Ask any clinical investigator, “What is the most difficult aspect of completing a research project,” and he or she will likely reply with “recruitment,” or successfully meeting target accrual of the study. Institutional review boards (IRBs) support this notion with evidence that between 20% and 30% of all approved research protocols never enroll a single participant. Failure to meet enrollment goals could potentially result in studies that lack sufficient power to answer the primary research question and may increase the risk of inaccurate interpretations of study findings.
The most common reason for under-enrollment or difficulty recruiting for studies is high screen failure, which can stem from prohibitive inclusion or exclusion criteria, limiting the ability to locate potential research subjects. For example, I was involved in a study with the goal of enrolling patients with heart failure who were naive to β-blocker therapy. In reality, it was challenging to find potential subjects who had not been placed on a β-blocker because evidence of the beneficial mortality-reducing effects of β-blockers was widely accepted at that time. It was necessary to modify the study’s enrollment criteria to include those who were maximized on current doses of β-blocker for at least 1 month rather than naive to this therapy.
Instead of modifying enrollment criteria to help improve recruitment, an alternative option is to add study sites. Transforming an investigation from a single-center study to a multicenter study carries many advantages. Aside from increasing recruiting opportunities through additional sites, researchers can take advantage of differing racial or ethnic demographics in various recruiting locations. For one study, I was interested in examining genetic markers of microalbuminuria in Hispanic/Latino patients compared with age-matched white patients with diabetes mellitus. The primary recruiting site was fairly homogeneous from an ethnicity standpoint, and it was taking many months to recruit, especially with the age-matching component. We were fortunate to work with an organization housed within our university’s Institute for Clinical and Translational Research. This network offered the opportunity to collaborate with additional sites and had a built-in infrastructure to aid in the logistics, such as a consortium IRB and multi-site data management services. We added two other sites and met our accrual goal much more quickly. My positive experience with the additional sites led to another collaborative effort with one of the sites for a study investigating immune responses to influenza vaccine in patients with heart failure.
In summary, expanding clinical studies from single- to multi-site designs can help meet accrual and ensure adequate power to investigate the hypothesis of interest. Moreover, a multi-site approach increases the potential to form fruitful collaborations for additional research and expands enrollment to include often-underrepresented populations, thus increasing the generalizability of clinical studies.
Orly Vardeny, Pharm.D.
Member, ACCP PBRN Community Advisory Panel