Performance Tracker 2025: Hospitals

When the Labour government came to power in July 2024, it inherited a hospital^* system in England that had been through a torrid 14 years. Previous governments had starved hospitals of capital investment, constantly reshuffled NHS structures, and held down hospital staff pay for years. The result was a system that was ill-equipped to deal with a global pandemic, and which has never fully recovered.

Hospital performance on urgent and emergency, elective, and cancer care – which had been declining for years before 2019 – reached record lows in the years after the Covid pandemic. The public noticed. In 2024, public satisfaction with accident and emergency, outpatient and inpatient services was the lowest on record. ¹ Taylor B, Lobont C, Dayan M, Merry L, Jefferies D and Wellings D, Public Satisfaction with the NHS and Social Care in 2024 (BSA), The King’s Fund, 2025, retrieved 31 October 2025, www.kingsfund.org.uk/insight-and-analysis/reports/public-satisfaction-nhs-social-care-in-2024-bsa  According to YouGov polling, health was the second most important issue to voters at the general election, behind only the cost-of-living crisis. ² Smith M, ‘General election 2024: what are the most important issues for voters?’, YouGov, 1 June 2024, retrieved 27 October 2025, https://yougov.co.uk/politics/articles/49594-general-election-2024-what-are-the-most-important-issues-for-voters

Immediately after the election, Labour commissioned Lord Ara Darzi to review the drivers of poor performance in the NHS. His report deemed the NHS to be “in serious trouble”. ³ Darzi A, Lord, Independent Investigation of the NHS in England, GOV.UK, 2024, https://assets.publishing.service.gov.uk/media/66f42ae630536cb92748271f/Lord-Darzi-Independent-Investigation-of-the-National-Health-Service-in-Englan…  The government’s plan to repair the system came in July 2025 in the form of the 10 Year Health Plan for England. ⁴ Department of Health and Social Care, 10 Year Health Plan for England: Fit for the future, CP 1350, The Stationery Office, 2025, p. 9, retrieved 31 October 2025, www.gov.uk/government/publications/10-year-health-plan-for-england-fit-for-the-future  That document made clear that the government wants the NHS to shift care to settings outside hospitals. But it is inevitable that hospitals will remain core to the government’s ambitions for the NHS. Most obviously, the government chose the 18-week elective waiting-time target – that 92% of patients should wait less than 18 weeks between a referral for elective care and the start of their treatment – as the measure of success for its ‘NHS mission’. There are pros and cons to using that milestone. But it is certain that the government will fail to reduce elective waiting times enough without substantial improvement in the way hospitals operate.

Hospitals’ performance also has fiscal ramifications. The scale of spending on hospital trusts is such that a failure to improve productivity in line with expectations in the 10 Year Health Plan for England would risk destabilising the finely balanced public finances, if more money then had to be channelled to the NHS.

What happens in hospitals in the coming years matters for the success of the government’s NHS mission, the nation’s access to care, for delivering the government’s wider agenda, and for its chances of retaining office at the next general election.

_{*    By ‘hospitals’, we mean acute, community, mental health, and ambulance trusts in the English NHS. We will refer to these as ‘hospital trusts’ throughout the report. At times throughout this report, we focus the analysis on particular types of trusts, mostly acute trusts and independent sector providers. We will make it clear in the text and in charts when we do so.}

Spending

Day-to-day spending continued to grow after the pandemic

Since the creation of the NHS in 1948, spending on health has grown at one of the fastest rates of any service in the British state. Total spending on health across the UK^* grew at an average annual rate of 3.9% in real terms between 1950/51 and 2009/10. The last Labour government raised health spending by an average of 5.9% per year in real terms between 1996/97 and 2009/10, second only to growth during the (much shorter) Heath government between 1969/70 and 1973/74, when health spending grew at an average annual rate of 6.1% in real terms.

There is no data for spending on hospital trusts before 2009/10, so our time series starts in that year. As Figure NHS 3.2 shows, spending growth on hospital trusts was low in the 2010s but never fell. As a result, in 2023/24, the government spent 50.4% more on hospital trusts in real terms than in 2009/10, an average annual increase of 3.0%. Accounting for a growing population, per-capita spending grew at an average rate of 2.2% per year in real terms across that time.

_{*    This is different from the measure of spending in Figure NHS 3.2, which shows day-to-day spending on hospital trusts in the English NHS.}

Much of the growth in spending on hospitals trusts has happened since 2019, when the May government launched the NHS Long Term Plan. ⁵ NHS, NHS Long Term Plan, NHS England, 2019, retrieved 31 October 2025, https://webarchive.nationalarchives.gov.uk/ukgwa/20230418155402/https:/www.longtermplan.nhs.uk/publication/nhs-long-term-plan  To fund the ambitions of that plan, the government estimated that it would entail spending £33.9billion more per year on the whole NHS by 2023/24 compared to 2018/19. ⁶ HM Treasury, Spending Round 2019, CP 170, The Stationery Office, 2019, p. 29, retrieved 31 October 2025, www.gov.uk/government/publications/spending-round-2019-document/spending-round-2019  Using GDP deflator estimates from the time shows that the government therefore expected spending on the NHS to rise by 3.3% in real terms per year over that period.

Spending on all NHS hospital trusts in fact grew faster than that estimate between 2018/19 and 2023/24, an average of 4.9% per year in real terms. Most of that spending went on hiring additional staff, as discussed in more depth below, and on meeting Covid pressures in 2020/21 and 2021/22. ⁷ Comptroller and Auditor General, Managing NHS Backlogs and Waiting Times in England, Session 2022–23, HC 799, National Audit Office, 2022, p. 20, www.nao.org.uk/wp-content/uploads/2022/11/managing-nhs-backlogs-and-waiting-times-in-england-summary.pdf  Growth since 2018/19 is well above the annual average increase of 1.9% per year on all NHS hospital trusts between 2010/11 and 2018/19, and also above the average annual real-terms spending increase on all of health in the UK of 3.9% between 1949/50 and 2009/10. ⁸ Stiebahl S, NHS Expenditure, House of Commons Library, 2024, retrieved 31 October 2025, https://commonslibrary.parliament.uk/research-briefings/sn00724

Spending on acute hospital trusts has followed almost exactly the same pattern as spending on all types of hospital trust. This is unsurprising. The majority of spending on NHS trusts is on acute trusts, accounting for approximately 75% of spending on NHS trusts in 2023/24. ⁹ NHS England, ‘Consolidated NHS provider accounts 2023/24’, ENGLAND.NHS.UK, 6 January 2025, retrieved 5 November 2025, www.england.nhs.uk/long-read/consolidated-nhs-provider-accounts-2023-24/

Spending is due to grow further in this parliament

After coming to power in July 2024, the Starmer government set budgets for 2024/25 through to 2028/29.

Day-to-day spending on the entire NHS is due to rise at an average real-terms rate of 3.0% per year between 2023/24 and 2028/29. The largest annual increase in that time has already happened. Between 2023/24 and 2024/25 spending was due to rise by 5.2%. The NHS’s budget is then expected to grow more slowly in 2025/26 – by 2.0% in real terms – before growing consistently by 3.0% in real terms in every year between 2026/27 and 2028/29. But it would not be a surprise if the government increased spending by more than that amount over the course of the spending review period. Governments have tended to provide the NHS with short-term, emergency funding when performance declines or becomes particularly salient in the media; for example, during the now annual ‘winter crisis’. Indeed, this pattern seems likely to be repeated in 2025/26, as NHS leaders have requested that the government provides the service with an additional £3bn worth of funding, though this top-up is arguably due to factors that are legitimately outside the control of NHS leaders, as discussed in more detail below. ¹⁰ Campbell D, ‘NHS leaders warn of longer waiting times if demand for extra £3bn not met’, The Guardian, 27 October 2025, retrieved 28 October 2025, www.theguardian.com/society/2025/oct/27/nhs-leaders-demand-extra-funding-waiting-times

The 10 Year Health Plan for England said that the government intends to “shift the pattern of health spending”, meaning that “the share of expenditure on hospital care will fall”. ¹¹ `Department of Health and Social Care, 10 Year Health Plan for England: Fit for the future, CP 1350, The Stationery Office, 2025, p. 9, retrieved 31 October 2025, www.gov.uk/government/publications/10-year-health-plan-for-england-fit-for-the-future  This implies that spending on hospitals will grow less quickly than the NHS total of 3.0% per year. But it is unclear how great this shift in resources will be or whether it will happen.

Annual growth of 3.0% would be higher than in the period between 2009/10 and 2023/24, roughly in line with the average annual increase in spending on health between 1979/80 and 1996/97, and well below the average growth of 5.9% per year between 1996/97 and 2009/10 under the New Labour governments, as shown in Figure NHS 3.1. It will also be below the average annual growth over the course of the last parliament, implying that Starmer is planning to increase spending on hospitals by less than the May to Sunak governments.

