11% of the NHS dentistry commissioned in England last year was not delivered, down from 16%. Across the 42 integrated care boards it runs from 3% to 51%
Integrated care boards commissioned 80.4 million units of NHS dental activity from high street dental contracts in England in the year to March 2026. A unit is how the NHS counts a dentist's work: a check-up is one and a crown is twelve. 71.7 million were delivered. The 11% that was not delivered is down from 16% the year before and 17% the year before that, mostly because more was delivered. Across England's 42 integrated care boards it runs from 3% in Mid and South Essex to 51% in Somerset, where ten of the board's 53 contracts delivered less than a tenth of their units and hold 23% of what it commissions. Here is the figure for every board, what it is made of, what the National Audit Office found behind it, and what the contract rules mean for a practice that falls short.
Practice Radar shows a dental practice where it ranks against every practice nearby, and does the work to move up.
Thirty days free. No card, and nothing starts automatically when it ends.
Key findings
- 10.9% of the NHS dental activity commissioned from high street dental contracts in England in 2025/26 was not delivered: 8.8 million of 80.4 million units of dental activity, the measure NHS contracts count work in. The year before it was 16.1%, and in 2023/24 it was 17.1%.
- More was delivered. Practices delivered 4.8% more units than the year before, and boards commissioned 1.3% fewer. About 4.0 of the 5.2 points of improvement came from the first and about 1.2 from the second.
- 2.8% to 50.9% is the range across England's 42 integrated care boards, from Mid and South Essex to Somerset. In 15 boards, more than 15% of what was commissioned was not delivered.
- In Somerset, 10 of the board's 53 contracts delivered less than a tenth of their units. They hold 22.6% of what the board commissions. Without them, 37.1% of what it commissioned was not delivered, which is still the highest figure in England.
- 36.7% of everything not delivered in England was in the ten boards with the lowest delivery. They commission 17.5% of England's NHS dental activity.
- 40 of the 42 boards delivered a larger share than the year before. In 38 of them more units were delivered. In Devon and Cornwall and the Isles of Scilly the share rose because fewer units were commissioned, and the units delivered fell.
- 43.7% of contracts open all year delivered less than 96% of their units, the level below which NHS England's rules say the commissioner recovers the money for the shortfall. Two years earlier it was 50.8%.
- 1,791 contracts, 30.7% of those open through all of the last three years, delivered less than 96% of their units in every one of them. These files leave out credits that some of them earned in 2024/25, so fewer may have fallen short under the contract rules.
Most NHS dentistry in England is bought in advance. An integrated care board (ICB), the NHS body that plans health services for an area, agrees a contract with a dental practice. The contract sets a yearly payment and a yearly number of units of dental activity, or UDAs, that the practice must deliver in return. A unit of dental activity is how the NHS counts a dentist's work: a check-up is one unit, a filling is three, five or seven depending on how much work it is, and a crown, denture or bridge is twelve. What the ICB has agreed to buy is what is commissioned. What practices then carry out and claim for is what is delivered.
The NHS Business Services Authority (NHSBSA), which pays dentists, publishes both numbers for every contract, every month. We added them up for the NHS year from April 2025 to March 2026, which this study calls last year or 2025/26, and for the two years before it. The sum is the one NHSBSA's own guidance for the data sets out: add up the units delivered, add up the units commissioned, divide one by the other. No practice is named or singled out here. Every figure is for England, an integrated care board, or a group of at least five contracts from at least three providers.
One thing to hold on to throughout. A unit that was not delivered is not a judgement on a dentist. Nor does it mean the money was lost: the rules say a commissioner recovers what it paid for units that were not delivered, as section 4 explains. The National Audit Office, whose findings are set out in section 5, lists two things behind poor access to NHS dentistry that bear on these figures: a contract that the Department of Health and Social Care and NHS England themselves see as needing reform, and how hard it has become to attract dentists to NHS work.
1. 10.9% of what was commissioned last year was not delivered, down from 16.1%
In 2025/26, ICBs had 80,438,166 units of dental activity commissioned from 6,340 contracts at the end of the year. Those contracts delivered 71,685,543 units, or 89.1%. The other 10.9%, 8,752,623 units, was not delivered. For scale, a check-up is one unit, so that is the same amount of activity as 8.8 million check-ups.
