Recently, we were treated to a series of articles by ProPublica promoting the nefarious physician and hospital ties with industry. The organization behind the Dollars for Docs website, wrote an article for the Boston Globe about how the data behind the website was compiled. The article starts by listing the goal of the website, “to compare US hospitals based on the percentage of their affiliated physicians who receive payments of various sizes from pharmaceutical and medical device companies.”
Gathering Data
The group turned to the federal government’s Open Payments system to find information on industry payments to doctors. They used a file called “General payments,” which includes categories like promotional speaking, consulting, meals, travel, gifts and royalties. This category does not include payments for research.
Next, to determine which physicians practiced at which hospitals, they used Medicare’s Physician Compare data, which includes up to five hospital affiliations for each physician. The researchers then used archived Physician Compare data to obtain a snapshot of physician hospital affiliations as of December 2014. However, this data does not include doctors who do not participate in Medicare and those who do not admit may patients to the hospital. The file included records on 600,348 physicians who had at least one hospital affiliation listed.
To get information on a hospital’s characteristics, including the address and ownership of the hospital, the researchers relied on Medicare’s Provider of Services file from 2014 and data from the American Hospital Association Annual Survey. The analysis included just under 5,000 hospitals (4,815) and of those, 2,241 are included in the Dollars for Docs website because they had at least fifty affiliated physicians.
The last step in collecting data was accessing ratings that were assigned to select teaching hospitals in 2014 by the American Medical Student Association, known as the AMSA Scorecard. AMSA reviewed the conflict-of-interest policies of 204 teaching hospitals and graded them: A (the best grade) through C. Some teaching hospitals received an “Incomplete” because their policies were “insufficient for evaluation.” ProPublica researchers obtained this information through the Internet Archive.
Combining the Data
Here is where the fun comes in: combining and cleaning out the data. ProPublica reached out to “several hospitals” to verify the accuracy of hospital affiliation of physicians. After consulting with the hospitals, they made the decision to only include each doctor’s primary hospital affiliation, as well as include only the top 100 medical specialties in the Physician Compare and Open Payments data. Such a move excluded nurse practitioners and physician assistants who do not have to report industry payments, along with doctors that are not enrolled in Medicare.
ProPublica then matched the physicians in the Open Payments data (which uses a unique ID) to their National Provider Identifier (NPI) numbers. According to ProPublica, their methods were able to match more than 99.7 percent of physicians to their NPI numbers. From there, researchers matched the 2014 Physician Compare data to the 2014 Open Payments data.
For every hospital, ProPublica calculated the number of physicians who listed it as the primary affiliation in December 2014, the number of those who received a payment (and percentage), the total number of payments to doctors at that hospital, the number who received at least $100 in payments, the number who received at least $1,000 in payments, the number who received at least $5,000 in payments, and the number who were promotional speakers.
Conclusion
In conclusion, we see that ProPublica went to great lengths to try to shed light on industry payments to physicians: pulling from several related (and non-related) databases to come up with the information. While the statistical correlation is high (.93 according to ProPublica), it is hard to be certain that the database is not only reliable, but a beneficial resource for patients.