Hospital trusts’ deficits appear to have worsened since the pandemic

NHS hospital trusts have a long record of spending beyond their budgets and have reported a combined deficit since 2013/14, ¹² Comptroller and Auditor General, NHS Financial Management and Sustainability, Session 2024–25, HC 124, National Audit Office, 2024, p. 20, retrieved 31 October 2025, www.nao.org.uk/reports/nhs-financial-management-and-sustainability-2024  though as Figure NHS 3.3 shows, acute trusts are almost entirely driving this, with the other types of trusts combined (community, mental health, ambulance and specialist trusts) running surpluses for every year until 2023/24. Nuffield Trust work shows that, in 2015/16, acute trusts spent 4.3% more than their budgets, a pattern that continued (albeit less severely) for every year between then and 2019/20. ¹³ Gainsbury S and Julian S, ‘NHS provider deficits are back: how bad is the situation?’, Nuffield Trust, 12 February 2025, retrieved 27 October 2025, www.nuffieldtrust.org.uk/news-item/nhs-provider-deficits-are-back-how-bad-is-the-situation  Acute trusts ran small surpluses in both 2020/21 and 2021/22 (0.6% and 0.4% respectively).
 

When the Labour government came to power in July 2024, it inherited a system in which acute trusts had returned to overspending their budgets. In 2022/23, acute trust overspent their budgets by £589million ¹⁴ NHS England, Consolidated NHS Provider Accounts 2022/23, HC 469, The Stationery Office, 2025, p. 6, retrieved 31 October 2025, www.england.nhs.uk/long-read/consolidated-nhs-provider-accounts-2022-23  – equating to 0.6% of their income. This increased to 1.2% in 2023/24. ¹⁵ Gainsbury S and Julian S, ‘NHS provider deficits are back: how bad is the situation?’, Nuffield Trust, 12 February 2025, retrieved 27 October 2025, www.nuffieldtrust.org.uk/news-item/nhs-provider-deficits-are-back-how-bad-is-the-situation

Systems continued to run deficits in 2024/25, though there is no outturn data at a trust level, as for 2023/24 and years before those. There are, however, initial returns^* that show that 17 out of 42 (40.5%) integrated care boards (ICBs, areas that incorporate several NHS trusts) reported a deficit in 2024/25. For all of England, this equated to a deficit of 0.4%.^** Lancashire and South Cumbria ICB reported the largest deficit, at 2.4%. A total of 21 ICBs reported very small surpluses, though none larger than 0.04% of their allocation.

One month into the 2025/26 financial year, NHS England estimated that there would be a £2.2bn deficit for the full year, representing 1.4% of income. ¹⁶ O’Mahony E, ‘2025/26 operating plan position’, NHS England, 28 May 2025, retrieved 31 October 2025, www.england.nhs.uk/long-read/2025-26-operating-plan-position  There is a wide range of anticipated deficits across ICBs.

Eleven of the 42 ICBs estimate that they will have either no deficit for the year, or else a small surplus, down from 25 in 2024/25. Nine ICBs account for more than half (51.9%) of the total projected deficit. The ICB with both the largest overall deficit and the largest deficit as a proportion of its allocation is Shropshire, Telford and Wrekin ICB, which expects that its deficit will be 5.7% of its total allocation for 2025/26. ¹⁷ O’Mahony E, ‘Financial performance update’, NHS England, 28 May 2025, retrieved 31 October 2025, www.england.nhs.uk/long-read/financial-performance-update-27-march-2025

Of the 17 ICBs that reported a deficit in 2024/25, only one (Lincolnshire) has forecasted that it will have either no deficit or a surplus in 2025/26. There are 15 ICBs that did not report a deficit in 2024/25 but forecast that they would move into such a position in 2025/26. Shropshire, Telford and Wrekin ICB has the largest year-to-year worsening of its financial position, moving from a 1.2% deficit in 2024/25 to a 5.7% forecast deficit in 2025/26.

Even before the onset of this year’s winter crisis, NHS leaders are requesting that the government provides them with an additional £3bn worth of funding for 2025/26. ¹⁸ Campbell D, ‘NHS leaders warn of longer waiting times if demand for extra £3bn not met’, The Guardian, 27 October 2025, retrieved 28 October 2025, www.theguardian.com/society/2025/oct/27/nhs-leaders-demand-extra-funding-waiting-times  That would amount to a 1.5% increase in the NHS’s day-to-day budget. The leaders argue that there is financial pressure coming from:

•    renewed strike action by resident doctors

•    the redundancies associated with cutting ICBs’ headcount by half (as discussed in the NHS overview chapter)

•    higher drug costs.

System leaders have a legitimate claim in this instance, as some of these costs are the result of decisions that central government has taken. It was Wes Streeting, secretary of state for health and social care, who decided that ICBs should cut their workforce in half. ¹⁹ Gault B, ‘ICBs asked to reduce workforce by over 12,000’, Healthcare Leader, 13 March 2025, retrieved 28 October 2025, https://healthcareleadernews.com/news/icbs-asked-to-reduce-workforce-by-over-12000  And it is within the gift of central government – not the NHS – to settle industrial action with striking doctors. There is also a suggestion that pharmaceutical companies are cutting drug prices in the US, while raising them in other countries, in response to a threat from the president, Donald Trump, that he would put a 100% tariff on pharmaceutical imports unless companies lowered drug prices. ²⁰ Kollewe J and Courea E, ‘NHS could pay 25% more for medicines under plan to end row with drugmakers and Trump’, The Guardian, 8 October 2025, retrieved 29 October 2025, www.theguardian.com/business/2025/oct/08/nhs-could-pay-25-more-for-medicines-under-plan-to-end-row-with-drugmakers-and-trump  If so, that is also a cost that is outside the control of the NHS.

If the government does not fund these cost pressures, it will mean that trusts have to cut spending elsewhere to balance their books. This will likely mean a reduction in things like paid overtime for staff, and therefore less elective activity carried out in hospitals, hurting the government’s ambition to reduce elective waiting times.

_{*    These may be amended when final accounts are filed.
**    This includes all trusts and ICBs themselves, not just acute trusts, so is not directly comparable to the 1.2% deficit among acute trusts for 2023/24 noted earlier.}

There are indications that the cost of delivering care is outstripping economy-wide inflation

On the face of it, it is quite difficult to reconcile relatively generous settlements for the NHS since 2019 – and since 2010 when compared to other services – with current levels of poor performance in hospitals, and trusts running deficits.

Some commentators argue that the NHS suffers from poor financial discipline because it knows that the government will step in with financial support when trusts run deficits or when performance deteriorates. ²¹ Black S, ‘The mythbuster: an ill-disciplined NHS is to blame for the lack of capital spending’, HSJ, 5 February 2024, retrieved 27 October 2025, www.hsj.co.uk/daily-insight/the-mythbuster-an-ill-disciplined-nhs-is-to-blame-for-the-lack-of-capital-spending/7036499.article  And it is true that the government tends to provide more funding to the NHS during winter crises or when performance reaches politically unpalatable levels.

Two other explanations are lower hospital productivity and the costs of providing care, which rise more quickly than income. As we and others have shown, ²² Warner M and Zaranko B, NHS Funding, Resources and Treatment Volumes, Institute for Fiscal Studies, 2022, https://ifs.org.uk/sites/default/files/2022-12/NHS-funding-resources-and-treatment-volumes-Institute-for-Fiscal-Studies.pdf ^{, 23} Freedman S and Wolf R, The NHS Productivity Puzzle: Why has hospital activity not increased in line with funding and staffing?, Institute for Government, 2023, retrieved 31 October 2025, www.instituteforgovernment.org.uk/publication/nhs-productivity  hospital productivity has declined since 2019, meaning that any additional pound of spending is not going as far as it was then. There has been an improvement in productivity since the worst of the pandemic, but productivity is still below 2019 levels, as discussed in more depth below.

But there is also evidence that the costs of delivering hospital care have outstripped economy-wide inflation in recent years. Pay settlements, including for 2025/26, have often been both higher than budgeted for and higher than economy-wide inflation. The extent to which the government has compensated the NHS with the additional money needed to fund those pay settlements is mixed.

In 2022/23, the previous government said that it would fund pay awards through reallocation away from some central programmes, implying cuts to those services. ²⁴ Kelly J, ‘2022/23 pay award’, NHS England, 20 July 2022, www.england.nhs.uk/wp-content/uploads/2022/07/B1863-2022-23-Pay-Award.pdf  In 2023/24, the National Audit Office (NAO) reported that the government did compensate the NHS for at least some – though it is unclear whether all – of the higher-than-budgeted-for pay awards. ²⁵ Comptroller and Auditor General, NHS Financial Management and Sustainability, Session 2024–25, HC 124, National Audit Office, 2024, p. 22, retrieved 31 October 2025, www.nao.org.uk/reports/nhs-financial-management-and-sustainability-2024  The NAO also reported that inflation exceeded NHS expectations by £1.4bn (1% of the NHS’s budget) in that year, ²⁶ Comptroller and Auditor General, NHS Financial Management and Sustainability, Session 2024–25, HC 124, National Audit Office, 2024, p. 23, retrieved 31 October 2025, www.nao.org.uk/reports/nhs-financial-management-and-sustainability-2024  requiring more spending on inputs such as medicines and energy.

