93 times a day a child or teenager in England has decayed teeth taken out in hospital. The figure for every council

Every year the Office for Health Improvement and Disparities publishes the number of times children and young people were admitted to an NHS hospital in England to have teeth taken out. The latest edition, published on 17 February 2026, covers the financial year ending March 2025. It gives a figure for every council. This page sets out the national picture, puts every council's figure in one searchable chart and one free file, and tests what a council's figure does and does not tell you, because the differences between councils only loosely follow the differences in children's teeth found by the national school survey.

First, what is counted. A hospital episode here is one stay or day case under one consultant in which taking out teeth was the main procedure, for a patient aged 0 to 19. It counts operations, not children and not teeth: a child who has six teeth out in one visit is one episode. For decay means tooth decay was recorded as the primary diagnosis. Extractions for other reasons are most often for impacted teeth, teeth in the wrong position and problems with teeth coming through, and are more common among 10 to 19-year-olds. Children are counted in the council where they live, not where the hospital is. A rate is episodes for every 100,000 residents aged 0 to 19, so councils of different sizes can be compared.

1. 93 times a day, and still the most common reason a 5 to 9-year-old is admitted

Hospitals recorded 56,143 episodes of tooth extraction among under-20s in the year to March 2025. In 33,976 of them, 60.5%, the primary diagnosis was tooth decay. That is 93 a day. The source says it is likely that most of these episodes involved a general anaesthetic. More than half of the decay episodes, 19,347 or 57%, were for children aged 5 to 9. Among 5 to 9-year-olds, 86.5% of all extractions were for decay. Among 15 to 19-year-olds, 21.4% were.

Hospital tooth extractions among 0 to 19-year-olds in England, by year

The number for decay is 11% up on the year before, when it was 30,587, and has risen in three of the four years since the pandemic year, when it fell to 14,645. It is still 3% below 2019/20 and 20% below the 42,209 recorded in the year to March 2015, the highest in the series. Extractions for other reasons rose faster in the latest year, by 20%, to 22,167. The statisticians say the rise since 2021 is likely to be services continuing to recover after the pandemic. They do not say that children's teeth have got worse.

Tooth decay remains the most common reason for a hospital episode among children aged 5 to 9, as it was the year before, when there were 19,381 such episodes. The table is the source's own. It counts hospital episodes of every kind for this age group by the diagnosis recorded, so it is not limited to extractions. Each row is one diagnosis code, so acute tonsillitis and chronic diseases of the tonsils and adenoids are counted separately.

The source puts the cost to the NHS of these admissions at £51.2 million for decay-related extractions and £87.7 million for all tooth extractions among under-20s in the year. And it finds a steep gradient by neighbourhood. Children and young people in the most deprived fifth of neighbourhoods had 403 decay episodes for every 100,000, against 121 in the least deprived fifth, 3.3 times as many.

2. Every council: from fewer than 8 episodes to 1,046 for every 100,000

Search for a council to read its figures: episodes of extraction for decay in the year to March 2025, the rate, the year before, 2019/20, and extractions for any reason. District councils also show their county's figure. Every count below England level is rounded to the nearest 5 by the source, and any count under 8 is withheld.

Hospital tooth extractions for decay for every 100,000 residents aged 0 to 19, by council, year to March 2025

Among the 263 councils with a published rate, the middle one recorded 173 episodes for every 100,000 under-20s, well below England's 251, because a small number of councils record very high rates. 26 councils are above 500, twice England's rate. 32 are below 50, and another 31 recorded too few episodes to publish at all. The ten councils with the most episodes account for 22% of England's total and hold 9% of its under-20s.

Five of the ten are in Yorkshire and the Humber. Two, Stoke-on-Trent and Newcastle-under-Lyme, show why the list needs care: in 2019/20 they recorded 45 and 15 episodes, and in 2024/25 they recorded 545 and 195. A change of that size in five years, in two councils side by side, points to a change in what reaches the hospital record rather than a tenfold change in children's teeth.

The low end cannot be ranked council by council, because the numbers are too small. Councils with fewer than 25 episodes are shown together.

19 of the 31 councils with too few episodes to publish are in the East Midlands, and 30 of that region's 35 councils recorded fewer than 25. In that region extractions for other reasons are recorded at 127 for every 100,000, about 78% of England's rate, while extractions for decay are recorded at 73, 29% of England's. Birmingham, with 333,000 under-20s, recorded 95 episodes for decay.

3. Why a council's figure is not, on its own, a measure of its children's teeth

The statisticians who publish these figures urge caution themselves. Their commentary warns that the data "may be an underestimation" because "there are tooth extractions conducted by community dental services in hospital settings, and that this activity is not always included" in the hospital record. It adds: "It is possible that different clinical coding protocols are applied at some sites and this could explain some of the variation in different geographical areas". Their guidance on small numbers adds that even a zero is withheld, "because there may well be activity occurring" that the hospital system does not record.

