Give flawed payments database time to improve
A "Viewpoint" published in JAMA urges readers to be patient with the new federal Open Payments Program database. The site, designed to report drug and device industry payments to physicians, debuted substantially incomplete, the authors wrote, but it is too important to dismiss before its shortcomings are addressed.
"Viewed in the abstract, the value inherent in the transparency offered by the OPP database is beyond dispute," wrote Dr. Eli Y. Adashi, former dean of medicine and biological sciences at Brown University, and Sachin Santhakumar, a student in the Warren Alpert Medical School. "However, viewed in the light of day, the true value of the OPP database remains uncertain and probably too early to ascertain."
That's because when the database produced by the Centers for Medicare and Medicaid Services went online Sept. 30, it was incomplete in four main respects:
- Only five months of data: Payment data were limited to August to December of 2013 because of a delay in CMS issuing its final rules.
- Incomplete vetting: Only 4.8 percent of doctors and only 29.8 percent of teaching hospitals were vetted before the site's debut. This left more than 12,000 disputed records unresolved and unpublished.
- De-identified payments: Because of possible inaccuracies, 39 percent of payments (totaling $2.2 billion) were de-identified, including 90 percent of research payments.
- Limited context: Database users may not be able to infer the proper context of payments because the field for annotation is limited to 500 characters.
"This is a bad start, but we need to give it a chance," Adashi said. "It's quite amazing to realize that industry pays physicians $4 billion to $5 billion a year. It's a relationship between parties that really ought to be, on many fronts, at arm's length."
Improvements are indeed underway, Adashi said. Recently the U.S. Department of Health and Human Services released new tools to ease navigation of the site. In June 2015, the database will add year-round data instead of just five months worth.
And even in its present form the site has yielded new degrees of public oversight, including a New York Times story about a company's ties to doctors who prescribe its painkillers.
But the true potential of the site for journalists, policymakers, health administrators, doctors, patients, and medical educators remains hindered by its incompleteness, Adashi said.
Ultimately the OPP would be most useful to patients if its data were available within other sites that report other indicators of medical practice. That way consumers could assess their doctors in an all-encompassing, user-friendly site, Adashi said.
"With all that information under one roof that's easy to navigate, you wouldn't have to spend too much valuable time looking it up before decide on seeing somebody," Adashi said.
That is unlikely to happen any time soon and even the problems with the rollout of OPP may take time to fully fix, Adashi said.
But problems notwithstanding, he and Santhakumar conclude: "The OPP initiative must be viewed as the most potent antidote for potential conflicts of interest engendered by financial interactions between industry and the health care delivery enterprise. As such, it should be given the benefits of doubt and time."