustment data until after the end of the payment year. After the payment year ends, CMS recalculates the risk score for any members for whom the MA plan made a retroactive submission. {00043348; 1} 22 343281.2 Case 1:12-cv-00299-WMS Document 32 Filed 02/05/16 Page 25 of 58 63. Thus, for example, the capitation rates for 2010 are based on the MA plans’ members’ health status (diagnosis codes) from 2009. The initial submission deadline for the 2009 diagnosis codes was September 4, 2009, and the final submission deadline was January 31, 2011. Thus, CMS calculated members’ initial risk factors for 2010 based on the September 4, 2009 data, but MA plans were allowed to continue to submit 2009 diagnoses until January 31, 2011. After that date, for every member with a newly-submitted diagnosis, CMS recalculated the risk score and reconciled the member’s payments in 2010 with the amount it would have paid at the new score. 64. To test the validity of MA plan risk adjustment data, CMS conducts Risk Adjustment Data Validation (“RADV”) audits after the MA plan’s final deadline for submitting risk adjustment data for the payment year. During such audits, CMS “validates” some of the MA plan’s HCC scores by reviewing medical records the plan contends support the claimed diagnosis codes. Id. at 7-1. To facilitate RADV audits, MA plans are required to submit to CMS medical records and coversheets for each sampled enrollee. Until February 2012, MA plans were required to include the “one best medical record” supporting each HCC. Id. at 7-9. Beginning with the forthcoming RADV audit, CMS will allow audited MA contracts to submit multiple medical records for each HCC being validated. CMS, Notice of Final Payment Error Calculation Methodology for Part C Medicare Advantage Risk Adjustment Data Validation ContractLevel Audits, February 24, 2012. {00043348; 1} 23 343281.2 Case 1:12-cv-00299-WMS Document 32 Filed 02/05/16 Page 26 of 58 65. Historically, CMS has not extrapolated RADV audit results to the plan as a whole, although CMS has proposed moving toward extrapolation of RADV results. Instead, CMS has merely sought repayment for those risk adjustment claims found to be false during the RADV audit. Because RADV audits generally use relatively small samples—a few hundred risk adjustment claims—the potential risk to MA plans, should they be found to have submitted false risk adjustment claims, has been relatively small. Without meaningful financial penalties, MA organizations have generally seen little incentive to conform to CMS’s risk adjustment rules. The fraudulent practices described in this Complaint are a product of the belief, common among MA organizations, that the law can be violated without meaningful consequence. D. CMS Requires MA Plans To Certify the Validity of Their Bid Rates and Risk Adjustment Data To Prevent Fraud 66. Recognizing that the integrity of the capitation rates depends on the integrity of the actuarial information MA plans use in developing their bid rates, and to otherwise guard against fraud, CMS requires MA organizations to submit attestations, each signed by the CEO or CFO (or their authorized, direct subordinate). These attestations are a condition that the MA plans must meet to be eligible to receive any capitation payments from CMS. 67. The first attestation, submitted annually, requires the MA organization to attest that the risk adjustment data it submits annually to CMS is “accurate, complete, and truthful.” The attestation acknowledges that risk adjustment information “directly affects {00043348; 1} 24 343281.2 Case 1:12-cv-00299-WMS Document 32 Filed 02/05/16 Page 27 of 58 the calculation of CMS payments . . . and that misrepresentations to CMS about the accuracy of such information may result in Federal civil action and/or criminal prosecution.” The regulations also provide that if claims data are generated by a “related entity, contractor, or subcontractor of an MA organization,” that entity must similarly certify the “accuracy, completeness, and truthfulness of the data.” 42 C.F.R. § 422.504(l)(2). 68. In addition, the MA organization (and any third-party submitters) must sign an Electronic Data Interchange (“EDI”) Enrollment Form before submitting risk adjustment data to CMS. The EDI Enrollment Form is a contract between the MA organization and CMS attesting to the accuracy of the data submitted. Participant Guide at 4.1. The MA organization attests on the Form “[b]ased on best knowledge, information, and belief, that it will submit risk adjustment data that are accurate, complete, and truthful.” 69. The next attestation is the MA organization’s certification “that the information and documentation comprising the bid submission proposal is accurate, complete, and truthful and fully conforms to the Bid Form and Plan Benefit Package requirements; and that the benefits described in the CMS-approved proposal bid submission agree with the benefit package the MA Organization will offer during the period covered by the proposal bid submission.” {00043348; 1} 25 343281.2 Case 1:12-cv-00299-WMS Document 32 Filed 02/05/16 Page 28 of 58 70. MA organizations must also submit bid submission attestations, certifying “that the information in its bid submission and assumptions related to projected reinsurance and low income cost sharing subsidies is accurate, complete, and truthful and fully conforms to the [bid submission regulations].” E. The False Claims Act Contains a Duty to Correct Known Errors 71. The False Claims Act contains an independent requirement to correct errors that will cause, or have caused, a government overpayment. The Act attaches liability to anyone who knowingly makes, uses, or causes to be made or used, a false statement or record material to an obligation to pay or transmit money to the government, or who knowingly conceals or knowingly and improperly avoids or decreases an obligation to pay or transmit money to the government. 31 U.S.C. § 3729(a)(1)(G). 