So before any council's figure was treated as a measure of decay, we tested it four ways.

Against extractions for other reasons. If children in one region simply had far more operations of every kind recorded, both kinds of extraction would move together. They do not. Extractions for decay run from 73 for every 100,000 in the East Midlands to 504 in Yorkshire and the Humber, 7.0 times as many. Extractions for other reasons run from 105 in the West Midlands to 202 in the South West, about twice as many. Across councils the pattern is the same: the council a tenth of the way from the top has 10 times the decay rate of the council a tenth of the way from the bottom, and 2.3 times the rate for other reasons.

Extractions for decay differ sevenfold between regions. Extractions for other reasons differ twofold

Against decay measured at school. The national survey of 5-year-olds examines children's teeth in school, so it does not depend on which hospital or clinic treats them. If the hospital figure tracked decay, councils with more decay in the survey would have higher hospital rates. Across the 202 councils with both figures, the link is weak: a rank correlation of 0.27, where 0 is no relationship and 1 is a perfect one. Once each council is compared only with others in its own region, the link disappears (a correlation of -0.04).

Councils with the same level of decay at age 5 record very different hospital extraction rates

Against deprivation. Decay follows deprivation closely: across councils, the school survey has a rank correlation of 0.70 with the 2025 deprivation index. The hospital figure has 0.25. Neighbourhood by neighbourhood, the source finds a decay rate 3.3 times as high in the most deprived fifth as in the least, so across England as a whole the deprivation gradient is plain in the hospital record. Between councils, something else outweighs it.

Against the same council a year earlier. The differences are not noise. A council's rate this year is very close to its rate last year (rank correlation 0.93), and seven of this year's ten highest were in last year's ten. About 37% of the variation in councils' rates is variation between regions. Stable differences that follow regions and not decay are what you would expect if they come from how local services are organised and recorded.

None of this means the high figures are wrong. Where a council's rate is high, that many episodes took place. What the data cannot say is whether a council with a low rate has fewer children losing teeth to decay, or as many whose operations are not in the hospital record.

4. The year's rise is concentrated in a few places

England recorded 3,389 more decay episodes than the year before. Four councils, Leeds, Cumberland, Shropshire and Telford and Wrekin, recorded about 1,210 more between them, on their rounded counts. That is equal to about 36% of the national rise, from 2.8% of England's under-20s. Leeds went from 610 to 1,065. Cumberland went from 20 to 320. Shropshire and Telford and Wrekin, which share an integrated care board, both roughly doubled. The source does not say why. A rise of that size in twelve months is more likely to reflect how many operations were carried out or recorded than a sudden change in children's teeth, and the statisticians say the national rise since 2021 is likely to be services continuing to recover.

Of the 249 councils with a published figure in both years, 29 have a rate that rose by more than chance would explain and 11 a rate that fell. For the other 209, the change is within what small numbers produce on their own. Against 2019/20, 38 of 238 comparable councils have risen and 55 have fallen. The largest falls in episodes are in Bradford (1,080 to 720), Liverpool (740 to 485) and Doncaster (855 to 630).

Episodes rose in eight of the nine regions in the year. Over five years they are higher in the North East, the West Midlands, the South East and the East of England and lower in the rest. By integrated care board, the bodies that commission this care, the recorded rate runs from 21 for every 100,000 in Leicester, Leicestershire and Rutland to 964 in South Yorkshire. All 42 boards are in the file.

What to do about it

If you report on a council or represent one, find it in the chart above or in the council file. Quote the count as hospital episodes, not children, and the rate against England's 251 for every 100,000. Before calling a low figure good news or a jump a crisis, put two questions to the integrated care board: which service takes out children's teeth under general anaesthetic for this area, and is that work in the hospital figures. The council's school survey figure, in the same file, is the independent check on decay itself.

If you scrutinise local health services, the figure counts operations done. It says nothing about how long children waited or how many are waiting now, which the source does not publish. A council whose figure has jumped is worth asking about for that reason: more operations in a year can mean a list being cleared.

If your practice sees children, more than half of these episodes are for children aged 5 to 9. If you have room for children, say so plainly where families look: on your NHS website profile, which lists children separately from adults, on your own site and on your Google listing. The NHS's own advice for families is on its children's teeth page, and it is the one to link.

If you quote a figure in a funding bid or a campaign, England's is the safer one, because it does not turn on how one area records its work. Name the year.

How we did this

Source. Office for Health Improvement and Disparities, Hospital tooth extractions in 0 to 19 year olds 2025, official statistics published on 17 February 2026: the data tables (version 1.1) and the short statistical commentary. The figures come from Hospital Episode Statistics, the record of admitted patient care in NHS hospitals in England. The year before is from the 2024 edition, and 2019/20 from the 2021 edition, which carries that year by council. All are published under the Open Government Licence v3.0.