The current government says that the pay award in 2025/26 will be “fully funded”, ²⁷ Department of Health and Social Care Media Centre, ‘NHS pay: everything you need to know about the 2025 pay award’, blog, GOV.UK, 27 May 2025, retrieved 1 November 2025, https://healthmedia.blog.gov.uk/2025/05/27/nhs-pay-everything-you-need-to-know-about-the-2025-pay-award  though it is unclear at this point whether that means it will provide further funding or reallocate money away from other parts of the health service.

Another common claim is that the requirement to spend more on new treatments and improved technology is driving higher NHS costs. ²⁸ Gainsbury S, ‘Down payment or making ends meet? NHS financial pressures in the run-up to the spending review’, Nuffield Trust, 5 June 2025, retrieved 27 October 2025, www.nuffieldtrust.org.uk/resource/down-payment-or-making-ends-meet-nhs-financial-pressures-in-the-run-up-to-the-spending-review  This is a view that one trust chief executive officer who we interviewed supported. ²⁹ Institute for Government interview.  The Office for Budget Responsibility concurs that the introduction of new technology in health tends to push up the costs of delivering care. ³⁰ Office for Budget Responsibility, Fiscal Risks and Sustainability, CP 1142, The Stationery Office, 2024, p. 71, https://obr.uk/docs/dlm_uploads/Fiscal-risks-and-sustainability-report-September-2024-1.pdf  This is in contrast to the introduction of technology in other sectors, where it often results in cost savings. ³¹ Office for Budget Responsibility, Fiscal Risks and Sustainability, CP 1142, The Stationery Office, 2024, p. 71, https://obr.uk/docs/dlm_uploads/Fiscal-risks-and-sustainability-report-September-2024-1.pdf  Spending on new treatments/ technology is intended to improve care for patients, but there is evidence that it does not always result in better outcomes. One study compared the value of the NHS’s spending on new drugs to the value it would have created if the NHS had instead chosen to spend that money on existing treatments between 2000 and 2020. The researchers found that the NHS spent £75.1bn on a selection of new drugs in that time, resulting in an estimated 3.8 million additional quality-adjusted life years (QALYs – 1 QALY represents one year lived in perfect health ³² National Institute for Health and Care Excellence, ‘Glossary’, (no date), retrieved 1 November 2025, www.nice.org.uk/glossary?letter=q ) for patients, compared to 5.0 million QALYs if the NHS had instead spent that money on existing treatments. ³³ Naci H, Murphy P, Woods B and others, ‘Population-health impact of new drugs recommended by the National Institute for Health and Care Excellence in England during 2000–20: a retrospective analysis’, The Lancet, vol. 405, no. 10472, pp. 50–60, retrieved 1 November 2025, www.thelancet.com/journals/lancet/article/PIIS0140-6736(24)02352-3/fulltext  In other words, the NHS could have been better off not spending money on those new treatments and instead prioritising existing treatments.

Another argument that interviewees raised is that the complexity of people’s health needs has increased in recent years. As a result, hospitals spend more to treat them. There is some evidence that supports this. One study found a steadily increasing incidence of multi-morbidity (people living with multiple long-term health conditions) between 2004 and 2019 and increasing likelihood of multi-morbidity as people age. ³⁴ MacRae C, Mercer S, Henderson D and others, ‘Age, sex, and socioeconomic differences in multimorbidity measured in four ways: UK primary care cross-sectional analysis’, British Journal of General Practice, vol. 73, no. 729, pp. e249–e256, retrieved 1 November 2025, https://bjgp.org/content/73/729/e249  If true, then this is a pattern that will continue as the population ages over the coming decades. ³⁵ Hoddinott S, Adult Social Care Across England, Institute for Government, 2025, p. 42, www.instituteforgovernment.org.uk/sites/default/files/2025-06/adult-social-care-across-england_1.pdf

Poorly timed planning contributes to financial instability

The government sets priorities and targets for the NHS through the annual publication of Operational Planning Guidance. This tells NHS systems what they are expected to achieve in the following financial year and what their financial framework will be. Given the parameters outlined in the planning guidance, NHS systems then spend months finalising their own plans for that year, which they then agree with NHS England.

NHS England aims to publish the planning guidance at least three months before the start of each financial year (i.e. at the end of December), to allow for effective planning in systems before the start of the year. But as the NAO points out, the NHS England “has not achieved that aim since 2017”. ³⁶ Comptroller and Auditor General, NHS Financial Management and Sustainability, Session 2024–25, HC 124, National Audit Office, 2024, p. 29, www.nao.org.uk/wp-content/uploads/2024/07/nhs-financial-management-and-sustainability.pdf  This means that systems are frequently forced to finalise their plans months into the financial year. The financial year 2024/25 was a particularly bad one, with the Sunak government publishing planning guidance just seven days before the start of it. ³⁷ Comptroller and Auditor General, NHS Financial Management and Sustainability, Session 2024–25, HC 124, National Audit Office, 2024, p. 29, www.nao.org.uk/wp-content/uploads/2024/07/nhs-financial-management-and-sustainability.pdf  The current government published the 2025/26 planning guidance at the end of January 2025, which is an improvement on the previous year, but still some way off the ambition of three months’ notice. ³⁸ NHS England, 2025/26 Priorities and Operational Planning Guidance, 2025, retrieved 1 November 2025, www.england.nhs.uk/long-read/2025-26-priorities-and-operational-planning-guidance

The NAO’s work also shows that ICBs report that NHS England pushes them to agree to undeliverable savings targets at the start of each financial year. Only five out of the 19 (26.3%) respondents to the NAO’s survey of ICBs felt that their savings targets were realistic and achievable when they were agreed with NHS England. ³⁹ Comptroller and Auditor General, NHS Financial Management and Sustainability, Session 2024–25, HC 124, National Audit Office, 2024, p. 30, www.nao.org.uk/wp-content/uploads/2024/07/nhs-financial-management-and-sustainability.pdf

The timing of pay awards in recent years has also made it difficult for trusts to plan effectively. Pay review bodies – the organisations that make independent recommendations about public sector staff pay – have typically returned their reports to government in July in recent years, more than a quarter of the way through the financial year. As previously discussed, those pay awards have tended to be both higher than inflation and higher than budgeted for. This has required trusts to increase in-year staff spending by more than expected. There was some progress in 2025, with pay review bodies publishing their reports in May. But this still leaves room for improvement, and the government should push pay review bodies to publish their reports before the start of the financial year.

This approach to financial management by recent governments makes it more difficult for trusts to plan effectively and likely contributes to some of the difficulties discussed above. The start of the first multi-year spending settlement since 2021 is the perfect opportunity for the current government to introduce more stability in the financial planning process. It is welcome, therefore, that NHS England published its Medium Term Planning Framework, ⁴⁰ NHS England, ‘Medium Term Planning Framework – delivering change together 2026/27 to 2028/29’, 24 October 2025, retrieved 30 October 2025, www.england.nhs.uk/publication/medium-term-planning-framework-delivering-change-together-2026-27-to-2028-29  which sets targets for the NHS for the next three years (2026/27 to 2028/29) in October 2025. This is an improvement in two ways. First, NHS England published it around five months before the start of the 2026/27 financial year. And second, it sets ambitions for three years, allowing trusts far more certainty over what is expected of them (though the NHS may revise these plans throughout the next three years).

Productivity

Since the end of the pandemic, activity in NHS hospitals has recovered slowly, despite substantially more funding and staffing. There are several reasons for this ‘productivity puzzle’, including:

•    historic underinvestment in capital

•    a lack of management capacity

•    increasing levels of inexperience among staff, among others. ⁴¹ Freedman S and Wolf R, The NHS Productivity Puzzle: Why has hospital activity not increased in line with funding and staffing?, Institute for Government, 2023, retrieved 1 November 2025, www.instituteforgovernment.org.uk/publication/nhs-productivity

Reduced productivity in hospitals should be of major concern to the government. The government spends more on hospitals than any other public service. A sluggish return to pre-pandemic productivity levels means the government has to spend much more to achieve the same outcomes. Given the tight fiscal environment and the desire for more spending on a wide range of other priorities, turning hospital productivity around should be a focus of the government’s health policy.

Hospital productivity is improving, though is still below pre-pandemic levels

For a while after the worst of the pandemic, hospitals struggled to return activity to pre-pandemic levels. That phase has now past and activity in elective care, diagnostics, cancer appointments and outpatient appointments is well above 2019 levels. But most activity is still behind where it should be if it had risen in line with staffing levels, and waiting times continue to be far longer for most care than in 2019. Activity and performance levels are discussed in more depth below.

The most recent annual release on public sector health care productivity from the Office for National Statistics (ONS) shows that productivity improved in 2022/23 compared with 2021/22: the quality-adjusted index rose by 1.8%. But by this measure, productivity was still 7.2% lower than in 2018/19 (the last year that was not affected by the pandemic) and approximately in line with the level between 2012/13 and 2013/14.

More recent data from an NHS England board meeting in February 2025 indicates that productivity has continued to improve. The productivity update report estimated that it increased by 2.4% in March to October 2024, compared to the same period in 2023. ⁴² Kelly J, ‘NHS productivity update’, NHS England, 6 February 2025, retrieved 1 November 2025, www.england.nhs.uk/long-read/nhs-productivity-update-feb-25  However, it also acknowledged that productivity at the end of that period was still 8% lower than in 2019/20.

