ePolicy News Special Edition: CMS Provider-Level Medicare Data Release: Helpful or Harmful to Patients?
Monday, May 5, 2014
CMS Provider-Level Medicare Data Release: Helpful or Harmful to Patients?
ASCP cautions that partial transparency may be worse than no transparency at all; Fears that incomplete provider-level information may misguide patient decision-making just as incomplete patient health information misguides appropriate diagnosis and treatment
On April 9 the Center for Medicare and Medicaid Services (CMS) released the Medicare Provider Utilization and Payment Data: Physician and Other Supplier Public Use File. The raw claims data covers approximately 880,000 distinct health care providers who collectively received $77 billion in Medicare Part B payments in CY 2012. Its release was intended to mark the pinnacle of the Medicare program’s movement into an age of unprecedented information transparency in the post Affordable Care Act (ACA) era. However, this unparalleled exposure to provider-level billing data appears to be sparking more public controversy than it is public awareness of fraud, waste, and abuse in the Medicare program.In fact, the public is left questioning both the method and mode of CMS’s historic data release – citing that the data is incomplete and expressing concern with CMS’s decision to partner exclusively with the Wall Street Journal and New York Times in its unveiling. Accordingly, not only did CMS release the data early to the two media outlets but it also embargoed them from reaching out to the provider community for context and then granted them exclusive privileges to host provider search databases.
Why Did CMS Release the Provider-Level Medicare Data?
CMS’s decision to publicly release provider-level Medicare data stems from growing public distrust in our country’s health care system. This distrust was particularly exacerbated in 2010 with the Wall Street Journal’s publication of a series of articles entitled“Secrets of the System,”which highlighted fraud and abuse in the Medicare program. Accordingly, the fraudulent activity exposed in these articles eventually prompted Dow Jones, the Wall Street Journal’s parent organization, to seek to overturn the 1979 injunction granted to the American Medical Association, which barred the public release of provider-level Medicare information. Then in May 2013 a federal judge finally vacated the injunction, citing that the public interest in disclosure now outweighed the privacy interest of individual physicians.
Does ASCP Agree with CMS’s Intent in Deciding to Release the Data?
In its April 9th Press Release, CMS asserted that the purpose of the data release was “to assist the public's understanding of Medicare fraud, waste, and abuse, as well as shed light on payments to physicians for services furnished to Medicare beneficiaries.”Accordingly, while ASCP supports CMS’s effort to expose and extinguish fraudulent activity and its effort to better inform and empower patients via provider-level data, the Society does not believe that the two efforts should be combined. The release of raw and incomplete Medicare Part B payment and utilization data does not adequately aid public understanding of the presence of fraud, waste, and abuse across the entire Medicare program. Moreover, even if the data was complete, it remains extremely complex and its interpretation often requires supplementary knowledge regarding conceptual, contractual, and clinical caveats. As such, it is highly unlikely that the average Medicare beneficiary is equipped with the tools and resources needed to identify and decipher evidence of fraudulent activity within the large, incomplete, and complicated data sets provided.
Conversely, CMS is already equipped with these tools and resources and is actually statutorily required under the Social Security Act to identify fraud, waste, and abuse within the Medicare and Medicaid programs. This statute-based responsibility was only further expanded under both the ACA and the Small Business Jobs Act in 2010, as evident by CMS’s establishment of the Fraud Prevention System and the Center for Program Integrity. Accordingly, ASCP encourages CMS to focus less on involving patients in the process of identifying fraud, waste, and abuse and more on partnering with the Office of the Inspector General and the Department of Justice to inform them of its existence.
Nonetheless, ASCP understands that, just as a provider’s lack of patient-level information threatens the appropriateness of treatment delivered, so can a patient’s lack of provider-level information threaten the quality, efficiency, and integrity of treatment received. Accordingly, ASCP acknowledges the value in bidirectional patient health information exchange as a mechanism for ensuring appropriate treatment and thus encourages the same bidirectional transmission of provider-level information as a mechanism for ensuring the quality, efficiency, and integrity of service delivery. However, ASCP cautions that oftentimes partial transparency can be worse than no transparency at all. As such, just as incomplete patient health information can misguide patient diagnosis and treatment, so can incomplete provider-level information misguide patient decision-making regarding his/her own health.
What Are ASCP’s Concerns Regarding Data Content and Utility?
In a response to CMS’s Request for Public Comments on “The Potential Release of Medicare Physician Data,” ASCP joined 95 other medical societies in requesting that any provider-level payment and utilization information released be accompanied by risk-adjusted quality information and be structured in a complete, accurate, and actionable format.However, it does not appear that CMS heeded this advice and now,contrary to the providers’ recommendations, the release of raw and incomplete Medicare Part B payment and utilization data in no way indicates the quality of provider-level services. At best, the data allows patients to identify whether or not a provider is furnishing multiple services for a single patient as a potential inefficiency indicator. However, because the data is not risk-adjusted, it is impossible to determine whether a provider is providing multiple services for a single patient due to the severity of patient condition, inefficient practice, or mere financial incentive.
Worse yet, the data set released does not fulfill CMS’s baseline intentions of conveying provider-level profit and depicting overutilization. It does not convey provider-level profit because the individual providers are tied to payment amounts that have not been extracted of the costs of equipment, personnel, and malpractice insurance.Additionally, even if the provider-level payments were stripped of these overhead costs, the claims were not necessarily tied to the appropriate provider that actually furnished and or/received payment for a given service. Accordingly, the provider may be a salaried employee of a group practice that collects Medicare payments on his/her behalf.
