Clinical trial recruitment is usually lost in the first hour after the referral form, not in the ad budget or the protocol. The referral arrives at 4:40 on a Friday, the coordinator is in a visit, and by Monday the person who raised their hand has stopped answering. Speed is the cheapest lever a site has and the one almost nobody measures. This post is about that hour: what the research says about it, how to see your own number, and what to do about it.
In short
- Tufts CSDD (2013): across more than 150 studies and nearly 16,000 sites, 11% of sites enrolled nobody and 37% under-enrolled.
- MIT and InsideSales.com (2011): odds of qualifying a lead were about 21 times higher inside five minutes than inside thirty.
- Harvard Business Review (2011): across 2,241 companies audited, the average first response to an inbound enquiry was 42 hours.
- SCRS (2023): sites report patient-facing staff turnover of 35% to 61% a year, which is why "the coordinator will call back" keeps failing.
Why clinical trial recruitment fails at the site, not the campaign
The number every site director knows is the Tufts one. In 2013 the Tufts Center for the Study of Drug Development analysed patient recruitment across more than 150 clinical studies and nearly 16,000 sites. Eleven per cent of sites in a typical trial never enrolled a single patient. Thirty-seven per cent enrolled fewer than they had committed to. Only 39% hit their target and 13% exceeded it. The trials themselves mostly got there in the end, at the cost of timelines that ran to nearly double the original plan.
What that report does not say, and what a site can find out for itself in an afternoon, is where its own referrals went. Pull the last thirty. For each one write down when it arrived, when a person first made contact, and what happened. Most sites we talk to have never done this, and the pattern is the same every time: the referrals that were called within the hour mostly turned into screening visits, and the ones that waited past a day mostly turned into nothing.
What the response-time research actually shows
The clearest work on response time was not done in medicine. It was done on sales enquiries, and it transfers because the mechanism is human, not commercial.
In 2011 a Harvard Business Review audit of 2,241 US companies found the average first response to an inbound enquiry was 42 hours, and 23% never responded at all. The same year, a study by MIT and InsideSales.com of several million contact attempts found the odds of qualifying a lead were roughly twenty-one times higher when the first contact came inside five minutes than inside thirty. After an hour the odds fell off a cliff.
A trial referral is the same shape as those enquiries. Somebody saw a Facebook ad for a migraine study, or their physician mentioned a trial, or they filled in a form on ClinicalTrials.gov, and in that moment they were ready to talk. An hour later they are at work, or picking up the kids, or wondering whether it was a good idea. Two days later the form is one of a dozen things they started and did not finish.
How to fix clinical trial recruitment response time at your site
- Measure time to first contact for the last thirty referrals. Arrival time is on the form, the email or the portal. First contact is in the phone log. Do not estimate. The median is your number.
- Put every referral channel into one queue. The study website, the ClinicalTrials.gov contact, the recruitment vendor's portal, the physician fax, the phone. If a referral can arrive somewhere nobody is watching, one will, at 4:40 on a Friday.
- Reply the same minute, even if the reply is a question. A text that says "Thanks for your interest in the migraine study. Is now a good time for four quick questions?" holds the person while a coordinator is in a visit. The reply does not need to be the prescreen. It needs to exist.
- Run the prescreen exactly as the IRB approved it. Same questions, same wording, same order, whether a person or a system is asking. Record the answers. Do not add a helpful extra question; that is a protocol deviation waiting to be found.
- Book the screening visit in the same conversation. A candidate who has just passed the prescreen and is told "someone will call you to schedule" has been handed back to the 42-hour average. Offer two times, take one, send the confirmation.
- Report the funnel per campaign. Time to first contact, contact rate, prescreen completion, show rate. The ad that produced 80 referrals and four screening visits is worse than the physician who sent six and five came in. And once they are in, keeping them is decided one visit at a time.
The mistake most sites make at step 3
Waiting until they can do the whole thing properly. The coordinator wants to make the first call personally, with the protocol open and twenty minutes clear, and that instinct is right about the prescreen and wrong about the first contact. The first contact is not a clinical conversation. It is an acknowledgement that a human being has been heard, and its whole value is that it happens now.
The second mistake is assuming the problem is staff who do not care. It is staff who are not there. The Society for Clinical Research Sites' 2023 workforce white paper reports patient-facing turnover at sites of 35% to 61% a year, up from a typical 10% to 37% before the pandemic, and puts the recovery time after a coordinator leaves a study at six to twelve months. A recruitment process that depends on one person remembering to call back will break every time that person changes.
Michigan, specifically
Metro Detroit has a dense research footprint for its size: the academic centres in Ann Arbor and Detroit, the hospital systems across Oakland, Wayne and Macomb, and a long tail of independent sites in suburban office parks. The candidate who sees your ad is usually within a short drive of three other sites running a study in the same indication. When two of them call back inside the hour and one calls back Tuesday, the candidate does not weigh the protocols. They go with the one who answered.
What we build for this
Our speed-to-lead screening and scheduling system for research sites does the first hour. It answers a new referral within seconds by text or voice, runs the IRB-approved prescreen verbatim and records every answer, books qualified candidates into the site's own scheduling system with a confirmation and reminders, and reports time to first contact, contact rate, prescreen completion and show rate for every campaign. It runs under a business associate agreement. It never decides eligibility; the study team does, at the screening visit, against the protocol.
Pull your last thirty referrals first. If the median time to first contact is under an hour, you do not need us for this. If it is not, talk to us.
