During our recent panel discussion with Beth Harper of Clinical Performance Partners and Kelly Anastasio of Yale University, we addressed some of the common challenges of recruiting participants to clinical trials. Beth and Kelly shared insight on the components that make up a successful recruitment campaign and discussed how organizations can encourage greater participation. Here the two presenters address more attendee questions about recruitment materials, empowering coordinators, projecting recruitment outcomes and more.
From a tactical perspective, do you have any thoughts on targeting potential participants via referrals, online media, flyers, letters, social media, etc.? Where have you seen the most success? Are there any specific tactics to avoid?
This is a really important, challenging and the million dollar question; what tactics work best with what audience? As with many outreach and media programs, even the best upfront research still requires a bit of experimentation. That’s why it’s so critical to capture the performance metrics, in a systematic way, for each type of tactic for each study population or indication. Over time, you will start to see the patterns for example, that a given tactic (say online media advertising) generates a ton of interest or responses but the conversion ratio in terms of how many of those respondents pass the pre-screening process, show up for their first visit, sign and consent and ultimately enroll could be very low. Or that a community health event doesn’t generate a huge volume of interest, but the quality of the candidates is very high in terms of the percentage that go on to enroll in a trial. This allows you to find the general types of tactics that work for your community and the patient populations you serve as well as to fine tune the tactics for particular trials.
How does a large institution like Yale empower CRCs to track these recruitment strategies or campaigns at the individual level?
Yale utilizes the recruitment call center where they track directly into OnCore; if it is a department not utilizing the recruitment team they still ask that they track the referral source for our CTSA reporting and, since they are mandated to manage their subjects in OnCore and the referral tracking is a click away, it becomes a part of their workflow.
What is the impact of the increased complexity of clinical trials on patient recruitment?
This is hard to quantify. In general, as an industry we have struggled to get true and accurate benchmarks in terms of the number or percentage of the population (at large) or of a given patient population (with a particular conduction) that participate in clinical trials. The often-quoted data of 3-5% hasn’t changed in decades. That said, there is a clear recognition that as protocols become more complex, there is greater burden on both the investigative site and the potential subjects, their families, caregivers, etc. to participate. What we don’t know is the extent to which the complexity actually translates into lower participation rates. Partly because this is difficult to track, but partly because with the added complexity, we are also seeing amazing potential new treatments where the hope of finding a cure could outweigh the burdens of participation. Certainly with all of the FDA’s patient-focused drug development, the pharmaceutical industry’s focus on patient centricity and the various support services (such as payment debit cards, transportation support and so forth), we see a push towards trying to minimize some of the burdens that come with the added complexity of trials.
When you don’t have historical data, what is the best practice to make funnel projections?
A good starting point is to plan for a 10:1 ratio. In other words, based on long standing funnel metrics Beth has tracked, a “typical” trial requires 10 times as many potential subjects to be reached in order to randomize one. If a site is planning to randomize 10 subjects in a trial, they will likely need to reach about 100 subjects to account for the loss of patients across the “leaky pipe.” Beth’s data suggests that about two-thirds of the patients will be lost at pre-screening (e.g., call center, website pre-screening, EMR or chart review). Of those who pass pre-screening about 50% will decline the opportunity to participate and then of those, who consent, about one-third will fail the full screening process. Obviously the percentages’s vary by therapeutic area and indication and a more rare disease might see much different ratios (e.g., 200:1) compared to more common diseases. The more systematically the site can capture and track the funnel metrics, over time you can learn the specific funnel data for your types of patient populations which can inform future decisions to commit to trials, to determine the realistic number of subjects you can enroll, and the extent you need to supplement your “top of the funnel” efforts above and beyond the known pool of patients that you have access to.