The money a trial investigator has to disclose

Why the rule exists, in the regulator's own words

The opening section of the rule states its reasoning directly. The agency may consider clinical studies inadequate, and the data inadequate, if appropriate steps have not been taken in the design, conduct, reporting and analysis of the studies to minimise bias. It then names one potential source of that bias: a financial interest of the clinical investigator in the outcome of the study.

The rule gives three examples of how that interest arises. Because of the way payment is arranged, such as a royalty. Because the investigator has a proprietary interest in the product, such as a patent. Or because the investigator has an equity interest in the sponsor of the study.

The consequence is a duty on the company, not on the investigator alone. An applicant whose submission relies in part on clinical data must disclose certain financial arrangements between the sponsors of the covered studies and the clinical investigators, and certain interests of those investigators in the product or the sponsor. The agency then uses that information, together with information about the design and purpose of the study and information obtained through on-site inspections, in assessing the reliability of the data.

That last clause is the whole frame. This is a data-reliability rule wearing the clothes of a conflict-of-interest rule. The question it exists to answer is not whether an investigator was paid but whether the numbers can be trusted.

Four interests, two dollar figures, and a one-year tail

The rule defines four categories, and the definitions carry the thresholds.

Compensation affected by the outcome of clinical studies means compensation that could be higher for a favourable outcome than for an unfavourable one. The rule gives examples: compensation explicitly greater for a favourable result, compensation in the form of an equity interest in the sponsor, or compensation tied to sales of the product, such as a royalty interest. There is no dollar threshold on this one. The structure of the payment is the problem, at any size.

A significant equity interest in the sponsor means any ownership interest, stock options or other financial interest whose value cannot readily be determined through reference to public prices — generally, interests in a non-publicly traded corporation — or any equity interest in a publicly traded corporation that exceeds fifty thousand dollars. The word generally is doing work there: an interest in a private company counts regardless of size, because nobody can price it.

A proprietary interest in the tested product means property or other financial interest in the product, including but not limited to a patent, trademark, copyright or licensing agreement. Again no threshold.

Significant payments of other sorts means payments by the sponsor to the investigator or the institution supporting the investigator's activities, with a monetary value of more than twenty-five thousand dollars, exclusive of the costs of conducting the clinical study or other clinical studies. The rule lists examples: a grant to fund ongoing research, compensation in the form of equipment, retainers for ongoing consultation, or honoraria.

Two features of the timing matter. Both dollar-threshold definitions run during the time the investigator is carrying out the study and for one year following its completion. And the definition of clinical investigator includes the spouse and each dependent child of the investigator. A holding that moves from the investigator to a spouse is still inside the definition.

A covered clinical study is narrower than "a study"

The rule does not reach every trial. It reaches covered clinical studies, and the definition is specific: any study of a drug or device in humans submitted in a marketing application or reclassification petition that the applicant or the agency relies on to establish that the product is effective, including studies showing equivalence to an effective product, or any study in which a single investigator makes a significant contribution to the demonstration of safety.

The definition then names what it generally does not include: phase one tolerance studies, pharmacokinetic studies, most clinical pharmacology studies unless they are critical to an efficacy determination, large open safety studies conducted at multiple sites, treatment protocols, and parallel track protocols. An applicant may consult the agency about which studies count.

The applicant is the party that submits a marketing application and is responsible for the certification and disclosure statements. The sponsor of a covered study is the party supporting that particular study at the time it was carried out — which need not be the same organisation as the applicant, and often is not.

The mechanics run through two forms. For every clinical investigator who participated in a covered study, the applicant submits either a certification on Form FDA 3454 attesting to the absence of the financial interests and arrangements described in the rule, signed and dated by the chief financial officer or another responsible corporate official, or a disclosure statement on Form FDA 3455. The applicant must also submit a list of all clinical investigators who conducted covered studies, identifying which of them are full-time or part-time employees of the sponsor of each study.

The disclosure form has five required contents: any outcome-dependent financial arrangement, any significant payments of other sorts, any proprietary interest in the tested product, any significant equity interest in the sponsor, and — the one readers usually miss — any steps taken to minimise the potential for bias resulting from the disclosed arrangements, interests or payments.

What the agency does with the answer, including deciding it does not matter

A disclosed interest is not a verdict. The rule directs the agency to evaluate the disclosed information to determine the impact of any disclosed financial interests on the reliability of the study, considering both the size and nature of the interest — including the potential increase in its value if the product is approved — and the steps taken to minimise the potential for bias.

