How often people miss appointments in the UK
The fullest UK figures come from the NHS, which records whether each appointment was kept.
- GP appointments: patients did not attend 16 million of the 376 million GP appointments in England in 2025, about 1 in 23, or 4.3%, according to NHS England.
- Hospital outpatient appointments: 8.2 million of 150.3 million were missed in the year to March 2026, 5.4%, down from 6.2% in 2019 to 2020, in NHS England's Hospital Outpatient Activity statistics.
In a Censuswide survey for NHS England of 2,000 people in England aged 16 or over, 12% said they had forgotten about an appointment in the past, and 11% that they had arrived too late.
Rates vary with the setting. A 2018 review of 105 studies in Health Policy found an average of about 23% in healthcare worldwide, from 13.2% in Oceania to 43.0% in Africa. The factors most often linked to no-shows were a long wait between booking and appointment, and having missed before.
We found no official UK figure for salons, physiotherapy or private dentists. Fresha, which sells booking software, surveyed over 200 UK beauty and wellness businesses in December 2025: hair and beauty owners estimated they lose almost 7% of monthly revenue to cancellations. That is vendor data, from owners' own estimates.
The studies at a glance
What each source tested, what it found, and how much weight it bears. Details and links follow.
| Source | What it tested | What it found | How strong |
|---|---|---|---|
| How common no-shows are | |||
| NHS England, 2025 and 2025 to 2026 | All GP and hospital outpatient appointments in England | 4.3% and 5.4% missed | Official figures, NHS |
| Health Policy review, 2018 | 105 healthcare studies | About 23% on average | Rates vary by setting |
| Reminders | |||
| BMJ Open review, 2016 | Text reminders against none, 21 studies pooled (16 comparisons for no-shows) | 25% fewer no-shows: 15% against 21% | Strong. All texts, all healthcare |
| Cochrane review, 2013 | Texts against none or a phone call, 8 trials | 78.6% attended after a text, 67.8% with none. Calls about the same | Moderate, on its own rating |
| Health Services Research review, 2012 | Text reminders, 18 studies | Attendance up. No clear difference between 24, 48 and 72 hours or more | Mixed: 8 trials, 10 other studies |
| Cochrane review, 2012 | Email reminders | No study met its standards | A gap in the evidence |
| PLOS ONE, 2015 | Text wording, two NHS trials in London of about 10,000 patients each | Stating the cost cut missed appointments from 11.1% to 8.4% | Strong, one NHS trust |
| Fines | |||
| BMJ Open trial, 2018 | A €34 fine, 6,746 Danish hospital appointments | About 5% missed with or without it | Strong, low starting rate |
| Israel Journal of Health Policy Research, 2023 | Charging for no-shows | Limited research, mixed results | A narrative review |
| Deposits | |||
| Resos No-Show Index, 2026 | 3.8 million restaurant bookings, 54% in the UK | 0.80% no-shows with a deposit or card hold, 2.47% without | Vendor data, not proof of cause |
Reminders: the strongest evidence
The largest pooled result is a 2016 systematic review in BMJ Open. It found 26 studies of text reminders for healthcare appointments, published from 2005 to 2015, and combined 21 of them: 8,345 people who got a reminder and 7,731 who did not.
- People sent a reminder were 25% less likely to miss their appointment: 15% did not turn up, against 21%.
- They were 23% more likely to attend: 67% against 54%.
- More than one reminder raised attendance more than a single reminder, but made no significant difference to no-shows.
A 2013 Cochrane review of eight trials with 6,615 people agreed: 78.6% attended after a text reminder, against 67.8% with none, and texts worked about as well as phone calls.
The limit: all of these studies were in healthcare, so the size of the effect for a salon or a physio is not known. The direction is consistent: reminders help.
Text or email?
Almost all of that evidence is for text messages: every study in the BMJ Open review used them. When Cochrane reviewed email reminders in 2012, it found no study that met its standards, so it could say nothing either way.
In the Resos No-Show Index, restaurant bookings that got a text reminder had a 16% lower recorded no-show rate. Resos, which sells restaurant booking software, calls this an observed association, not a measured effect.
That does not show that email reminders fail. It means the evidence is for text.
When to send a reminder, and what to say
Timing. A 2012 review of 18 studies in Health Services Research found no significant difference between text reminders sent 24, 48, or 72 or more hours before. In the BMJ Open review, most of the studies that said when sent theirs 48 hours or less before. The day before is a sensible default, not a proven ideal.