72. Accordingly, MA plans not only have a duty to submit correct data to CMS, but also, for data they have already submitted, must delete records known to be incorrect from CMS’s database using a “delete code.” V. BACKGROUND 73. As a non-profit, consumer-run organization, GHC has traditionally catered to the public interest, often highlighting its efforts to support low-income patients and provide affordable, quality care. GHC has received consistently high marks for the quality of care offered through its own facilities and its network of health care providers. {00043348; 1} 26 343281.2 Case 1:12-cv-00299-WMS Document 32 Filed 02/05/16 Page 29 of 58 Its Medicare Advantage plans have also traditionally been well regarded, receiving accolades from industry groups and Medicare itself. A. GHC’s Internal Risk Adjustment Review Department and Proactive Management of Risk Adjustment Claims Submission 74. As described above, because risk adjustment has a significant impact on capitation rates, Medicare Part C Plans must carefully monitor risk adjustment claims to ensure the completeness and accuracy of claims submitted. At GHC, the Insurance and Health Data Analysis (“IHDA”) department is primarily responsible for reviewing and verifying all risk adjustment codes submitted to CMS. Relator was head of the department, and carefully developed a risk adjustment methodology consistent with the applicable coding standards and CMS regulations. 75. As part of the internal procedures developed by IHDA, reviewers look for both diagnoses that were present in the medical records but were not coded by providers when they submitted their claims for reimbursement, and diagnoses included on the provider-submitted claims that are unsupported by documentation in the member’s medical record. IHDA fixes both types of errors. 76. GHC has several mechanisms to ensure proper coding and documentation. Primarily, GHC utilizes a software algorithm that mines its claims data to match previously submitted risk adjustment claims with existing documentation and to flag potential problems. Once the algorithm identifies a member for review, GHC’s team of three full-time resident nurses and a network of physicians reviews the documentation {00043348; 1} 27 343281.2 Case 1:12-cv-00299-WMS Document 32 Filed 02/05/16 Page 30 of 58 and coding supporting GHC’s risk adjustment claims for that member. Chart reviews are generally performed within seven days of a provider visit if data indicates the need to follow-up with the physician. 77. For example, the algorithms and chart reviewers look for situations in which a member’s past medical records suggested that he or she may have a condition, such as diabetes, but had not yet been treated for that condition in the current year. Similarly, if the reviewer noticed that the doctor prescribed a treatment for a chronic condition, but did not describe the treatment in the record (for example, prescribing insulin but not indicating treatment for diabetes), the provider would be contacted within seven days to amend the record, if appropriate. This ensures that a typographical oversight does not prevent the inclusion of diagnosis codes for members who are receiving active treatment. The prompt contact also ensures that the provider’s memory of the visit is fresh at the time of any amendment to the patient’s chart. 78. The algorithms and chart reviewers also look for situations in which a provider has submitted a claim suggesting a member is currently being treated for a condition, but the member’s overall medical record does not support that conclusion. For example, if a provider claims a member has an active cancer diagnosis, but the member received neither chemotherapy nor a surgical intervention recently, the algorithm flags that risk adjustment claim for review as it is likely the provider mistakenly diagnosed the member with active cancer instead of properly recording a history of cancer. In this {00043348; 1} 28 343281.2 Case 1:12-cv-00299-WMS Document 32 Filed 02/05/16 Page 31 of 58 situation, the diagnosis code would be reviewed and then corrected (through the submission of a “delete” code to CMS), resulting in a smaller monthly capitation payment from CMS to GHC. 79. Relator and her team carefully developed this process of retrospective review of the risk adjustment data for GHC. While strictly following CMS and industry coding standards, the IHDA team significantly increased GHC revenues. B. GHC Executives Try To Boost Risk Adjustment Revenues by Hiring Outside Vendor to Bypass IHDA Department 80. Despite the success of the IHDA department, GHC executives have twice hired outside consultants who promised to substantially increase GHC’s risk adjustment scores. In both cases, the outside vendors planned to do this largely by disregarding CMS coding and risk adjustment claims submission rules. Relator and other members of GHC’s Documentation and Coding Core Team (“DC Team”) (which team included Relator and other key GHC personnel for billing, coding, and related issues) were able to stop the first such effort (by vendor Leprechaun LLC), but unfortunately were unable to stop DxID’s fraudulent efforts. 81. As early as 2007, employees in the IHDA department began hearing complaints from GHC leadership that the department was not generating enough money. Relator was criticized by her superiors for being too “conservative” in her approach to risk adjustment. {00043348; 1} 29 343281.2 Case 1:12-cv-00299-WMS Document 32 Filed 02/05/16 Page 32 of 58 82. Around October 2008, GHC hired Leprechaun LLC to help GHC “improve” its risk adjustment scores. Leprechaun provides risk adjustment data review services to MA plans, and was known for producing increased revenues through the chart review process. 83. Leprechaun worked with Relator’s department in its retrospective chart review and coding process. Leprechaun tried to introduce coding and documentation standards that conflicted with CMS regulations. Specifically, it proposed submitting risk adjustment claims based on documentation that was clearly inadequate to support such claims under CMS rules. Relator and others on the DC Team complained to GHC executives and otherwise resisted these efforts by Leprechaun. They demanded the company change its review policies to conform to traditional coding methodologies and avoid compliance risks. Leprechaun and GHC leadership eventually complied. 84. Relator and her compatriots prevented Leprechaun from submitting the false risk adjustment claims to CMS that it had proposed submitting. Leprechaun, however, did identify a number of legitimate risk adjustment claims that GHC had previously missed. These claims were submitted to CMS. 85. Despite an increasingly positive financial performance by the IHDA department, GHC’s financial condition deteriorated, due largely to poor business decisions by company management. From 2008 through 2010, GHC went from an operating income of almost $57 million to an operating loss of $60 million. Concerned {00043348; 1} 30 343281.2 Case 1:12-cv-00299-WMS Document 32 Filed 02/05/16 Page 33 of 58 with the company’s trajectory, GHC leadership made promises to the board of directors regarding the financial performance of the non-profit institution. Because of the substantial financial impact of risk adjustment, GHC CEO Scott Armstrong and CFO Ric Magnuson turned their attention to this area as a potential source of additional revenue — even if doing so meant breaking CMS rules. 