There is no data about productivity at a subnational level. One (simplistic) way to assess whether the national picture is broadly similar around England is to compare changes in staffing levels at an ICB level to changes in different activity levels.

Figure NHS 3.7 plots the change in completed elective cases to changes in staffing levels between January 2020 and January 2025 by ICB.^* There were only three ICBs – Gloucestershire, Devon and the Black Country – where the number of completed elective cases grew more quickly than staffing levels. Fourteen out of the 42 ICBs (33.3%) saw declines in completed elective cases in that time, even though all ICBs employed more staff in 2025 than in 2019. In total across England, the number of completed elective cases rose by 7.0% compared to an increase in staff of 21.7%.

_{*    In the case of completed elective cases, this shows the level in activity in the previous 12 months at each date.}

Diagnostic activity fares better under this analysis. There were no ICBs that carried out fewer diagnostic tests in the year to January 2025 than in the year to January 2020. In this instance, 15 out of the 42 ICBs (35.7%) increased their diagnostic activity by more than their staffing levels. The England average increases in staffing and diagnostic activity were 21.7% and 19.1% respectively. The relatively fast increase in diagnostic activity compared to elective activity could be because the previous government chose to direct investment towards setting up community diagnostic centres, intended to increase the number of diagnostic tests carried out in the health service. ⁴³ Comptroller and Auditor General, NHS England’s Management of Elective Care Transformation Programmes, Session 2024–25, HC 766, National Audit Office, 2025, p. 26, retrieved 1 November 2025, www.nao.org.uk/reports/nhs-englands-management-of-elective-care-transformation-programmes

It was not necessarily the same areas that increased both diagnostic activity and completed elective cases by more than staffing. But of the three ICBs where completed elective cases outstripped staffing gains, two (Gloucestershire and the Black Country) achieved the same feat with diagnostic tests, with Gloucestershire having the third fastest and fourth fastest rate of growth in electives and diagnostics respectively.

Trusts are looking to make considerable efficiency savings in 2025/26

The Starmer government is clearly aware that it needs money to go further in hospitals. The 2025 spending review included ‘departmental efficiency plans’, which government departments signed up to for the period of the spending review. ⁴⁴ HM Treasury, Spending Review 2025: Departmental efficiency plans, GOV.UK, 2025, p. 8, https://assets.publishing.service.gov.uk/media/68492799d0ca5d7801e4e709/Efficiency_delivery_plans_-_supplementary_document_-_FINAL.pdf  If the Department of Health and Social Care (DHSC) is to be successful in delivering its efficiency plan, it will require hospitals to realise savings. To this end, the NHS has asked systems to “collectively deliver £11.0 billion” in 2025/26 – 7.1% of their total funding allocation. ⁴⁵ O’Mahony E, ‘Financial performance update’, NHS England, 28 May 2025, retrieved 1 November 2025, www.england.nhs.uk/long-read/financial-performance-update-27-march-2025  In comparison, they achieved £8.7bn worth of ‘efficiencies’ in 2024/25. The definition of ‘efficiencies’ that the NHS uses is unclear. An efficiency saving should be when a system can deliver the same level of performance for less money. But when speaking to trusts that have been tasked with delivering these efficiencies, they say that cutting spending by that much could harm performance, implying that they are not truly efficiencies, but rather cuts.

There are a range of ways in which trusts are looking to balance their budgets in 2025/26. One interviewee from an acute hospital trust reported a range of steps they intend to take. ⁴⁶ Institute for Government interview.  They said that as staffing is the largest spending line for every trust, it is the logical place where efficiency savings would start. The first target would be a reduction in agency and bank staff. This is in line with the government’s ambition to reduce temporary staffing. ⁴⁷ Streeting W and Mackey J, ‘Letter: Further action to reduce NHS spending on temporary agency staffing’, NHS England, 2 June 2025, retrieved 1 November 2025, www.england.nhs.uk/long-read/letter-further-action-reduce-nhs-spending-temporary-agency-staffing  But that is not costless. The interviewee argued that it would require permanent staff to cover gaps in rostering. They described a situation in which a senior nurse working as a ward manager might be required to cover half or three quarters of the work that a temporary staff member would have done, in addition to their own duties. They pointed out that this relies on goodwill from staff and increases the risk of burnout.

After agency staff, they reported that headcount reductions of permanent staff would be the next step. In the case of non-clinical staff working in hospitals, it appears as though many trusts have already implemented a hiring freeze, with the number of those staff remaining flat through 2024 and 2025 after years of growth, as discussed in more detail below. Reducing headcount only works as a true efficiency saving if performance is not harmed in the process. The Institute for Government has previously argued that the approach to cutting management and administrative staff that the coalition government took in the early 2010s was a false economy and ultimately contributed to worsening public service performance. ⁴⁸ Hoddinott S, Fright M and Pope T, ‘Austerity’ in Public Services: Lessons from the 2010s, Institute for Government, 2022, p. 7, www.instituteforgovernment.org.uk/sites/default/files/publications/austerity-public-services.pdf  Budgets are tighter now, and it is very likely that cutting non-clinical support staff would harm performance as an increasing ratio of clinical to non-clinical staff would likely mean that clinical staff have to spend more time carrying out administrative work.

The same interviewee said that they would then be looking to deliver existing services more efficiently. This would include things like increasing operating theatre use and investing in tools that streamline administrative tasks.

The interviewee ⁴⁹ Institute for Government interview.  described a “triangle of priorities” that trusts face, with financial control, quality of care and access to care at each corner. They said that it is close to impossible for a trust to be high achieving on all three priorities. If financial control becomes the priority for most trusts this year, it will likely lead to reductions in either access to or quality of care. This may make it more difficult for the government to achieve some of its objectives for the NHS. As an example, another interviewee ⁵⁰ Institute for Government interview.  pointed out that increased elective activity over the past year has been possible due to trusts spending more on things like weekend clinics and overtime for staff. They said that some of that may stop in a more constrained financial environment.

Bed numbers are growing slowly but occupancy remains at unsafe levels

Previous Institute for Government work showed that a lack of bed capacity in trusts was a key driver of poor patient flow and, ultimately, poor productivity. ⁵¹ Freedman S and Wolf R, The NHS Productivity Puzzle: Why has hospital activity not increased in line with funding and staffing?, Institute for Government, 2023, p. 15, retrieved 1 November 2025, www.instituteforgovernment.org.uk/publication/nhs-productivity  There was a long-term decline in general and acute bed numbers before the pandemic (these beds make up the majority of capacity in the acute hospital sector). Much of that was for good reason. The NHS managed to treat more people who would have been admitted to hospital as outpatients and reduce the average length of stay for those who were admitted to hospital. ⁵² Ewbank L, Thompson J, McKenna H, Anandaciva S and Ward D, NHS Hospital Bed Numbers, The King’s Fund, 2021, retrieved 1 November 2025, www.kingsfund.org.uk/insight-and-analysis/long-reads/nhs-hospital-bed-numbers  The number of beds dipped further during the pandemic, as the NHS reduced the number of available beds to reduce the spread of Covid in hospitals. ⁵³ Nuffield Trust, ‘Hospital bed occupancy’, Nuffield Trust, 26 June 2025, retrieved 27 October 2025, www.nuffieldtrust.org.uk/resource/hospital-bed-occupancy

The number of beds in NHS hospitals has increased steadily over the past few years, after falling to a historic low during 2020/21. General and acute bed numbers are now above pre-pandemic levels. There was an average of 105,694 general and acute overnight beds in hospitals in 2024/25, compared to 101,432 in 2019/20 – an increase of 4.2%. There was a slight dip again in the first quarter of 2025/26, with only 104,198 general and acute overnight beds available in that quarter. But despite that increase since the pandemic, bed occupancy in the sector remains stubbornly high, indicating that demand outstrips supply. There was no month in 2024/25 in which general and acute bed occupancy was below 91.4%, well above the Royal College of Emergency Medicine’s recommendation that trusts should have an upper ceiling of 85% for bed occupancy. ⁵⁴ Royal College of Emergency Medicine, ‘‘The system has to change’ as staff and patients continue to suffer due to overcrowding’, press release, 5 April 2024, retrieved 5 November 2025, https://rcem.ac.uk/press-release/the-system-has-to-change-as-ae-staff-and-patients-continue-to-suffer-due-to-overcrowding-rcem/

Some of that high demand for bed space is because there are large numbers of people who remain in hospital even though they are eligible for discharge. Throughout 2024/25, an average of 12,663 patients remained in hospital each day despite meeting the criteria for discharge – 12.0% of general and acute beds each day, on average. In September 2025, NHS England published details of the financial cost of each delayed bed day. It estimated that the NHS spent £220m on people who were in hospital but eligible for discharge in that month. ⁵⁵ NHS England, ‘Acute discharge situation report’, 2025, retrieved 1 November 2025, www.england.nhs.uk/statistics/statistical-work-areas/discharge-delays/acute-discharge-situation-report  If that is a typical monthly cost, then it implies the NHS spends £2.6bn per year on delayed discharge – only slightly less than the £2.8bn that the government spent on the courts and tribunals system in 2024/25.