Share of commissioned NHS dental activity in England that was not delivered
That is a clear improvement. In 2024/25, 16.1% was not delivered, and in 2023/24, 17.1%. Two things moved. Practices delivered 4.8% more units than the year before, and ICBs had 1.3% fewer commissioned at the year end. More delivery did most of the work. Set against the units commissioned the year before, last year's delivery comes to 88.0%, which accounts for about 4.0 of the 5.2 points. The other 1.2 or so came from commissioning less. Looking only at the 5,937 contracts open through both years, units commissioned fell 1.0%, units delivered rose 4.9% and the share delivered rose from 84.2% to 89.1%, so the improvement is not a result of which contracts are in the list. Against 2023/24, units delivered are up 5.2% and units commissioned down 2.1%.
One caution about 2024/25. From March 2024 to March 2025 the government's dental recovery plan offered a new patient premium, which the National Audit Office describes as a credit of units of dental activity, worth £15 or £50, for each eligible new patient a practice saw. Those credits are not in NHSBSA's delivered figure. We tested that: in every one of the 36 monthly files, the units delivered are within 1% of the units that the month's courses of treatment earn by their bands, in the premium months as in the others. So the figures here count treatment carried out, on the same basis in all three years. What the files cannot show is how many contracts reached their target in 2024/25 with the help of premium credits. That matters for the counts of contracts in section 4.
Our figure for 2023/24 can be checked against an official one. The National Audit Office reported that 83% of contracted units were delivered across England that year, and that 41% of contracts with a target of more than 100 units delivered less than 90% of it. Run the same way on the same year, our sum gives 82.9% and 41.0%.
The headline is a net figure. Some contracts deliver more than they were commissioned for, which the rules allow within limits, and that offsets part of what others fall short by. Counting only the contracts that fell short, their shortfall was 12.2% of everything commissioned. Contracts that went over delivered 1.4% more than commissioned between them.
See this for your own town.
Every practice near yours on one screen: what the CQC found, what its NHS contract pays and who owns it. Beside them, who lives nearby, how well off the area is, children's teeth and the new homes planned. The sample town shows the whole screen now, with no sign-up.
Run on our own practice first: ninth of the 14 practices measured in its town in May, first of 18 in September, on UK-wide search value. Page-one keywords: 6 in May, 194 in September.
Thirty days free. No card, and nothing starts automatically when it ends.
2. From 2.8% to 50.9%: every integrated care board
The national figure hides a very wide range. In Mid and South Essex, 2.8% of what was commissioned was not delivered. In Somerset it was 50.9%. In 15 of the 42 boards more than 15% went undelivered, and in three more than a quarter: Somerset, Devon and Cornwall and the Isles of Scilly. In only 6 boards did contracts between them deliver 96% or more (Birmingham and Solihull, at 95.99%, falls just short).
Share of commissioned NHS dental activity delivered in each integrated care board, 2025/26
Somerset's figure needs taking apart, because half is a great deal. It is not the work of two or three contracts: leave out the three with the largest shortfalls and 44.8% was still not delivered. It is not new either. The board's own Dental Recovery Workplan, in May 2024, put average contract delivery in Somerset at 49% against 69% for the South West, and the National Audit Office found that Somerset had the fewest courses of NHS dental treatment for its population of any board in 2023/24. But the figure is concentrated. Ten of the board's 53 contracts delivered less than a tenth of their units last year, three of them nothing at all, and seven of the ten had been below a tenth the year before as well. Those ten hold 22.6% of the units the board commissions. Across England, contracts like that hold 1.0%. These files cannot say why a contract stands almost idle. The practice may have closed or lost its dentists, or the contract may be waiting to be changed, ended or offered to another practice. Leave those ten out and 37.1% of what Somerset commissioned was not delivered, which is still the highest figure in England. The middle contract in Somerset delivered 60.7% of its units. In Devon, the next board, contracts like that hold 0.9% of the units.
