Click here for free access to our latest coronavirus/COVID-19 research, commentary, and news.

Support nonprofit science journalism

Science’s extensive COVID-19 coverage is free to all readers. To support our nonprofit science journalism, please make a tax-deductible gift today.

Will sunshine database clarify or cloud researchers' industry ties?

A federal database publicly disclosing companies’ payments to doctors went live this week, drawing a splash of attention to the $3.5 billion given to physicians over 5 months last year. Although one purpose of the site is to shine light on potential conflicts of interest in research, the site may fall short of that goal.

The database’s origins go back to an investigation by Senator Charles Grassley (R–IA) into drug company consulting by academic researchers. Starting in 2008, Grassley staffers alleged that about a dozen psychiatrists and others failed to fully disclose to their employers hundreds of thousands of dollars of consulting income they were receiving from companies. Public Health Service (PHS) rules for National Institutes of Health (NIH) grants required that the researchers disclose such payments to their institutions; but the reported totals fell short of those reported to Grassley by companies. In some cases, the payments were from companies whose drugs the researchers were studying, Grassley alleged (see the full backstory from a former Grassley staffer here).

The Institute of Medicine (IOM) recommended broader disclosure of potential financial interests, and in 2011 PHS followed through with tighter rules. IOM also endorsed a Grassley proposal to require companies to report those payments for posting in a public database. Besides revealing payments that might influence a physician’s prescribing decisions, medical colleges said such a database could help them crosscheck what their faculty (those who are M.D.s, anyway) report under the PHS rules.

The Grassley proposal became part of the 2010 Affordable Care Act, and the Open Payments database, run by the Centers for Medicare & Medicaid Services (CMS), went live on 30 September. For now, it contains 4.4 million payments reported by drug and device companies to 546,000 physicians and 1360 teaching hospitals from August 2013 to December 2013. They range from free trips, meals, speaking fees, and gifts to research funding. The ties “do not necessarily signal wrongdoing” and may include “beneficial ones,” CMS said in a press release; but disclosing them may “discourag[e] inappropriate relationships.” 

So far, Open Payments has gotten a mixed response. The investigative reporting group ProPublica warns that the data are incomplete and funds for ongoing research may not be included. (In addition, about 40% of payments are deidentified because CMS has not yet been able to match them with an individual physician.) The site contains little more than enormous raw spreadsheets that are awkward to search. CMS has promised better search tools soon.

Other problems involve the research data itself. For example, the payments include the dollar value of drugs donated by companies for clinical trials. For some drugs for cancer and hepatitis C that cost upward of $100,000 a year, the total dollar value for just a few patients could be “staggering,” wrote several Johns Hopkins University researchers in a 30 September editorial online in the Annals of Internal Medicine. They say the public may not realize that the payments represent donated drugs, not money going to individual researchers.

It’s also not yet clear that the database will be much help for institutions to audit faculty reports, says Heather Pierce of the Association of American Medical Colleges in Washington, D.C. Besides the fact that the data are incomplete and from 2013, the payments appear as individual line items, whereas institutions normally ask faculty for total fees from a company. There will also be discrepancies because the companies may include the dollar value of expenses such as hotel and transportation for a speaking engagement—money the researcher never sees.

“There are a lot of potential issues that may create confusion or misunderstanding in the research payments side,” Pierce says.

She says that concerns about such discrepancies have deterred some institutions from following through on a suggestion in the PHS rules that they post researchers’ consulting income online. Many are instead releasing it only upon request.

At the same time, Pierce says, even with flawed data in Open Payments, “the fact that it’s there can be a starting point for conversations” about why clinical researchers interact with industry, which “can be a very positive thing.”

*Correction, 3 October, 8 a.m.: Drug costs can be $100,000 a year, not a month.