High bed occupancy is remarkably consistent across England and across time. Throughout 2024/25, almost all trusts had bed occupancy averaging above 85%. Yet the Royal College of Emergency Medicine (RCEM) recommends that a safe level of bed occupancy is no more than 85%. ⁵⁶ Royal College of Emergency Medicine, ‘‘The system has to change’ as staff and patients continue to suffer due to overcrowding’, press release, 5 April 2024, retrieved 5 November 2025, https://rcem.ac.uk/press-release/the-system-has-to-change-as-ae-staff-and-patients-continue-to-suffer-due-to-overcrowding-rcem/  If occupancy exceeds this number, the RCEM argues there is no capacity for surges in demand, which leads to patients having to wait longer for care. ⁵⁷ Royal College of Emergency Medicine, ‘RCEM Explains: Hospital Beds MP Briefing’, blog post, {no date), retrieved 5 November 2025, https://rcem.ac.uk/rcem-explains/mp-briefing-rcem-explains-hospital-beds-june-2021/  Of the 124 trusts for which there is monthly data, only four had an average bed occupancy beneath the recommended safe level. Of those, three – Alder Hey Children’s NHS Foundation Trust, Birmingham Women’s and Children’s NHS Foundation Trust and Sheffield Children’s NHS Foundation Trust – are small, paediatric trusts with different patterns of demand. The remaining one – Harrogate and District NHS Foundation Trust – is a relatively small district hospital with only approximately 300 general and acute beds.

There is also very little variation between months. January and February 2025 were the joint highest months when occupancy reached 94.2%, compared to the annual average of 93.4%. The lowest average month was August 2024, during the summer lull, in which occupancy dipped to ‘only’ 91.4%.

There was also little variation across regions. The region with the highest average occupancy in 2024/25 was the Midlands with 94.2% and the lowest was the North East and Yorkshire with 92.1%.

No matter which hospital you walk into in England, in whichever month, you are likely to find that almost every bed is occupied.

The NHS is trying to use ‘virtual wards’ to reduce pressure on hospital beds

The previous government pushed the expansion of ‘virtual wards’, which are made up of ‘virtual beds’ as a way of alleviating pressure on physical beds. The NHS says that virtual wards “allow patients to get hospital-level care at home safely and in familiar surroundings, helping speed up their recovery while freeing up hospital beds for patients that need them most”. ⁵⁸ NHS England, ‘What is a virtual ward?’, (no date), retrieved 1 November 2025, www.england.nhs.uk/virtual-wards/what-is-a-virtual-ward  It means that a patient can return home from hospital or avoid admission in the first place, while staff in a hospital care for them.

The previous government set and met a target to roll out 10,000 virtual ward beds (approximately 10% of the general and acute bed capacity in hospitals) by September 2023. ⁵⁹ NHS England, ‘NHS delivers 10,000 virtual ward beds target with hundreds of thousands of patients treated at home’, 12 October 2023, retrieved 1 November 2025, www.england.nhs.uk/2023/10/nhs-delivers-10000-virtual-ward-beds-target-with-hundreds-of-thousands-of-patients-treated-at-home

There was continued growth in the number of virtual beds between September 2023 and September 2024, from 10,421 to 12,497, an increase of 19.9%. But since then, the number of virtual beds has stagnated somewhat. In September 2025, there were 12,522 virtual beds, an increase of only 0.2% on a year earlier and a drop of 2.4% from the high of 12,825 in March 2025.

However, occupancy has never reached the same level as physical beds. The highest level was 80.5% in January 2025, but the average occupancy between July 2023 and April 2025 was 72.0%. Occupancy has fallen further in recent months, dipping below 70% (68.2%) in September 2025.

One interviewee told us that the measure of virtual beds may be largely illusory. They pointed to the fact that an unoccupied virtual ward bed does not exist in reality, but is more a reflection of the estimated capacity of a hospital to accommodate patients who could occupy such beds. ⁶⁰ Institute for Government interview.  There are clear criteria that trusts have to meet to provide a virtual bed, which include the following:

•    There must be clear clinical oversight.

•    Staffing must be for a minimum of 12 hours a day.

•    Patients must have access to hospital-level treatments, diagnostic tests and pharmacies.

Uptake of virtual beds has varied across England. In 2024/25, Northamptonshire Integrated Care Board had the most virtual beds per 1,000 people at 0.42, and made use of a high proportion of those beds – 88.6% were occupied throughout 2024/25. At the opposite end of the scale, Birmingham and Solihull Integrated Care Board had 0.10 virtual beds per 1,000 people.

There has also been considerable variation in occupancy. Mid and South Essex Integrated Care Board reports that its 0.12 virtual beds per 1,000 people were occupied 100% of the time in 2024/25, though this is likely a reporting issue, as the NHS acknowledges that there are still some problems with data quality. In contrast, less than half (48.0%) of Somerset Integrated Care Board’s virtual beds were occupied on average throughout 2024/25.

It is unclear what effect virtual wards are having on hospital performance

At a macro level, there has been no identifiable decline in physical bed occupancy as the number of virtual beds increased. But that could be because there is such a large backlog in demand that as a physical bed is emptied when someone moves into a virtual bed, it is immediately backfilled. If that is the case, then it is a positive outcome as it means that someone who previously had unmet need is now being treated.

More thorough evaluations of the efficacy of virtual wards have returned mixed results. An NHS England evaluation found that every two and a half admissions to a virtual ward in the South East was associated with one less non-elective admission and generated a positive financial benefit for systems. ⁶¹ NHS England, ‘Summary of South East region virtual wards evaluation’, 16 May 2024, retrieved 1 November 2025, www.england.nhs.uk/long-read/summary-of-south-east-region-virtual-wards-evaluation  The same evaluation also pointed out that successful implementation is context-dependent, with more rural locations experiencing greater difficulties in having a single centralised team that works on virtual beds.

Another evaluation found that keeping someone in a virtual ward for one day cost the NHS £935, compared to £536 if they had been treated as an inpatient in a hospital bed. ⁶² Jalilian A, Sedda L, Unsworth A and Farrier M, ‘Length of stay and economic sustainability of virtual ward care in a medium-sized hospital of the UK: a retrospective longitudinal study’, BMJ, 2024, vol. 14, no. 1, p. e081378, retrieved 1 November 2025, https://bmjopen.bmj.com/content/14/1/e081378  The same report found that a patient treated in a virtual ward had a shorter length of stay in hospital on average, but was more likely to be readmitted and then had a lower chance of surviving than other patients.

One review of the published evidence shows that clinical outcomes – including around mortality and admission to residential care – either remained the same or improved in virtual wards compared to inpatient care. ⁶³ Normal G, Bennett P and Vardy E, ‘Virtual wards: a rapid evidence synthesis and implications for the care of older people’, Age and Ageing, 2023, vol. 52, no. 1, retrieved 1 November 2025, https://academic.oup.com/ageing/article/52/1/afac319/6974849

Work from The Health Foundation shows that the public are slightly more supportive than not of virtual wards, though that differs between different groups. Those in more deprived socioeconomic groups are less supportive than less deprived cohorts. ⁶⁴ Thornton N, Horton T and Hardie T, ‘How do the public and NHS staff feel about virtual wards?’, The Health Foundation, 29 July 2023, retrieved 1 November 2025, www.health.org.uk/reports-and-analysis/analysis/how-do-the-public-and-nhs-staff-feel-about-virtual-wards  And there seems to be high support for the approach among those who have used virtual wards. ⁶⁵ Lyndon H, Viney T and Slade V, ‘A service evaluation of virtual wards in Cornwall, UK’, Oxford Open Digital Health, 2025, vol. 3, retrieved 1 November 2025, https://academic.oup.com/oodh/article/doi/10.1093/oodh/oqaf008/8113185

The NHS Confederation points out in its review of virtual wards that effectively staffing virtual wards requires trusts to either hire more people or to reallocate existing staff, reducing capacity elsewhere. ⁶⁶ Hakim R, Realising the Potential of Virtual Wards, NHS Confederation, 2023, p. 20, www.nhsconfed.org/system/files/2023-05/Realising-the-potential-of-virtual-wards.pdf

In all, virtual wards seem to be a useful tool to deliver some types of intermediate care (either before admission to or after discharge from hospital). But the evidence does not indicate that in their current format they are an innovation that will fundamentally change the way the NHS delivers care. However, that is exactly what the government intends. One of the five “big bets” in the 10 Year Health Plan for England is that, by 2035, “wearables are your personal health custodians”, which will “enable virtual wards to become the norm for managing many conditions at home”. ⁶⁷ Department of Health and Social Care, 10 Year Health Plan for England: Fit for the future, CP 1350, The Stationery Office, 2025, p. 120, retrieved 1 November 2025, www.gov.uk/government/publications/10-year-health-plan-for-england-fit-for-the-future  This may become possible as trusts increase the adoption of existing technology and there are further advances that improve the quality of care. But as it stands, many of the barriers to wider use of virtual wards are related to sufficiency of staffing and the cost effectiveness and appropriateness of virtual wards compared to inpatient care.