The undelivered activity is concentrated. The ten boards with the lowest delivery commission 17.5% of England's NHS dental activity and account for 36.7% of everything that was not delivered. Five of the ten are in the South West, where 26.1% of what was commissioned across the region was not delivered. In London it was 4.8%.
These differences are not the luck of the draw, and they are not new. We dealt every contract in England out at random across the 42 boards a few hundred times. Chance alone puts the middle 80% of boards within 5.9 points of each other. The real spread is 18.6 points. The order barely changes from year to year either. Eight of this year's ten lowest boards were in the ten lowest the year before.
Nearly every board's share rose. 40 of the 42 delivered a larger share than in 2024/25, and 40 a larger share than in 2023/24. But a share can rise because more is delivered or because less is commissioned, and two of the three largest rises are of the second kind. In Devon the share delivered went from 54.9% to 68.3%, while the units commissioned fell by 24.8% and the units delivered fell by 6.5%. In Cornwall and the Isles of Scilly the share delivered went from 61.5% to 73.2%, while the units commissioned fell by 17.3% and the units delivered fell by 1.6%. In the other 38 boards whose share rose, more units were delivered than the year before. Among boards whose commissioned units moved by less than 2%, the biggest rises were in Bristol, North Somerset and South Gloucestershire, from 68.2% to 77.7%, and Staffordshire and Stoke-on-Trent, from 82.1% to 91.5%. Two boards slipped, both by a point or less, and both delivered more units than the year before.
| Integrated care board | Units commissioned | Units delivered | Delivered | Not delivered | Delivered in 2024/25 | Contracts below 96% |
|---|---|---|---|---|---|---|
| Somerset | 731,190 | 359,205 | 49.1% | 50.9% | 45.4% | 80.0% |
| Devon | 1,210,088 | 826,225 | 68.3% | 31.7% | 54.9% | 66.4% |
| Cornwall and the Isles of Scilly | 619,055 | 453,222 | 73.2% | 26.8% | 61.5% | 67.2% |
| Dorset | 1,056,244 | 798,720 | 75.6% | 24.4% | 67.0% | 59.1% |
| Bristol, North Somerset and South Gloucestershire | 1,456,283 | 1,131,631 | 77.7% | 22.3% | 68.2% | 59.6% |
| Cambridgeshire and Peterborough | 1,033,844 | 814,045 | 78.7% | 21.3% | 68.0% | 55.9% |
| Lincolnshire | 1,063,386 | 840,189 | 79.0% | 21.0% | 73.5% | 66.7% |
| Norfolk and Waveney | 1,300,558 | 1,032,367 | 79.4% | 20.6% | 69.7% | 44.8% |
| Northamptonshire | 1,024,900 | 826,416 | 80.6% | 19.4% | 76.9% | 62.7% |
| North East and North Cumbria | 4,590,002 | 3,795,217 | 82.7% | 17.3% | 75.9% | 65.9% |
| Cheshire and Merseyside | 4,522,172 | 3,764,952 | 83.3% | 16.7% | 84.2% | 67.0% |
| Hampshire and Isle of Wight | 2,540,951 | 2,121,051 | 83.5% | 16.5% | 74.6% | 53.8% |
| Gloucestershire | 700,513 | 587,834 | 83.9% | 16.1% | 73.0% | 56.9% |
| Humber and North Yorkshire | 2,367,170 | 1,992,303 | 84.2% | 15.8% | 79.4% | 42.3% |
| Bath and North East Somerset, Swindon and Wiltshire | 1,071,250 | 903,406 | 84.3% | 15.7% | 81.3% | 56.7% |
| Herefordshire and Worcestershire | 1,073,557 | 913,363 | 85.1% | 14.9% | 77.7% | 40.0% |
| Nottingham and Nottinghamshire | 1,890,182 | 1,660,645 | 87.9% | 12.1% | 80.1% | 49.6% |
| Shropshire, Telford and Wrekin | 775,543 | 683,299 | 88.1% | 11.9% | 79.8% | 52.4% |
| South Yorkshire | 2,465,480 | 2,206,018 | 89.5% | 10.5% | 86.5% | 44.3% |
| Derby and Derbyshire | 1,445,776 | 1,294,654 | 89.5% | 10.5% | 85.5% | 54.5% |
