Failure to find enough clinical trial participants is more than just a logistical problem. There also are important ethical concerns. “If the study doesn’t finish, it really does violate the principle of beneficence,” says James Riddle, MCSE, CIP, CPIA, CRQM, vice president of research services and strategic consulting at Advarra.
Often, participants know they will not benefit individually, and are choosing to participate specifically for the benefit of others. “If the study is never completed, we put participants at risk [and] their altruistic gift doesn’t benefit the rest of society,” Riddle explains.
When IRBs review study protocols, they are not required to consider how likely it is the study will be completed. “Many IRBs do concern themselves with that, however — for reasons of not wanting to put participants at risk,” Riddle says. When IRBs conduct the risk/benefit analysis for human research subjects, decisions are based on the assumption the study will happen.
“If you aren’t going to finish the study, that risk/benefit ratio becomes unbalanced very quickly,” Riddle cautions.
Some IRBs explore researchers’ plans to ensure they meet recruitment targets. For example, IRBs might examine whether financial resources, staffing resources, or availability of participants are sufficient. “But those questions are not universally asked by all IRBs,” Riddle laments. “In some respects, the responsibility for ensuring that recruitment targets are met falls to the funder, who may have a vested interest in making sure the research finishes.”
When it comes to recruitment and completion, researchers should consider these key questions:
• Are there adequate resources to finish the study? Shortages of staff or funding commonly hinder the ability to recruit. “I find that many protocols run short on resources because researchers didn’t plan for how involved it would be to get enough participants,” Riddle observes.
If these problems exist, researchers should be realistic. It is possible the study might be better left to another team or to conduct the work later when sufficient resources are available. “If you can’t do it right, it’s probably not worth doing,” Riddle says.
• Are there enough participants available? “You need to do some really good due diligence that there are indeed enough people to recruit to make the study go,” Riddle says.
That might mean working with the research site, an IT department, or a clinical trials office. “Most learning healthcare systems will be able to query their patient database to determine how many folks may be eligible based on treatment history at the facility,” Riddle notes.
The key is to really understand the true number of available people who might be interested in participating, before the trial starts. If there are not enough potential participants at the researcher’s institution, it is important to know up front. “You can go back to the funder and perhaps ask for more resources,” Riddle suggests. “It might be a good time to find a collaborator at another institution who might be interested in partnering with you.”
• Of the people who want to participate, are there enough who meet the eligibility criteria? Realistic exclusion and inclusion criteria are needed. “If your protocol has criteria that are so lofty that you inadvertently exclude everybody, that’s a problem,” Riddle explains.
• What, if anything, should people be told about the possibility the study might not finish? Participants should know there is a possibility the research might not finish. “It does seem like a reasonable thing to include in the consent form,” Riddle says.
Participants usually are informed that they may be removed from the study for various reasons. The next sentence could state there is a possibility the research may end early. Then, people can make an informed choice on enrollment.
“Participants would probably care less about that if the study had the prospect of directly benefiting them,” Riddle offers.
• If investigators do not meet their recruitment targets, what is the next step? Researchers can always go ahead with the study — with a smaller sample size. “But if [the sample] gets too small, you actually won’t learn anything useful because you don’t have a big enough statistical power,” Riddle warns.
It is unethical to put participants at risk if there is nothing meaningful to be learned. “It’s not fair to the participants,” Riddle says. “It’s a big waste of resources that could have been used for some other trial that would have actually generated useful data.”
Instead, researchers can redouble recruitment efforts. “They can work with a partner to help design the protocol and help to hit recruitment targets, whether it’s an internal office, a contract research organization, a protocol development expert, or consultant,” Riddle suggests.
Research staff also can expand recruitment activities beyond the walls of their institutions to actively seek places in the community where participants might live. This can help engage people who otherwise would not participate. That could include community gathering locations, other local healthcare systems, or even companies that specialize in participant recruitment and matching of potential participants with trials. “The key is to think outside the halls of your own institution,” Riddle says.
Putting these contingency plans in place to hit recruitment targets is more ethical than abandoning a study that is underway. In addition to putting people at needless risk, there are broader concerns about diminishing trust in research. “There is the reputational damage that happens if people participate only to find out that the study never finished,” Riddle explains.
Researchers estimate one in four cancer trials fail to complete, with about 10% related to poor accrual.1 “Poor accrual also affects other types of pathologies, including rare disorders and benign disorders,” says Habib Rizk, MD, MSc, chair of the clinical and translational research ethics consultation service at the Medical University of South Carolina.
Recently, a clinical trial about metoprolol as a possible treatment for vestibular migraine ended early after researchers recruited only 130 subjects, half of the planned accrual goal of 266.2 The investigators still published their interim results as “lessons learned.”
Poor participant accrual can happen with any type of studies or disease states. “It is typically a problem of interventional clinical trials, especially when there is a placebo arm,” Rizk explains.
Physicians may not know about the trial, so they do not refer patients. Patients may not be willing to go through a tough protocol, or may want only the standard of care treatment. “It may be more of a problem when we are dealing with a nonlethal disease,” Rizk offers.
During protocol preparation, researchers determine the number of the study sample necessary to demonstrate the postulated effect of the intervention. “Lack of accrual will lead to results that have poor significance since they are not powered enough to demonstrate clinical effect,” Rizk says. “Depending on the scope of the study, this could be a big problem.”
To guard against this possibility, investigators must create a backup plan to tackle lack of accrual. This pre-hoc mental exercise allows researchers to determine the chances of it happening and potential solutions. “Funding is, after all, not limitless,” Rizk notes.
The sponsor or the Data Safety and Monitoring Board could terminate studies early if researchers do not reach recruitment targets. Adaptive clinical trial design may allow the study to go forward. This approach relies on an interim analysis that could change the outcome measure of the study, according to predefined rules.
For instance, if an interim analysis shows there is a larger difference than initially thought between a study drug and placebo, a repeat power analysis may show that a smaller sample size is needed. This will allow the study to proceed without relaxing eligibility criteria.
“If there is clear informed consent about an adaptive clinical trial design, ethical issues are less problematic, since the study subject is aware of the possibility of the trial changing course,” Rizk explains. The informed consent form can specify data will be analyzed throughout the trial, and changes to certain aspects of the trial may occur (e.g., changes to the number of subjects recruited, changes to the number of subjects going in the placebo arm or the study drug arm, or changes to the number of doses of medication administered).
From both ethical and regulatory standpoints, there are good reasons to analyze possible responses to recruitment pitfalls in study protocols. “Surprises may always happen,” Rizk says. “But changes could be done using predefined rules that sponsors, investigators, and regulatory bodies have agreed upon — and that study subjects have signed an informed consent about.”
REFERENCES
1. Stensland KD, McBride RB, Latif A, et al. Adult cancer clinical trials that fail to complete: An epidemic? J Natl Cancer Inst 2014;106:dju229.
2. Bayer O, Adrion C, Al Tawil A, et al. Results and lessons learnt from a randomized controlled trial: Prophylactic treatment of vestibular migraine with metoprolol (PROVEMIG). Trials 2019;20:813.
Failure to find enough clinical trial participants is more than just a logistical problem. There also are important ethical concerns. If the study remains incomplete, investigators risk violating the principle of beneficence.
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content. TESTING
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content