86. In late 2011, Armstrong attended a conference held by the Alliance of Community Health Plans (“ACHP”) and that at the conference, he spoke with a colleague from Independent Health. (Relator believes Armstrong spoke with IHC’s Chief Executive Officer.) 87. The IH executive told Armstrong about an exciting opportunity with a new IHC subsidiary, DxID. He reported that IH had made a lot of money using DxID’s risk adjustment methodology and algorithms to conduct a retrospective review of its risk adjustment claims. 88. Upon returning from the conference, Armstrong approached GHC’s Chief Financial Officer, Ric Magnuson, and instructed him to hire DxID. After a number of meetings with senior management and with the DC Team, DxID was officially hired in November 2011 to perform a risk-adjustment data review for 2010 dates of service. 89. Relator has heard and believes, and on that basis alleges, that DxID’s contract with GHC was based, in whole or in part, on an incentive-based payment model, whereby DxID was paid a percentage of the value of the new risk adjustment claims it {00043348; 1} 31 343281.2 Case 1:12-cv-00299-WMS Document 32 Filed 02/05/16 Page 34 of 58 submitted. Relator understands that DxID submitted approximately $12 million in new risk adjustment claims for 2010 for GHC, and was paid approximately $1.5 million in incentive compensation. VI. DEFENDANTS DEFRAUD THE UNITED STATES 90. Defendants have engaged in a deliberate scheme to defraud the United States by submitting thousands of false claims for risk adjustment payments on behalf of both GHC and IH. Defendants submitted and caused the submission of these false claims (and conspired to do the same) knowing that the patients upon whom the claims were based did not have the claimed diagnoses, had not been treated for those diagnoses in that year, or were otherwise ineligible for risk adjustment payments under CMS rules. 91. Defendants have “upcoded” the risk adjustment claims they submitted to Medicare, claiming that a patient had been treated, in the relevant time period for: (a) a diagnosis that the patient did not have; (b) a more severe diagnosis than the one the patient had; and/or (c) a diagnosis the patient may have previously been treated for, but which was not treated in the relevant year. 92. Contrary to Medicare rules, Defendants have also submitted risk adjustment claims even though the member’s physician did not diagnose the patient as having the condition in question or did not, according to his or her own records, treat the patient for the condition in question during the relevant year. First, Defendants submitted risk adjustment claims supported only by vague references in the patient’s chart or {00043348; 1} 32 343281.2 Case 1:12-cv-00299-WMS Document 32 Filed 02/05/16 Page 35 of 58 documents that CMS rules plainly state may not be used to substantiate risk adjustment claims. Second, and worse, in some cases Defendants submitted risk adjustment claims even though the patient’s physician explicitly stated in his or her treatment notes that the patient did not have and/or had not been treated for the exact condition Defendants claimed. 93. Defendants DxID and IH have also submitted risk adjustment claims based on documents generated by DxID and IH, which were not reviewed or approved by the treating physician until months or years after the date of the medical service in question. 94. Defendants have also refused to correct previously submitted risk adjustment claims even though Defendants knew, or should have known, those claims were false. 95. In this manner, Defendants have fraudulently caused CMS to pay thousands of false claims for risk adjustment payments worth millions of dollars. A. DxID’s “Audit” of GHC’s Risk Adjustment Claims and Subsequent Submission of Thousands of False Claims to CMS on GHC’s Behalf 96. As noted above, in approximately November 2011, GHC hired DxID to conduct a retrospective review of its risk adjustment claims for 2010 dates of service. Under CMS rules, any changes to claims for 2010 dates of service were due by January 31, 2012. Thus, DxID worked for less than three months to identify as many additional diagnosis codes as possible. {00043348; 1} 33 343281.2 Case 1:12-cv-00299-WMS Document 32 Filed 02/05/16 Page 36 of 58 97. Early on, it became clear to Relator and others on the DC Team that DxID’s review process did not comply with CMS rules. DxID proposed using invalid documentation sources, such as “problem lists” (a part of the medical record that often contains notations about diagnoses that a patient may have, may once have had, or are otherwise of interest to the provider), to support new risk adjustment claims. DxID also proposed submitting risk adjustment claims whenever a patient had a diagnosis, regardless of whether the patient had been treated for the condition, as CMS rules require. 98. To prevent Relator or others connected with her IHDA department from enforcing CMS rules (as they had done with the prior vendor, Leprechaun), GHC leadership directed that DxID bypass IHDA when submitting its new risk adjustment claims. Instead of using GHC’s standard process for risk adjustment claims — which included review by IHDA — GHC’s leadership directed DxID to create a file of the new risk adjustment claims in a format ready for submission to CMS, and to then submit these claims through GHC’s Finance and Decision Support (“FDS”) department, in effect creating an end-run around IHDA and GHC’s established channels for submission of risk adjustment data to CMS. The FDS department is run by GHC’s head of Medicare Finance & Decision Support, Debbie Sather, who has long complained that IHDA was being “too conservative” by following CMS rules for coding and risk adjustment claims submission. {00043348; 1} 34 343281.2 Case 1:12-cv-00299-WMS Document 32 Filed 02/05/16 Page 37 of 58 99. Before conducting its review, DxID asked GHC’s leadership to approve or reject certain coding policies — policies that were largely in violation of CMS rules. As part of its presentation to GHC advocating for each policy, DxID included an estimation of the rate of return GHC could expect if it adopted the policy in question. 100. On December 27, 2011, GHC and DxID employees participated in a conference call to address coding policies for three specific conditions: chronic kidney disease (“CKD”), hypoxemia, and old myocardial infarctions (“old MI”). See Exhibit 4, GHC-DxID Conference Call Notes, incorporated herein. 