How to reduce patient no-shows: an evidence-based guide

Most clinics can reduce patient no-shows, but not by as much as the marketing material suggests, and not with any single measure. The best evidence available says the realistic lever set is narrow: reminders raise attendance by about 14% in relative terms — a risk ratio of 1.14 (Cochrane, 2013) — the wording of the reminder is worth another two to three percentage points (Hallsworth et al., PLOS ONE, 2015), shortening the gap between booking and visit helps, and open-access scheduling produced a mean reduction in no-show rates of 2.78% across the studies that measured it (Ansell et al., BMC Health Services Research, 2017). Stacked and applied consistently, those add up to a real improvement. Applied one at a time, they mostly disappoint.

This guide is written for clinicians and clinic owners who want to know what the published evidence actually supports — the honest baseline rates, the effect sizes, the limits, and the one finding that should change how you think about repeat non-attenders altogether. Every figure below is attributed and linked.

What is a normal patient no-show rate?

In general practice, expect something in the low-to-mid teens. A 2021 systematic review in the British Journal of General Practice pooled 26 studies and found a mean missed-appointment rate of 15.2% with a median of 12.9%, and reported that rates were broadly similar between countries. Anything consistently above the high teens is worth investigating as a system problem.

Spread matters more than the average, because the average hides specialty and setting effects:

• General practice: mean 15.2%, median 12.9% across 26 studies (Parsons et al., BJGP, 2021).
• Secondary and subspecialty clinics: a 12-year analysis of 10 clinics in a large US Veterans Affairs medical centre found a mean no-show rate of 18.8% (SD 2.4%), ranging from 25.7% in gastroenterology to 12.6% in audiology (Kheirkhah et al., BMC Health Services Research, 2016).
• A well-run system can sit far lower: a rural US health system analysing 1,260,083 appointments across 263,464 patients recorded an overall no-show rate of 6.0% (Shour et al., BMC Health Services Research, 2023).
• Nationally, at scale: around 5.5 million NHS outpatient appointments were missed in England in 2012–13 — 9.3% of the total (Hallsworth et al., PLOS ONE, 2015).
• The published range is enormous: the VA authors noted earlier studies reporting rates as low as 3% and as high as 80%, which is why a borrowed benchmark is close to useless.

The practical conclusion: stop benchmarking against an industry average you read in a vendor blog, and start benchmarking against your own clinic last quarter, broken down by clinician, session, day of week and appointment type. That single change in measurement usually reveals where the losses actually are.

The cost side is also easier to quantify than most clinics assume. The VA analysis estimated the average cost of a no-show at $196 per missed appointment in 2008, derived from the full cost per encounter on the reasoning that staff costs remain whether or not the patient arrives. Your own figure will differ, but the method — cost per encounter multiplied by unfilled slots — is reproducible in an afternoon.

Why do patients miss appointments?

Mostly for practical reasons, not indifference. The BJGP review found the most commonly reported reasons were work or family and childcare commitments, forgetting the appointment, and transport or weather difficulties. Patients most likely to miss an appointment were younger (under 21), those of minority ethnicity, and those of lower socioeconomic status.

Two scheduling factors are within your control and show up clearly in the data. In the rural US analysis, appointments booked more than 60 days in advance had a 7.7% no-show rate, and patients aged 21–30 had the highest rate at 11.8% against 2.9% for those over 60. Lead time is a design choice; age is not. That asymmetry is where the actionable work is.

It also explains why reminders alone underperform. A reminder solves exactly one of the reported causes — forgetting. It does nothing about a shift that cannot be swapped, a child with no cover, or a bus route that does not run. If your reminder programme has plateaued, the remaining non-attendance is probably not a memory problem. Language can be a barrier here too: a reminder a patient cannot read is not a reminder, which is part of the broader problem covered in our guide to language barriers in healthcare.

Do appointment reminders actually work?

Yes, but modestly. The 2013 Cochrane review of mobile phone messaging reminders found moderate-quality evidence from seven randomised trials (5,841 participants) that text reminders improved attendance compared with no reminders, with a risk ratio of 1.14 (95% CI 1.03 to 1.26). Pooled attendance was 67.8% with no reminder, 78.6% with text reminders and 80.3% with phone call reminders.