Staffing

Some key staffing indicators are trending in the right direction

Doctor and nurse numbers continue to grow

The number of full-time equivalent (FTE) doctors and nurses working in hospital trusts^* grew between June 2024 and June 2025, by 5.1% and 2.8% respectively. This is a continuation of a trend of staff growth that started in 2019 with the introduction of the NHS Long Term Plan. ⁶⁸ NHS, NHS Long Term Plan, NHS England, 2019, retrieved 31 October 2025, https://webarchive.nationalarchives.gov.uk/ukgwa/20230418155402/https:/www.longtermplan.nhs.uk/publication/nhs-long-term-plan  The average annual growth of doctors and nurses between 2010 and 2019 was 1.8% and 1.9% respectively. Between 2019 and 2025, this rose to 4.8% and 4.7% respectively. In January 2025, there were almost a third more doctors and nurses working in hospitals than in January 2019 (32.1% and 31.5% respectively).

The growth rate for nurses has slowed in the past year. Growth of 2.8% is lower than any year other than 2021 since the start of the NHS Long Term Plan in April 2019. As discussed in more detail below, high leaver rates heavily affected 2021, while leaver rates are low in 2025, implying that trusts are slowing the rate at which they hire new nurses.

_{*    We exclude nurses working in community settings from these numbers. But we include doctors working in community trusts here because there is no easy way to separate them from the headline total.}

Fewer staff are leaving their posts than before the pandemic

In 2019, before the Covid pandemic, 10.7% of NHS staff left the NHS.^* There was then a decline in leaver rates during the first year of the pandemic, as staff opted not to leave the security of their current role or felt that it was important to continue working during the pandemic. From mid-2021 onwards, leaver rates spiked as pandemic restrictions eased and people who would have moved roles in 2020 did so in that year. At the peak of this post-pandemic movement, in the 12 months to September 2022, one in eight NHS staff left their role.

Since then, leaver rates have gradually fallen. In the 12 months to June 2025, the rate of all staff leaving their role was 9.9%, which was lower than any 12-month period since September 2010 except during the Covid pandemic (9.2%).

But there are differences between staff groups. Consultants have always left their posts far less frequently than other staff. But even by those standards, the 12 months to June 2025 were particularly low: 4.8%, lower than at any point, including the pandemic years.

In contrast, the rate at which ambulance staff leave their roles has increased since 2010. In 2010, 6.2% of ambulance staff left their posts, a rate comparable with consultants. In the 12 months to June 2025, this had risen to 9.4%, above nurses and health visitors and closer to the level for all staff in the NHS.

There are not particularly large differences between regions in leaver rates. At the lowest end, 9.4% of staff left their posts in the Midlands in 2024. At the highest end, 10.9% of staff left their posts in the South West.

This reduced leaver rate has happened despite continued dissatisfaction with pay, as discussed in more detail below. Lower leaver rates is a positive development for the NHS. As staff stay in post longer, they become more experienced and generally more effective at carrying out their work. This could therefore contribute to rising productivity rates over the coming years.

_{*    The NHS defines a leaver as someone “leaving… active service, including those going on or returning from maternity leave or a career break, for example”. Someone who moves between trusts within the English NHS is not counted as a leaver.}

Vacancy rates are at their lowest levels on record for some staff groups

As with leaver rates, vacancy rates – which is the number of full-time equivalent (FTE) vacancies as a proportion of the planned FTE workforce – fell and then rose during the pandemic. Since the peak in late 2022, they have been falling across the NHS.

Nursing vacancies were at the lowest level on record (including the pandemic) in the 12 months to June 2025, reaching 6.4%. The next lowest 12-month period before the pandemic was 10.6%, in 2017/18, when the NHS first published data for vacancies. The fall in medical vacancies has not been as dramatic as for nursing vacancies, but they too were at the lowest level (5.2%) than at any other point on record, including the pandemic. The lowest 12-month period before the pandemic was 2019, when medical vacancies sat at 7.8%.

It is difficult to interpret changes in vacancy rates. It could reflect the fall in leaving rates – as discussed above – or greater ease of recruiting staff. But it could also be because trusts reduced the number of staff that they planned to hire as their deficits increased throughout 2023/24 and 2024/25. This would tally with the slow down in the growth rate of nurses discussed above. Interviewees also pointed out that the NHS’s People Plan (published in 2020 and which set out how the NHS intended to improve culture and therefore the recruitment and retention of staff) had a particular focus on retaining nursing staff, which would in turn contribute to lower vacancy rates. ⁶⁹ NHS England, We Are the NHS: People Plan for 2020/21 – action for us all, 2020, retrieved 28 October 2025, retrieved 1 November 2025, www.england.nhs.uk/publication/we-are-the-nhs-people-plan-for-2020-21-action-for-us-all

The NHS is reducing its reliance on international staff

There has been a pattern in recent years of the NHS growing the workforce by recruiting international staff.^* This is true both in general practice and in hospitals. But this pattern shifted in 2024. Among both doctors and nurses, the growth in the workforce – measured as the number of joiners minus the number of leavers – has increasingly come from British staff, while the number of net joiners from the rest of the world has gradually fallen. In 2025, British staff overtook staff from the rest of the world as the largest source of net joiners.

_{*    This does not necessarily mean that the staff have come straight from their country of origin to work in the UK. Someone hired into the NHS from the ‘rest of the world’ may have been living in the UK for years before the NHS employed them.}

Among doctors, there were 8,011 net joiners in the NHS in the 12 months to June 2025. A net increase in doctors from outside the UK and European Economic Area (EEA) accounted for 44.4% of the increase in doctors. In comparison, in the 12 months to March 2023, almost the entire growth of the doctor workforce (94.0%) came from the rest of the world.

The pattern is even more extreme among nurses. In the 12 months to March and 12 months to June 2025, the net number of joiners from the UK exceeded those from the rest of the world for the first time since December 2020. At its peak, in the 12 months to December 2021, there were 14,740 net joiners from the rest of the world compared to only 2,279 in the 12 months to June 2025 – the lowest level on record.

The same broad trend was true for both nurses and doctors: as the rate of British staff joining and leaving the NHS fell and rose respectively between March 2021 and December 2023, the NHS hired more staff from the rest of the world. When those trends in British staff reversed from 2023, the NHS then reduced the number of staff it hired from the rest of the world.

There are likely several reasons for these changes. One could be a reversion to pre-pandemic trends. The NHS was a stressful place to work during the pandemic, and it could be that the NHS has become relatively more attractive for British staff as the pandemic has abated. Another explanation could be trusts’ financial pressures. When budgets are tight, trusts often choose to stop international recruitment first. This is because it is expensive to find and hire a staff member from outside the UK. Work from the Nuffield Trust shows that the up-front typical fees for hiring a nurse from abroad are between £10,000 and £12,000, a cost that trusts do not incur if they hire a British-trained nurse. ⁷⁰ Palmer B, Leone C and Appleby J, Return on Investment of Overseas Nurse Recruitment: Lessons for the NHS, Nuffield Trust, 2021, p. 8, www.nuffieldtrust.org.uk/sites/default/files/2021-10/1633336126_recruitment-of-nurses-lessons-briefing-web.pdf

In part, lower leaver rates reflect improved pay since 2022

In December 2022, before industrial action started in the NHS, pay for all staff in the NHS was 10.4% lower in real terms than it was in 2010. There was variation between staff groups. Nurses’ and health visitors’ pay had fallen by 12.2% and ambulance staff’s by 13.9%. Doctors’ pay had fallen by a lot more: 23.5% for consultants and 23.7% for resident doctors.

Pay has improved for all staff since then. But salaries for Agenda for Change^* staff – who include ambulance staff, nurses and health visitors – have not improved substantially. This is partly because they negotiated smaller pay rises to end their industrial action. But it is also because they accepted non-consolidated (that is, one-off) payments as part of their settlements, while doctors ensured that almost all their settlements were consolidated.

Pay changes since late 2022 have been inversely proportional to the amount of time that staff groups went on strike. Agenda for Change staff ended their industrial action quickly and received the lowest settlement, while resident doctors were persistent with their strikes for much longer, and have had the largest pay increase since 2022.

Despite receiving larger increases, doctors’ pay has still ended up lower in real terms compared to 2010 levels than that for other staff. In June 2025, resident doctors’ and consultants’ earnings were 15.3% and 19.4% lower in real terms respectively. Nurses’ and health visitors’ average earnings were 9.4% lower in real terms than they were in 2010. In comparison, earnings across the entire economy and in the private sector had increased by 3.2% and 4.4% respectively in real terms.