101. DxID encouraged GHC to adopt policies that would allow claims to be submitted for these three conditions in cases that failed to meet CMS standards. For example, DxID encouraged GHC to submit risk adjustment claims for CKD if “[i]t is documented in the Snapshot and labs are ordered overtime and fall in the range of CKD level 3 or higher.” Exh. 2 at 1. DxID also coded from labs not taken in the payment year. This suggestion violates CMS rules for risk adjustment claims in several ways. 102. First, the medical record’s “Snapshot” is merely an automatic entry generated from previous visits. It provides a general overview of the patient’s past condition, but does not reflect the treatment the patient received during the visit in question. Accordingly, it may not be used as the basis for a risk adjustment claim. {00043348; 1} 35 343281.2 Case 1:12-cv-00299-WMS Document 32 Filed 02/05/16 Page 38 of 58 103. Second, test results may not be used, by themselves, as the basis for a risk adjustment claim. Thus, in this instance a lab report without any interpretation of the results by the treating physician is an invalid basis for submission of a claim. 104. GHC adopted DxID’s flawed CKD coding proposal. 105. For hypoxemia, DxID proposed submitting a claim for the diagnosis: “if COPD or emphysema is listed in active problem list and member is on oxygen.” DxID justified this violation of CMS rules by arguing: “[w]e would say that the doctor has recognized Hypoxemia and ordered the oxygen.” Id. at 3. Again, though, diagnoses listed on problem lists may not be used as the basis for a risk adjustment claim unless the problem list includes documentation that the diagnosis was treated or affected treatment provided to the patient on the day in question. Moreover, documentation concerning the provision of medical supplies and equipment (such as oxygen) may not be used as the basis for a risk adjustment claim. Id. at 3. 106. GHC adopted DxID’s flawed hypoxemia coding rule. 107. Next, DxID proposed a very aggressive approach to coding old MIs. DxID encouraged GHC to submit risk adjustment claims for old MI in cases in which there was no mention at all of the condition by the physician in the medical record for the relevant date of service. DxID proposed submitting a claim for old MI so long as there was mention of an old MI at some point in the patient’s chart, regardless of how long ago, and even if that reference appeared in unconfirmed test results. See Exh. 2 at 3. {00043348; 1} 36 343281.2 Case 1:12-cv-00299-WMS Document 32 Filed 02/05/16 Page 39 of 58 108. CMS coding rules for risk adjustment claims bar claims for old MIs supported exclusively by EKGs for three reasons: (1) they are not drawn from a face-toface visit; (2) they are not from an approved provider type, in that they are coded from a visit with a diagnostic radiologist; and (3), in some cases, they are not documented in the treatment year. 109. DC Team Member and Director of Health Information Management, Rhona Moses, opposed this unlawful stance, stating “[t]here needs to be some verbiage to indicate a heart attack in chart note in the [date of service] year.” Exh. 2 at 3. 110. GHC nonetheless adopted DxID’s bogus old MI policy. 111. GHC performed a risk-reward analysis in assessing each policy decision proposed by DxID. In the December 27, 2011 conference call, “Dr. Tarnoff suggested DxID estimate value of what CKD, Old MI, and Hypoxia would be worth and the frequency within member population. GHC would use this information internally to assess the risk and decide the next step.” Exh. 2 at 2. The financial impact of the coding policies was mentioned numerous times during the call. (e.g., “DxID has provided, for Debbie [Sather] and Ric [Magnuson], a sample of financial impact for CKD and Old MI.” Id. at 5) 112. Prior to January 31, 2012, DxID submitted 4,578 new diagnosis codes for risk adjustment claims to CMS on GHC’s behalf. DxID reviewed 15,875 patient charts to find those 4,578 new claims. {00043348; 1} 37 343281.2 Case 1:12-cv-00299-WMS Document 32 Filed 02/05/16 Page 40 of 58 113. In a February 2012 PowerPoint, DxID presented its results to the DC Team. See Exhibit 1, PowerPoint Presentation, incorporated herein. In that presentation, DxID identified the top “finds” from its review — meaning the highest volume codes that risk adjust — including: (1) old MI; (2) polyneuropathy; (3) vascular disease; (4) CKD; and (5) Chronic Obstructive Pulmonary Disorder (“COPD”). 114. In a corresponding document, “DxID Post-2010 Chart Review Outcomes,” DxID expanded on the procedures it used during the 42-day chart review cycle for GHC. Specifically, it listed by HCC the number of GHC members affected by the review. This list summarized the number of new diagnosis codes within each HCC that were submitted to CMS as a result of the review. See Exhibit 2, DxID Chart Review Outcomes, incorporated herein, page 4. 115. After DxID disclosed its results and methods to the DC Team, Relator became concerned about the findings. Given her role overseeing GHC’s own internal audit and review processes for risk adjustment claims, she knew the error rate in the charts was not as high as DxID represented. DxID found a new diagnosis code for every three to four charts it reviewed — a number completely inconsistent with Relator’s years of experience performing chart reviews for GHC. 116. Moreover, in the Chart Review Outcomes document, DxID also outlined the coding and audit rules it used to identify new risk adjustment claims for GHC. Relator identified numerous errors in DxID’s application of CMS coding rules. For {00043348; 1} 38 343281.2 Case 1:12-cv-00299-WMS Document 32 Filed 02/05/16 Page 41 of 58 example, CMS guidelines state that a “problem list” may only be used as the basis for a risk adjustment claim if it shows “evaluation and treatment for each condition that relates to an ICD-9 code on the date of service, and it must be signed and dated by the physician.” Participant Guide at 7-17; Exhibit 2 at 17. DxID, on the other hand, applied a rule whereby the mere presence of a diagnosis in the problem list was sufficient to justify a risk adjustment claim, even where the providers’ notes explicitly state that the patient did not have that condition. 