Three findings from that review are worth acting on directly:

• Text messages are about as effective as phone calls — three studies (2,509 participants) found a similar impact, risk ratio 0.99 (95% CI 0.95 to 1.02). You are not sacrificing effect by dropping the call.
• And they are far cheaper. In two included studies, the cost per text message per attendance was 55% and 65% lower than the cost per phone call reminder.
• Layering helps a little. One small study (291 participants) found text plus postal reminders beat postal alone, risk ratio 1.10 (95% CI 1.02 to 1.19) — low-quality evidence, so treat it as a hint rather than a plan.

The honest counterweight: reminders are not magic, and badly implemented ones barely register. The VA study found that introducing a centralised phone reminder moved the no-show rate only from 16.3% to 15.8%, and its authors concluded that the reminder system "only modestly reduced the no-show rate". The Cochrane authors were similarly cautious, judging the evidence low to moderate quality and insufficient to conclusively inform policy. Expect a real but bounded gain — and a far smaller one if the implementation is poor.

Does the wording of a reminder change the no-show rate?

More than most clinics expect, and at no extra cost. Two randomised controlled trials in an English NHS trust tested rephrased SMS reminders against the trust's existing message. Stating the cost of a missed appointment to the health system produced a did-not-attend rate of 8.4% versus 11.1% for the existing message (odds ratio 0.74, 95% CI 0.61–0.89).

The second trial replicated the effect at 8.2% — and added the more useful finding: expressing the same idea in general terms was significantly less effective (9.9%, odds ratio 1.22, 95% CI 1.00–1.48). Specificity is doing the work, not guilt. The figure the researchers used was not invented for effect either: they took the 2012–13 NHS tariff for initial and follow-up appointments across all five specialties involved, found a mean of £161, and presented it as an approximation. Because none of the new messages exceeded the 160-character SMS limit, the change cost nothing to make — and the authors estimated it would avoid about 5,800 missed appointments a year at that trust alone.

Can changing how you schedule reduce no-shows?

Yes, and this is the lever clinics most often skip. A 2017 systematic review of interventions to reduce primary-care wait times found open-access scheduling — keeping a substantial share of each day's capacity free for same-day booking — to be the most commonly used intervention. Of the four included studies that measured no-shows before and after, every one showed a decrease, with a mean reduction in no-show rates of 2.78%.

The mechanism is intuitive once you pair it with the lead-time data above. A patient who books this morning for this afternoon has almost no opportunity to forget, fall ill, change their mind or have the reason for the visit resolve itself. One of the included studies also reported that no-shows declined under open access because patients were more likely to be scheduled with their own family physician — continuity and attendance pull in the same direction.

Scheduling changes worth testing, roughly in order of evidence strength:

• Shorten the booking-to-visit gap wherever clinically safe; treat any routine appointment booked more than two months out as a flagged risk.
• Reserve same-day capacity rather than filling the book weeks ahead — the open-access principle, even partially applied.
• Protect continuity: book patients with their usual clinician where possible.
• Confirm and re-offer long-lead appointments a week or two out, which gives you time to refill the slot rather than discovering the gap on the day.
• Run a short-notice waiting list so a cancellation becomes a filled slot instead of an empty one.

Same-day flow is the other half of this. In clinics that run walk-in or token-based outpatient sessions rather than fixed appointment slots, the equivalent problem is the patient who checks in and then disappears before being called. A live queue that marks unanswered tokens and keeps the line moving — which is how the patient queue in Shifaa AI is designed to work, with the no-show rate tracked in analytics — converts that from a stalled session into a measured number you can act on. Shifaa AI is a coming-soon product; the point here is the metric, not the tool — and the wider shift it sits inside is covered in our look at how AI is changing outpatient practice.

Should you overbook or charge a fee for no-shows?

Overbooking has reasonable support as a way to protect clinic volume; no-show fees have much weaker evidence and real equity costs. A study of US family medicine residency practices identified as exemplars found that practices keeping no-show rates at 10% or below all used multiple strategies together — patient education, patient reminders, patient sanctions and some degree of open-access scheduling — and that practices managing no-shows well encouraged walk-ins and work-ins and overbooked schedules, either uniformly or based on individual no-show rates.