| West Yorkshire | 3,816,397 | 3,432,383 | 89.9% | 10.1% | 87.5% | 36.3% |
| Leicester, Leicestershire and Rutland | 1,640,672 | 1,493,853 | 91.1% | 8.9% | 87.0% | 42.9% |
| Suffolk and North East Essex | 1,286,590 | 1,172,211 | 91.1% | 8.9% | 78.8% | 51.4% |
| Staffordshire and Stoke-on-Trent | 1,821,966 | 1,667,594 | 91.5% | 8.5% | 82.1% | 52.3% |
| Sussex | 2,419,815 | 2,217,647 | 91.6% | 8.4% | 84.1% | 41.9% |
| Lancashire and South Cumbria | 2,673,268 | 2,452,333 | 91.7% | 8.3% | 88.5% | 54.1% |
| Surrey Heartlands | 1,120,918 | 1,030,940 | 92.0% | 8.0% | 88.6% | 46.2% |
| South West London | 1,812,063 | 1,672,103 | 92.3% | 7.7% | 89.3% | 35.2% |
| Kent and Medway | 2,395,649 | 2,231,141 | 93.1% | 6.9% | 86.5% | 32.4% |
| Buckinghamshire, Oxfordshire and Berkshire West | 2,012,962 | 1,895,519 | 94.2% | 5.8% | 90.1% | 43.0% |
| Black Country | 2,012,911 | 1,905,617 | 94.7% | 5.3% | 87.6% | 30.1% |
| North West London | 3,133,243 | 2,980,415 | 95.1% | 4.9% | 92.7% | 29.3% |
| South East London | 2,712,948 | 2,584,139 | 95.3% | 4.7% | 91.7% | 29.7% |
| Greater Manchester | 4,727,644 | 4,522,243 | 95.7% | 4.3% | 95.8% | 30.2% |
| North Central London | 1,904,285 | 1,823,535 | 95.8% | 4.2% | 91.1% | 34.2% |
| Birmingham and Solihull | 2,045,996 | 1,963,900 | 96.0% | 4.0% | 91.3% | 27.5% |
| Hertfordshire and West Essex | 2,091,102 | 2,012,882 | 96.3% | 3.7% | 92.4% | 35.5% |
| Frimley | 929,732 | 896,452 | 96.4% | 3.6% | 90.8% | 33.7% |
| Coventry and Warwickshire | 1,374,034 | 1,327,397 | 96.6% | 3.4% | 91.1% | 31.2% |
| North East London | 2,544,696 | 2,460,837 | 96.7% | 3.3% | 93.1% | 20.2% |
| Bedfordshire, Luton and Milton Keynes | 1,189,882 | 1,155,004 | 97.1% | 2.9% | 92.5% | 25.0% |
| Mid and South Essex | 1,833,249 | 1,782,636 | 97.2% | 2.8% | 94.0% | 33.3% |
The last column is the share of each board's contracts, among those open for the whole year, that delivered less than 96% of their units. It runs from 20.2% in North East London to 80.0% in Somerset. The file also has each board's units commissioned and delivered in 2024/25, so a rise in its share can be told apart from a cut in what it commissioned.
3. Look up the contracts based in a council area
NHS dentistry is commissioned by the integrated care board, not the council, and NHSBSA's guidance for this data says the board is the level it is meant for. So this study does not rank councils. Many readers will still want to know about the contracts nearest to them, and ICBs are large: 36 of the 42 cover three or more councils. So we placed every contract in the council area where its practice is, using the practice postcode, and added up the same two numbers. Type the name of a council to read them. Each result starts with the number of contracts it rests on and gives the figure for the whole board beside it.
Find the NHS dental contracts based in a council area, 2025/26
Read a council area's figure with three cautions. First, it describes the contracts whose practices are based there. It says nothing about the council itself, which neither buys nor provides NHS dentistry, and patients cross council boundaries to see a dentist.
Second, it rests on few contracts. Half of the council areas shown have 17 contracts or fewer, and the smallest have five. One large contract that delivers very little can pull such a figure far below its board's. So a council area's figure should never be quoted without the number of contracts beside it.
