MelloJam Listen now

7 Clinic No-Show Statistics and Reminder Results Every Practice Should Know in 2026

No-show statistics are easy to quote and easy to misuse. A patient who cancels two days before an appointment is not the same as a patient who never arrives, and neither outcome has the same effect as a released slot that is filled by someone else. Reminder studies also measure different outcomes: attendance, no-show, cancellation, rescheduling, reattendance, or revenue.

We read the original behavioral intervention review, the targeted primary-care phone trial, the ophthalmology reattendance trial, and the digital notification review. We extracted the most useful control and intervention figures, preserved each study's denominator and follow-up window, and translated the evidence into local capacity and revenue formulas.

Back to Clinic statistics

1. No-shows, cancellations, reschedules, and unfilled slots are different statistics

A no-show is a patient who does not present for a scheduled appointment and did not cancel ahead of time when cancellation was possible. A cancellation is an appointment ended before the scheduled start, and a late cancellation is a cancellation inside a clinic-defined threshold such as 24 or 48 hours. A reschedule moves the patient to another appointment, while an arrival or kept appointment means the patient attended. A clinic-bumped appointment should be tracked separately from a patient-initiated cancellation.

The denominator changes the result. In the targeted primary-care trial, the no-show rate was calculated as no-shows divided by no-shows plus arrived appointments. The trial used all appointment statuses for its other rates, defining all appointments as arrivals, cancellations, reschedules, and no-shows. A dashboard that uses all scheduled appointments as its denominator will therefore not be directly comparable with the trial's primary no-show rate.

The evidence table keeps those distinctions visible. It also shows why there is no universal outpatient no-show benchmark: the studies cover different specialties, risk groups, appointment types, reminder channels, and follow-up windows.

Study Setting Appointment type Sample Baseline or control rate Intervention Outcome rate Follow-up Limitation
Behavioral intervention review Ambulatory and outpatient services across multiple countries Mixed outpatient appointments 61 studies in final analysis; 56 evaluated reminders No single pooled baseline rate in the review summary SMS, telephone, mail, incentives, and other behavioral approaches Evidence supported reminders, but delivery methods had mixed results Varied by study Most studies tested reminders, leaving other behavioral approaches under-studied
Digital notification review Healthcare settings in Europe, Asia, Africa, Australia, and America Prescheduled appointments 26 articles; 21 in primary meta-analysis; 8,345 notification patients and 7,731 controls Control attendance 54%; control no-show 21% Electronic text notifications, mostly SMS Attendance 67% and no-show 15% in intervention groups Varied by study Search covered 2005 through April 2015 and pooled heterogeneous settings
Targeted primary-care phone trial Academic hospital-based primary-care clinic Appointments for patients predicted to have at least 15% no-show risk within 7 days 2,247 enrolled; 1,129 intervention and 1,118 control No-show 29.2% in control Coordinator phone call 7 days before the visit plus usual automated call No-show 22.8% in intervention; cancellations 13.1% versus 11.5%; reschedules 14.2% versus 12.2% Appointment outcome; cancellation lead time Single center, high-risk population, and only 72.9% received every planned call
Ophthalmology reattendance trial Academic ophthalmology department Adult return visits after a no-show; active patient portal required 362 no-show patients; 189 intervention and 173 control Attendance within 30 days was 11.6% in control; health-system average no-show rate was 18.8% Patient portal message within one business day plus standard mailed letter, versus mailed letter alone Attendance within 30 days 22.2% versus 11.6%; rescheduling within 30 days 37.0% versus 22.5% 30 days after the missed visit Single academic eye center; new visits and patients without an active portal were excluded

The table is a comparison map, not a ranking of clinics. A rate from a high-risk primary-care cohort should not be used as the expected rate for every outpatient practice, and a 30-day reattendance result should not be described as a reduction in the original no-show rate.

2. A 61-study review found strong reminder evidence but little evidence for alternatives

The 2023 behavioral economics systematic review started with 1,225 records and included 61 studies after screening and risk-of-bias assessment. Every included study evaluated ambulatory or outpatient care. The most common intervention was a reminder, appearing in 56 studies. The review included randomized controlled, non-randomized experimental, and other controlled or quasi-experimental designs that measured an objective behavior rather than an attitude.

The evidence supports reminders delivered by SMS, telephone, or mail across different settings, but the review did not identify one universally best delivery method. The review found mixed results for the way reminders were delivered and very little evidence for many other behavioral mechanisms. Almost all studies used non-financial interventions, while only two studies examined incentives or disincentives.

The practical conclusion is narrower than "reminders solve no-shows." A clinic should test a defined intervention against a defined control, record the appointment type and timing, and measure both attendance and the alternatives that patients choose. The review provides evidence that reminders are worth testing, not a universal percentage to apply to a new clinic.

