United States Court of Appeals FOR THE DISTRICT OF COLUMBIA CIRCUIT
January 30, 2009
CONSUMERS' CHECKBOOK, CENTER FOR THE STUDY OF SERVICES, APPELLEE
UNITED STATES DEPARTMENT OF HEALTH AND HUMAN SERVICES ET AL., APPELLANTS AMERICAN MEDICAL ASSOCIATION, INTERVENOR
Appeal from the United States District Court for the District of Columbia (No. 06cv02201).
The opinion of the court was delivered by: Karen Lecraft Henderson, Circuit Judge
Argued October 8, 2008
Before: HENDERSON, RANDOLPH and ROGERS, Circuit Judges.
Separate opinion filed by Circuit Judge ROGERS, concurring in part and dissenting in part.
Consumers' Checkbook, Center for the Study of Services (CSS) filed this action under the Freedom of Information Act (FOIA), 5 U.S.C. §§ 552 et seq., seeking from the United States Department of Health and Human Services (HHS) records for all Medicare claims submitted to HHS by physicians in several localities during 2004. The district court granted summary judgment in CSS's favor, concluding that the records are not exempt from disclosure under FOIA Exemption 6, id. § 552(b)(6). See Consumers' Checkbook, Ctr. for Study of Servs. v. U.S. Dep't of Health & Human Servs., 502 F. Supp. 2d 79, 83-86 (D.D.C. 2007) (Memorandum Opinion). For the reasons set forth below, we reverse the judgment of the district court.
On March 27, 2006, CSS submitted a FOIA request to the Centers for Medicare and Medicaid Services (CMS), a division within HHS, seeking a subset of data elements from all Medicare claims submitted by certain physicians in 2004. The data elements include the diagnosis, the type and place of service and the Unique Physician Identifying Number (UPIN) of the physician who performed the services. CSS limited its request to physicians in the District of Columbia, Illinois, Maryland, Washington and Virginia. It did not request data that identifies Medicare beneficiaries. At the time of the request, every physician was assigned a UPIN when he enrolled in Medicare.*fn1 A physician's name, office zip code, medical or surgical specialty and UPIN are publicly available on the internet. The fees a physician receives from Medicare for performing a specific service or procedure are also publicly available on the internet. Combined with the publicly available fee schedule, the data requested by CSS can be used to calculate the total payments Medicare made to any individually identified physician for claims submitted in 2004.
CMS denied the FOIA request and CSS appealed to the CMS Deputy Administrator. On December 26, 2006, CSS filed a complaint in district court under FOIA seeking injunctive relief. Both parties moved for summary judgment. HHS argued that the requested records are exempt from disclosure under FOIA Exemption 6. Alternatively, it argued that a twenty-nine- year-old permanent injunction issued by the United States District Court for the Middle District of Florida bars disclosure of the requested data from physicians who are American Medical Association (AMA) members. See Fla. Med. Ass'n v. Dep't of Health, Educ. & Welfare, 479 F. Supp. 1291 (M.D. Fla. 1979). In an opinion and order filed August 22, 2007, the district court granted summary judgment in CSS's favor and this appeal followed.*fn2 Memorandum Opinion at 81, 89.
HHS appeals the district court's grant of summary judgment as to the requested Medicare records the court held were not exempt from disclosure under FOIA Exemption 6.*fn3 We review the district court's grant of summary judgment de novo. Judicial Watch, Inc. v. FDA, 449 F.3d 141, 145 (D.C. Cir. 2006). FOIA provides that an agency must disclose all records upon request by "any person," 5 U.S.C. § 552(a)(3), unless a statutory exemption applies. Id. § 552(b). FOIA Exemption 6 provides that FOIA "does not apply to matters that are . . . personnel and medical files and similar files the disclosure of which would constitute a clearly unwarranted invasion of personal privacy." Id. § 552(b)(6). It is undisputed that the requested Medicare records are personnel, medical, or "similar files." See Memorandum Opinion at 83. Accordingly, we must determine whether "disclosure would compromise a substantial, as opposed to a de minimis, privacy interest." Nat'l Ass'n of Retired Fed. Employees v. Horner, 879 F.2d 873, 874 (D.C. Cir. 1989). If a substantial privacy interest is at stake, then we must balance the privacy interest in non-disclosure against the public interest. Id. (citing Ripskis v. HUD, 746 F.2d 1, 3 (D.C. Cir. 1984)). Disclosure is not required if it "would constitute a clearly unwarranted invasion of personal privacy." 5 U.S.C. § 552(b)(6). The agency bears the burden to persuade the court that the exemption applies. Ripskis, 746 F.2d at 3.
We have consistently held that an individual has a substantial privacy interest under FOIA in his financial information, including income. In Multi AG Media v. Department of Agriculture, 515 F.3d 1224 (D.C. Cir. 2008), we found that the disclosure of information on "irrigation practices, farm acreage, and the number and width of rows of tobacco and cotton" implicated substantial privacy interests because it would "in some cases allow for an inference to be drawn about the financial situation of an individual farmer" receiving federal subsidies. 515 F.3d at 1226, 1230. In Painting and Drywall Work Preservation Fund, Inc. v. HUD, 936 F.2d 1300 (D.C. Cir. 1991), we found that contractors on federal construction projects had a substantial privacy interest in their names, addresses, hourly pay, hours worked and wages. 936 F.2d at 1301-02; see also Sheet Metal Workers Int'l Ass'n, Local No. 9 v. U.S. Air Force, 63 F.3d 994, 995, 998 (10th Cir. 1995) (government contractors on federal construction projects have substantial privacy interest in payroll records); Painting Indus. of Haw. Market Recovery Fund v. U.S. Dep't of Air Force, 26 F.3d 1479, 1484 (9th Cir. 1994) (same); Hopkins v. HUD, 929 F.2d 81, 86-87 (2d Cir. 1991)(same). The Congress has also recognized the privacy interest an individual taxpayer has in his tax return information, including the "nature, source, or amount of his income," 26 U.S.C. § 6103(b)(2)(A), and has prohibited the disclosure of tax return information with limited exceptions. Id. § 6103(a).