Third, inside a board the differences between council areas are no wider than chance would make them. When we dealt every contract out at random among the council areas of its own board, the middle 80% of council areas came out 21.4 points apart. The real spread is 21.4 points. The board's figure is the reliable guide to what is happening locally. A council area's figure is a detail within it.
4. 43.7% of contracts delivered less than 96% of their units, down from 50.8% two years earlier
Behind every area figure are individual contracts. Of the 6,115 contracts open for the whole of 2025/26, 32.1% delivered everything they were commissioned for or more. 43.7% delivered less than 96%, 32.2% less than 90%, and 6.5% less than half.
NHS dental contracts by the share of their commissioned activity they delivered, 2025/26
The 96% line matters because of how the year end works. NHS England's Policy Book for Primary Dental Services sets it out. A contract that delivers between 96% and 100% is within tolerance: no money is taken back, provided the practice agrees to make up the missing units early in the next year (section 8.4.5). Below 96%, the commissioner recovers the money paid for the units that were not delivered, usually by deductions from the practice's payments the following autumn, and may also issue a breach notice (sections 8.4.3 and 8.4.4). Commissioners can waive or spread recovery in exceptional circumstances (section 8.4.9).
The share of contracts below that line was 50.8% in 2023/24, 62.3% in 2024/25 and 43.7% last year. The middle figure should be read with care. 2024/25 was the year of the new patient premium, whose credits are not in these files, and 24.8% of contracts finished between 90% and 96% that year, against 10.1% the year before and 11.5% last year. Some of those contracts may have been within tolerance once their credits were counted. The comparison with 2023/24 is the safer one.
The shortfall is concentrated at the bottom. The 6.5% of contracts that delivered less than half account for 33.4% of all the units contracts fell short by.
It is also persistent. Of the contracts below 96% in 2024/25, 58.9% were below it again last year. And 1,791 contracts, 30.7% of the 5,834 open through all of the last three years, delivered less than 96% of their units in every one, as these files count them. Between them they hold 34.5% of the units commissioned from those contracts. That matters because of another rule. Where a contract has delivered less than 96% for three years in a row, the Policy Book says the commissioner should seek a permanent reduction in the contract (sections 6.6 and 8.4.6). These files cannot show which contracts had exceptional circumstances, an agreed temporary reduction, or premium credits in 2024/25 that lifted them to 96%. So 1,791 is the most the rule could reach among these contracts, not the number it will. Setting 2024/25 aside, 1,939 contracts were below 96% in both 2023/24 and 2025/26.
NHSBSA does not publish how much money was recovered for 2025/26, and we have not estimated it. The National Audit Office reported that commissioners recovered £479 million from dental contracts for 2022/23, and an average of £149 million a year for the three years from 2017/18 to 2019/20, before the pandemic.
5. Why NHS dentistry goes undelivered: what the National Audit Office found
The National Audit Office's investigation into the NHS dental recovery plan, published in November 2024, is the most recent independent account. It is about access to NHS dentistry, and what it describes is the system practices work in. It does not put under-delivery down to individual dentists. Three of its points bear on the figures here.
Recruiting dentists to NHS work has become harder. The report quotes the Review Body on Doctors' and Dentists' Remuneration, which advises government on dentists' pay and found in 2024 that “attracting dentists to NHS work is increasingly difficult, resulting in significant under-provision of NHS dentistry” (paragraph 1.21). A contract's units can only be delivered if there is a dentist to do the work. When an associate leaves and cannot be replaced, the units they would have delivered go undelivered, however busy the rest of the practice is.
Some areas find it harder than others. The people the auditors spoke to said the issue is not necessarily a lack of dentists in the country, but that not enough are given reason to do enough NHS work, and that some areas find it harder to recruit than others (paragraph 1.22). The delivery figures fit that. The South West has the lowest delivery of any NHS region, 73.9%, and the fewest dentists doing NHS work for its population of any region: 40.5 for every 100,000 people in NHSBSA's workforce figures for 2025/26. London has 53.4. Across the 42 boards, those with fewer NHS dentists for their population tend to deliver less of what is commissioned (rank correlation 0.63, where 0 is no relationship and 1 a perfect one). Eight of the ten boards with the lowest delivery are also among the ten with the fewest NHS dentists per head. That is an association, not proof of cause.