Industrial action by doctors lasted longer and was therefore more impactful on performance than that carried out by Agenda for Change staff. Resident doctors first walked out in March 2023 and only ended their industrial action after the new government came to power in July 2024. Consultants started striking in July 2023 and reached a deal with the Conservative government in April 2024. ⁷¹ Department of Health and Social Care, ‘End of NHS consultant strike action as government offer accepted’, press release, 5 April 2024, retrieved 1 November 2025, www.gov.uk/government/news/end-of-nhs-consultant-strike-action-as-government-offer-accepted

The British Medical Association (BMA) Resident Doctors Committee said that “the [2024] deal was not the end of the journey, but a chance to bank a pay uplift for doctors that represented a step towards our unchanged goal of full pay restoration”.^{**, 72} British Medical Association, ‘2024 pay deal for resident doctors working in England’, 2024, retrieved 1 November 2025, www.bma.org.uk/our-campaigns/resident-doctor-campaigns/pay-in-england/2024-pay-deal-for-resident-doctors-working-in-england  It followed through on that warning, winning a mandate for renewed strike action in July 2025. ⁷³ British Medical Association, ‘Resident doctors announce strike action in England’, press release, 9 July 2025, retrieved 1 November 2025, www.bma.org.uk/bma-media-centre/resident-doctors-announce-strike-action-in-england  Ongoing strike action poses a real threat to the government’s goal of reducing elective waiting times. Across all the resident doctors’ strikes, the NHS was forced to cancel more than 1.3 million elective appointments and procedures. But there is some evidence that these strikes may have been less impactful than previous rounds. Initial reports from the NHS indicate that fewer resident doctors walked out, resulting in fewer cancelled elective procedures. ⁷⁴ Campbell D, ‘NHS chiefs and BMA in row over patient safety during doctors’ strike’, The Guardian, 28 July 2025, retrieved 1 November 2025, www.theguardian.com/society/2025/jul/28/nhs-chiefs-and-bma-in-row-over-patient-safety-during-doctors-strike

_{*    These are all non-doctor staff who work in the NHS.}

_{**   Our estimates of the extent to which resident doctors’ pay has been cut since 2009 differs from those of the British Medical Association because it uses a different measure of inflation (the Retail Price Index) compared to us (the Consumer Price Index).}

The NHS also appears to have improved how it responds to resident doctors’ strikes, cancelling fewer appointments per doctor on strike in each consecutive walk-out. In the first resident doctors’ strike, starting on 13 March 2023, the NHS cancelled or rescheduled 2.0 elective appointments per doctor that walked out. That number fell relatively steadily throughout further strikes and reached a record low of 0.7 in the most recent strike, which started on 25 July 2025.

Strikes are still costly, however. Wes Streeting has claimed that resident doctors’ strikes in July 2025 cost the NHS £240m. ⁷⁵ Wells L, ‘BREAKING NEWS: BMA dismisses Streeting’s fresh offer to resident doctors’, Healthcare Management, 5 November 2025, retrieved 6 November 2025, www.healthcare-management.uk/breaking-news-bma-dismisses-streeting-makes-fresh-offer-resident-doctors-bid-stop-strikes  Those strikes are set to continue, as the government and the BMA failed to reach an agreement in early November 2025. ⁷⁶ Wells L, ‘BREAKING NEWS: BMA dismisses Streeting’s fresh offer to resident doctors’, Healthcare Management, 5 November 2025, retrieved 6 November 2025, www.healthcare-management.uk/breaking-news-bma-dismisses-streeting-makes-fresh-offer-resident-doctors-bid-stop-strikes  As a result, resident doctors will walk out for five further days from 13 November. ⁷⁷ Triggle N, ‘BMA rejects fresh offer to end doctor strikes’, BBC, 5 November 2025, retrieved 6 November 2025, www.bbc.co.uk/news/articles/cdrze1rzejlo  That will be the 13th time that resident doctors have gone on strike (either alone or with consultant colleagues) since March 2023.

Satisfaction with pay is still low among some staff groups

Staff satisfaction with pay dropped from 2020 onwards and particularly sharply among doctors in 2022. Since then, there has been some recovery. Total NHS satisfaction with pay increased slightly in 2024 compared to 2023: 32.0% up from 31.2%. But satisfaction among doctors increased substantially. Resident doctors’ satisfaction increased from 13.6% to 26.4% between 2023 and 2024, though this was still lower than any years apart from 2022 and 2023. Satisfaction among consultants increased by even more, by 20.3 percentage points, from 39.3% to 59.6%, returning their satisfaction to the level it was at in 2021, and higher than it was in 2018.

In contrast, satisfaction among nurses and midwives fell slightly between 2023 and 2024, from 27.7% to 27.1%, which was lower than any year apart from 2022.

Growth in the non-clinical NHS workforce has stopped since 2024

From the start of the NHS Long Term Plan in April 2019 through to the end of 2023, non-clinical (this includes management and admin staff, among others) hospital staff numbers grew at similar rates to those for clinical staff. In December 2023, there were 20.8% more clinical staff and 22.5% more non-clinical staff than in April 2019.

This pattern ended in 2024. Hospitals essentially stopped expanding the number of non-clinical staff, while clinical staff numbers continued to increase. There were only 0.1% more of the former in June 2025 than in January 2024, compared to 5.0% more of the latter.

Pausing non-clinical recruitment may have consequences for some of the government’s ambitions for hospitals. For example, the NHS hopes to reduce the elective waiting list by increasing the rate at which it removes patients who no longer need care. ⁷⁸ Institute for Government interview.  This is made easier with an effective and well-staffed administrative team who can, for example, follow up with patients to remind them to come in for appointments, or remove them from the waiting list if they no longer need care. They may also carry out administrative work that will otherwise fall to clinical staff.

This abrupt brake on non-clinical staff groups may reflect the financial pressures that trusts are under, as discussed in more depth above.

Staff absences due to sickness remain higher than pre-pandemic levels, partly due to growing mental health absences

Unsurprisingly, the pandemic led to accelerating rates of sickness absence in the NHS. 2021/22 and 2022/23 were the high points of sickness absence on record, when 5.4% of available staff days were lost to sickness absence. After a reduction in 2023/24 (4.9%), the rate of absence increased again in 2024/25, to 5.1%, higher than any year on record that was not affected by the pandemic. Rates of absence from sickness appear to be higher in the NHS than the rest of the economy. In 2024, 2.0% of available days were lost to sickness in the entire economy in England. ⁷⁹ Office for National Statistics, ‘Sickness absence in the UK labour market’, 4 June 2025, retrieved 1 November 2025, www.ons.gov.uk/employmentandlabourmarket/peopleinwork/employmentandemployeetypes/datasets/sicknessabsenceinthelabourmarket ^*

_{*    These rates may not be directly comparable.}

An increase in absences due to mental health issues partly drove the increase in absences in 2024/25 (1.36% of available staff days, up from 1.30% in 2023). This is higher than any year on record, including the pandemic. The ONS estimates that 0.3% of available days were lost to mental health reasons in the entire economy in 2024, although its data is not completely comparable to NHS data as it uses different categories of absence and its data is for the whole of the UK rather than just England. ⁸⁰ Office for National Statistics, ‘Sickness absence in the UK labour market’, 4 June 2025, retrieved 1 November 2025, www.ons.gov.uk/employmentandlabourmarket/peopleinwork/employmentandemployeetypes/datasets/sicknessabsenceinthelabourmarket  Mental health absences also make up a greater proportion of sick days in the NHS than in the wider economy: 26.4% compared to 13.8% in the UK economy in 2024.

There are large differences in rates of sickness absence between staff groups. Doctors (both consultants and resident doctors) have consistently lower rates of sickness absence. In 2024/25, 1.61% of consultants’ days and 1.96% of resident doctors’ days were lost to sickness absence, compared to 5.13% for all staff groups.

Nurses and health visitors tend to have slightly higher rates of sickness absence than the average for all staff: 5.49% in 2024/25.

In contrast, ambulance staff have by far the highest rates of sickness absence among the staff groups we looked at – 7.08% of their days were lost to sickness absence in 2024/25, almost two percentage points higher than the NHS-wide average.

Absence rates have increased for all staff groups both since 2016 and from 2019 (the year before the pandemic). Ambulance staff have had the largest increase, at 1.12% percentage points between 2019 and 2024/25.

If absence rates for all reasons across all staff groups had been the same in 2024/25 as they were in 2019, staff would have worked an extra 4.1 million days throughout the year – equivalent to an additional 18,200^* full-time equivalent staff a year (1.4% of the entire NHS workforce). The Labour government has identified reducing absence rates as a key contributor to productivity improvements.

In NHS England’s February 2025 board papers on NHS productivity, Julian Kelly (then chief financial officer of NHS England) said that reducing sickness absence was an enabler of productivity improvements. ⁸¹ Kelly J, ‘NHS productivity update’, NHS England, 6 February 2025, retrieved 1 November 2025, www.england.nhs.uk/long-read/nhs-productivity-update-feb-25  These improvements in productivity would come from:

•    staff working more consistently in their roles

•    trusts spending less on replacement staff to cover absences

•    reduced admin work due to not having to try to cover gaps in rotas.

Reducing sickness absence was also part of the DHSC’s plan for how it would deliver 4.5% efficiency savings by 2028/29. ⁸² HM Treasury, Spending Review 2025: Departmental efficiency plans, GOV.UK, 2025, p. 9, https://assets.publishing.service.gov.uk/media/68492799d0ca5d7801e4e709/Efficiency_delivery_plans_-_supplementary_document_-_FINAL.pdf  The NHS’s 10 Year Health Plan for England goes a step further and aims to “reduce sickness absence rates to the lowest recorded national average level (approximately 4.1%), saving a potential £200 million a year from reductions in temporary staffing costs”. ⁸³ Department of Health and Social Care, 10 Year Health Plan for England: Fit for the future, CP 1350, The Stationery Office, 2025, p. 105, retrieved 1 November 2025, www.gov.uk/government/publications/10-year-health-plan-for-england-fit-for-the-future  If the NHS had operated at this target in 2024/25, it would have meant that staff worked an extra 5.4 million days – equivalent to an additional 24,000 full-time equivalent additional staff working in the health service (1.8% of the entire workforce).