117. Similarly, DxID’s Chart Review Outcomes presentation to GHC states that a diagnosis may be submitted for a risk adjustment claim as long as there is a laboratory, radiology, or other diagnostic test result in the patient’s chart that has been signed by a physician, regardless of whether the physician treated the patient for that condition in the year in question. See id. at 6; see also id. at 16 (“laboratory results indicating hypoxemia that are documented in a dated medical record signed by a provider relevant for risk adjustment should be considered for the purposes of risk adjustment”). CMS rules plainly prohibit the use of laboratory, radiology, and other diagnostic test results as the sole support for risk adjustment claims. Instead, a qualified provider must actually treat the patient for the condition in question in the relevant year. 118. DxID policies (that GHC adopted) also direct the submission of risk adjustment claims based on claims for durable medical equipment, as long as the initial order for the equipment required a physician to document its medical necessity. See id. {00043348; 1} 39 343281.2 Case 1:12-cv-00299-WMS Document 32 Filed 02/05/16 Page 42 of 58 at 16 (“If a Medicare Advantage organization uses clinical guidelines that require clinical evidence of hypoxemia to provide home oxygen, then the use of continuous oxygen should be considered for the purposes of risk adjustment.”) Again, however, CMS rules provide that a risk adjustment claim may only be submitted if a physician or hospital treated the patient in a face-to-face visit in the year in question. See Participant Guide at 7.1.5. Without such a face-to-face visit, the existence of a prior certification of the existence of a diagnosis is not enough to support a risk adjustment claim, even when combined with current treatment through durable medical equipment. 119. Throughout the presentation, DxID makes clear that its approach looks only at whether the patient has the diagnosis, not whether they were treated for the condition by a qualified provider in the relevant time period. See, e.g., id. at 8 (“The Plan has to substantiate, from a clinical and coding standpoint, that the patient has the disease and that the medical record supports the fact that they do.”). 120. Relator asked for permission from her superiors to review the diagnosis codes identified by DxID. She was provided with a copy of the over four thousand codes that had been submitted to CMS as a result of the review process, and the corresponding patient number and dates of service. Together with her physician partner, Dr. Don Rappe, Relator began a code-by-code review of the new diagnoses to look for support in the medical record. {00043348; 1} 40 343281.2 Case 1:12-cv-00299-WMS Document 32 Filed 02/05/16 Page 43 of 58 121. Immediately, Relator and Dr. Rappe began finding systematic problems with the diagnosis codes. Claims were purportedly justified by improper documentation, if there was any documentation at all. Relator and Dr. Rappe initially reviewed 117 charts. Of the charts reviewed, Relator and Dr. Rappe agreed with only 27 (23%) of the codes submitted to CMS. An additional three percent of the codes reviewed had previously been coded and submitted by GHC. Based on their professional experience, Dr. Rappe and Relator concluded 74% of the codes submitted did not have sufficient documentation to justify submitting the diagnosis. The majority of these false claims (42 claims or 36% of the total) were based solely on the inclusion of the diagnosis in the problem list, where there was no other reference to the diagnosis in the medical record. In other instances, the new diagnosis codes had no support whatsoever (5%), were based solely on test results (4%), or were based on a problem list from a visit with an improper provider (3%). 122. In addition to fraudulently increasing new diagnoses, DxID failed to remove previously submitted incorrect diagnoses. The only codes DxID “deleted” were ones it had added, and later determined to be unsupported through a quality control process. GHC even submitted some of those codes despite DxID’s reservations. See Exhibit 3 (email and redacted spreadsheet of “delete” codes). DxID neither removed, nor recommended GHC remove, any codes for which the provider originally included an incorrect diagnosis. {00043348; 1} 41 343281.2 Case 1:12-cv-00299-WMS Document 32 Filed 02/05/16 Page 44 of 58 123. This failure to delete previously submitted codes that were not supported by the medical record is consistent with DxID’s general policy. When DxID first explained its procedures to the DC Team, a member of the team asked DxID whether DxID deleted previously submitted codes found to be incorrect. DxID stated it neither looked for, nor if it found did it delete, any previously submitted incorrect claims. 124. In March 2012, after her review of 117 new diagnoses from DxID, Relator presented her findings to her superiors at GHC. She specifically highlighted the lack of proper documentation and prevalence of clearly erroneous diagnoses. She also reported that DxID “added” codes that GHC had previously submitted to CMS. GHC took no action to inform CMS of the mistakes, nor did it suggest a broader review of the findings. The only concern Relator’s superiors expressed was that GHC had overpaid DxID for those codes that had already been submitted. It was due to this concern, not any concern for accuracy of the truly new claims, that Relator’s superiors suggested she continue her review of the 4,500 diagnoses. 125. Relator subsequently reviewed an additional 83 diagnosis codes that DxID used to submit risk adjustment claims for GHC. The error rate for these 83 codes was similar to rate for the original 117 codes – 74% of them did not meet CMS standards. 126. Relator continued to encourage GHC to reverse course and delete these false and fraudulent claims. On April 4, 2012, she submitted to DxID a list of 40 of the new risk adjustment claims for further review. Her superiors only allowed her to submit {00043348; 1} 42 343281.2 Case 1:12-cv-00299-WMS Document 32 Filed 02/05/16 Page 45 of 58 40 diagnosis codes for further review, because they decided to accept DxID’s coding policies for the remaining codes, even though those policies violated CMS rules. For example, notwithstanding CMS’s clear rules, GHC decided to adopt DxID’s policy of submitting claims based solely on a reference in the patient’s “problem list.” 127. Based on the results of Relator’s audit, over 40% of the diagnosis codes DxID submitted through its chart review of 2010 dates of service were coded at least in part from embedded problem lists. 128. On April 11, 2012, DxID responded to the list of 40 diagnosis codes Relator found were improperly submitted. Even under the most generous interpretation of CMS rules, DxID’s purported justification for 25 of the diagnosis codes failed. 