The information requested by CSS would reveal the total Medicare payments received by a physician for covered services. CSS notes that the information would not reveal a physician's gross revenue because it would not include income from non-Medicare sources. Nor would it reveal a physician's net income because it would not include business operating expenses. But the requested information need not reveal completely an individual's personal finances to implicate substantial privacy concerns. See Multi AG Media, 515 F.3d at 1228-29 (substantial privacy interests implicated because requested information "would necessarily reveal at least a portion of the owner's personal finances") (quoting Nat'l Parks & Conservation Ass'n v. Kleppe, 547 F.2d 673, 685 (D.C. Cir. 1976)). CSS also argues that a physician does not have a privacy interest in Medicare payments because the payments relate to business activities and not personal finances. We have, however, recognized substantial privacy interests in business-related financial information for individually owned or closely held businesses because the "financial makeup of the businesses mirrors the financial situation of the individual family members." Multi AG Media, 515 F.3d at 1229 (internal quotations omitted); cf. Wash. Post Co. v. U.S. Dep't of Justice, 863 F.2d 96, 100 (D.C. Cir. 1988) (information related to employees' business decisions in developing and marketing medication does not implicate privacy interests under FOIA Exemption 7); Wash. Post Co. v. U.S. Dep't of Health & Human Servs., 690 F.2d 252, 261-62 (D.C. Cir. 1982) (no substantial privacy interest in list of organizations in which scientific consultants have financial holdings related to their consulting duties, but not dollar amounts of holdings). Accordingly, we conclude that physicians have a substantial privacy interest in the total payments they receive from Medicare for covered services.
We next examine the public interest in disclosure. The only relevant public interest in disclosure "is the extent to which disclosure would serve the 'core purpose of the FOIA,' which is 'contribut[ing] significantly to public understanding of the operations or activities of the government.'" U.S. Dep't of Def. v. FLRA, 510 U.S. 487, 495 (1994) (quoting U.S. Dep't of Justice v. Reporters Comm. for Freedom of Press, 489 U.S. 749, 775 (1989)). The requested information must "shed light on an agency's performance of its statutory duties." Reporters Comm. for Freedom of Press, 489 U.S. at 773. "[I]nformation about private citizens . . . that reveals little or nothing about an agency's own conduct" does not serve a relevant public interest under FOIA. Id. The requesting party's intended use for the information is irrelevant to our analysis. See id. at 771 ("[T]he identity of the requesting party has no bearing on the merits of his or her FOIA request.").
CSS claims that disclosure of the requested records will serve the public interest by revealing information about "(i) HHS's performance in maintaining and enhancing the quality and efficiency of services provided under the Medicare program, (ii) the agency's ability to root out Medicare fraud and waste; and (iii) the agency's compliance with various transparency initiatives." Appellee's Br. 21. We examine each contention in turn.
The Congress has charged HHS with "promoting the effective, efficient, and economical delivery of health care services, and of promoting the quality of services of the type for which payment may be made" by contracting with peer review organizations.*fn4 42 U.S.C. § 1395y(g). In 2003, the Congress directed HHS to establish a demonstration program to "examine health delivery factors that encourage the delivery of improved quality in patient care." Id. § 1395cc-3(b). While CMS has certain responsibilities to promote quality healthcare for Medicare beneficiaries, it is not authorized to "exercise any supervision or control over the practice of medicine or the manner in which medical services are provided, or over the selection, tenure, or compensation of any officer or employee of any institution, agency, or person providing health services." Id. § 1395.
CMS is also responsible for enrolling health care providers and suppliers, including physicians, in the Medicare program. A "provider" or "supplier" must be enrolled in the Medicare program to receive payment for covered services. 42 C.F.R. §§ 424.505, 424.510. The term "supplier" refers to "a physician or other practitioner, or an entity other than a provider, that furnishes health care services under Medicare." Id. § 400.202. The term "physician" includes "a doctor of medicine or osteopathy legally authorized to practice medicine and surgery by the State in which he performs such function or action." 42 U.S.C. § 1395x(r). CMS may exclude from participation in any Federal health care program, including Medicare, "[a]ny individual or entity . . . whose license to provide health care has been revoked or suspended by any State licensing authority . . . for reasons bearing on the individual's or entity's professional competence [or] professional performance." Id. § 1320a-7(b)(4).
CMS may also exclude an individual or entity that "has furnished or caused to be furnished items or services to patients . . . substantially in excess of the needs of such patients or of a quality which fails to meet professionally recognized standards of health care." Id. § 1320a-7(b)(6)(B).
CSS makes three arguments regarding how the requested data will shed light on HHS's performance of its mission to promote quality healthcare for Medicare beneficiaries. First, it claims that the requested data will indicate the quality of care Medicare patients are receiving. The claim rests on the assumption that the frequency with which a physician performs a medical procedure indicates the quality of the procedure. The medical community has not reached a consensus on whether the number of procedures performed by a physician correlates to the quality of those procedures. Compare John D. Birkmeyer et al., Surgeon Volume and Operative Mortality in the United States, 349 New Eng. J. Med. 2117, 2117 (2003) ("Patients can often improve their chances of survival substantially, even at high-volume hospitals, by selecting surgeons who perform the operations frequently."), with Ethan A. Halm et al., Is Volume Related to Outcome in Health Care? A Systematic Review and Methodologic Critique of the Literature, 137 Annals Internal Med. 511, 517 (2002) ("Twenty years of research have established that, for some procedures and conditions, higher volume among hospitals and physicians is associated with better outcomes. However, the magnitude of the relationship varies greatly among individual procedures and conditions. . . . Even when a significant association exists, volume does not predict outcome well for individual hospitals or physicians."). Even assuming a strong correlation between volume and quality, the data CSS requests will not indicate total volume because it does not include procedures performed by physicians for non-Medicare patients.