3. A digital reminder meta-analysis found attendance of 67% versus 54% and no-shows of 15% versus 21%

The digital notification review included 26 articles, with 21 studies and 8,345 notification recipients versus 7,731 no-notification controls in the primary meta-analysis. Pooled attendance was 67% for electronic text notifications and 54% for controls. The pooled risk ratio was 1.23, with a 95% confidence interval of 1.10 to 1.38. The pooled no-show rate was 15% with notifications and 21% without them, with a risk ratio of 0.75 and a 95% confidence interval of 0.68 to 0.82.

The pooled risk difference was 13 percentage points for attendance and 5 percentage points for no-shows. The no-show result had lower heterogeneity than attendance, but the studies still covered different settings and message programs. The pooled cancellation rates were 11% with notifications and 8% in controls, and the difference was not statistically significant. This is a useful warning: an intervention can reduce no-shows without creating a statistically detectable change in cancellations.

Message frequency also mattered for one outcome. In meta-regression, multiple notifications increased attendance by 25% compared with 6% for one notification, but multiple notifications did not significantly reduce no-shows. The review's search ended in 2015, so these results are a historical evidence base for digital reminders rather than a current benchmark for a specific patient portal or texting platform.

4. A targeted phone call reduced high-risk primary-care no-shows from 29.2% to 22.8%

The randomized trial enrolled 2,247 primary-care patients who had a predicted no-show risk of at least 15% for an appointment in the next 7 days. The intervention group was placed in a calling queue seven days before the appointment and received a phone call from a trained patient service coordinator. Both groups also received the clinic's usual automated call three days before the appointment. The coordinator used concrete planning questions rather than only repeating the date and time.

No-shows were 22.8% in the targeted-call group and 29.2% in the control group, an absolute difference of 6.4 percentage points with p less than 0.001. The trial authors described this as a 22% relative reduction in no-shows. Arrival rates were 54.3% versus 51.9%, cancellations were 13.1% versus 11.5%, and reschedules were 14.2% versus 12.2%; none of those differences was statistically significant.

The intervention did change when patients cancelled or rescheduled. Those events occurred an average of 0.35 days earlier in the phone-call group, with a 95% confidence interval of 0.07 to 0.64 days and p = 0.01. That earlier notice may be operationally valuable even when the cancellation rate itself does not change, because a clinic can have more time to release and refill the slot.

5. An ophthalmology portal message doubled 30-day reattendance after a no-show

The ophthalmology randomized trial studied 362 adult return patients who had missed an appointment and had an active patient portal. The control group received the department's standard mailed letter within two weeks. The intervention group received the same letter plus an electronic health record message within one business day of the missed visit, asking the patient to reschedule.

Within 30 days, 22.2% of the intervention group attended a follow-up appointment compared with 11.6% of controls. The odds ratio for attendance was 2.186, with a 95% confidence interval of 1.225 to 3.898 and p = 0.008. Rescheduling within 30 days was 37.0% in the intervention group versus 22.5% in the control group, with an odds ratio of 2.021 and p = 0.003.

The same study reported an average no-show rate of 18.8% across scheduled visits in its health system. That is a useful ophthalmology health-system benchmark, not a universal clinic rate and not the trial's control rate. The study excluded new patients and patients without an active portal, and it took place at one academic eye center. The higher 28.4% attendance among the 74 intervention patients whose messages were recorded as read is an exposure subgroup, not a randomized comparison, so it should not be interpreted as a separate causal effect.

6. Earlier cancellations can recover capacity even when cancellation rates do not fall

The digital notification meta-analysis found no significant reduction in cancellations, while the targeted primary-care trial also found no significant difference in cancellation or reschedule rates. Yet the primary-care phone intervention moved cancellations and reschedules 0.35 days earlier. These results point to a more useful operational question than "Did the reminder reduce cancellations?": did it create enough notice to make the released appointment usable?

That question requires a slot-level outcome. A cancellation two days before a 60-minute appointment may be recovered, while a cancellation 20 minutes before the start may leave the slot empty. A no-show may be impossible to refill after the patient fails to arrive, but a reminder may still convert some no-shows into earlier cancellations or reschedules. Record the event time and the time the slot was released instead of treating every non-arrival as the same capacity loss.

The original studies do not establish a universal recovered-slot rate. A clinic should calculate it from its own scheduling and fill data, then report it alongside the patient-level outcome. This keeps a statistically unchanged cancellation rate from hiding a meaningful change in how early the clinic learns about the cancelled visit.

7. A local dashboard should convert attendance events into recovered slots, hours, and revenue at risk

Use separate fields for appointment status, who initiated the change, the time of the change, the original duration, and whether another patient filled the slot. Report results by specialty, appointment type, new versus returning patient, lead time, provider, location, daypart, and reminder channel. A single clinic-wide no-show rate can hide a large difference between a short routine follow-up and a long procedure slot.

The core calculations are:

The primary-care trial reported 1.97 work RVUs per intervention patient versus 1.84 in controls, a difference that was not statistically significant at p = 0.10. Its sensitivity analyses estimated 8.8% higher per-patient billing using total billing and 7.8% higher modeled revenue using standard Medicare charges, but the authors did not track the fixed cost of building the intervention. Those figures are evidence about one study's financial modeling, not a revenue multiplier for every clinic.

Sources