CRIO’s Benchmarking Data: Differences in Small and Large Sites

By Raymond Nomizu

This blog post is the second in a series on site performance KPIs based on an analysis CRIO performed for the Site Accreditation and Standards Institute (SASI), a nonprofit organization that serves as a quality accreditation organization for clinical research sites. Click here to learn more about how this analysis was performed.

Growing a site can be challenging, and for many sites, passing a certain revenue threshold is critical to achieving scale and profitability. Our metrics paint a picture of what a “small” (1st quartile), “medium” (median) and “large” (3rd quartile) site looks like.

Top quartile sites are over $2M in annualized revenue, while the median site is roughly $800K

Let’s start with the most critical metric: annualized revenue. In our database, a 1st quartile site has about $300K or less of annual revenue – just enough to cover a few dedicated FTEs. These sites might be newer ones, or sites that function as an adjunctive operation within a larger health care practice. A median site has about $800K of annual revenue – enough to cover a team of 5 or 6 dedicated employees, and turn a small profit. A 3rd quartile site has $2.5M or more of annual revenue – large enough to cover 15-20 employees, earn profit, and position itself for potential acquisition. As we’ve described here in one of our e-books covering the site consolidation trend, investors usually look for $5M in revenue or more for “platform” investments, but can do $2M or more for “tuck-ins”.

Higher-earning sites earn more through enrollment, not necessarily through higher rates

The question is how the top quartile sites earn more. To break this down, we calculated the average revenue of a bottom-quartile site (ie, below 1st quartile), a second/third quartile site, and a top-quartile site (ie, above 3rd quartile), and then, holding these cohorts constant, the average weekly patient visit count and resulting revenue per visit metric. The following is the result:

The key difference for a top quartile site is that they see more patient visits per week – 92 vs. 18, on average. They earn only 5% more per visit than a middle-quartile site ($1395 vs. $1325). Bottom quartile sites see 3 patient visits per week, and earn just $713 per visit, likely reflecting either lower-paying studies or inability to negotiate effectively due to lack of a track record.

Overall, site enrollment varies significantly across sites

Looking at randomizations per month, the top quartile site averages 11 randomizations per month (almost 3 per week), while medium sites average 4 per month (roughly 1 per week), and small sites average 2 per month. Enrollment is highly correlated with overall patient visit volume:

• Small sites conduct 5 visits per week and randomize roughly 0.5 patients
• Medium sites conduct 12 visits per week and randomize roughly 1 patient
• Large sites conduct 38 visits per week and randomize roughly 3 patients

Note the ratios are pretty consistent across sites – for every 10-12 visits, a site can expect 1 randomization to occur. This suggests that most studies are at least 10-12 visits per subject, if not longer to compensate for the subjects that drop out. For our purposes, a “visit” includes not just the Scheduled Visits per the protocol, but also any prescreening visits or “sub visits” a site may have created in their template (e.g., a site may split out the X ray portion of V1 into a separate stand-alone V1a). (It does exclude Unscheduled visits).

On average, CRIO sites achieve 80% of their target enrollment goal; higher producing ones achieve 125%, and lower producing ones achieve 40%. This wide range in enrollment performance aligns well with what is consistently observed across trials – that a small minority of sites often account for a significant chunk of enrollment, and a good sized tail of sites account for zero.

Revenue per visit can vary significantly

At the same time, 3rd quartile sites on average earn total revenue of $2060 per visit completed (Note: this revenue includes study level invoices), while medium sites average about $1310, and smaller sites only $830. The larger per-visit revenue could reflect a variety of factors, including (1) a more lucrative therapeutic mix, (2) the ability to negotiate higher budgets, or (3) more complete collection of revenue earned. The first factor may be outside a site’s control, but the second and third, in our experience, could easily account for 30% of the revenue difference. Often, the sites with higher enrollment can negotiate higher budgets – and sites that are thorough in chasing down revenue can unlock significant hidden revenue (see here for a case study). That said, it’s likely that most of the discrepancy observed in this metric is the result of the types of studies that these sites do.

Large sites have more studies, and put more personnel on their studies

Large sites have 18 active studies going on at once, of which 10 are enrolling and 8 are in “maintenance” stage (active patients but enrollment closed). Medium sites have 8 and 5 studies respectively, while small sites have 4 and 2 respectively.

Large sites will have more personnel, and that may explain why the third-quartile number of users assigned to a study is 10 – i.e., 10 CRIO users with non-Investigator (data collection or view-only) rights on a study. For investigators, the third quartile number comes to 4. By contrast, small sites average about 4 Coordinators per study, and 2 Investigators per study. Clearly, even for smaller sites and studies, redundancy is required.

Back to Blog