Second, CSS claims that the requested data will enable the public to determine if Medicare is paying physicians with insufficient certifications, disciplinary histories or poor evaluations for a large quantity, or any number, of procedures. The public can determine through publicly available information whether physicians with insufficient certifications, disciplinary histories or poor evaluations are enrolled in the Medicare program. See U.S. Dep't of Def. Dep't of Military Affairs v. FLRA, 964 F.2d 26, 29-30 (D.C. Cir. 1992) (recognizing that "alternative sources of information available that could serve the public interest in disclosure" diminish public interest value of disclosure). One can infer without the requested data that an enrolled physician is performing at least some procedures. See 42 C.F.R. § 424.540(a)(1) (CMS may deactivate physician's Medicare billing privileges if no claims submitted for 12 consecutive months). Currently available information does not enable an individual to know whether a physician with an insufficient certification is performing a large number of (or any) specific procedure, but this added knowledge does not shed any additional light on whether CMS is following its enrollment procedures contained in 42 C.F.R. § 424.510. Cf. Multi AG Media, 515 F.3d at 1231 (requested data indicated entity's eligibility vel non to participate in federal benefits program).
Third, CSS claims the "requested records can also be analyzed in conjunction with other treatment records to determine whether individual Medicare doctors are providing all services required to reach standards of recommended care." Appellee's Br. 24. CSS does not explain how the requested data can be used to perform this analysis. Rather, it cites articles noting that the quality of care delivered to Medicare beneficiaries has room for improvement and greater access to information is necessary for improvement. See Stephen F. Jencks et al., Change in the Quality of Care Delivered to Medicare Beneficiaries, 1998-1999 to 2000-2001, 289 JAMA 305, 305 (2003); Elizabeth A. McGlynn et al., The Quality of Health Care Delivered to Adults in the United States, 348 New Eng. J. Med. 2635, 2643-44 (2003).
Even if the requested data could be used to measure the quality of care provided by Medicare-enrolled physicians, it would not shed light on the "agency's performance of its statutory duties." Reporters Comm. for Freedom of Press, 489 U.S. at 773. CSS argues that the requested data will indicate the quality of care being provided by Medicare-enrolled physicians and thereby permit the public to assess how well CMS is fulfilling its statutory duty to promote quality. But we fail to see how the requested data will allow the public to evaluate the performance of any specific quality-promoting programs CMS has a statutory duty to undertake. The data will not reveal how well the peer review organizations with which HHS contracts to promote quality healthcare are performing their duties, see 42 U.S.C. § 1395y(g), or how well the demonstration program "examine[d] health delivery factors that encourage the delivery of improved quality in patient care." Id. § 1395cc-3(b). The data will not assist the public in determining whether Medicare is enrolling physicians who do not meet the enrollment requirements. Nor will it enable the public to determine whether CMS is properly excluding physicians who "fail to meet professionally recognized standards of health care," id. § 1320a-7(b)(6)(B), because nothing indicates that a physician who performs a procedure less often fails to meet recognized standards of health care.
CSS next contends that disclosure of the requested data will serve the public interest by revealing fraudulent Medicare claims made by physicians. For example, CSS notes that physicians who submit claims for procedures outside their specialties or who submit unusually high numbers of claims in general or for specific procedures may be committing fraud. But CSS has not provided any evidence of alleged fraud the requested data would reveal. In United States Department of State v. Ray, 502 U.S. 164 (1991), the Supreme Court rejected the respondents' "asserted [public] interest [under FOIA Exemption 6] in ascertaining the veracity of the interview reports" prepared by the State Department based on interviews with Haitian nationals involuntarily returned to Haiti. 502 U.S. at 179. The respondents had not presented "a scintilla of evidence . . . that tends to impugn the integrity of the reports." Id. The Court noted: "If a totally unsupported suggestion that the interest in finding out whether Government agents have been telling the truth justified disclosure of private materials, Government agencies would have no defense against requests for production of private information." Id. Similarly, if an unsupported suggestion that an agency may be distributing federal funds to a fraudulent claimant justifies disclosure of private information, the agency would have no defense against FOIA requests for release of private information.
CSS points to a Government Accountability Office (GAO) report as general evidence that Medicare is especially susceptible to fraud. In 1990, the GAO "designated the Medicare program as high risk for fraud, waste, abuse, and mismanagement, in part because of its sheer size and complexity." U.S. GAO,GAO 06-813, Medicare Integrity Program: Agency Approach for Allocating Funds Should Be Revised 1 (2006), http://www.gao.gov/new.items/d06813.pdf. In 1997, the Congress established the Medicare Integrity Program, under which CMS contracts with eligible entities to safeguard Medicare payments, including investigating potential fraud cases. Id. at 1, 11-12; 42 U.S.C. § 1395ddd(a), (b)(1). Without more, the GAO's report does not raise a cognizable public interest under FOIA in verifying that CMS is adequately detecting fraud.*fn5 In Multi AG Media, we did find that release of the requested data would serve the public interest by allowing the public to "more easily determine whether [the United States Department of Agriculture] is catching cheaters and lawfully administering its subsidy and benefit programs." 515 F.3d at 1232. The specific public interest in Multi AG Media, however, was enabling the public "to look at the information the agency had before it when [determining whether a particular farm is eligible to participate in the benefit programs in the first place] so that the public can monitor whether the agency is correctly doing its job." Id. at 1231. In contrast, the Medicare claims data is irrelevant to whether physicians meet the Medicare enrollment requirements.