129. In five cases, DxID submitted a diagnosis code for a condition that was: (1) listed on a problem list embedded in a patient’s medical record; and (2) from a visit with a type of physician or other provider who would be highly unlikely to treat the condition. For example, for one patient, DxID submitted a claim alleging the patient had been treated for major depression (diagnosis code 296.33) during a January 19, 2010 visit with an ophthalmologist. The physician had copied the patient’s problem list, including the past diagnosis of depression, into the body of his note. Unsurprisingly, there was no suggestion in the treatment notes that the ophthalmologist had treated or considered the patient’s prior depression during the course of the eye care visit. {00043348; 1} 43 343281.2 Case 1:12-cv-00299-WMS Document 32 Filed 02/05/16 Page 46 of 58 130. To justify its submission of a diagnosis code for major depression, DxID provided GHC with an “audit trail,” identifying the specific information in the medical record it used to support its code submission, and additional comments generated to respond to the inquiry. In this case, DxID attempted to justify the submission of a risk adjustment claim for major depression by noting “[p]atient active problem list - major depressive disorder-recurrent-w/o psychotic [296.33].” 131. Despite Relator’s warnings and GHC’s knowledge of the remarkable error rate in DxID’s submissions to CMS for the 2010 dates of service, GHC hired DxID to conduct a more extensive review of risk adjustment claims for 2011 dates of service. 132. For the review of 2011 services, GHC instructed DxID to submit risk adjustment claims based on a source of documentation previously rejected by GHC and DxID for 2010 dates of service: incidental findings. 133. “Incidental findings” are diagnoses found (by coincidence) in the results of tests ordered by a treating physician, but neither treated by the physician at the time the tests were ordered, nor subsequently. It is inappropriate to submit risk adjustment claims for “incidental findings” because CMS rules prohibit the submission of diagnosis codes based solely on diagnostic radiology services or laboratory services, and often such diagnoses have never been confirmed by a physician. 134. On May 30, 2012, the Office of the Inspector General published a report of an enforcement action taken against an MA plan, PacifiCare Texas. The OIG found {00043348; 1} 44 343281.2 Case 1:12-cv-00299-WMS Document 32 Filed 02/05/16 Page 47 of 58 that 43 of the 100 risk adjustment claims it audited were invalid. The report published information about the coding rules and established CMS standards applied by the auditors in reviewing PacifiCare’s risk adjustment claims. Several of these rules have also been routinely violated by GHC and DxID. 135. In June 2012, GHC employees circulated the report. Members of the DC Team were seriously concerned about the results. 136. In response, Debbie Sather suggested that GHC would review the new coding policies adopted in connection with the DxID review. She indicated that she would lead this review, and would include GHC leadership in the discussion. 137. Sather announced in July 2012 that GHC decided to reverse only one coding policy: the policy of submitting diagnosis codes for old MIs based on EKGs. Significantly, she also indicated that this policy change would only affect future DxID reviews. Thus the bogus claims submitted for calendar year 2010 service dates would not be corrected. Relator specifically offered to correct this problem for 2010 dates of service and was instructed by Sather not to do so. 138. Upon information and belief, DxID’s review process for 2011 likely generated more unsubstantiated codes than the previous review because DxID had more time to conduct its review. 139. GHC leadership has also pressured its internal coding team to adopt DxID’s lax and improper documentation standards for its own work. GHC conducted {00043348; 1} 45 343281.2 Case 1:12-cv-00299-WMS Document 32 Filed 02/05/16 Page 48 of 58 internal meetings regarding the coding policies DxID applied in its review, to determine whether to adopt the policies within GHC’s internal review procedures. Although Debbie Sather advocated for adoption of DxID’s improper coding rules, other individuals within GHC have refused to adopt the new policies. Specifically, Rhona Moses, a certified coder, in a meeting in early March 2012, expressed dissatisfaction with the new coding procedures and suggested they were inconsistent with CMS coding guidelines, with which MA plans are required to comply. She has refused to code according to DxID’s standards. B. Examples of Specific False Claims Submitted By DxID and GHC 140. At the end of January 2012, DxID, with GHC’s consent, submitted false risk adjustment claims to CMS on GHC’s behalf, for the following patients. (These patients are identified in this complaint by anonymous references, Patient A, Patient B, etc. to protect their confidentiality. Relator has in her possession and has provided to the United States, as part of her statutory disclosure, more specific identifying information for each of these patients. Such additional detail is incorporated into this complaint by reference to the extent required by Federal Rule of Civil Procedure 9(b) and as allowed by federal and any relevant state privacy statutes and rules.) 141. Patient A had a routine medical examination on September 2, 2010. In the course of his visit, the doctor used a common function of the Electronic Medical Record (“EMR”) software to insert the text of the patient’s “Problem List” into the {00043348; 1} 46 343281.2 Case 1:12-cv-00299-WMS Document 32 Filed 02/05/16 Page 49 of 58 treatment notes for that visit. In Relator’s experience, doctors often use this function to add the problem list into the text of their notes in the EMR, because doing so allows them to review the problem list while composing their treatment notes, rather than requiring the more onerous process of closing their notes and opening the problem list each time they wish to reference it. The problem list mentioned a diagnosis of “Major Depressive Disorder, Recurrent Episode, Mild [diagnosis code 296.31].” This indicated that at some point in the past (in this patient’s case, in May 2004) a doctor diagnosed this patient with the specified condition. The problem list does not necessarily correspond to the conditions that patient currently has, nor to conditions the doctor actively treated in the face-to-face visit. 