I once helped teach medical statistics. Much dark humor and many aphorisms were to be heard. For example: “If you torture the data long enough, it will tell you what you want to hear.” Another, “Garbage in — Garbage out,” is a shorthand way of stressing the importance of data integrity and reliability. If the data is not well defined, collected, accessible, or verifiable then any subsequent conclusions are correspondingly suspect. In the age of “big-data” and transparency of medical information, this latter concept of data reliability should be guarding the door of accountability. I suggest that the standards being applied so far by CMS to data collection would not be acceptable to editors of scientific journals and yet publication is going forward. Here is what CMS is saying about its Open Payment System:
Open Payments System Reopens, Extends Physician Registration and Review Period Aug 15, 2014
CMS announced today that the Open Payments system is once again available for physicians and teaching hospitals to register, review and, as needed, dispute financial payment information received from health care manufacturers. The system was taken offline on August 3 to resolve a technical issue. To account for system down time, CMS is extending the time for physicians and teaching hospitals to review their records to September 8, 2014. The public website will be available on September 30, 2014.
“CMS takes data integrity very seriously and took swift action after a physician reported a problem,” said CMS Deputy Administrator and Director of the Center for Program Integrity Shantanu Agrawal, M.D. “We have identified the root cause of the problem and have instituted a system fix to prevent similar errors. We strongly encourage physicians to review their records before the deadline and before the data are posted publically to identify any discrepancies.”
A full investigation into a physician complaint found that manufacturers and group purchasing organizations (GPOs) submitted intermingled data, such as the wrong state license number or national provider identifier (NPI), for physicians with the same last and first names. This erroneously linked physician data in the Open Payments system.
CMS has implemented system fixes to resolve the issue, and revalidated all data in the system to verify that the physician identifiers used by the applicable manufacturer or GPO are accurate, and that all payment records are attributed to a single physician. Incorrect payment transactions have been removed from the current review and dispute process and this data will not be published. [Emphasis added.]
CMS remains committed to ensuring the data made public from the system is as accurate as possible and extended the time for review and dispute to provide physicians and teaching hospitals with a full 45 days.
This series of articles began with my stymied attempt to register as a physician to see if any payments had been made to me in error. There should have been none. Earlier today I was able to log on to the Open Payment System and create a profile. When asked to provide the code representing my medical specialty, I was referred to a 359 page “User Guide.” [I have no additional hair to pull out and I long ago lost my hari-kari knife!]
Note also that CMS indicates that it is going ahead with publication of the data it has collected so far without the “incorrect” payment transactions already identified. I am informed that as many as one third of the reported payments will be excluded. I have no confidence that all incorrect payment transactions have been identified and I suspect no physician would disagree.
I am as interested as anyone in seeing the results. I believe the public will be stunned at the extent to which Big-Pharma and Big-Device have captured our health care system including its institutions of medical education. Nonetheless, without confidence that the system of reporting is fair and accurate, the benefits of such reporting will be compromised. With- or without torture, the confessions of the data will always be suspect.
I have no conflicts of interest or bias to report other than that the public be fairly served by our medical and educational systems.
Peter Hasselbacher, MD
Emeritus Professor of Medicine, UofL
August 26, 2014