Note what that study does and does not say. It is a survey of exemplar practices, not a randomised comparison, so it tells you what successful practices do rather than proving which component caused the result. Its clearest signal is the one the authors emphasised: the practices that succeeded were diligent about following their own policies consistently. A no-show policy that is enforced inconsistently is roughly equivalent to no policy.

On fees specifically, be careful. The populations most likely to miss appointments in the BJGP review were younger, poorer and from minority ethnic groups — precisely the patients a financial penalty hits hardest, and the ones for whom missing care carries the most clinical risk. Blanket overbooking has its own cost too: it lengthens waiting times and overruns sessions when everyone turns up. Both tools work better aimed than applied flat.

Can AI predict which patients will miss their appointment?

It can rank risk usefully, but not reliably enough to act on individual patients punitively. A 2023 model built on 1,260,083 appointments at a rural US health system achieved an area under the curve of 0.84 on the training set and 0.83 on the test set, and the authors translated it into an operational rule: one overbook for every six at-risk appointments per provider per day.

The number nobody quotes is the one that matters most. At a cut-off of 0.4, the model's sensitivity was 0.71 and its positive predictive value was 0.18 — meaning that of every 100 appointments flagged as high-risk, roughly 82 patients actually turned up. That is still commercially useful for capacity planning, where you are playing averages across a session. It is nowhere near good enough to justify telling an individual patient they are a predicted no-show, denying them a prime slot, or charging them in advance.

So use risk scores the way they work: aggregate, not individual. Point them at outreach — an extra confirmation call, a transport conversation, a telephone or video alternative — rather than at penalties. That framing is the same one we apply to clinical AI generally: it is decision support, the clinician decides, and a confident-looking score is not a verdict. If that distinction interests you, our guide to clinical decision support systems covers where it holds and where it breaks.

Why is a repeated no-show a clinical safety signal?

Because it is a risk marker for premature death. A national retrospective data-linkage study of 824,374 patients in Scotland — about 15% of the population, followed over three years with a 16-month mortality follow-up — found that patients with long-term mental health conditions who missed more than two appointments a year had a greater than eight-fold increase in the risk of all-cause mortality compared with those who missed none, with a dose-based response as missed appointments rose.

The authors found that patients with more long-term conditions were more likely to miss appointments even after controlling for the number of appointments made, and that those who died prematurely commonly did so from non-natural external causes including suicide. Their conclusion was blunt: for these patients, existing primary care appointment systems are ineffective.

What does a practical plan look like?

The following sequence is illustrative — it stacks the measures above in rough order of evidence strength and effort, for a clinic starting from no formal programme. No study has tested this combination as a package, so treat the sequence as a structure rather than a forecast; what the evidence does show is that the family medicine practices holding no-show rates at 10% or below got there with the whole stack applied consistently, never with one measure. Judge progress over quarters, not weeks.

• Weeks 1–2 — measure properly. Define a no-show precisely (the VA study counted both non-attendance and same-day cancellations too late to refill). Report the rate monthly by clinician, session, weekday, appointment type and lead time. Calculate your own cost per missed slot from your cost per encounter.
• Weeks 3–4 — fix the reminder basics. Get valid mobile numbers (the UK trials reached only 20% of eligible patients, which caps any reminder programme hard). Send SMS as the default channel on cost grounds, and make confirming or rescheduling a one-step reply.
• Month 2 — rewrite the message. Make it specific rather than general, per the UK trials, and make any figure you quote true for your setting. Send in the patient's own language where you can.
• Month 2 — shorten the lead time. Pull routine bookings in, and confirm anything booked more than 60 days out a fortnight before the date.
• Month 3 — open some same-day capacity. Reserve a share of each session for same-day booking and protect continuity with the usual clinician where possible.
• Month 3 — run a short-notice list so cancellations become filled slots, and allow work-ins and walk-ins to absorb the gaps.
• Month 4 — target the repeat non-attenders by name. This is the clinical-safety list, not the billing list. Prioritise patients with long-term mental health conditions and multimorbidity, and ask what the barrier actually is.
• Ongoing — review the number monthly with the whole team, and only then consider risk-based overbooking, aimed at sessions rather than individuals.