142. During the course of Patient A’s September 2, 2010 visit, the doctor specifically evaluated the patient’s mood, noting “[the patient] does not have much in the way of depression in fact has an amazingly sunny disposition.” (emphasis added). The doctor concluded the major depressive disorder was “resolved.” On the day of that visit, the doctor removed the diagnosis from the problem list. 143. Despite this clear notation by the treating physician that the patient was not suffering from major depression, and the fact the doctor removed the diagnosis from the problem list, DxID coded Patient A for Major Depressive Disorder [diagnosis code 296.31]. Because the patient neither had, nor was treated for Major Depressive Disorder in 2010, the risk adjustment claim DxID submitted for that diagnosis, on GHC’s behalf {00043348; 1} 47 343281.2 Case 1:12-cv-00299-WMS Document 32 Filed 02/05/16 Page 50 of 58 and with its permission, was false and fraudulent within the meaning of the False Claims Act. 144. Patient B was treated on March 22, 2010 for diabetes. When diabetes is linked with nerve issues in the extremities, it is called “diabetic neuropathy.” A patient with diabetes without other complications is grouped into HCC 19. A patient with diabetes so serious that it has also caused diabetic neuropathy falls into the substantially more lucrative HCC 16. For Patient B, diabetic neuropathy appeared on the problem list. However, the doctor examined the patient and determined that “She does not have significant diabetic neuropathy” (emphasis added) — which is the standard language inserted by the EMR when the doctor indicates that the patient does not have diabetic neuropathy. DxID, despite this express statement, and its knowledge that this language was automatically inserted when the doctor has determined the patient does not have diabetic neuropathy, submitted a risk adjustment claim to CMS for Patient B for diagnosis code 357.2 Polyneuropathy in Diabetes. Because Patient B was not diagnosed with (let alone treated for) diabetic neuropathy in 2010, the risk adjustment claim DxID submitted for that diagnosis, on GHC’s behalf and with its permission, was false and fraudulent within the meaning of the False Claims Act. 145. Patient C was treated on November 17, 2010 for diabetic neuropathy. In addition to diabetes, the problem list also indicated the patient had Chronic Kidney Disease (“CKD”). Diabetes can cause kidney disease, resulting in a condition called {00043348; 1} 48 343281.2 Case 1:12-cv-00299-WMS Document 32 Filed 02/05/16 Page 51 of 58 diabetic nephropathy, but a diabetic patient with CKD does not necessarily have diabetic nephropathy. In many cases, a patient’s CKD and diabetes are independent of each other. Where a patient’s diabetes is so severe that it causes CKD, the patient is so ill that he or she is likely to be substantially more expensive to treat. Accordingly, the HCC for diabetes with diabetic nephropathy is significantly more lucrative than the HCC for diabetes alone. 146. In Patient C’s case, the doctor tested the patient specifically for albumin, a protein indicating the presence of diabetic nephropathy. The doctor explicitly found the patient “[did] not have significant diabetic nephropathy.” (emphasis added). Despite the fact that the diagnosis had specifically been ruled out by a physician on the referenced visit date, DxID added and submitted code 250.4 Diabetes with Renal Manifestations. Because Patient C was not diagnosed with (let alone treated for) diabetic nephropathy in 2010, the risk adjustment claim DxID submitted for that diagnosis, on GHC’s behalf and with its permission, was false and fraudulent within the meaning of the False Claims Act. 147. Patient D was treated on February 11, 2010 by a family practice physician for cardiology-related issues. During the visit, there was no mention of or reference to an old myocardial infarction (“old MI”). Patient D received several ECGs in 2009 that revealed inconsistent results as to the presence of an old MI. Submitting a risk adjustment claim for old MI based on the February 2010 visit is triply wrong, because: (1) the treating physician gave no indication she had diagnosed the patient with an old {00043348; 1} 49 343281.2 Case 1:12-cv-00299-WMS Document 32 Filed 02/05/16 Page 52 of 58 MI, let alone that such a diagnosis had affected her treatment of the patient; (2) the 2009 ECG that revealed an old MI was a diagnostic test performed by a technician, not treatment provided in a face-to-face encounter with a qualified provider as required to support a risk adjustment claim; and (3) the 2009 ECG report fell outside of the relevant time period (calendar year 2010). Nevertheless, DxID submitted a risk adjustment claim for diagnosis code 412, old MI. Because the patient was never diagnosed with (let alone treated for) an old MI during 2010, the risk adjustment claim DxID submitted for that diagnosis, on GHC’s behalf and with its permission, was false and fraudulent within the meaning of the False Claims Act. 148. These are but a few select representative examples of the types of false risk adjustment claims that DxID and GHC conspired to submit and did submit to CMS. 149. GHC estimated that CMS would pay it more than $12 million for the risk adjustment claims DxID submitted on GHC’s behalf for the 2010 service year. If 74% of those claims were erroneous (consistent with the error rate Relator has found during her review), GHC and DxID have submitted and conspired to submit more than $8 million in false claims to the United States. 150. Based on the most recent reports Relator possessed before leaving GHC, she estimates that DxID submitted in excess of $23.3 million in risk adjustment claims on GHC’s behalf for the 2011 service year. {00043348; 1} 50 343281.2 Case 1:12-cv-00299-WMS Document 32 Filed 02/05/16 Page 53 of 58 C. DxID and IH Submitted and Conspired to Submit False Risk Adjustment Claims on IH’s Behalf 151. As noted above, GHC learned about DxID when an Independent Health executive told GHC’s CEO that Independent Health had been using DxID’s algorithms and methodologies in the past and had seen a significant positive financial impact. 152. Because DxID was created under IHC in the Fall of 2011 — mere months before it began working for GHC — it is highly likely the algorithms and procedures DxID used to review IH’s risk adjustment claims are the same as the procedures used to review GHC’s claims. 153. IH and DxID also submitted unsupported risk adjustment diagnosis codes to CMS through the use of physician “queries” and addenda to medical records. In short, in situations where DxID (and IH) suspect a patient may have a condition — notwithstanding the fact the patient’s treating physician neither diagnosed nor treated the condition in the year in question — DxID (and IH) send the treating physician a form — often many months or more than a year after the fact — asking them to claim the diagnosis was treated. 