Michael Joyce is a writer and producer with HealthNewsReview.org. He tweets as @mlmjoyce
The well-respected medical journal, JAMA Internal Medicine, has published a special collection of articles highlighting blatant conflicts of interest between healthcare and industry.
It’s a topic we’ve covered hundreds of times dating back to the spring of 2005 and we’re glad to see it getting special attention.
Before launching into the JAMA articles I wanted to talk with someone who could set the stage for me. Perhaps provide a sense of the depth and breadth of conflicts of interest (COI) in our current medical climate, and maybe teach me about the types of conflicts that might be particularly ubiquitous, pernicious or dangerous. I turned to Roy Poses MD, an internist at Brown University with a special interest in evidence-based medicine and who is President of FIRM — The Foundation for Integrity and Responsibility in Medicine. Roy got back to me in a matter of minutes and – shall we say – pulled no punches:
- ‘drug, device, biotech, medical supplies and equipment (distributors of such too), hospitals, other provider organizations, health insurance, health care IT, consulting, contract research organizations, medical education and communication companies … etc …’
“with every sort of person who can make decisions about …
- ‘medicine, health care, public health, health policy, health education and research …’
“and every sort of group that would deal with all the above including …
- ‘academic health care institutions, their parent universities, medical societies, patient organizations, health care foundations, think tanks, accrediting organizations, medical journals, the media (albeit they are for-profit), non-profit insurers — and of course — local, state, and US governmental organizations, as well as global health organizations.’
Which (types of COI) are most important is hard to tell, and depends on how one conceptualizes important. But they are all bad in a sense. “
Thanks Dr. Poses.
You’ve certainly given us some context with which to approach this very timely collection of articles.
Researchers wanted to know if the doctors involved in writing guidelines for the management of cholesterol and hepatitis C had any conflicts of interest. In 2011, the Institute of Medicine (IOM) published standards for what they believed constituted conflict of interest (COI). The IOM standards call for no commercial COI among guideline committee chairs and co-chairs. Commercial COI should exist in less than 50% of regular committee members. This study found that both groups — the chairs and co-chairs, as well as the regular committee members — not only had conflicts of interest that they disclosed, but also conflicts of interest they did not disclose (but were revealed in other publications).
Chair & co-Chair group: about one-third of the doctors involved with the cholesterol guidelines disclosed COI, and a review of publications found more COI that was not reported. In the hepatitis C group two-thirds of the doctors disclosed COI and, again, more COI was found in outside sources. (again, according to IOM standards, none should have COI).
Regular committee group: 44 percent of the members involved with the cholesterol guidelines disclosed COI, but other conflicts of interest were identified that were not disclosed. In the hepatitis C group almost 3 out of 4 members disclosed COI but further conflicts of interest were identified in other publications. (again, according to IOM standards, no more than half of the regular committee members should have COI).
Other conflicts of interest that were most typical included industry-sponsored research, consultancy fees, and honoraria.
Therefore, neither group met the IOM standards for commercial conflict of interest. Furthermore, there was often a discrepancy between self-reporting by doctors and conflicts of interest discovered in contemporaneous articles.
From September 2013 through June 2014 researchers looked at a random sample of 439 PAO leaders, representing about six percent of the nearly 8,000 PAO’s identified in the United States. They were asked about the nature of their activities, their financial relationships with industry, and the perceived effectiveness of their conflict of interest policies. Of the 439 surveys mailed to PAO leaders, two-thirds were returned with at least 80 percent of the questions answered. The survey results were:
The median amount received by those who got industry funding was $50,000.
Nearly half the industry support came from the pharmaceutical, device, and/or biotechnology sectors.
8 percent of PAO leaders perceived pressure to conform their positions to the interests of corporate donors.
Conflicts of interest with patient advocacy groups are particularly disconcerting because of the potential to directly misguide patients. These groups now have — for better and for worse — an extended reach and influence on the Internet. They also can author and dictate guidelines, fundraise, sponsor research, influence and guide media coverage, and shape public policy and perception.