Finally, CSS argues that the requested data will shed light on whether HHS is complying with its own transparency initiatives. HHS recently proposed a new system of records "to assist in projects that provide transparency in health care on a broad-scale enabling consumers to compare the quality and price of health care services so that they can make informed choices among individual physicians, practitioners and providers of services." Privacy Act of 1974; Report of New System of Records, 72 Fed. Reg. 52,133, 52,133 (Sept. 12, 2007). Since 2001, HHS and CMS have launched quality initiatives "to assure quality health care for all Americans through accountability and public disclosure," including publicly reporting certain quality measures to aid consumer decision-making. CMS, Quality Initiatives - General Information - Overview, http://www. cms.hhs.gov/QualityInitiativesGenInfo/ (last visited Jan. 23, 2009).
Contrary to CSS's assertion, the requested data will not assist the public in determining whether CMS is complying with its transparency initiatives to provide consumers with more information about service providers. First, the public is already familiar with the type of data contained in the Medicare claims database, which includes the diagnosis, the type and place of service and the physician's UPIN, as evidenced by CSS's FOIA request. The public does not need the data itself to evaluate whether CMS's failure to disclose it constitutes a failure to comply with CMS's transparency initiatives. Nor does the public need the data to evaluate whether the steps already taken by CMS are in fact assisting consumers in making informed decisions. Second, according to CSS's logic, CMS must disclose any information possibly relevant to consumer health care decision making, regardless of privacy interests, simply because CMS stated its intention to provide more information relevant to consumer health care decisions. CMS has undertaken certain transparency initiatives but at no point has it pledged, or been directed by the Congress, to disclose any information to the public that could possibly assist consumers in health care decisions without regard to any countervailing interest, including the FOIA-recognized privacy interest. See HHS, Value-Driven Health Care Home: Transparency Leads to Change, http://www.hhs.gov/valuedriven/ (last visited Jan. 23, 2009) ("Transparency is a broad-scale initiative enabling consumers to compare the quality and price of health care services, so they can make informed choices among doctors and hospitals. In cooperation with America's largest employers and the medical profession, this initiative is laying the foundation for pooling and analyzing information about procedures, hospitals and physician services. When this data foundation is in place, regional health information alliances will turn the raw data into useful information for consumers."). CSS in fact seeks to use FOIA to compel CMS to comply with its transparency initiatives as CSS views them, not to evaluate whether CMS is fulfilling its duties.*fn6
In sum, the requested data does not serve any FOIA-related public interest in disclosure. Accordingly, we need not balance the non-existent public interest against every physician's substantial privacy interest in the Medicare payments he receives. See Nat'l Ass'n of Retired Fed. Employees v. Horner, 879 F.2d 873, 879 (D.C. Cir. 1989) ("We have been shown no public interest in . . . disclosure . . . . We need not linger over the balance; something, even a modest privacy interest, outweighs nothing every time."). Accordingly, disclosure of the requested data "would constitute a clearly unwarranted invasion of personal privacy." 5 U.S.C. § 552(b)(6). And even were we to find a FOIA-related public interest in disclosure, it would be negligible at best and insufficient to outweigh the significant privacy interest in non-disclosure. See Painting & Drywall Work Preserv. Fund, Inc. v. HUD, 936 F.2d 1300, 1303 (D.C. Cir. 1991)("attenuated public interest in disclosure does not outweigh the construction workers' significant privacy interest in the requested information").
For the foregoing reasons, we conclude that the requested Medicare claims data CSS seeks is exempt from disclosure under FOIA Exemption 6, 5 U.S.C. § 552(b)(6). Accordingly, the judgment of the district court is reversed and the case remanded for further proceedings consistent with this opinion.
ROGERS, Circuit Judge, concurring in part and dissenting in part
In holding that Exemption 6 of the Freedom of Information Act ("FOIA"), 5 U.S.C. § 552(b)(6), is dispositive, Op. at 15, the court has assessed a strong privacy interest for private physicians who receive Medicare reimbursements while ignoring the commanding public interest in disclosure of information that would enhance the public's ability to evaluate how well the Department of Health and Human Services ("HHS") is performing its statutory duties under the Medicare program. As subjective evaluations should not affect the balancing of these interests, it is significant that two district courts here, now and in 1979, and HHS itself have reached a different conclusion about the importance of disclosing such data to the public.
The district court concluded that the data requested by the Consumers' Checkbook, Center for the Study of Services ("the Center") would enhance the public's ability to understand whether HHS is effectively policing reimbursements and physician practices. Consumers' Checkbook, Ctr. for the Study of Servs. v. Dep't of Health & Human Servs., 502 F. Supp. 2d 79, 85-86 (D.D.C. 2007) ("Consumers' Checkbook I"). Another district court judge in this district reached much the same conclusion in 1979. Pub. Citizen Health Research Group v. Dep't of Health, Educ. & Welfare, 477 F. Supp. 595, 604 (D.D.C. 1979) ("Public Citizen"), rev'd on other grounds, 668 F.2d 537 (D.C. Cir. 1981). HHS, in turn, has previously released such data to at least one private enterprise unrelated to the agency's own initiatives, see Alley v. Dep't of Health & Human Servs., No. CV-07-BE-0096-E, slip op. at 3 (N.D. Ala. May 8, 2008) ("Alley, N.D. Ala. 2008"), and also announced as recently as 2007 that it would release much of the requested data to research entities outside of the federal government as part of a program of transparency, Privacy Act of 1974; Report of New System of Records, 72 Fed. Reg. 52,133 (Sep. 12, 2007) ("2007 Records System"). HHS also has not appealed the waiver of FOIA fees for the Center premised on the determination that the requested data is "in the public interest because it is likely to contribute significantly to public understanding of the operations or activities of the government," 5 U.S.C. § 552(a)(4)(A)(iii). Perhaps this is because, as Judge Gerhard Gesell wrote in 1979, "[p]ractitioners who contract with the government to provide medical services in exchange for federal payments perform a quasi-public function," and given "Congress' abiding concern to deliver cost-efficient public health care and physicians' clear prerogative to avoid government business," important public interests are at stake. Public Citizen, 477 F. Supp. at 604.
