Patient retention in clinical trials is decided one visit at a time, and the participant who withdraws almost always missed a visit first. Dropout in phase 3 trials is often substantial, sometimes more than 30%, as Alexander's 2013 review in Pharmacy and Therapeutics put it, citing the National Academies' 2010 report on missing data, and most sites treat that number as weather. It is not. The missed visit is the earliest, cheapest signal a site gets, and the reminder research says what to do with it.
In short
- Alexander (2013), citing the National Academies (2010): phase 3 dropout is often very substantial, sometimes more than 30%.
- Cochrane (Gurol-Urganci et al., 2013): text-message reminders increased attendance at healthcare appointments compared with no reminder, risk ratio 1.10 across 3,547 participants.
- Junod Perron et al. (2013), 6,450 patients: text reminders and phone reminders produced similar missed-visit rates, 11.7% and 10.2%, at a fraction of the cost per message.
- Kaiser Permanente (2022): for high-risk primary care visits, a second reminder text cut no-shows by 7% and same-day cancellations by 6%.
Why a missed visit matters more than a withdrawal
A withdrawal is a form. A missed visit is a choice the participant has just made, usually for a reason they could tell you if anyone asked that day. Between the two is a window, often a week or two, in which the participant is still enrolled, still reachable and still persuadable. Sites that win on retention are the ones that live in that window, the same way sites that win on recruitment live in the first hour after a referral.
There is a second reason to care, and it is regulatory. When the FDA broke down its FY2024 clinical investigator observations by theme, the protocol compliance category included missed visits, missed assessments, missed labs, and visits, assessments and labs performed out of window. A visit that slips past its window is a protocol deviation whether or not the participant ever comes back, and it lands in the same file as the expiring documents in the regulatory binder. The reminder that keeps the participant is the same reminder that keeps the visit inside its window.
What the reminder research shows
The evidence on reminders comes from routine healthcare rather than trials, and it transfers, because a study visit is an appointment with a longer preparation list.
The 2013 Cochrane review by Gurol-Urganci and colleagues pooled four randomized trials with 3,547 participants and found that text-message reminders increased attendance compared with no reminder, with a risk ratio of 1.10. Text and phone reminders performed about the same, and the text cost less. A randomized trial at an academic primary care clinic in Geneva the same year, Junod Perron and colleagues with 6,450 patients, put numbers on the comparison: 11.7% missed with a text reminder against 10.2% with a phone call, close enough that the cheaper one wins for routine visits and the phone call is kept for the visits that matter most.
The most useful recent result is from Kaiser Permanente Washington in 2022. For primary care and mental health visits flagged as high no-show risk, adding a second text reminder, sent three days out on top of the one sent two days out, cut no-shows by a further 7% and same-day cancellations by 6%. One more message, targeted at the visits most likely to be missed, moved the number. That is the whole strategy in a sentence.
How to run visits so participants stay
- Count your own dropout rate on the last three closed studies. Withdrawals divided by enrolments, per study. The industry's "sometimes more than 30%" tells you nothing about your site; your number does.
- Remind every visit, twice, with the preparation. Three days out and the day before. Say what to bring, whether to fast, how long it will take, where to park. A reminder that only says "you have a visit" gets a participant who arrives having eaten breakfast before a fasting draw, which is a visit that has to be repeated.
- Ask for a confirmation and treat silence as a flag. A participant who has not replied by the afternoon before is the one most likely not to come. That name goes to a coordinator for a call that day, not a note for the morning.
- Call the same day after a miss, and reschedule inside the window. Not "we missed you, call us when you can." Two times offered, one taken, confirmation sent. The visit window is in the protocol; the reschedule has to land inside it.
- Keep one voice. The participant should hear from the same coordinator, by name, from consent to close-out. Every handover is a small reason to stop coming.
- Report no-show and reschedule by study, monthly. A study with a 20% miss rate has a burden problem to raise with the sponsor. A site with a 20% miss rate across every study has a process problem to fix at home.
The mistake most sites make at step 3
Sending the reminder and considering the job done. The reminder is the easy half. The hard half is noticing, the afternoon before, that a participant has not replied, and having a person whose job it is that afternoon to pick up the phone. Most sites have the reminder and not the person, and the missed visit is discovered when the exam room sits empty at 9:15.
The second mistake is sending reminders in wording that was never approved. Every message a participant receives about a study is study communication, and the IRB approved a version of it. A coordinator improvising a friendlier text is doing the right thing for the wrong file.
Michigan, specifically
A site in Oakland or Macomb County runs its winter visit schedule against I-75 and M-59 in January. The participant who was going to make the 8:00 draw from Rochester Hills is looking at a windshield at 7:10 and deciding. The reminder that went out the afternoon before, with a reply that said "if the roads are bad, text this number and we will move you to Thursday," is the difference between a rescheduled visit inside its window and a deviation. Michigan sites that build the weather into the reminder keep more participants in February than the ones that do not.
What we build for this
Our visit adherence messaging for research sites runs the schedule above. Reminders and confirmations by text, email or voice across the whole visit lifecycle, with visit-specific preparation; non-responses flagged to a named coordinator the same day; rescheduling outreach after a miss, inside the protocol window. It runs under a business associate agreement and uses only wording the IRB has approved. It never changes a visit window, and a coordinator sees every non-response.
Count the last three studies first. If your dropout rate is well under 30% and your no-show rate is single digits, keep doing what you are doing. If not, talk to us.