The contract itself. The contract dates from 2006 and pays by units of activity. The report says it is widely perceived as needing reform, including by the Department of Health and Social Care and NHS England, and that many in the sector view it as a disincentive to provide NHS care when a practice can offer private care too (summary, paragraph 11, and paragraph 1.23).
Two things that might be expected to explain the pattern do not. Delivery is no lower in deprived neighbourhoods: contracts in the most deprived fifth of England delivered 89.7% of what was commissioned and those in the least deprived fifth 89.2%. And it is not a matter of contract size: contracts under 2,000 units delivered 85.9% between them, those of 20,000 or more 87.4%, and the sizes in between 87.0% to 92.5%. Allowing for both leaves the range across boards almost unchanged, at 49.6% to 97.6% delivered.
There is also a limit to what these files can see. They count units of dental activity only. Some ICBs now pay practices for other work out of the same contracts, such as urgent-care sessions paid by the session, under what NHS England calls flexible commissioning. The Policy Book says such work can be paid for by setting it against a contract's existing units (section 6.10). Where the unit target was reduced to match, the percentage here is unaffected. The National Audit Office notes that this dataset does not include units awarded for flexible commissioning (footnote 16). So in a board that uses it widely and credits the work in units outside these files, delivery will look lower here than it was. These files give no way to measure that for any board.
What to do about it
If you hold an NHS contract and expect to finish below 96%, it helps to act early. The Policy Book says the commissioner and the contractor may agree at any time to vary the number of units and the payment, and either side can ask (section 8.3). A temporary reduction for one year needs evidence of exceptional circumstances and is decided case by case (section 6.9). NHSBSA works out every contract's April to September delivery by 31 October for the mid-year review, so the summer is the time to look at your own numbers. If a dentist has left and you cannot replace them, tell your ICB when it happens and keep the evidence that you tried to recruit.
If you have missed 96% three years running, expect your ICB to open a conversation about a permanent reduction, because the Policy Book tells it to. A contract set at the level you can staff means no money taken back each autumn. Go in knowing your delivery in each of the last three years. The Policy Book says the contract should be rebased at the highest level delivered across them.
If you deliver in full and have room for more, say so. Where it has the money, an ICB can agree that a contract delivers and is paid for up to 110% of its units in a year, and it is expected to ask for expressions of interest and to look at the previous year's delivery (Policy Book section 8.4.8). In a board where a fifth of what was commissioned went undelivered, there is commissioned activity that nobody delivered. The table in section 2 shows where that is.
If you are private or mixed, the same table is a map of where patients are most likely to be looking for a practice that can see them. In those areas, being the practice people find when they search for a dentist matters all the more. Say plainly on your website and your Google listing whether you take new patients and what a first appointment costs.
If you report on or represent an area, find its integrated care board in section 2 or in the file. Two questions for the board are worth asking. What happened to the commissioned activity that was not delivered: how much was offered to other practices in the same year, and how much was moved to other dental services. And, where the share delivered has risen, whether more dentistry was delivered or less was commissioned. The file has the units for both years.
How we did this
Source. NHSBSA's English Contractor Monthly General Dental Activity, published on its Open Data Portal under the Open Government Licence v3.0. There is one file for each month, with one row for each contract that provides general dental services: its commissioner, practice postcode, the units of dental activity it is contracted to deliver in the year (UDA_PERF_TARGET) and the units delivered in that month (UDA_DELIVERED). We used the 36 files from April 2023 to March 2026, downloaded on 12 November 2026. The files for 2024/25 and 2025/26 are identical to the copies we took in September.
Commissioned and delivered. NHSBSA's guidance for the data says the commissioned figure should not be added across months and that the latest month of the year shows the latest contractual position. So commissioned is each contract's target in the March file, which includes any change agreed during the year, and delivered is the sum of its twelve monthly figures. The share delivered for an area is all the units delivered by its contracts divided by all the units commissioned from them. Measured against the targets as they stood in April instead, England's figure is 89.0% delivered.