When the 2016 opioid prescription guidelines were published by the CDC they received substantial criticism. Some claimed the guidelines were based on weak evidence, others found the process by which the guidelines were formed to be ‘too secretive,’ and there were those who found them overly strict or, at least, restrictive for doctors who wished to prescribe opioid pain medications.
In response, a one-month public comment period was opened and 158 organizations responded. Of those respondents 44 groups (28%) got money from opioid manufacturers, 15% had ties to other life science companies, and 40% had no funding from life science companies.
Of the 44 groups who did have ties to opioid manufacturers, none disclosed this to the CDC. However, the CDC did not prompt or require organizations to disclose their financial associations in their comments.
Approximately 80 percent of the respondents supported the CDC guidelines with or without recommendations; including many that received funding from opioid manufacturers.
But the most disconcerting finding was that opposition to the guidelines was more common from organizations who got money from opioid manufacturers (38%) when compared to those who did not (6%). Funding was defined as “receipt of grants, contracts, gifts, advertising revenue, exhibition fees, or other material support.” Evidence for funding relationships came from “self-reports, public website disclosures, annual reports, and federal tax returns.”
In their discussion the authors offer the following closing comment:
“US Senate committees and investigative journalists have scrutinized the financial associations of opioid manufacturers and patient advocacy and professional organizations. A major concern is that opposition to regulatory, payment, or clinical policies to reduce opioid use may originate from groups that stand to lose financially if sales of opioids decline. Our findings demonstrate that greater transparency is required about the financial relationships between opioid manufacturers and patient and professional groups.”
The authors looked for potential conflicts of interest among 634 hematologist-oncologists who use Twitter. Using the open payments website, they found that about 80 percent of these tweeting doctors had at least one documented financial conflict of interest.
Roughly half of the hematologist-oncologists were recipients of research funding. Nearly 3 out of 4 received personal payments (ie. checks made out personally to them and not tied to their employer). About 60 percent received more than $100 in personal payments and about 45 percent got more than $1000.
Hematologist-oncologists on Twitter received a median of $1,644 in personal payments and a median of $11,064.21 in research payments.
“There were really two motivations for looking into this,” says Dr. Vinay Prasad, a hematologist-oncologist from Portland, Oregon who co-authored the study. “I would see these doctors tweeting about cancer drugs who were only giving the upside to drugs I knew were not necessarily the safest or most affordable option. When I would bring this up they would tweet ‘it’s not a problem’ or ‘it’s not our concern.’ I’d then look them up and they usually had FCOI. So they’re tweeting about drugs and not telling people they get money.
“Disclosure is not enough. Would any of us accept a judge in the courtroom who is paid by the prosecutor? If these doctors want to shape public opinion about a drug they shouldn’t take money. You either recuse yourself from the money, and if you don’t, recuse yourself from rendering opinions directly to the public.”
The authors say there are limitations to the study:
“We were unable to examine use of Twitter over time, whether non-hematologist-oncologists are active on Twitter and have conflicts, whether physician tweets are associated with FCOI, and whether hematologist-oncologists are tweeting about products or companies about which they are conflicted. These limitations may form the basis of future work.”
They do go on to conclude:
“Our results raise the question of how FCOIs should be disclosed and managed in an age in which information, interpretation, and criticism associated with cancer products and practices are increasingly available on social media. As a minimum standard, physicians who are active on Twitter should disclose FCOIs in their 5-line profile biography, possibly with a link to a more complete disclosure. For tweets regarding specific products that cause an FCOI, we advise users to include the hashtag #FCOI. Policies beyond disclosure should also be considered.”
Recently there has been considerable controversy over whether we have become overly concerned with conflicts of interest and the need for doctors to disclose them. Some will argue that it’s virtually impossible to find a doctor without a conflict of interest. Others claim zero tolerance of COI is the only way to ensure the integrity of medicine.
Regardless of your stance, the concrete examples of conflicts of interest in this special issue of JAMA Internal Medicine make it clear that the problem is pervasive, worrisome, and clearly has the potential to not only change healthcare delivery, but impact those who receive it.
It also underscores how difficult it has become for healthcare providers and journalists to keep up with an increasingly entangled web of COI. It makes me wonder: how can we expect the public to navigate it?