Abstract
It is accepted that participants in research should be compensated for any undue harm they suffer as a result of participation. Ensuring that appropriate insurance systems are in place to provide compensation or treatment to the research participant in the event of injury, disability or death attributable to participation is the responsibility of research ethics committees. 1 But how does an ethics committee know whether the sponsor’s insurance arrangements are adequate or appropriate? This article describes two different systems of insurance that may be available to compensate participants. It is difficult to know whether these systems work well for participants as little information is available about successful claims or the experience of claiming. However, we do know that in making claims participants face considerable barriers.
Two different ways of obtaining compensation
One way for the participant to obtain compensation for harm suffered as a result of participation in research is through a claim for negligence against the sponsor or other organizations involved in the study. Put simply, negligence means causing harm to another person through doing something or failing to do something to provide a proper level of care where a duty of care exists. A successful claim requires the adjudication of a court and therefore involves a legal process. Alternatively, compensation may be available through a no-fault compensation scheme. If such a scheme is available, then the claim can be settled out of court between the claimant, the sponsor and insurer. How do research participants fare under such these two schemes? Unfortunately, there is little or no information to answer this question.
There have been no court cases relating to negligence in research since 1996, when the Medical Research Council and Department of Health were successfully sued for failing to realize that growth hormone extracted from pituitaries could potentially transmit variant Creutzfeldt-Jakob disease (vCJD).
Similarly, there is no published information about claims settled by no-fault compensation or by any other private arrangement between the research participant, the insured and the insurer. One conclusion that could be drawn from the lack of claims is that research participants come to little harm in research and therefore there are no claims. But it may not be so simple. The lack of claims may be due to the difficulties or barriers that research participants face when they attempt to claim compensation.
Barriers to a successful claim for negligence
Did the interventional harm the patient?
If research participant is harmed, then they may have a claim for negligence if an individual or an institution with a duty of care has engaged in conduct that is culpable because it falls short of what a reasonable person would do to protect another individual from foreseeable risks of harm. So, for example, if the protocol states that 10 mg of the investigational drug should be administered and a hospital administers twice that amount, causing harm to the research participant, then the hospital has been negligent. But the process of proving negligence is a difficult affair. A review for the Scottish Government notes:
It is not enough for a pursuer to show that he or she has suffered a legally recognised injury, that the defender had a duty of care towards the pursuer, and that the defender fell below the standard of reasonable care – that is, that the duty of care was breached. Negligence alone does not entitle a pursuer to damages in compensation for harm or loss. In addition to negligence (breach of a duty of care), the pursuer must show a direct causal connection between the breach of duty (the fault) and the injury suffered (the damage or harm). There is, therefore, no entitlement to compensation unless causation can be established. 2
In any successful claim for negligence, not only does the research participant need to prove that there has been a breach of a duty of care, and that a harm has occurred, they also need to prove that that breach has led to that harm or injury. Proof of causation is very difficult when the research participant may be suffering from a long-term or terminal illness. For example, a patient with cancer in a trial may suffer extreme diarrhoea that leads to heart failure. It would be extremely difficult to prove that the investigational drug has caused this effect. The cancer may have affected their gastrointestinal system or one of the other drugs they have been taking may have caused the problem. It may be impossible to disentangle the effects of research treatment or procedures from the illness trajectory. While this particular issue may not be a stumbling block for healthy volunteers, volunteers still face barriers in claiming as they may have difficulty establishing which organization is responsible for the harm.
Who is to blame?
The arrangements for conduct of some clinical research, in particular clinical trials, can be extremely complex, making any claim for negligence a difficult and very costly affair. Trials may now use many different types of organizations or subcontractors to provide services to the trial. For example, the chief investigator may not write the protocol. This might be contracted to a company. Trials may also use the services of pharmaceutical suppliers, trial management organizations and so on. Some trials are conducted on an international if not global basis with subcontractors based in many different countries. A rule of thumb is that if the organization undertook the activity, it is liable for any claims for negligence arising from that activity. Two things follow from this. First, the subcontractors must have adequate cover for their own negligence and secondly, the sponsor must have agreements in place with all subcontractors that clearly identify who is responsible for a particular activity.
While the trial regulations are framed in such a way that the buck stops with the sponsor, in any claim the sponsors’ legal advisors and insurers may wish to bring subcontractors into any action where there is evidence that they have been negligent. Given the complexity of the studies and the burden of proof required, the legal expenses of dealing with negligence claims can be considerable.
Who has sufficient funds to pay for compensation claims?
Irrespective of any insurance cover, the United Kingdom’s National Health Service (NHS), large academic institutions and large pharmaceutical companies have very deep pockets or many assets to defend claims and provide compensation. But small start-up companies with few other assets may be entirely reliant on insurance, which can easily become exhausted. The manufacturer of TGN 1412, Tegenero, was insured for £2m but declared insolvency rather than face the costs of defending the claims of the research participants damaged in that trial. 3
Sponsors such as the NHS and universities are unlikely to declare insolvency but the depth of their pockets is not the only issue. Cover for negligence in NHS is provided through a pooled risk scheme called Clinical Negligence Scheme for Trusts (CNST). This covers all liabilities for clinical negligence, including those that occur in research. But the NHS Clinical Negligence Scheme only applies to NHS patients. NHS sponsors may recruit healthy volunteers as comparators in clinical studies. But should there be a claim from these participants, then the NHS trust would be uninsured and would have to meet any successful claim from its own funds. Similarly, the situation for private patients at NHS sites is complex as they would not be covered by NHS indemnities. Many private hospitals do have cover for research but there are still complex issues of who will foot the bill for any sequelae of serious adverse events.