Which contracts. Contracts in the March file with a target of more than 100 units, 6,340 in 2025/26. This is the group the National Audit Office used. The 665 lines left out have a target of 100 or fewer, most of them 0 or 1, and look like lines for work commissioned on other terms. They delivered 556,018 units against targets adding up to 1,098. That is real dentistry, but with no target to set it against it cannot go into a share of what was commissioned. NHSBSA's simple sum over every row, which counts it, gives 90.1% delivered. Contracts that closed before March are not in the March file and are not counted on either side. They delivered 267,001 units during the year. Contracts open for only part of the year are in the totals but not in the shares of contracts, which use the 6,115 with a row in all twelve months. Using only those contracts for the totals as well gives 89.2%.
How much that choice matters to a board. For most boards, very little. In seven boards the units on those left-out lines and closed contracts come to two points or more of what the board commissioned: Kent and Medway (98.1%), Derby and Derbyshire (93.5%), South Yorkshire (93.2%), North East and North Cumbria (85.9%), Buckinghamshire, Oxfordshire and Berkshire West (97.1%), Norfolk and Waveney (81.7%) and Sussex (93.6%). The figure in brackets is the share delivered if those units are counted. Counting them, Kent and Medway has the highest share delivered in England, and nine of the ten lowest boards are still the ten lowest. For the five lowest boards they add 0.4 points at most.
Foundation dentists. NHSBSA's delivered figure leaves out units delivered by foundation dentists, who are newly qualified dentists in their training year, and reports them in a separate column. We followed NHSBSA and left them out. Adding them back, England delivered 90.9% of what was commissioned, 41.5% of contracts were below 96%, and the order of the boards is almost unchanged. Our September study of contract values gave 41.2% of contracts below 96% in 2025/26, against 43.7% here. The two differ for two reasons. That study counted foundation dentists' units in, and it looked at a narrower group: the 5,666 contracts with a target of 1,000 units or more, leaving out NHS trusts and lines priced under £10 or over £100 a unit. Rebuilt from these files on that basis, the figure is 41.2%, the one published in September. Leaving foundation dentists out of that same group gives 43.5%.
Places. The integrated care board is the contract's commissioner in the March file. These are the 42 boards as they stood in March 2026. Several merged or changed their boundaries in April 2026, and there are now 36. The council is the lower-tier council of the contract's practice postcode, from the ONS National Statistics Postcode Lookup for August 2026. Every postcode matched. The address is the contract's main practice, and a contract with more than one surgery is counted where that practice is.
Keeping practices anonymous. Each contract's figures are public in NHSBSA's files. We still publish only pooled figures. A council is shown only if it has five or more contracts from at least three providers, and no provider holds half the contracts or half the units. The 9 council areas below that are pooled in one row. In the file, a council area's row has its number of contracts and its percentage, and no units. Each board's units are in its own row, so council units would let a reader take a board's council areas away from its total. In seven boards that would leave the figures of a single contract, and in seven those of the one council area that is hidden. A council area's row has no share of contracts below 96% and no figure for the year before either. On a handful of contracts a change of tens of points from one year to the next is routine.
Controls we ran. The same sum on 2023/24 reproduces the National Audit Office's published figures. NHSBSA's annual Official Statistics count units by a different route and include contracts these files leave out, and they show a similar rise in units delivered between 2024/25 and 2025/26: 5.1%, against 4.6% in the monthly files when every row and foundation dentists' units are counted. For the spread between boards we tested chance by dealing contracts out at random, and standardised each board for the size of its contracts and the deprivation of their neighbourhoods (English Indices of Deprivation 2025). For council areas we repeated the chance test within each board. Leaving out NHS trusts and contracts under 1,000 units gives 89.2% delivered for England.
What this is not. It is not NHS England's year-end reconciliation, which is done contract by contract after manual adjustments for missing claims, exceptional circumstances and changes of ownership. These figures count treatment by the month it was completed and are taken before any of that. It does not measure private dentistry, or NHS dentistry in hospitals and community services. It does not show how many patients went without care, because a unit not delivered at one practice may have been delivered to the same patient at another. And it does not include work paid for outside the unit system, or the credits of the new patient premium.