154. Relator learned about DxID and IH’s use of the “addendum” process because DxID tried to convince GHC to do the same. Specifically, DxID pitched the practice during a May 22, 2012 meeting. DxID reported that Independent Health had been doing it for five years, and that it was lucrative. {00043348; 1} 51 343281.2 Case 1:12-cv-00299-WMS Document 32 Filed 02/05/16 Page 54 of 58 155. According to a policy document DxID gave to GHC, (see Exhibit 5), DxID reviews patient medical records to find conditions that DxID believes the patient likely has, but for which he or she had not yet been treated during the relevant treatment year (or for which treatment documentation is absent). DxID then sends the patient’s treating physician(s) a standardized letter that essentially asks the doctor to claim — again, often months after the fact and contrary to the contemporaneous medical records — that the patient was treated for the condition DxID wants to submit a claim for. See also Exhibit 6, Sample Addendum Form. 156. The addendum forms are drafted to lead physicians to certain diagnoses. Rather than pose an open ended question regarding the meaning of a particular entry in the patient’s medical record, a physician is provided with a list of possible conditions and told that DxID (or the health plan) believes the patient has those conditions. See Exhibit 6. The conditions on the form have check boxes next to them. The instructions direct physicians to “select any that apply.” All conditions listed on the sample form correspond with diagnoses that may be used for the submission of a risk adjustment claim. 157. Medical records may only be changed after the date of service when, and to the extent, the physician remembers the service and can attest to what s/he actually considered and treated at the time. {00043348; 1} 52 343281.2 Case 1:12-cv-00299-WMS Document 32 Filed 02/05/16 Page 55 of 58 158. Through the addendum process, IH and DxID pressure physicians to improperly amend medical records to include diagnoses for conditions that were not supported in medical record documentation. IH and DxID systematically pressure physicians to improperly amend medical records with new, high value HCCs. This has resulted in the submission of fraudulent risk adjustment claims to CMS. 159. For these reasons, Relator believes, and on that basis alleges, that IH and DxID have submitted and conspired to submit false risk adjustment claims to CMS on IH’s behalf using the same or substantially similar fraudulent processes and procedures as those DxID used in connection with its fraudulent review and submission of risk adjustment claims for GHC. Relator also believes, and on that basis alleges, that IH and DxID have submitted and conspired to submit false risk adjustment claims using the query and addendum process to document additional risk adjustment diagnosis codes. COUNT I Violations of the Federal False Claims Act 31 U.S.C. §§ 3729(a)(1)(A)–(C), (G) 160. Relator realleges and incorporates by reference the allegations made in Paragraphs 1 through 159 of this Complaint. 161. This is a claim for treble damages and forfeitures under the Federal False Claims Act, 31 U.S.C. §§ 3279–33, as amended. 162. Through the acts described above, Defendants, their agents, employees, and co-conspirators, knowingly presented, or caused to be presented, to the United States {00043348; 1} 53 343281.2 Case 1:12-cv-00299-WMS Document 32 Filed 02/05/16 Page 56 of 58 false and fraudulent claims, and knowingly failed to disclose material facts, in order to obtain payment or approval from the United States and its contractors, grantees, and other recipients of its funds. 163. Through the acts described above, Defendants, their agents, employees, and co-conspirators, knowingly made, used, and caused to be made and used false records and statements, which also omitted material facts, in order to induce the United States to approve and pay false and fraudulent claims. 164. Through the acts described above, Defendants, their agents, employees, and co-conspirators, knowingly made, used, and caused to be made and used false records and statements material to an obligation to pay and transmit money to the United States, and knowingly concealed and improperly avoided and decreased an obligation to pay and transmit money to the United States. 165. Through the acts described above, Defendants, their agents, employees, and other co-conspirators knowingly conspired to submit false claims to the United States and to deceive the United States for the purpose of causing the United States to pay or allow false or fraudulent claims. 166. The United States, unaware of the falsity of the records, statements, and claims made and submitted by Defendants, its agents, employees, and co-conspirators, and as a result thereof, paid money that it otherwise would not have paid. {00043348; 1} 54 343281.2 Case 1:12-cv-00299-WMS Document 32 Filed 02/05/16 Page 57 of 58 167. By reason of the payments made by the United States, as a result of Defendants’ fraud, the United States has suffered millions of dollars in damages and continues to be damaged. PRAYER WHEREFORE, qui tam plaintiff Teresa Ross prays for judgment against the defendants Group Health Cooperative, Independent Health Association, Independent Health Corporation, DxID LLC, Dr. John Haughton, and Betsy Gaffney (“Defendants”) as follows: 1. That Defendants cease and desist from violating 31 U.S.C. §§ 3729–33. 2. That the Court enter judgment against Defendants in an amount equal to three times the amount of damages the United States has sustained as a result of Defendants’ actions in violation of the Federal False Claims Act, as well as a civil penalty of $11,000 for each violation of 31 U.S.C. § 3729; 3. That Relator be awarded the maximum amount allowed pursuant to 31 U.S.C. § 3730(d) of the Federal False Claims Act; 4. That Relator be awarded all costs of this action, including attorneys' fees and expenses; and 5. That the United States and Relator receive all such other relief as the Court deems just and proper. {00043348; 1} 55 343281.2 Case 1:12-cv-00299-WMS Document 32 Filed 02/05/16 Page 58 of 58 JURY DEMAND Pursuant to Rule 38 of the Federal Rules of Civil Procedure, Relator hereby demands trial by jury. DATED: February 5, 2016 Respectfully submitted, By: /s/ Brian M. Melber Rodney O. Personius Brian M. Melber Personius Melber LLP 2100 Main Place Tower Buffalo, NY 14202 Tel: (716) 855-1050 Fax: (716) 855-1052 rop@personiusmelber.com bmm@personiusmelber.com Timothy P. McCormack Constantine Cannon LLP 1001 Pennsylvania Ave. NW Suite 1300N Washington, DC 20004 Tel: (202) 204-3516 Fax: (202) 204-3501 tmccormack@constantinecannon.com Attorneys For Qui Tam Plaintiff-Relator [Under Seal Relator A] {00043348; 1} 56 343281.2