Wording. The clearest UK evidence is two randomised trials at Barts Health NHS Trust in London, published in PLOS ONE in 2015. Outpatients got a text five days ahead, worded one of four ways. In trial one, of 10,111 patients:
| What the text said | Missed |
|---|---|
| The usual: call the number on your letter | 11.1% |
| The number to call, in the text | 9.8% |
| That 9 out of 10 people attend | 10.0% |
| "Not attending costs NHS £160 approx." | 8.4% |
A second trial, of 9,848 patients, repeated it: 8.2% missed with the cost message, against 9.9% with a vaguer "wastes NHS money". The limits: one trust, and the cost was to the NHS, not the patient. Whether it works the same way for a small business has not been tested.
Deposits: what the evidence shows
We found no randomised trial of deposits for salons, clinics or other personal services. The 2023 review of charging for no-shows lists upfront payments as a method in use, but gives no results for them.
The most detailed figures come from booking software. The Resos No-Show Index covers 3,768,761 restaurant bookings at 2,417 restaurants from August 2025 to July 2026, 54% of them in the UK, with 2.33% no-shows overall:
- With a deposit or card hold: 0.80% (1,907 of 238,539).
- Without: 2.47% (85,589 of 3,461,164).
That is about a third of the rate, but Resos itself calls this an observed association, not a measured effect, because restaurants that ask for a deposit differ from those that do not. They may, for example, be busier, dearer or taking bigger tables.
A deposit has a second job that needs no research: when someone does not come, you keep what your cancellation policy says. Against that, asking for money up front may put some people off booking, and we found no good study of how many.
Fines: limited and mixed evidence
The strongest test of a fine is a randomised trial at a Danish hospital's orthopaedic clinic, published in BMJ Open in 2018. Half of 6,746 appointments came with a letter saying that missing it without notice would cost DKK 250, about €34.
- No-shows did not change: 130 with the fine, 131 without, about 5% of uncancelled appointments in each group.
- Collecting was hard: of the 130 fines issued, 27 were paid.
The authors suggest that at a rate of about 5%, the people who still miss often have reasons a fine does not touch, such as illness.
A 2023 review in the Israel Journal of Health Policy Research sums up the field: the research on fines is limited and its results are mixed. One study it cites, at a mental health clinic in 1995, saw no-shows fall from 20.1% to 9.27% after a $30 fee among patients who had missed repeatedly, but the review notes that such studies each covered one specialty, date from the mid-1990s and have had no recent follow-up. The authors expect small charges to make little difference.
What this means for a small business
- Count your own rate: missed bookings divided by all bookings, over a month. The NHS figures above are 4.3% and 5.4%, the restaurant data about 2%.
- Send a reminder, by text if you can, since that is where the evidence is. The day before is fine.
- Make it easy to cancel or change, so you hear in time to fill the slot. The 2023 review points this way, though it is not proven.
- Watch for repeat no-shows and long waits, the factors the 2018 review found most often.
- Think about a deposit where a no-show costs most, such as long appointments. It protects your income, though the evidence that it changes who turns up is booking-company data.
- Do not count on fines after the event. The one large trial found no effect, and most of its fines went unpaid.
How we checked
We read each paper through the journal or the Europe PMC archive, not articles quoting it: the full text of the two BMJ Open papers, the PLOS ONE trials and the 2023 review, and the abstracts of the Health Policy, Health Services Research and two Cochrane reviews. NHS figures are from NHS England's own pages, and vendor figures from each company's own page. We left out figures found in downloadable spreadsheets alone, and vendor figures with no published method. Rates from counts were worked out by script.
We have no connection with Resos or Fresha. subba sells booking software with reminders, so follow the links and judge for yourself.
Last checked on 12 November 2026. Corrections to [email protected].
Where subba fits
subba is an AI receptionist and booking system for small UK businesses, from £9.99 a month. It sends each customer one reminder about a day before their booking, with a link to change or cancel, by email, or by text if you add the texts add-on (from £9.99 a month for 250 texts) and choose text in your settings. A customer with no email address on their booking, such as a caller who booked by phone, gets no reminder unless reminders go by text. You can switch on a second reminder on the day, two or three hours before.
You can ask for a deposit (a set amount for each service) or the full price when people book. It goes to your own Stripe account and subba takes no fee on it (Stripe charges its usual card fee), and the cancellation policy you set decides what is refunded when a customer cancels. You can mark no-shows in the calendar, and the dashboard counts this month's. subba does not charge a card after a no-show, and its reminder wording cannot be changed. See the pricing page.