The Freedom of Information Act requires agencies to disclose all requested agency records, 5 U.S.C. § 552(a), unless a statutory exemption applies, id. § 552(b). It is designed to "'pierce the veil of administrative secrecy and to open agency action to the light of public scrutiny.'" Dep't of Air Force v. Rose, 425 U.S. 352, 361 (1976) (quoting Rose v. Dep't of Air Force, 495 F.2d 261, 263 (2d Cir. 1974)). Consistent with "the basic policy that disclosure, not secrecy, is the dominant objective of the Act," the statutory exemptions are "narrowly construed." Id. at 361. Under Exemption 6 an agency may withhold "personnel and medical files and similar files the disclosure of which would constitute a clearly unwarranted invasion of personal privacy." 5 U.S.C. § 552(b)(6) (emphasis added). If a court determines that a substantial privacy interest is at stake, the court must then consider whether the "public interest in disclosure outweighs the individual privacy concerns." Nat'l Ass'n of Home Builders v. Norton, 309 F.3d 26, 35 (D.C. Cir. 2002). FOIA's "strong presumption in favor of disclosure places the burden on the agency" to justify nondisclosure. Dep't of State v. Ray, 502 U.S. 164, 173 (1991). This "presumption favoring disclosure . . . is at its zenith under Exemption 6." Nat'l Ass'n of Home Builders, 309 F.3d at 37.
In March 2006, the Center requested HHS to release certain Medicare claims data for health care providers in the District of Columbia and the States of Illinois, Maryland, Washington, and Virginia. This data includes the providers' Medicare identification number, procedure codes, diagnosis codes, and geographic codes, but not patient identities. Although this data does not directly reveal annual Medicare reimbursement amounts for particular Medicare providers, the district court found that the information can be used, in conjunction with free, public information that Congress requires HHS to disseminate to the public, see Omnibus Budget Reconciliation Act of 1990, Pub. L. No. 101-508, § 4164, as amended, ("OBRA 1990"), 42 U.S.C. § 1395u note, and a free website for which HHS contracts, to determine those amounts. See Consumers' Checkbook I, 502 F. Supp. 2d at 84-85; Appellant's Br. at 18.
The crux of the court's determination today that physicians' privacy interests outweigh the public interest in disclosure is its conclusion that the requested data cannot assist the public in assessing either the quality of Medicare services or HHS's efforts to combat fraud and waste. Op. at 14-15. In reaching this conclusion the court overstates the inviolability of the privacy interest and overlooks the near undeniable fact that the requested data can be of some assistance to the public's evaluation of how HHS is carrying out its initiatives aimed at measuring and improving health care quality and its efforts to combat Medicare fraud and waste.
There is little doubt that the disclosure of the requested material would implicate more than a de minimus privacy interest. Privacy encompasses "the individual's control of information concerning his or her person," Dep't of Justice v. Reporters Comm. for Freedom of the Press, 489 U.S. 749, 763 (1989) ("Reporters Comm."), and extends to protect an individual's name and address, see Nat'l Ass'n of Retired Fed. Employees v. Horner, 879 F.2d 873, 876 (D.C. Cir. 1989); Fed. Labor Relations Auth. v. Dep't of the Treasury, 884 F.2d 1446, 1453 (D.C. Cir. 1989), as well as information that would "in some cases allow for an inference to be drawn about the financial situation of an individual," even where the data reveals only a portion of an individual's financial situation, Multi AG Media LLC v. Dep't of Agric., 515 F.3d 1224, 1230 (D.C. Cir. 2008); see also Lepelletier v. FDIC, 64 F.3d 37, 47 (D.C. Cir. 1999); Nat'l Parks & Conservation Ass'n v. Kleppe, 547 F.2d 673, 685-86 (D.C. Cir. 1976). Indeed, "[w]hen there is a substantial probability that disclosure will cause an interference with personal privacy, it matters not that there may be two or three links in the chain." Nat'l Ass'n of Retired Fed. Employees, 879 F.2d at 878. HHS suggests that Medicare reimbursements represent, on average, a quarter of a physician's income, and can account for a "large percentage" of total income for some physicians. Reply Br. at 16.
Nevertheless, as the Center points out, the physicians' privacy interest is "particularly limited," Appellee's Br. at 14-15, because the requested data pertains to receipt of government funds and would not reveal physicians' take-home earnings. Although the fact of federal government funding may not be dispositive in assessing the privacy interest, see Multi AG Media, 515 F.3d at 1230; Nat'l Ass'n of Retired Fed. Employees, 879 F.2d at 876; Painting & Drywall Work Pres. Fund v. Dep't of Housing & Urban Dev., 936 F.2d 1300, 1302-03 (D.C. Cir. 1991), it cannot be denied that there is an element of public service involved when physicians participate in the Medicare program. See Public Citizen, 477 F. Supp. at 604. Further, because the requested data does not directly reveal total income and because HHS has not shown that one can predictably determine total income using the Medicare reimbursement amounts, "the privacy interest that may exist is [not] particularly strong," Multi AG Media, 515 F.3d at 1230; see also Getman v. NLRB, 450 F.2d 670, 675 (D.C. Cir. 1971); Public Citizen, 477 F. Supp. at 603-04. The extent of that privacy interest varies according to how much of the physician's income is derived from Medicare; a doctor whose patients are mostly Medicare beneficiaries has a greater privacy interest in her Medicare reimbursement amounts than a doctor who treats only a few Medicare patients. Additionally, Congress has already required HHS to publish some private information about Medicare-participating physicians, see, e.g., OBRA 1990, 42 U.S.C. § 1395u note (requiring release of physicians' identification number, address, and related information), and HHS has released data much as the Center seeks, see, e.g, Alley (N.D. Ala. 2008). Thus, regardless whether a practitioner has many or only a few Medicare patients, HHS fails to meet its burden to show that the privacy invasion at issue would be overly intrusive.