One footnote on capacity: a recovered slot is only worth having if the visit it enables does not cost you the evening in documentation. That is the other half of clinic efficiency, and the reason the administrative burden numbers matter alongside the attendance numbers; both feed the same outcome discussed in our guide to physician burnout. If you are redesigning your session structure anyway, our free SOAP note template is a reasonable place to standardise what each recovered slot produces.

Frequently asked questions

What is a good no-show rate for a clinic?

There is no universal target, but a systematic review of 26 general-practice studies found a mean missed-appointment rate of 15.2% and a median of 12.9%, and rates were similar across countries. Published rates range from 3% to 80% depending on setting, so your own trend over time is a far more useful benchmark than any industry average.

How much do SMS appointment reminders reduce no-shows?

The 2013 Cochrane review found moderate-quality evidence that text reminders improved attendance versus no reminders, risk ratio 1.14 (95% CI 1.03 to 1.26), with pooled attendance of 78.6% against 67.8%. They performed about as well as phone calls while costing 55–65% less per attendance. Expect a useful but modest gain, not a fix.

Should clinics charge a fee for missed appointments?

The evidence is weak and the equity risk is real. Exemplar practices that achieved no-show rates of 10% or below did use sanctions, but always alongside education, reminders and open-access scheduling, and the study was observational. Because non-attendance concentrates in younger, poorer and minority ethnic patients, a flat fee penalises those at the highest clinical risk.

Does a patient who misses appointments need clinical follow-up?

Often, yes. In a Scottish study of 824,374 patients, those with long-term mental health conditions who missed more than two appointments a year had a greater than eight-fold increase in all-cause mortality risk, with a dose-response pattern. Repeated non-attendance is best treated as a risk marker warranting active outreach rather than discharge.

References

• Parsons J, Bryce C, Atherton H. Which patients miss appointments with general practice and the reasons why: a systematic review. British Journal of General Practice, 2021;71(707):e406-e412. bjgp.org/content/71/707/e406
• Gurol-Urganci I, de Jongh T, Vodopivec-Jamsek V, Atun R, Car J. Mobile phone messaging reminders for attendance at healthcare appointments. Cochrane Database of Systematic Reviews, 2013;(12):CD007458. pmc.ncbi.nlm.nih.gov/articles/PMC6485985
• Hallsworth M, Berry D, Sanders M, et al. Stating appointment costs in SMS reminders reduces missed hospital appointments: findings from two randomised controlled trials. PLOS ONE, 2015;10(9):e0137306. journals.plos.org/plosone/article?id=10.1371/journal.pone.0137306
• Kheirkhah P, Feng Q, Travis LM, Tavakoli-Tabasi S, Sharafkhaneh A. Prevalence, predictors and economic consequences of no-shows. BMC Health Services Research, 2016;16:13. pmc.ncbi.nlm.nih.gov/articles/PMC4714455
• Ansell D, Crispo JAG, Simard B, Bjerre LM. Interventions to reduce wait times for primary care appointments: a systematic review. BMC Health Services Research, 2017;17:295. pmc.ncbi.nlm.nih.gov/articles/PMC5397774
• Johnson BJ, Mold JW, Pontious JM. Reduction and management of no-shows by family medicine residency practice exemplars. Annals of Family Medicine, 2007;5(6):534-539. annfammed.org/content/5/6/534
• Shour AR, Jones GL, Anguzu R, Doi SA, Onitilo AA. Development of an evidence-based model for predicting patient, provider, and appointment factors that influence no-shows in a rural healthcare system. BMC Health Services Research, 2023;23:966. pmc.ncbi.nlm.nih.gov/articles/PMC10503036
• McQueenie R, Ellis DA, McConnachie A, Wilson P, Williamson AE. Morbidity, mortality and missed appointments in healthcare: a national retrospective data linkage study. BMC Medicine, 2019;17:2. pmc.ncbi.nlm.nih.gov/articles/PMC6329132