Conflicts of interest in oncology

Dr Bishal Gyawali - Queen's University, Kingston, Canada

The problem of conflicts of interest exist in all of medicine but as a practitioner and as a clinician in oncology I wanted to touch upon the issue of conflicts of interest in oncology. We have lots of empirical data now that unambiguously show that receipt of conflicts of interest has effects on how guidelines are being made, how recommendations are being made in presentations, in journal articles, how there is a bias in publication based on conflicts of interest, how there is misinterpretation of data based on conflicts of interest, and ranging from an individual physician’s behaviour in recommending low-value practices up until formulating guidelines. So this is a big problem that cannot be ignored.

However, on the other side, at least we have these data, because there are some databases that we can look into to see whether a certain physician has been receiving money from the industry, whether it is research money or whether it is personal money, whether it is honorarium or travel or food. We have this research available because that database exists. So we criticise a lot of these conflicts of interest issues based on this data is coming from the US.

So for all its faults, at least the United States has a database that you can look into to check the conflict of interest of any particular physician. Other countries, like Canada itself, we do not have a database so there is no way to check whether a certain physician has a certain conflict of interest or not. There is no way to test that. So although we criticise the US cancer care system for being heavily influenced by pharma and for the problem of conflicts of interest, at least they have maintained a database and you can look it up. Canada does not and most of the high-income countries in the world do not. There are some countries in Europe, the Netherlands for example, who have started doing it, but many countries in the world, they do not have such a database.

Interestingly, the problem of conflict of interest is not associated with high-income countries alone. There are low- and middle-income countries’ physicians who also have conflicts of interest. This is particularly troubling because it affects individual physician behaviour. In the case of low- and middle-income countries, at least the countries where I have my personal first-hand experience, a pharma rep directly approaches the physician with recommendations to use their particular product and with several offers that will lead to conflict of interest. There is no authority looking over these issues. And forget about authority: there is no database. The lack of database is not only about low- and middle-income countries; there is no database to look up conflict of interest for any physicians except the United States and a few other countries in the world.

So this has been an issue. You must have watched several news – there used to be some high-profile news about this famous person did not disclose conflict of interest in their publication. But how do people find it out? At least there is a database and they can check and they can see that, okay, this particular individual had received so much of personal money from this particular industry, but he’s not disclosing it in the papers. But if there is no such database, how do you even check it?

The other thing about conflict of interest, some people think that disclosure is enough, and disclosure is the solution. I would argue that disclosure is not the solution because of 100 oncologists out there, maybe one or two people will be writing papers where they will disclose it. But there are the other 98 oncologists who will not be an author in the article, who are just practising oncologists, but there is the issue of conflict of interest interfering with everything. So, yes, when people write an article, they have to disclose that they have conflict of interest but is the solution to conflict of interests just disclosure? What about several oncologists who never write a paper, and what about their conflict of interest? So unless there is a database we can’t just rely on someone will probably publish a paper one day, and then we’ll look up conflict of interest.

So that is about database. The other thing that I find a little bit strange is about whenever these articles are published in journals, sometimes it’s so difficult to find the conflict of interest information if you don’t have access to the full text of the article. So that should not be happening. For example, lots of readers in the world, they will not have access to the full text of these paywalled articles, but they will have access to the abstract. But if the conflict of interest part is kept behind the paywall, then that is a disservice to the audience, to the readers, because many readers will formulate their opinions just by reading abstracts. They don’t want to have access to the full text, so if the conflict of interest information is not available in the abstract, and if it is behind the paywall, then how would they even know about it?

It does not make much sense to expect the readers to pay for the full text just to know whether the authors have conflict of interest or not. So I would highly recommend journals to put conflict of interest information with the abstract; when people are looking an article up on PubMed the conflict of interest information should appear in the public page.

The third thing is about writing editorials for certain trials or inviting discussants for plenaries of the most important oral presentations in these big meetings. At least the experts who were writing editorials for a drug, they should not be receiving money from the same company whose drug is being discussed. At least the discussant who is discussing about a particular trial in big oncology meetings should not be receiving money from the exact same company whose product is being discussed. That is an obvious huge conflict of interest, and we have published a paper, actually, showing that the editorialists who receive money from the same company, they write more favourable editorials regarding the product that is being discussed. So I just just wanted to highlight these discrepancies in the conflict of interest.

To summarise: first, disclosure alone should not be considered enough. Maybe only one or two percent of practitioners, they end up writing papers. What about other practitioners who don’t write papers, and how do you even check whether what is being disclosed in the paper is actually true? Second, all countries should work towards making an open database that shows how much conflict of interest, how much money someone is receiving from any individual company and under what domain. For all its criticism, at least the United States has this open payment database which allows us to interrogate that data, so kudos to United States for doing that. And third, whenever there is an opportunity to do an editorial, or do a discussion of a major trial, at least we need to be asking experts who do not have any financial incentive to bias the results in either direction. So that’s my view about conflict of interest.