Conflict of Interest Disclosure, and What Counts

What has to be disclosed

Authors are responsible for disclosing all relationships and activities that might bias, or be seen to bias, their work.¹ The second half of that sentence is doing as much work as the first.

The financial category is enumerated. It covers employment, consultancies, stock ownership or options, honoraria, patents, and paid expert testimony.¹ Those are the ones authors remember without prompting.

The rest of the list is where disclosures are usually incomplete. It also covers personal relationships, academic competition, and intellectual beliefs.¹ None of those involves money, and all of them can bias a judgment or appear to.

ICMJE has developed a disclosure form to standardize the process. Its member journals require the form, and other journals are encouraged to adopt it.¹

Why "there is no conflict" is the wrong test

The instruction is to disclose relationships and activities rather than to disclose conflicts. Those are two different questions, and only one of them asks an author to judge their own work.

That distinction does most of the work in practice. An author asking themselves whether a relationship influenced their judgment will usually, and often correctly, answer no. An author asking whether a relationship exists, and whether a reader might see it as bearing on the work, will answer differently.

ICMJE is explicit that the presence of a relationship or activity does not always indicate a problematic influence.¹ Disclosure is not an admission. Withholding, on the other hand, is the thing that damages a paper when it emerges later.

What an editor can check without judging anything

Four things can be checked, and none of them requires a view about anyone's integrity. Editors Canada's fundamentals ask that such findings be raised as clear, tactful, and actionable queries.²

The first is whether a disclosure statement exists at all, and whether it takes the form the destination requires. Journals publish that requirement, and it varies between them.

The second is whether the statement covers every author. On a multi-author paper the corresponding author often completes one form and assumes the rest, and a statement listing eleven authors' disclosures for a paper with fourteen authors is visible without any inquiry.

The third is whether the statement agrees with the rest of the manuscript. A funding acknowledgment naming a company, a methods section naming a proprietary instrument, and an author affiliation at a commercial laboratory are all facts the paper already states. Each is a relationship a reader will connect to the disclosure.

The fourth is whether the statement agrees with the paper's own record elsewhere. Editors Canada's copy editing standards cover checking for missing information and resolving inconsistencies across a document.³

Where the author decides, and it is not close

Whether a relationship exists, whether it is material, how to describe it, and whether to disclose something not strictly required are the authors' decisions. An editor who characterized a relationship on an author's behalf would be making a representation about that person to a journal.

An editor also does not decide whether a disclosure is adequate. That is the journal's judgment, and increasingly the reader's.

What an editor supplies is the reconciliation. The statement is present, complete for every author, in the required form, and consistent with the facts the paper states elsewhere.⁴ Each item can be resolved by the authors in an afternoon before submission, and none of them can be resolved comfortably after publication.

Everything a client sends is treated in confidence, including manuscripts under review, funding correspondence, and disclosure forms. We work on academic and research documents and journal articles against whatever the destination publishes, and we query rather than complete anything an author must attest to.

Investigators who keep a standing list of their own relationships, updated when one changes, spend minutes on this rather than an evening. The list is the same for every paper. Only the form it is copied into changes.