Additionally, though CMS set out to tie payments to individual providers based on the “performing provider” listed on claims, the Agency was often left tying payments to the“billing provider”listed on claims when no performing provider was specified. Hence, if a claim did not contain a performing provider, and the clinical laboratory’s group NPI or Laboratory Director’s individual NPI was listed as the billing provider, then there was no way to tie the payments on these claims to the individual providers that actually furnished the services being billed. As such, since LabCorp and Quest Diagnostics both listed billing providers on claims using the clinical laboratories’ group-level NPIs, the performing providers employed by these organizations cannot be individually linked to individual claims and thus do not appear in the data set released.
Nonetheless,ASCP cautions that, even if the data represents accurate and complete provider-level payments for a given provider, CMS must consider whether or not the provider is referral-based when holding the provider accountable for potential overutilization. Accordingly, it is inappropriate for CMS to hold the referral-based pathology and laboratory community accountable for the potential overutilization of services ordered and/or billed by another provider.
What is ASCP Doing to Better Inform Its Members of the Limitations in the Data?
In consideration of the above concerns, ASCP asserts that the partial transparency provided by CMS’s data release lacks meaningful context, benchmarks for provider comparison, and adequate explanation of data limitations. Hence, the Society fears and the data may actually distort patients’ rational decision-making regarding their own health and potentially harm patients in the process.
As such, while CMS provided a methodology document outlining some of the limitations in the data, these limitations were not clearly and fully communicated in the primary communication materials announcing the release. Worse yet, media coverage surely did not emphasize these limitations and instead often relied on them to generate controversial stories that lacked accurate foundations and even threatened individual providers’ reputations. For example, though only three of the top ten highest billing Medicare providers are being further investigated for fraud, the Washington Post published an article grouping together all ten. Accordingly, Dr. Frank Cockerill, Laboratory Director at the Mayo Clinic, was not charged with fraud but appeared on this list. Like many laboratories, the Mayo Clinic does not list a performing provider on its Medicare Part B claims. However, less common among laboratories, the Mayo Clinic listed the Lab Director’s individual NPI on the billing provider line of its claims as opposed to the laboratory’s group NPI. As such, all of the pathologists at the Rochester, NY-based Mayo Clinic billed for their services with claims that listed Dr. Cockerill as the billing provider, as opposed to listing the Mayo Clinic’s group-level NPI. Hence, even though this billing practice is not prohibited, it created a very suspicious illusion that Dr. Cockerill was performing and billing for $11.1 million worth of Medicare services in 2012.
In response to the unfair negative media attention across the provider community resulting from this data release, and in an effort to better inform our members and their patients, ASCP is taking responsibility to expose the major limitations within the CMS data released.Accordingly, the Society identifies twelve major data limitations as impediments to meaningful and actionable public interpretation, informed patient decision-making, and accurate and fair provider assessment. The limitations are segmented into the following four categories:
- Data is inappropriately raw
- Data is incomplete in volume and value
- Data is not controlled for non-fraudulent variables to enable effective provider-level payment comparisons
- Data is limited in scope and interpretive utility
Click here to access a chart listing each major data limitation, identifying the corresponding missing data element/item, and providing an example of the likely misinterpretations to ensue.
Reference AMA’s list of limitations in the CMS data released here
What is ASCP Recommending to CMS for Future Consideration?
ASCP supports CMS’s efforts to provide patients with information regarding the cost of Medicare services prior to consumption. Accordingly, the Society notes that while CMS already provides a Physician Fee Schedule Look-Up Tool, it is geared primarily toward the “health care professional, supplier, or provider,” rather than the patient. As such, ASCP encourages CMS to simplify this tool, and the related information accessible, so that it is geared more toward informing and empowering patients.
Moreover, ASCP encourages CMS to work with Congress to close the in-office self-referral loophole so that providers can no longer bill for self-referred anatomic pathology tests. While the Society is confident in CMS’s Anti-fraud/waste/abuse tools and resources, we understand that overutilization resulting from the financial incentives underlying a provider’s ability to self-refer may be difficult to detect. However, we note that many experts in quantitative analysis have already utilized Medicare data to demonstrate how this harmful practice can be detected. Accordingly, we direct CMS to reputable studies out of the Government Accountability Office (GAO),MedPAC, and Journal of Health Affairs that successfully used provider-level Medicare data to identify the presence and negative byproducts of harmful self-referrals for anatomic pathology services.
Lastly, ASCP encourages CMS to focus less on cost data and more on clinical data when seeking to improve the payment and delivery of Medicare services. Accordingly, the Society supports CMS’s efforts to expand upon quality reporting programs so that they better reflect the unique practice patterns of a broader array of provider specialties. Similarly, ASCP supports CMS efforts to develop these pay-for-performance initiatives so that they not only improve the delivery of Medicare services within the current fee-for-service reimbursement structure, but also aid in the necessary transformation of the fragmented, volume-driven payment system into a patient-centered, value-based payment system. Finally, the Society calls upon CMS to develop and guide the use of interoperable health information exchange as a mechanism for facilitating bidirectional patient and provider-level information. It is crucial that both patients and providers have access to relevant databases and the ability to update and/or correct their information contained in these databases.
What Can Providers Do to Mitigate the Negative Media Associated with this Data Release?
ASCP calls upon the provider community to take the initiative to understand and explain the data if patients should question its content or utility. In turn, ASCP remains committed to working with CMS to better communicate and remedy the limitations in the publically available data set.
For Additional Questions Regarding the CMS Data Release, Contact Kaitlin Cooke, Senior Manager of Advocacy & Public Policy at ASCP: Kaitlin.Cooke@ascp.org or 202-347-4450