Mostly Cloudy with a Little Bit of Sunshine—Accuracy of CMS’s Release of “Sunshine” Data QuestionedOctober 2, 2014
On September 30, 2014, the Centers for Medicare & Medicaid Services (“CMS”) released for public review the first round of Open Payments (more commonly known as “Sunshine”) data. According to CMS, the Sunshine data is intended to “help consumers understand the financial relationships between the health care industry, and physicians and teaching hospitals.”
Unfortunately, this initial release has been marred by numerous problems pertaining to data accuracy and the lack of information about the potential benefits that can arise from relationships between the drug and device industry and physicians.
Due to a short window for physicians to review and request correction of their data (45 days), combined with—surprise!—continuing website shutdowns, a complicated registration process, and missing data, many physicians were unable to confirm the accuracy of the data reportedly associated with their practice. The American Medical Association (“AMA”) stated that “[o]nly 26,000 physicians out of the nearly 550,000 physicians affected by the Sunshine Act were able to register to review their data and seek correction of any inaccuracies.”
CMS acknowledges problems with data accuracy, stating, “During the review and dispute period, CMS identified payment records that had inconsistent physician information, such as National Provider Identifier (“NPI”) for one doctor and a license number for another.” CMS’s fix was to de-identify 40% of the published data, calling into question how these data will help achieve the above-mentioned goal.
In addition to issues with data accuracy, both physician and industry trade groups requested that CMS provide context on the Sunshine website about the payments. In a letter to CMS, PhRMA, AdvaMed, and BIO stated that “providing context for reported payments and other transfers of value is critical to ensuring patients do not form mistaken impressions that all payments to physicians are suspect.” CMS apparently rejected this suggestion. The Deputy Administrator and Director of the Center for Program Integrity at CMS stated: “Open Payments does not identify which financial relationships are beneficial and which could cause conflicts of interest. It simply makes the data available to the public. So while these data could discourage payments and other transfers of value that might have an inappropriate influence on research, education, and clinical decision-making, they could also help identify relationships that lead to the development of beneficial new technologies.”
CMS seems to overestimate the ability of the public to determine, based solely on the information provided on the website, into which category certain payments fall. For example, members of the public are not likely to know that research payments are reported as a single payment to the primary investigator. This could appear to be a large, multi-million dollar payment to one physician, and, without context, someone viewing the data would not understand that the payment reported is intended to cover any number of years worth of research and a variety of costs associated with the research.
Facts matter. Context matters. And accuracy matters. Unfortunately, the data published by CMS, intended to shed light and provide transparency, do not adequately account for any of these factors. No doubt people will review the data and reach their own conclusions about the meaning. What will CMS do if it learns these conclusions are wrong? Based on its actions thus far, the answer is most likely to be nothing.