Rebooting the Statistic That 5% of Eligible Patients Participate in Clinical Trials

For three decades, the widely-accepted and frequently-cited statistic that only 5% of eligible patients participate in clinical trials has served as a clear indication that the clinical research enterprise is failing to attract study participants.

It is true that patient recruitment has been a top challenge that has increased in difficulty over the past 30 years. And we can thank the statistic for helping to bring attention to this challenge. But the figure itself is superficial and flawed and reflects poorly on our growing awareness and commitment to patient-engaged clinical research.

Put simply—the 5% figure needs a serious reboot.

It is based on a simple descriptive statistic that fails to take into account the many factors we now know to be associated with a patient’s decision to participate in a clinical trial. The figure is also derived from a very specific patient population participating in a small subset of clinical trial activity and should not be generalized to all patients in all disease conditions.

Moreover, the figure is a static and outdated estimate that fails to recognize the evolving nature of investigational therapies in the drug development pipeline, and the highly customized strategies and executional models now required and implemented to engage study volunteers and their support network.

Back to the source

The statistic was first introduced by National Cancer Institute (NCI) researchers Michael Friedman and Dennis Cain at a 1989 workshop of the American Cancer Society and published the following year in the journal Cancer.¹ Friedman and Cain looked at NCI-funded cooperative group studies of select oncology diseases actively enrolling study volunteers. They divided the number enrolled by the number of new cancer cases referred to the NCI in the prior year and showed referral capture rates ranging from 0.7% to 3.7%.

The intent of the study was to convey the level of clinical research activity and infrastructure among NCI’s Cooperative Group system. But the referral capture rates attracted attention and stimulated a number of initiatives to improve physician referral into, and raise public and patient education and awareness about, NCI-funded clinical trials.

Soon after the statistic was published, it began resurfacing in the literature and in meetings and conferences, gradually morphing into a figure embraced by professionals across therapeutic areas and involved in not only government-funded but also pharmaceutical industry-funded clinical trials.

The calculation is deeply flawed—in part a function of the time when it was conducted. It does not factor protocol screening and inclusion and exclusion criteria in its assessment of "patient eligibility." It is based on the assumption that if a patient has been diagnosed and referred then surely she or he will participate in the clinical trial. The role that competing clinical trials play in limiting the availability of study participants was also not part of the calculation.

But our interpretation of the 5% figure is perhaps the most notable shortcoming. Historically, the primary explanations of the statistic included the failure to execute an effective recruitment campaign, patient distrust, and low awareness among referring physicians. In the early 2000s, the explanations broadened to include public and patient awareness and literacy, and patient attitudes and perceptions about clinical research.

More recently, the burden of participation, access to clinical trials, and the important role that patients’ unique needs and preferences play have received more recognition as factors that influence self-determination of eligibility. What concerns do patients have about the risks associated with the investigational treatment and study assessments and how do they vary by disease condition and participant subgroups? What role does the informed consent process play in discouraging the decision to participate? How easy is it for patients to find and join clinical trials, how convenient is participation based on the protocol requirements, and how do these experiences and needs vary by participant subgroup? This is but a partial list of areas we are now examining that influence eligibility.

The sophisticated, subjective, and active role that clinical care providers—physicians and nurses—play in determining eligibility and facilitating patient engagement is one area that has been largely absent from interpretation of the statistic. It also represents an untapped opportunity, particularly in light of growing interest in leveraging clinical care infrastructure and data to support clinical research.

Eligibility and participation enablers

Studies in the literature have consistently demonstrated that clinical care providers hold unprecedented influence: they are the most trusted sources for health, medical, and clinical trial information. The vast majority of patients report that they would consider participating in clinical trials if their physician or nurse recommended it. A similarly high percentage of global study volunteers confirms that they spoke with their physician prior to making the decision to participate in a clinical trial. And patients who receive information about clinical trials from their healthcare provider are significantly more likely to participate.

My team at the Tufts Center for the Study of Drug Development (Tufts CSDD) surveyed 755 physicians and 1,255 nurses primarily in North America and in Europe and found high self-reported levels of general awareness about clinical research, familiarity with the clinical research process, and high reported willingness and confidence in referring patients. But the reported rate of referral into industry-funded clinical trials was remarkably low: Physicians reported referring a median of five (5) patients into clinical trials annually, a referral rate that is less than 0.2% of their annual clinical care patient volume. Nurse referral volume was even lower—a median of two (2) patients annually—representing a .04% referral rate.

The reasons for referring were in large part associated with the subjective determination of a patient’s ability to participate and confidence in where and who is conducting the clinical trial. Our study found, for example, that eight out-of-10 physicians (80%) and two-thirds of nurses (68%) are most likely to refer their patients to familiar colleagues and research centers. A high percentage of physicians and nurses indicated feeling that the focus and demands of the clinical trial may not be appropriate for a given patient, or may be too difficult for a given patient to adhere to the protocol requirements. Contrary to commonly-held beliefs, the least mentioned reason—only 9% of physicians and 2% of nurses—was the fear of losing patients to another clinical care provider.

Having established more intimate relationships with their patients, physicians, and nurses determine eligibility based on their own subjective assessment of each patient's individual ability and predisposition to enroll and participate.

Closing thoughts

Given the high and rising complexity of protocol designs, the growing volume of active clinical trials underway, and the increasingly customized ways that clinical trials are executed, it is extremely difficult—perhaps impossible—to derive an accurate and meaningful statistic that characterizes the proportion of eligible participants overall and by disease condition.

What figures should we use? Those that are disease specific; that recognize the unique experiences of patients living with that disease (e.g., percent enrolled to screened; percent complete to enrolled); and those that are particular to a given protocol (e.g., actual to planned timelines).

It is also extremely difficult to end the use of a flawed statistic once it has gone into circulation.

That said, I’m hopeful that this column may encourage colleagues to scrutinize that statistic and to continuously push our understanding of factors influencing the determination of eligibility to participate in clinical trials.

Ken Getz, MBA, executive director and professor, Tufts University School of Medicine

Reference

1. Friedman M.A.; Cain D.F. National Cancer Institute Sponsored Cooperative Clinical Trials. Cancer. 1990, 6 (10), 2376-2382. https://acsjournals.onlinelibrary.wiley.com/doi/10.1002/1097-0142(19900515)65:10+%3C2376::AID-CNCR2820651504%3E3.0.CO;2-A