Each year Insight Pharma Reports, a sister publishing arm to Clinical Informatics News, conducts a survey of 100, qualified respondents seeking insight into the trends in clinical trial patient enrollment and recruitment. The third annual survey, which will be released soon, reveals subtle shifts in clinical trials startup challenges.
Our pool of qualified responses comes from sponsors, CROs, PROs (Patient Recruitment Organization), vendors, and site. Sites, CROs, and sponsors represent the majority of the responses. One hundred percent of the respondents currently work in patient recruitment and enrollment.
Responses suggest remaining challenges with enrollment planning and perhaps a slight shift toward modeling and simulation to predict enrollment. Still only half of respondents report using social media to recruit patients, but users are pleased with the performance. On the site side, though, personal relationships and history still seem to outweigh other indicators.
Finding the Right Patients and Keeping Them
Enrollment planning is challenging. We gave respondents eight options for their biggest challenges and year after year all the options are flagged. Hitting enrollment goals and recruiting deadlines topped the list in both this year’s survey and last year’s. More than 35% of respondents also cited unrealistic protocols, short or unrealistic time frames, and unrealistic budget.
Concern about short enrollment timelines fell a bit in 2017. In 2016, it was the second most frequently mentioned concern, listed by 45% of respondents. In 2017, short enrollment timelines fell to the fourth highest concern at 39%.
Unfortunately, predicting enrollment remains elusive. Investigators and key opinion leaders lost some prestige in terms of enrollment prediction, though both are still heavily used. Respondents using investigator feedback to predict enrollment fell from 64% to 50%; those using key opinion leader recommendations fell from 41% to 31%. Simulation models, meanwhile, gained a bit of traction with 29% of respondents citing their use, a rise of 5 points.
On the whole, most respondents felt their enrollment predictions were accurate. The percentage reporting predictions that were “accurate” or “somewhat accurate” was over 70% both in 2016 and 2017. But, worryingly, the percentage calling predictions “inaccurate” nearly doubled, from 9% to 17%.
Going Social
The number of companies using social media to recruit patients for clinical trials was perfectly split: 50 respondents were using it; 50 were not. Of the half that were using social media for patient recruitment, a vast majority—88%—said it was effective. These responses mirror those from 2016.
Facebook was by far the most popular social media site used for patient recruitment, with 86% of companies using it. Health and wellness blogs came in second. CraigsList and Twitter were closely ranked as the third and fourth most-used platforms. No platform that we asked about—including Pinterest, YouTube, and Instagram—was not used by at least a few companies, and several were using avenues we’d not considered.
When asked to rank the top three most-effective recruiting avenues, social media came in first followed by online advertisements, posters at hospitals and clinics, and connections with advocacy groups. Again, of the 14 survey options that were offered for recruiting efforts, each one was named by at least four respondents, demonstrating the breadth of ongoing recruiting efforts and how varying opinions can be on their usefulness.
Closing the Deal
Unfortunately, sites continue to fall occasionally short of enrollment targets. Most respondents said it only happens 6-15% of the time, but more than 20% of respondents put their answers on the far edges of the range: reporting that they fail to meet enrollment targets either less than 5% of the time or up to 25% of the time.
Patient dropout rates continue to be low. The vast majority of responses—86—report losing less than 15% of patients enrolled in trials, and most of those—49—report losing less than 5% of trial patients, an improvement over last year’s 44.
Relationships With Sites
Existing site and investigator relationships still dictate much of site selection. In fact, the percentage of respondents reporting using third party resources for site selection fell from 25% to 14%. Sponsors and CROs continue to look to historical performance, specific clinical research focus, site experience in the indication, and a site’s ability to target patient population in determining which sites to work with again. These top four indicators of productive sites are unchanged from last year.
Interestingly, only 33% of respondents answered that they choose sites with the help of an internal database of site metrics. But all respondents did answer list their criteria for productive sites. Sites are considered productive when they have a successful historical performance, a specific clinical research focus, a proven ability to target a patient population, and experience in the indication. These four criteria described the most productive sites in both 2016 and 2017.