By contrast, there is a commanding and important public interest in disclosure of the data the Center seeks.The single relevant public interest in FOIA balancing is the "extent to which disclosure of the information sought would 'she[d] light on an agency's performance of its statutory duties' or otherwise let citizens know 'what their government is up to.'" Dep't of Def. v. Fed. Labor Relations Auth., 510 U.S. 487, 497 (1994) (quoting Reporters Comm., 489 U.S. at 773) (alteration in Reporters Comm.). The public interest inquiry focuses, not on the absolute value, but on "the incremental value of the specific information being withheld." Schrecker v. Dep't of Justice, 349 F.3d 657, 661 (D.C. Cir. 2003) (emphasis added).*fn7 Because Medicare "distributes extensive amounts of public funds," there is a "special need" for public oversight of HHS's activities in administering Medicare. Multi AG Media, 515 F.3d at 1232; see generally Government Accountability Office, Medicare Integrity Program: Agency Approach for Allocating Funds Should Be Revised, GAO-06-813 (Sep. 2006). As this court observed in Multi Ag Media, "Congress has recognized the importance of ensuring the responsible use" of Medicare funds, 515 F.3d at 1232; see, e.g., Inspector General Act of 1978, Pub. L. No. 95-452, § 2, 92 Stat. 101 (1978). Indeed, HHS itself has acknowledged both that it "shares [the Center's] broad policy goals," Appellant's Br. at 39, in public disclosure, and, in response to the Center's fee waiver request, that it "do[es] not dispute that the requested records pertains to operations or activities of the Federal Government and that the disclosure of the records would reveal meaningful information about government operations or activities," Letter from Herb B. Kuhn, Acting Deputy Adm'r, Dep't of Health & Human Servs., to Robert Krughoff, President, Consumer's Checkbook, Ctr. for the Study of Servs. 2 (Mar. 16, 2007) ("Kuhn letter of Mar. 16, 2007") (emphasis added).
There should be little dispute that the requested data would shed light on at least two types of HHS activities.*fn8 First, the data would enable members of the public to evaluate HHS's performance of its statutory duties regarding the quality of Medicare-provided services. Even if HHS does not have statutory authorization to supervise or control the practice of medicine, see 42 U.S.C. § 1395, Op. at 8, HHS has a statutory and regulatory duty to evaluate and work to improve the quality, cost, and efficiency of services delivered by Medicare providers. For instance, 42 U.S.C. § 1395y(g) requires HHS to enter into contracts with "utilization and quality control peer review organizations" in order to promote improved delivery and quality of health services. See 42 U.S.C. § 1320c et seq. (elaborating upon section 1395y(g)). Likewise, Congress has required HHS to "establish a 5-year demonstration program" for projects examining "health delivery factors that encourage the delivery of improved quality in patient care." Id. § 1395cc-3(b). Similarly, HHS's 2007 Records System is designed "to assist in projects that provide transparency in health care on a broad-scale enabling consumers to compare the quality and price of health care so that they can make informed choices among individual physicians, practitioners and providers of services."
72 Fed. Reg. at 52,133. Likewise, in 2006 President Bush signed Executive Order No. 13,410, requiring HHS to "implement programs measuring the quality of services supplied by health care providers to the beneficiaries or enrollees" of Medicare. See 71 Fed. Reg. 51,089, 51,090 (Aug. 22, 2006).
Studies show that releasing the data the Center seeks would enable members of the public to evaluate HHS's effectiveness in fulfilling its duties. One study utilized Medicare claims data in determining that the number of procedures performed by a surgeon was inversely related to patient mortality rates for each of eight studied procedures. See John D. Birkmeyer et al., Surgeon Volume and Operative Mortality in the United States, 349 NEW ENG. J. MED. 2117, 2122-23 (2003). Other studies indicate that information about the number of times a physician has performed a particular procedure would shed at least some light on that physician's success rate. See, e.g., Jim C. Hu et al., Role of Surgeon Volume In Radical Prostatectomy Outcomes, 21 J. CLINICAL ONCOLOGY 401 (2003); Deborah Schrag et al., Hospital and Surgeon Procedure Volume as Predictors of Outcome Following Rectal Cancer Resection, 236 ANNALS OF SURGERY 583 (2002). That there may not be unanimity within the medical profession about the closeness of the correlation between experience and quality does not diminish the public interest in disclosure, as the court implies, see Op. at 9 (citing Ethan A. Halm et al., Is Volume Related to Outcome in Health Care? A Systematic Review and Methodologic Critique of the Literature, 137 ANNALS OF INTERNAL MED. 511 (2002)).
Neither HHS nor intervenor the American Medical Association ("AMA") suggests there is no correlation between experience and quality. Even assuming the link between quality and the number of procedures a provider has performed is weak, and even though the requested data will only partially reveal physicians' experience levels, the data has "incremental value" for ascertaining the quality of services performed both at the provider level and program-wide. See Schrecker, 349 F.3d at 661.
The requested data also could be used to evaluate the disciplinary and board certification histories of Medicare providers and to study whether Medicare providers meet recommended standards of care for patients with different diagnoses. The data has value over and above currently available information because the public could use it to evaluate the extent of particular physicians' Medicare participation and to determine whether physicians have qualifications to provide the services for which they seek federal reimbursement; currently available information reveals only whether a physician participates in Medicare. Such independent assessments of the quality of Medicare-funded services, analyzed in the aggregate, would serve as a check on HHS's own quality-measuring projects, helping the public ascertain possible weaknesses in the Medicare program itself and in HHS's statutorily-required assessments of the program.