What is counted. Finished consultant episodes for patients aged 0 to 19 at the start of the episode, as ordinary admissions or day cases, where the main procedure is a tooth extraction (procedure codes beginning F09 or F10). An episode is "for decay" where the primary diagnosis is one of the source's eight codes: four for dental caries, and pulpitis, chronic apical periodontitis and two for periapical abscess. Outpatient appointments are not included, and nor is anything done in a dental practice or a community clinic. Patients are placed by the council they live in.

Rates. The rates are the source's own: episodes divided by the Office for National Statistics' mid-2024 estimate of residents aged 0 to 19, times 100,000. They are crude rates, not adjusted for age. The 2023/24 rates use mid-2023 estimates and the 2019/20 rates use the mid-2019 estimates published at the time, which were later revised after the 2021 census, so a five-year change in a rate is less exact than a one-year change.

Rounding and withheld figures. Below England level the source rounds every count to the nearest 5 and withholds any count under 8, zeros included. We have not estimated any withheld figure, by subtraction or any other way. In the file a withheld count reads "under 8" and its rate is blank. Where a council's count for decay is withheld, the file does not show its count for other reasons either, because taking that from the count for any reason would narrow the withheld figure down. Councils with fewer than 25 episodes are never ranked one by one. The published council counts add up to 98.7% of England's total. The rest is withheld, lost to rounding or has no council recorded.

Chance. The source gives no confidence intervals. To show how much a small count can move by chance, the file carries an approximate 95% interval for each council's rate, calculated by us with Byar's method on the published count and widened by 2 episodes either way for the rounding. On that basis 76 councils are above England's rate, 141 are below and 46 are not clearly different. A change between two years is called a rise or a fall only where the two rates differ by more than chance after the same allowance. The interval covers chance only. It says nothing about recording.

Councils. 294 unitary, metropolitan, London borough and district councils on April 2023 boundaries, and the 21 county councils, which have rows of their own in the file. The 2019/20 figures are on the boundaries of the time, so the six councils created since then (Cumberland, Westmorland and Furness, North Yorkshire, Somerset, North Northamptonshire and West Northamptonshire) have no 2019/20 figure. Each district's county is from the ONS district to county lookup.

Controls we ran. Each is described in section 3. The decay figure at age 5 is OHID Fingertips indicator 93563 for 2023/24, and the deprivation score is the 2025 English Indices of Deprivation averaged by council, both as published in our council study of children's decay. Using only the hospital rate for ages 5 to 9, the age nearest the survey, gives a rank correlation of 0.24 across 171 councils, no stronger. Against 2019/20, a council's rate has a rank correlation of 0.72 with its rate now.

What this is not. It is not a count of children, of teeth, or of general anaesthetics. It is not a waiting list. It does not include private treatment. And it is not a measure of how well any council, hospital, dental service or family looks after children's teeth: a low figure may mean the work is done elsewhere, and a high one may mean it is recorded in full.

Questions we get asked about this data

How many children have teeth removed in hospital because of decay?

NHS hospitals in England recorded 33,976 episodes of tooth extraction for decay among 0 to 19-year-olds in the year to March 2025, 93 a day. That counts operations, so a child admitted twice is counted twice, and the source says it may be an underestimate because some extractions by community dental services are not in the hospital record.

Does a low figure mean children in my council have healthy teeth?

Not on its own. Across councils the hospital figure has only a weak link with the decay found in 5-year-olds by the school survey. A low figure can mean fewer children needing teeth out, or that the operations are done by a service whose work is not in the hospital record. The school survey figure for your council is the better guide to decay.

Why is there no figure for my council?

The source withholds any count under 8, including zero, and 31 councils fall below that for decay in 2024/25. Search for the council in the chart in section 2 to see its county's figure and its count of extractions for any reason.

Why did my council's figure change so much in a year?

Small counts move by chance, and for 209 of 249 councils the change since 2023/24 is within that. Where a figure has doubled or more, that is more likely to be more operating lists being run, or a change in what is recorded, than a change in children's teeth. The source does not give reasons for individual councils.

For editors

The data is free to reuse with attribution to Practice Radar, under Creative Commons BY 4.0. Source: Office for Health Improvement and Disparities, Hospital tooth extractions in 0 to 19 year olds, 2025 edition (financial year ending March 2025), 2024 edition and 2021 edition, from Hospital Episode Statistics, under the Open Government Licence v3.0. The council file, children-hospital-tooth-extractions-by-council-2026.csv, has one row per council with its ONS code, and rows for each integrated care board, each region and England. Please carry the source's caveat with any council figure: extractions carried out by community dental services are not always in the hospital record.

Jack Henty, who compiled this analysis, co-owns Bollington Dental Practice in Cheshire and founded Practice Radar, which measures local dental markets for practices. He is available for interview.

Contact: hello@practiceradar.co.uk