Failure to register and report clinical trials is not a victimless crime. It harms patients, undermines public health, slows down medical progress, and wastes public funds.
Nonetheless, not a single institution has ever been fined, anywhere in the world, for violating universally recognised ethical obligations.
During a recent parliamentary hearing in the UK, the issue of sanctions was once again on the agenda.
The UK national ethics regulator, the Health Research Authority, has long opposed sanctions – even though trial registration has been mandatory in Britain since 2013, and posting drug trial results onto the European trial registry has been mandatory since 2014.
Following public and political pressure to end impunity for unethical medical research, Health Research Authority (HRA) representatives seemingly changed course. They recently told parliament that they had developed three options for sanctions (see Q192 here):
Publishing a ‘league table of performance’
Examining applicants’ past performance when reviewing new grant applications
“Ask[ing] Government for the power to fine in the case of non-compliance”
The Health Research Authority’s options for sanctions will fail to ensure that all UK clinical trials are registered and report their results.
Below a discussion of the HRA’s red herring options.
HRA option 1: League table - necessary but not sufficient
TranspariMED strongly supports the creation of a public annual league table that covers all interventional clinical trials and meets key transparency benchmarks. Such a league table would help to drive improvements in performance by flagging shortcomings, and provide patients and taxpayers with important information about the performance of public bodies and pharmaceutical companies.
However, experience shows that a league table by itself will not result in 100% compliance. A public league table for drug trials has been in existence in the form of the EU Trials Tracker since September 2018, but according to AllTrials, one year later, “33 NHS Trusts and six university sponsors have 0% of their due trials reported”. Note that in other sectors, it would be unthinkable to propose a public league table as an alternative to sanctions for health and safety violations that put public health at risk.
HRA option 2: Reviewing individuals’ performance - misdirected
Examining individual applicants’ past clinical trial registration and publication record when reviewing new applications for ethics approval is misdirected. True, examining past performance and refusing to grant ethics approval for serial offenders could theoretically help to drive positive behaviour at the individual level.
However, this measure has numerous weaknesses:
It will fail to act as an incentive for researchers who are approaching retirement or are about to move abroad; such individuals could neglect to register and report trials without suffering any adverse consequences. Thus, it cannot form the basis for ensuring that all trials are registered and reported.
It punishes people for their past behaviour, rather than focusing on improving future conduct. In addition, reporting the results of every clinical trial has only been a global ethics obligation since 2013, so punishing people for their conduct pre-2013 is arguably inappropriate. Similarly, foreign researchers from countries where the imperative to publish is not widely known could be unfairly discriminated against.
It will not result in any sanctions being imposed. Imagine a researcher failed to publish the results of a single trial, but otherwise has a stellar track record. Would you refuse her permission to launch a potentially life-saving cancer trial on that basis? Many waivers would have to be granted, and in the absence of clear criteria, all those waivers would be decided on by the Health Research Authority – i.e. the very body that has consistently opposed sanctions with teeth.
It lets institutions off the hook. Focusing on personal rather than institutional performance leaves pharma companies and universities off the hook for ensuring that their employees comply with regulations and act ethically towards patients.
It focuses on academic publication, rather than on the uploading of results onto trial registries. Going forward, as per WHO best practices, all trial results should be uploaded onto registries within 12 months. Looking back at past journal publications will not help to drive this message home. Also, legally it is institutions are the ‘sponsors’ of clinical trials and thus responsible for posting summary results onto registries. (For this reason, the European Medicines Agency sends reminder emails to sponsoring institutions, not to individuals.)
Examining the past track record of individuals cannot substitute for holding institutions to account for their compliance with regulatory requirements, stewardship of public funds, and organisational adherence to scientific good practices.
Note: A more constructive variant of this option would be to defer individuals’ ethics approval until all their past trials have been retrospectively registered and reported. This would could help to save some older trials from becoming research waste. However, this approach would not qualify as a sanction.
HRA option 3: Financial penalties - sidelined
Health Research Authority representatives told the Committee that “financial penalties require legislation, which does not seem very likely in the present climate”, and suggested that this was a valid reason for the Health Research Authority not to seek such legislation in the first place (see Q201 here). This is deeply disappointing, and underlines once again that the UK’s medical research ethics gatekeeper is inclined to put the interests of the researchers it is supposed to regulate above the interests of the patients it is supposed to protect – in defiance of parliament’s wishes.
Making sanctions work: The way forward
Achieving 100% clinical trial registration and reporting rates is achievable in the UK without changing any legislation.
The Health Research Authority should make ethics approval for new clinical trials contingent on the sponsoring organisations putting into place systems and processes that ensure that all new clinical trials are consistently registered and reported in a timely manner (irrespective of individual employees’ priorities and actions).
Full compliance could be achieved by 2022 without new laws or financial penalties:
As a first step, the Health Research Authority should use the first iteration of its league table in 2020 (see above) to identify all trial sponsors who have failed to register a clinical trial and/or upload its results to a trial registry within 12 months of trial completion.
The Health Research Authority should then inform such sponsors that their system and processes require strengthening, and give them a one-year deadline for doing so.
After one year, in 2021, the Health Research Authority should use the second iteration of its annual league table to again identify sponsors with unregistered and/or unreported trials. Any organisation that has still failed to adopt adequate systems and processes at that point, and is thus demonstrably not able to ensure that its medical research complies with ethical and regulatory requirements, should be barred from sponsoring any more clinical trials. The mechanism for preventing this would be to automatically refuse ethics approval for any research project listing such an institution as the trial sponsor.
Such a sanctions system would have multiple advantages.
First, it would be effective at getting all future trials registered and promptly reported, something the Health Research Authority has signally failed to achieve in the past. (Over 30% of non-drug trials in the UK are still not even being registered.)
Second, it would be more efficient than reviewing thousands of individuals’ publication records.
Third, by forcing institutions to put into place strong registry management systems and processes, it would ensure consistent compliance, avoiding the problems that currently arise when researchers retire or move jobs.
Fourth, it would set out clear standards, rather than vague criteria that could result in individuals being unfairly punished.
Finally, it would not prevent or delay life-saving research from taking place, or create additional red tape for researchers. Many universities and NHS Trusts have already set up joint trial units that sponsor trials for multiple institutions. Universities unable to meet stringent transparency expectations would lose the right to sponsor trials, but their staff could continue conducting research under an external sponsorship umbrella.
The alternative is to allow pharma companies and universities with insufficient safeguards to run even more clinical trials with even more UK patients, without any guarantee that their results will be made public within 12 months.
Do you disagree? Do you have a better solution? If yes, TranspariMED would love to hear from you.