Research institutions could improve their conflicts of interest (COI) management by devoting staff or departmental resources to the issue. Some larger research organizations place a COI manager within the regulatory affairs department or in a research integrity office. And they make COI review part of the research ethics review process.
“The way our system works is any human subjects application comes to our office before it goes to the IRB,” says June Insco, outside interests and conflict of interest manager in the office of regulatory affairs at the University of Michigan Medical School in Ann Arbor.
“If we want a disclosure on the informed consent document, we’ll ask for that before we send the application forward to the IRB,” Insco says. “Our plan gives the IRB discretion to change what subjects see.”
Monitoring COI requires help from the institution, says Kara Manning Drolet, PhD, associate vice president in the Oregon Health and Science University (OHSU) research integrity office in Portland. Drolet also is the chair of the conflict of interest in research committee.
“We have to make sure COI disclosures are being made to journals,” she says. “It’s difficult, given limited staff and with all the other things we need to do. We consider some of the higher-level conflicts and monitor people in those categories more than those with lower-level conflicts.”
Insco and Drolet provide the following tips on how to improve conflicts of interest management:
• Monitor publications and disclosures. “When we do an annual review of somebody’s disclosure form, we monitor publications,” Drolet says.
“We do a PubMed search and look at articles where investigators should have made a public disclosure statement, and if it’s not there, we talk to the investigator about it and find out why it’s not there,” Drolet says. “We require a correction to the journal if we find they haven’t disclosed.”
The University of Michigan also checks on publications each year. “Once a year, we ask them to provide us with copies of publications to see if they did, in fact, disclose as they’re required to do,” Insco says.
Researchers send letters to journal publishers, outlining their COIs. If the journal fails to publish the disclosure, the investigator will need to show the COI staff a copy of the letter, she explains.
“Then we contact the publication and tell them to put in the disclosure online,” she says. “One of the challenges for investigators is that journal-to-journal requirements for disclosure vary; there is no standard practice, and that makes it difficult to get the right disclosures out there.”
Even if the journal publishes the disclosure solely online, it is sufficient because people reading the article would have access to the disclosure, Insco notes.
Sometimes there are misunderstandings and mistakes that lead to problems with journal disclosures. “Last week, we found a disclosure where the lead author failed to disclose,” Insco says.
It turned out the journal used a disclosure form that asked for financial interests of the past 12 months. The University of Michigan’s own COI rules did not limit disclosures to the past year, so the researchers should have followed his institution’s rules, she says.
“The author said, ‘I haven’t worked for this company for a year, so I don’t have to report it,’” Insco explains. “But in our minds, the principal investigator had received many payments from this company over the years.”
• Include COI disclosures in IRB submission template. “We have a section in our template that asks, ‘Who could benefit financially?’” Insco says. “For example, if someone is a consultant to a company, we ask the study team to put in the sentence, ‘In interest of transparency, we would like you to know that Dr. Jones is a consultant to the sponsor of this study.’”
If the investigator’s consulting relationship is over, then the wording would include the past tense: “We would like you to know that Dr. Jones was a consultant to the sponsor of this study.”
• Focus on more significant COIs. “We’re looking at what we consider to be those higher-level conflicts, and are we capturing all the ones we want to capture?” Drolet says. “If investigators are doing higher-level consulting, we might start including those in the monitoring requirements we’re already using.”
Conflict management plans at the University of Michigan can be applied in a variety of conflict of interest circumstances, including significant financial COI and others.
For example, there might be a situation where a co-investigator on a study should not be the person who determines whether a participant’s adverse events are directly related to the study drug. “Someone else has to determine the relatedness,” Insco says.
“Over the years, the IRB and conflict of interest committee have had a general dialogue about these issues,” she adds. “Sometimes, a situation arises where someone is very conflicted and also is a senior faculty member, so we might say, ‘You cannot be the principal investigator because you’re too conflicted.’”
Then, the senior faculty member might return with a junior staff member and say this person could be the principal investigator and the senior person will be the co-investigator. If the junior person does not have the experience and influence to carry out that PI role, then the COI office will not agree to the arrangement or will arrange for an ombudsman to work with the junior researcher, she explains.
“That ombudsman would have to reach out to the junior person and talk to them a couple of times a year, to keep a comfort level and make sure they don’t feel coerced in any way by that conflicted party,” Insco says.
• Check Open Payments annually. Open Payments Data is a federal government website with a search tool to find payments made by drug and medical device companies to physicians and teaching hospitals.
Anyone can enter a physician’s (or teaching hospital’s) name and search for information about the general payments the person has received, research payments, ownership and investment interest, associated research transactions, and disputed payments. The tool also provides context, showing where the person’s payments stand on a line graph that lists the national mean and the specialty mean.
“Every year, when Open Payments Data comes out with new information, we download the data and we compare anybody who received $40,000 in payments to what they told us at the institution,” Insco says. “This gives us a systematic look at the people who are getting a higher level of payments, including travel.”
Journal editors could also check principal investigators’ disclosure information against what is listed on Open Payments Data, she notes.
“To rely on someone to write down all their conflicts is not as necessary as it was before Open Payments,” Insco says. “They have it at their fingertips to do themselves.”