Because "the purpose of FOIA is to permit the public to decide for itself whether government action is proper," Washington Post Co., 690 F.2d at 264, the existence of internal HHS quality-measuring programs does not diminish the public interest in disclosure. The Supreme Court recognized in Reporters Committee that "the FOIA's central purpose is to ensure that the Government's activities be opened to the sharp eye of public scrutiny." 489 U.S. at 773. In National Association of Home Builders, this court held even though the agency had released its method for designating owl habitats, there was still a public interest in disclosure of the data used in that determination. 309 F.3d at 363. Similarly, in Multi AG Media, this court observed that the data at issue "sa[id] everything about whether a particular farm is eligible to participate in the [federal] benefit programs in the first place and thus 'shed light on the agency's performance of its statutory duties.'" 515 F.3d at 1231 (quoting Reporters Comm., 489 U.S. at 773). The Center has requested several of the same data elements (including provider identification number, diagnosis information, and surgical procedures performed) that HHS has announced it plans to use in the 2007 Records System as part of its oversight and transparency initiatives. See 72 Fed. Reg. at 52,135. Thus, even assuming HHS does not use the requested data in determining eligibility or deciding whether to pay a claim, the public has an interest in disclosure of the requested data elements. And even if the data would not specifically shed light on a particular HHS initiative, see Op. at 11, the public interest in the data is strong with respect to HHS's fulfillment of the goal underlying many of its statutory and regulatory activities: improving the quality of health care provided under Medicare.
HHS suggests that disclosure of physician-identifying information (specifically, unique physician identification numbers), even if the requested data could shed light on the performance of HHS's statutory duties, would not contribute any additional public benefit and so any data release should redact physicians' names. To the contrary, physician-identifying information would enable the public to analyze the information in context. For example, using the data along with the physician's name, researchers would be able to ascertain whether a physician's low Medicare procedure volume is explained by the number of younger patients being treated.
Also, withholding identifying information would compromise the public interest in connection with HHS's reaffirmation of its goal of "enabling consumers to compare the quality and price of health care services so that they can make informed choices among individual physicians, practitioners and providers of services." 72 Fed. Reg. at 52,133. As HHS's stated policy is to facilitate consumer choices about which providers to patronize, it cannot credibly maintain that providing physician-identifying information could not shed light on its own activities. Moreover, because the data the Center seeks cannot be retrieved in any other way, the public interest in disclosure is not significantly diminished by the derivative nature of its proposed use.*fn9
Second, the requested data would shed light on HHS's fraud-detection and fraud-prevention efforts. For instance, the data could identify providers who perform "a suspiciously large number of procedures in a given time period" or "submit claims for procedures that are outside [their] own practice areas." Appellee's Br. at 29. The data could therefore facilitate public monitoring of HHS detection and prevention of fraud. Additionally, to the extent that consumer choice could be enhanced by knowing which physicians are potentially responsible for wasteful or even fraudulent claims, release of physician-identifying data is consistent with HHS's goal of improving consumers' decisions about which medical providers to patronize. See 72 Fed. Reg. at 52,133. The public could utilize the requested information in determining whether HHS is fulfilling this stated goal.
Again, that there may already be, as HHS and the AMA assert, significant government oversight of physicians that the public can oversee does not diminish the public interest in disclosure of the requested data. For instance, the public can currently use a public database to determine whether Medicare is reimbursing any providers who have already been excluded from Medicare for misconduct. However, FOIA's purpose is "to permit the public to decide for itself whether government action is proper." Washington Post Co., 690 F.2d at 264. Consequently, the public's interest in monitoring compliance is not limited to ensuring that once HHS identifies a provider who has engaged in misconduct, HHS does not reimburse that provider for services. Rather, the public also has an interest in monitoring the effectiveness of HHS's identification of providers responsible for misconduct. The court's suggestion that the Center failed to present evidence of alleged fraud that the requested data would reveal creates a heightened disclosure requirement that is without precedent. Op. at 11. Its reliance on Ray, 502 U.S. 164, is misplaced; the only issue in that case was whether the redaction of names and identifying information was lawful, see id. at 168, and the Supreme Court rejected the asserted public interest in release of names and other identifying information for studying the veracity of the released reports because there was no evidence suggesting that the reports lacked integrity, id. at 179. The Center is not asserting a public interest in disclosure of the physicians' identities for purposes of verifying the accuracy of the other requested data elements. Rather, release of the physicians' identities would enable the public to place the released information in context and better assess HHS's fulfilment of it statutory and regulatory goals. As noted by the Center, the Government Accountability Office report, Medicare Integrity Program, Agency Approach for Allocating Funds Should Be Revised, GAO 06-813 (Sep. 2006), indicated that HHS had estimated $12.1 billion net of improper payments to Medicare providers in 2005, offering support for the proposition that HHS makes some Medicare payments to fraudulent claimants.
In sum, Medicare providers' privacy interest in data that would reveal part of their annual income is more than de minimus but not particularly strong, especially given previous and planned disclosures by HHS. On the other hand, the requested data would shed light on at least two key HHS responsibilities under Medicare: (1) measuring and improving the quality of health care that is provided and (2) combating and detecting fraud and waste. To the extent that the requested data may shed little light on the quality of health care delivered by physicians with only a few Medicare patients, such physicians also have relatively weak privacy interests in Medicare reimbursement amounts, which likely represent a small portion of their annual income. For physicians treating many Medicare patients, the privacy interest is greater but so is the usefulness of the requested data. HHS has reached the same conclusion as the Center about the meaningfulness of the requested data in informing the public about HHS's Medicare activities, see, e.g., Kuhn letter of Mar. 16, 2007; 2007 Records System, 72 Fed. Reg. at 52,133. HHS, consequently, has not met its burden to show that release of the requested data "would constitute a clearly unwarranted invasion of personal privacy," 5 U.S.C. § 552(b)(6) (emphasis added).