With many different organizations involved – sites, device manufacturers, drug suppliers and sponsors – cause and effect can be very tangled, liabilities difficult to prove and costs high. The odds are stacked against participants in pursuing such claims, so perhaps it is not surprising that no claims for negligence have reached the courts in over 20 years.
No-fault compensation – time for setting standards?
When should it be required?
The alternative route through which research participants can claim compensation is through a no-fault compensation scheme. But no-fault arrangements are only available in particular types of studies and therefore are not available to all participants.
Non-negligent harm cover or no-fault compensation provides the possibility of settling a claim for harm where legal liability would be impossible or difficult to prove. An example would be where the participant suffers from a novel serious adverse reaction that has not been documented before or an adverse reaction which is significantly more severe than previously documented. No-fault compensation can also provide compensation for a research participant when a claim for negligence is impossible to pursue. When the sponsor in the Tegenero Trial declared insolvency, compensation for the victims was achieved through drawing on the no-fault compensation arrangements of the subcontractor – the clinical research organization – Parexcel. 4
The Clinical Trials Regulations 2004 require ethics committee to ensure:
provision for indemnity or compensation in the event of injury or death attributable to a clinical trial (Article 6.3(h) of 2001/20/EC); insurance or indemnity to cover the liability of the investigator and sponsor (Article 6.3(i) of 2001/20/EC).
This is interpreted as meaning that no-fault compensation is a requirement for clinical trials, but for other types of studies, the NRES guidance 1 states that it is up to ethics committees to decide on a case by case basis whether a no-fault scheme is required. But on what basis would an ethics committee make this decision?
It is hard to predict which studies are likely to be risky; commercial insurers, for example, will have differing views about this. Insurers usually work by insuring the whole portfolio automatically then picking out certain categories of studies which they regard as of greater risk. The insurance will review these studies separately. It is never very clear on what basis the insurers make these assessments as the sponsor probably knows more than anyone else about the risks of the trials, particularly where novel therapies are involved. For a study that has been referred, the insurer may agree cover, set an extra premium for the study or refuse cover. Where cover is refused, a broker will need to be employed to see if another insurer will provide cover. Without a doubt, drug trials where one of the inclusion criteria is pregnancy are universally seen as risky. Insurers have little appetite for them and will set very high premiums. For some insurers, any drug, device or surgery study enrolling children under 5 is risky. Others are only concerned with enrolment of large numbers of healthy children, for example in vaccine trials. Some are very concerned about new technology such as cellular therapies or products. Others regard these as low risk if they are undertaken on a few participants with life-threatening conditions with no other treatment options. Some have a risk calculus based on the size of the study and the health of participants – that is, any large-scale trials of entities with little track record in otherwise healthy individuals are inherent risky.
Access and the impartiality of no-fault schemes
Not all study sponsors have access to no-fault schemes. Academic institutions purchase commercial insurance and therefore can usually buy both types of cover. No-fault schemes are a requirement for pharmaceutical companies working under the Association of British Pharmaceutical Industry (ABPI) guidance. However, NHS trusts cannot provide non-fault compensation as it is ultra vires for NHS organizations to offer advance compensation to participants for harm where no liability arises. In order for the NHS pay compensation to apply, negligence must be established. Therefore, research participants in studies sponsored by NHS trusts may be at a considerable disadvantage if they are harmed.
There is no legal guidance or requirements about how no-fault schemes should operate, and such schemes can vary, giving rise to concerns about their impartiality. In settling claims, the amount of independence from the insurers and the insured varies. For example, schemes provided by commercial insurers to academic institutions require the appointment of an independent lawyer to decide the level of compensation. The insurance company will honour their decision. For commercial trials, the Association of British Pharmaceutical Industry has different guidelines for Phase 1 trials and other phase trials. As a result of the Tegenero scandal, the Association guidance for phase 1 trials was enhanced. The sponsor must agree to compensate the trial participants for any injury that arises through their participation in the trial. If there is dispute about the level of compensation, then it must be referred to an independent arbiter appointed by the President of the Royal College of Physicians. The guidelines also state that the provision of no-fault compensation for Phase 1 is contractual obligation with the research participant. But for trials other than Phase 1, the level of compensation is very much in the hands of the commercial sponsors. In the case of a dispute with a claimant about the level of compensation, there is no obligation for the company to seek an independent opinion or to honour it. The Association guidance merely recommends that the company does so.
The difficulties in pursuing successful claims for clinical negligence lead to the view that ethics committees should put more emphasis on no-fault schemes. But these vary in the access and the extent to which they provide an impartial adjudication of the claim. The difficulties faced by research participants in claiming compensation may mask a significant harm. Alternatively, trials may be safe, liabilities overestimated and the sector over-insured. Without centrally collected information, it hard to know whether all is well or whether the system needs to be reformed.