It then says something that changes how any disclosure should be read. In assessing the potential of an investigator's financial interests to bias a study, the agency will take into account the design and purpose of the study. Study designs using such approaches as multiple investigators, most of whom do not have a disclosable interest, blinding, objective endpoints, or measurement of endpoints by someone other than the investigator, may adequately protect against any bias created by a disclosable financial interest.

That is a regulator saying that architecture can neutralise incentive. A single-site, unblinded study measured by the investigator who holds the patent is a different object from a fifty-site blinded study with a central endpoint committee, even where the disclosure forms look identical.

Where the agency does conclude that financial interests raise a serious question about the integrity of the data, four responses are listed. Auditing the data derived from the investigator in question. Requesting further analyses, such as an evaluation of the effect of that investigator's data on the overall study outcome. Requesting additional independent studies to confirm the results. And refusing to treat the covered study as providing data that can be the basis for an agency action.

There is also a gate before any of that. The agency may refuse to file a marketing application that does not contain the required information, or a certification that the applicant acted with due diligence to obtain it but could not, stating the reason.

What the rule does not do, and what that means for reading a study

Part 54 is a submission requirement. It attaches when an applicant files a marketing application for a human drug, biological product or device, and it attaches to the studies inside that application that the applicant or the agency relies on for effectiveness or, for a single investigator, for safety.

It does not run a public register. The disclosures go to the agency. The recordkeeping section requires the applicant to keep complete records of the arrangements, the payments and the interests for two years after approval of the application, and to permit authorised agency officers to access, copy and verify them on request at reasonable times. That is an inspection right, not a publication duty.

It does not reach a study that was never submitted in a marketing application. A study run and published outside that pipeline — including any study of a compound whose sponsor never filed for approval at all — has no part 54 certification or disclosure behind it. That is not an accusation about such studies. It is a description of which paperwork does and does not exist.

It also does not run on the investigator alone. The investigator must give the sponsor sufficient accurate financial information to allow complete and accurate certification or disclosure, and must promptly update that information if relevant changes occur during the investigation or for one year after completion. But the filing duty, and the refusal-to-file consequence, sit with the applicant.

For a reader assessing a study cited on a product page, the useful takeaway is a question rather than a rule. The regulator's own position is that a financial interest matters most where the design leaves room for it to matter — where there is one site, no blinding, a subjective endpoint, or an endpoint measured by the interested party. Those are features visible in a published methods section, and they can be checked without any disclosure form at all.

Sources

Each document below is named as it names itself, with the date printed on that document rather than the day it was read.

1. 21 CFR 54.1 — Purpose, read in full for the statement that financial interest is a potential source of bias, the three examples given (royalty arrangement, proprietary interest such as a patent, equity interest in the sponsor), and the statement that the agency uses the information together with study design and inspection findings to assess data reliability — Electronic Code of Federal Regulations, National Archives and Records Administration, September 2026
2. 21 CFR 54.2 — Definitions, read in full for the fifty-thousand-dollar public-equity threshold, the twenty-five-thousand-dollar other-payments threshold, the treatment of non-publicly traded interests, the one-year tail on both, the inclusion of spouse and dependent children in the definition of clinical investigator, and the definition and exclusions of a covered clinical study — Electronic Code of Federal Regulations, National Archives and Records Administration, September 2026
3. 21 CFR 54.4 — Certification and disclosure requirements, read in full for Form FDA 3454 and Form FDA 3455, the requirement that the certification be signed by the chief financial officer or another responsible corporate official, the five required contents of a disclosure statement including steps taken to minimise bias, the investigator's duty to supply and update information, and the agency's power to refuse to file — Electronic Code of Federal Regulations, National Archives and Records Administration, September 2026
4. 21 CFR 54.5 — Agency evaluation of financial interests, read in full for the size-and-nature test, the effect-of-study-design paragraph naming multiple investigators, blinding, objective endpoints and independent endpoint measurement, and the four actions available where a serious question about data integrity arises — Electronic Code of Federal Regulations, National Archives and Records Administration, September 2026
5. 21 CFR 54.3 and 54.6 — Scope and recordkeeping, read for the applicant-side scope of the rule, the two-year retention period after approval, and the agency access, copying and verification right — Electronic Code of Federal Regulations, National Archives and Records Administration, September 2026