Although FOIA Exemption 6 would not bar release of the requested data, HHS contends that release is nonetheless barred by an injunction issued by the United States District Court for the Middle District of Florida in 1979.*fn10 If the injunction would bar release of the data that the Center seeks, then HHS would not "improperly" be withholding it and the court would lack jurisdiction to order disclosure. GTE Sylvania, Inc. v. Consumers Union of the United States, Inc., 445 U.S. 375, 384 (1980); see Dep't of Justice v. Tax Analysts, 492 U.S. 136, 142 (1989). The district court concluded that the 1979 Florida injunction was "immaterial" to its analysis because the Center was seeking "different records." Consumers' Checkbook I, 502 F. Supp. 2d at 86 n.1. Although this is true, the Center appears to understate the effect of releasing at least some of the requested data in urging a narrow construction of the injunction while HHS appears to overstate the scope of the injunction.
The data requested by the Center does not coincide precisely with the data elements that were covered by the list addressed in the Florida injunction. The Center has requested 29 data elements for claims submitted to HHS, including the physician provider's identification number, the patient's diagnosis, the procedures performed, and the time and place of service. It did not request the provider's name, address, or reimbursement amounts, the elements specifically covered by the Florida injunction, see supra note 4. However, the district court found that by combining the requested data with publicly available information the public could calculate the reimbursement amounts for particular procedures by individual physicians. Consumers' Checkbook I, 502 F. Supp. at 84. The Center has not shown that this finding is clearly erroneous.
The 1979 Florida injunction contains broad language and its purpose to apply broadly is evident from the accompanying declaratory judgment and opinion, see Haskell v. Kansas Nat. Gas Co., 224 U.S. 217, 223 (1912). It "permanently enjoin[s]" disclosure of "any list" of Medicare reimbursement amounts for "any years" that would identify members of the recertified class.
Florida Med. Ass'n v. Dep't of Health, Educ. & Welfare, No. 78-178-Civ-J-S, 1-2 (M.D. Fla. Oct. 22, 1979). The accompanying declaratory judgments states: "Any . . . disclosure of annual Medicare reimbursement amounts, for any years, in a manner that would personally and individually identify the providers of services under the Medicare program who are members of the recertified class in this case is declared to be contrary to federal law." Id. at 2. The accompanying opinion describes the issue presented as "whether the Secretary . . . of [HHS's predecessor] may disclose information concerning the annual amounts of reimbursements paid to Medicare providers in a way that would individually identify at least some of those providers." Florida Med. Ass'n, 479 F. Supp. at 1294. The opinion concludes that the list at issue was a "similar file" under FOIA Exemption 6, observing that "[c]courts must look past mere appearances and beneath labels, to the actual character and nature of the information in question." Id. at 1303. A 1982 modification provided that the injunction did not prohibit disclosure of annual Medicare payment information about individual physicians pursuant to the law enforcement exception under the Privacy Act, 5 U.S.C. § 552a(b)(7). Florida Med. Ass'n v. Dep't of Health & Human Servs., No. 78-178-Civ-J-S (M.D. Fla. Dec. 2, 1982).
In these circumstances, Schering Corp. v. Illinois Antibiotics Co., 62 F.3d 903, 906-07 (7th Cir. 1995), and ALPO Petfoods, Inc. v. Ralston Purina Co., 913 F.2d 958, 972 (D.C. Cir. 1990), on which the Center relies, may not counsel a contrary conclusion about the scope of the 1979 Florida injunction. Although injunctions are to be construed narrowly, "the rule of strict construction of injunctions should not be pressed to a dryly logical extreme," Schering, 62 F.3d at 906, and an injunction should be tailored to the harm redressed, ALPO Petfoods, 913 F.2d at 972. HHS views the injunction as protecting against invasions of privacy resulting from release of reimbursement amounts for members of the recertified class, presumably relying on this court's FOIA precedent regarding linkage, e.g., Nat'l Ass'n of Retired Fed. Employees, 879 F.2d at 878. The Center points out, however, that HHS's 2007 Records System for quality monitoring, designed to advance the public interest in transparency, would entail releasing to outside researchers the information covered by the 1979 Florida injunction. HHS has responded that such data might not include the reimbursement amounts. See Reply Br. at 11. Where this leaves HHS's view of the scope of the 1979 Florida injunction is unclear. In any event, the question remains whether any doubt about the scope of the injunction requires it to be read narrowly. See In re Baldwin-United Corp., 770 F.2d 328, 339 (2d Cir. 1985).
Furthermore, by its specific terms, the 1979 Florida injunction is limited to the recertified class, as HHS acknowledged during oral argument; it does not reach the release of data concerning other physicians. HHS responds that segregating such data would be an arduous, lengthy task, subject to error. Perhaps so, but HHS has yet to explain satisfactorily why this would be in an age of computerized record keeping. Given FOIA's presumption in favor of disclosure, HHS is obligated to segregate these records, see 5 U.S.C. §552(b), or at least to "provide a more detailed justification than the conclusory statements it has offered to date," Mead Data Centr., Inc. v. Dep't of Air Force, 566 F.2d 242, 260 (D.C. Cir. 1977), as to why segregation would be unreasonable.
Accordingly, I would affirm the district court's ruling that FOIA Exemption 6 does not bar release of the Medicare data that the Center seeks, at least as to records pertaining to physicians who are not members of the recertified class covered by the 1979 Florida injunction and perhaps with regard to others as well, and I would remand the case to the district court for further proceedings on the scope of the injunction, including an opportunity for HHS to explain why an order requiring segregation of the data would be unreasonable.