Questions we get asked about this data
What is a unit of dental activity?
It is how NHS dental contracts in England count work. A course of treatment earns a set number of units according to its band: one for a check-up, three, five or seven for fillings, extractions and root canal work depending on how much is done, and twelve for crowns, dentures and bridges. A contract pays a yearly sum for a yearly number of units.
Does not delivered mean the money was wasted?
No. Where a contract delivers less than 96% of its units, NHS England's rules say the commissioner recovers the money for the units that were not delivered. What the ICB then does with that money is a separate question, and one these files cannot answer.
How much of the NHS dentistry commissioned in my area was delivered?
Search for your integrated care board in section 2, or open the file. Across England it was 89.1% in 2025/26, and across the 42 boards it ranged from 49.1% to 97.2%. Section 3 has a lookup of the contracts based in each council area, to be read with the number of contracts and the board's figure beside it.
Did the figure improve because targets were cut?
Mostly not. Across England, units delivered rose 4.8% and units commissioned fell 1.3%, so about 4.0 of the 5.2 points of improvement came from more dentistry being delivered. It differs by board. In Devon and Cornwall and the Isles of Scilly the share delivered rose while the units delivered fell, because the units commissioned fell further.
Why is my council not in the lookup?
A council area is shown only where it has five or more NHS contracts from at least three providers and no provider holds half of them, so that no single practice can be picked out. 9 council areas fall below that and are pooled together in the file's last row.
Is a low figure the fault of the practices in my area?
That is not how the National Audit Office describes it. Among the things it found behind poor access to NHS dentistry are a contract that government itself sees as needing reform and the difficulty of attracting dentists to NHS work, which it was told is greater in some areas than others. A practice that loses a dentist and cannot replace them will under-deliver however hard everyone else works. These files show how much was delivered. They cannot show why.
For editors
The data is free to reuse with attribution to Practice Radar, under Creative Commons BY 4.0. Source: NHS Business Services Authority, English Contractor Monthly General Dental Activity, April 2023 to March 2026, with the 96% rule from NHS England's Policy Book for Primary Dental Services and the account of what lies behind it from the National Audit Office's Investigation into the NHS dental recovery plan (HC 308, November 2024). The file, nhs-dentistry-commissioned-but-not-delivered-2026.csv, has one row for England and one for each integrated care board, with ONS codes and the units for this year and last, and 286 rows giving the percentage delivered by the contracts based in each council area.
Three things to carry with the figures. Not delivered does not mean dentists failed or money was lost: section 5 gives the National Audit Office's account and section 4 the rule on recovery. The lowest board's figure is concentrated in a small number of contracts that delivered almost nothing, which section 2 sets out. And council areas are not ranked in this study, because NHS dentistry is commissioned by integrated care boards and the differences between council areas inside a board are no wider than chance. When quoting a council area's figure, please give the number of contracts it rests on and its board's figure, and say that it describes the NHS dental contracts based in the area.
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
Cite and reuse
To cite this study. Practice Radar, “11% of the NHS dentistry commissioned in England last year was not delivered, down from 16%. Across the 42 integrated care boards it runs from 3% to 51%”, 12 November 2026, https://practiceradar.co.uk/blog/nhs-dentistry-commissioned-but-not-delivered-2026
The data. NHS dentistry commissioned and delivered by area (CSV, 331 rows), with a README that explains each column.
Licence. The data is free to reuse under Creative Commons BY 4.0. Attribution means crediting Practice Radar and linking to this study. It is built from public sector information, so please also keep these lines: “Contains public sector information licensed under the Open Government Licence v3.0. Contains OS data © Crown copyright and database right 2026. Contains Royal Mail data © Royal Mail copyright and database right 2026. Source: Office for National Statistics licensed under the Open Government Licence v.3.0.”
See every NHS contract in your town
Practice Radar reads the NHS contracts around your practice every month: what each was commissioned to deliver and how far through its target it is. It sits beside the local search market you compete in.
Run on our own practice first: ninth of the 14 practices measured in its town in May, first of 18 in September, on UK-wide search value. Page-one keywords: 6 in May, 194 in September.
Thirty days free. No card, and nothing starts automatically when it ends.