Massive New Payments Database has Treasure for Data Miners, with Caveats

Within the Patient Protection and Affordable Care (PPAC) Act of 2010 is a provision, known as a “Sunshine Law”. The law requires that healthcare manufacturers (pharmaceutical, biotechnology and medical device organizations) disclose payments they make to healthcare professionals and organizations as gifts, consulting fees, honoraria, travel, research grants and other means. The payment information is given to the Centers for Medicare and Medicaid Services, CMS, who then aggregate and publish the data, in a program called “Open Payments”. There were several delays in implementing the program, and in the meantime the organization ProPublica stepped in with a program called Dollars for Docs, aggregating data from a subset of 17 manufacturers who were already reporting payments.

Last month, however, the first Open Payments database from CMS has become publically available on their website.  While the main intent of the sunshine law is to inform the public of financial relationships between their healthcare providers and the manufacturers, the database also provides healthcare companies with both competitive intelligence and insights on key clinicians and institutions. Snowfish has analyzed the database and developed a white paper describing the database, its strengths and weaknesses, and ideas on how the data can be used by companies. Here we give highlights from our analysis, but for a deeper dive please contact Snowfish for your complimentary copy of this white paper.

What the database contains… and doesn’t

This initial release of the Open Payments database details payments totaling $3.5 Billion, for just the five month period between August 1 and December 31, 2013. The payments are made to a total of 546,000 healthcare providers and 1,360 hospitals, by 1,419 manufacturers and Group Purchasing Organizations (GPOs).

That all sounds impressive for a dataset until we find that despite the long delays in implementing the program,  only $1.3 Billion of the payments have been properly assigned by CMS to a named healthcare provider or hospital. For the remaining $2.2 Billion CMS has had to “de-identify” the payments, making them near useless for analysis. It is quite alarming that in four years of preparation CMS and its contractor, CGI Federal, did not have the foresight to insist that manufacturers provide unique, unambiguous identifiers for each payee. One such immediately obvious unique identifier would be the NPI number used in CMS’s own National Plan and Provider Enumeration System. Such a simple measure would have avoided much work in trying to identify the payees, and the embarrassment of having mainly failed in that attempt.

In addition, the database is missing another $1.1 Billion of payments, because either there was some dispute about the payment, or the manufacturer exercised their right under the law to delay publication of a payment for up to 4 years.

So in fact the identified data published represents only 28% of the total of $4.6 Billion in payments for the period. Still, there are certainly many interesting insights from evaluating the $1.3 Billion of properly identified payments. Thus, we may have hope that CMS will do far better with respect to payee identification in future releases of the database.

Structure of the database

The database is divided into three sections, based on the overall type of financial relationship:

• General Payments: Payments or other transfers of value not made in connection with a research agreement or research protocol.
• Research Payments: Payments or other transfers of value made in connection with a research agreement or research protocol.
• Physician Ownership Information: Information about physicians who have an ownership or investment interest in an applicable manufacturer or GPO.

The General Payments section contains identified payments from 948 companies. This ranges from a single payment to as high as 165,154 payments per company and from a total spend between just $13 and $130 million per company.

In the separate “Research” section we find 294 companies reporting payments for research, ranging from just $100 to $18 million.

How Snowfish helps industry use this new this data.

For well over a decade, Snowfish has been the business of identifying, mapping and profiling KOLs and Centers of Excellence, providing detailed analyses to scores of healthcare companies. We have incorporated payment information into our research ever since it first became available, including the ProPublica’s datasets. We have now adapted our proprietary software systems to incorporate the new CMS data, and are able to offer you this enhanced capability. Such financial data is extremely valuable to providing competitive intelligence, using purpose-built software tools and extensive experience.  

David Fishman is President of Snowfish, a leader in commercial analytics for life science companies with products in all stages of the life cycle. Snowfish specializes in driving innovation and challenging companies to look in new directions through a unique collaboration of strategic vision and sophisticated analytics overlaid with solid domain expertise.  He can be reached at dave.fishman@snowfish.net.