_{*    Assuming there are 254 working days in 2025 (net of bank holidays). Staff have a range of annual leave entitlements. ‘On appointment’ staff receive 27 days a year, staff who have worked for five or more years receive 29 days, and staff who have worked for 10 or more years receive 33 days. Assuming the middle value of these means there are 225 working staff days for each FTE NHS staff member.}
 

Rates of absence are lowest in London trusts and highest in more deprived parts of England

Rates of absence are not uniform across England. Of the 32 trusts in London for which we have data, 30 have a rate of absence that is below the national average. The trust in London with the highest rate of absence is the London Ambulance Service NHS Trust. That is because there are higher rates of absence among ambulance staff, as previously mentioned.

The North West has the highest rate of sickness absence, with all but four of the 31 trusts in the region for which we have data reporting rates above the national average.

Differences in population likely explain some of the differences. The ONS reports absence rates in the wider economy at a regional level across England. NHS regions are slightly different but similar enough for comparison. In both the NHS and the wider economy, London had the lowest sickness absence rates in 2024: 4.5% in the NHS and 1.5% in the entire economy. The North West had the highest rate in the NHS and the second highest rate in England (6.0% and 2.3% respectively). One of the largest discrepancies between the NHS and the wider economy was in the South West. That region had only the fourth highest level of absences in the NHS (4.9% compared to an NHS average of 5.1%) but the highest absence rate in the wider economy (2.4% compared to an England average of 2.0%). ⁸⁴ Office for National Statistics, ‘Sickness absence in the UK labour market’, 4 June 2025, retrieved 1 November 2025, www.ons.gov.uk/employmentandlabourmarket/peopleinwork/employmentandemployeetypes/datasets/sicknessabsenceinthelabourmarket

There is also a strong relationship^* between the deprivation of a trust’s catchment area – the pool of people most likely to attend a trust – and absence rates. This is for both total absences and for absences due to mental health reasons. There is only a moderate relationship between absences for non-mental health reasons and deprivation of the catchment area. The strongest effect is for all reasons, where each increase in the Index of Multiple Deprivation score^** for a trust’s catchment area is associated with a 0.07 percentage-point increase in sickness absence. The size of the effect is smaller for mental health absences (0.05) and smallest for other reasons (0.03).

There could be multiple reasons for these relationships. First, staff themselves are likely to live in more deprived areas when they work in a trust that has higher rates of deprivation among the patient population. Given the higher prevalence of poor physical and mental health among more deprived communities, ⁸⁵ Marmot M, Fair Society, Healthy Lives: The Marmot review: Strategic review of health inequalities in England post-2010, The Marmot Review, 2010, www.instituteofhealthequity.org/resources-reports/fair-society-healthy-lives-the-marmot-review/fair-society-healthy-lives-full-report-pdf.pdf  this finding could be because the health of NHS staff reflects the health of the communities in which they live.

Second, due to the poor health of patients in those areas, NHS staff’s work could be more complex and more stressful than it is in less deprived parts of England. This might increase the risk of burnout and rates of absence due to mental ill health. This could explain why there is both a closer and stronger relationship between absences due to mental ill health and deprivation than there is for absences due to other reasons.

_{*    The R-squared value shows the amount of variation in one variable: sickness absence rate in this instance. This is explained by the independent variable: deprivation of the trust’s catchment area. R-squared values range between 0 and 1. Across this report, we interpret R-squared relationships in the following way: 0 = no relationship; 0–0.1 = weak; 0.1–0.35 = moderate; 0.35+ = strong. From now on, we do not refer to the numerical value of the R-squared in the text, only the strength of the relationship. The R-squared value is observable in the charts.}

_{**   For reference, the lowest Index of Multiple Deprivation score for a trust is 9.8 and the highest is 39.3.}

Performance

The government has taken a welcome approach of slimming down the number of targets that it asks the NHS to achieve. This follows years in which the number set out in the annual planning guidance had slowly crept up, making it hard for systems to know what to prioritise and where to spend limited resources.

The NHS has missed major national targets for nearly a decade, with performance declining years before the pandemic hit the system. The last time the NHS hit cancer, A&E, elective and diagnostic targets was in December 2015, July 2015, September 2015 and November 2013, respectively.

That made the government’s pre-election promise to return to targets for all NHS performance metrics an incredibly stretching commitment. ⁸⁶ Illman J, ‘Streeting commits to hit four-hour target in first term’, HSJ, 19 June 2024, retrieved 6 November 2025, https://www.hsj.co.uk/policy-and-regulation/streeting-commits-to-hit-four-hour-target-in-first-term/7037339.article  But in January 2025, Wes Streeting, secretary of state for health and social care, admitted that it would be impossible for the NHS to meet “national cancer, A&E, diagnostic, mental health and ambulance waiting time targets by the end of this Parliament”. ⁸⁷ West D, ‘Exclusive: Govt abandons commitment to hit cancer, mental health and A&E targets’, HSJ, 31 January 2025, retrieved 27 October 2025, www.hsj.co.uk/acute-care/exclusive-govt-abandons-commitment-to-hit-cancer-mental-health-and-aande-targets/7038590.article  That leaves the elective waiting-time target (that 92% of patients should wait less than 18 weeks between a referral for elective care and the start of their treatment) as the major metric by which the government wishes to be judged on the NHS and it is the milestone used for its ‘NHS mission’.

Elective care

Even returning just elective care to the national target will require a major turnaround in hospitals. The last time that the NHS hit the target for elective care was in September 2015 and the NHS reached the lowest level outside of the pandemic and since 2007 (56.6%) in the December before Labour won the general election. The NHS has a long way to go before meeting the 92% target.

Elective performance is improving gradually

The government’s inheritance in terms of elective care was therefore bleak. There have been both headwinds and tailwinds to improving performance since the 2024 general election. An undoubted initial tailwind was that Streeting acted quickly to (in the end, temporarily) resolve the strike action among resident doctors – a bout of industrial action that had seen 1.2 million appointments cancelled or rescheduled. That resolution likely contributed to the recent modest improvement in elective performance.

On the other hand, the government inherited an NHS where priorities had already been determined for 2024/25, meaning that the first time Labour was able to set the parameters and targets for the service was in the financial year starting in April 2025. One interviewee told us that, for that reason, they would not expect there to have been much progress on elective performance between July 2024 and March 2025. They said that under the previous government, the system was directed towards reducing the longest waits, rather than 18-week performance.

April 2025 was the first month in which the current government’s planning guidance for the NHS came into effect. ⁸⁸ Institute for Government interview.  That planning guidance outlines an ambition to have 65% of cases waiting less than 18 weeks by March 2026, with all trusts improving by at least 5 percentage points compared to March 2025. ⁸⁹ NHS England, 2025/26 Priorities and Operational Planning Guidance, 2025, retrieved 1 November 2025, www.england.nhs.uk/long-read/2025-26-priorities-and-operational-planning-guidance

There have been small signs of improvement towards that interim goal. Since that low point of 56.6% in December 2023, performance has gradually ticked up. In August 2025, 61.0% of people were waiting less than 18 weeks. This is broadly the same level of performance that the NHS last achieved in mid-2022.

While improvement to 65% by March 2026 would be a promising first step, it would still require the NHS to improve performance by a further 27 percentage points over the following three years (assuming a spring 2029 general election) to meet the 92% target. That would mean an average annual improvement of 9 percentage points.

Various organisations have made estimates about whether the NHS will return to standard by the next election. The Institute for Fiscal Studies’ central estimate (which assumes that the NHS can increase the number of completed elected cases by 3.5% a year, compared to 2.7% per year between 2010 and 2019) forecasts that performance will have improved to 74% by mid-2029 ⁹⁰ Harvey-Rich O and Warner M, Can the Government Achieve its 18-week Elective Waiting Time Target?, Institute for Fiscal Studies, 2025, retrieved 1 November 2025, https://ifs.org.uk/publications/can-government-achieve-its-18-week-elective-waiting-time-target  – a large improvement, but still a good distance from the government’s stated ambition. In its work on the same topic, The Health Foundation estimates that if current trends of additions and removals from the elective waiting list continue between now and the next general election, the NHS will miss the 92% target, though it will be much closer than under the Institute for Fiscal Studies’ central estimate. ⁹¹ Opie-Martin S, Cavallaro F, Marszalek K, Gardner T and Tallack C, ‘One year on: is the government on track to meet its waiting times pledge?’, The Health Foundation, 25 September 2025, retrieved 1 November 2025, www.health.org.uk/reports-and-analysis/analysis/one-year-on-is-the-government-on-track-to-meet-its-waiting-times

18-week performance has worsened in all acute trusts since the 2019

There is a wide range in trusts’ 18-week elective waiting-time performance. In March 2025, half of acute trusts^* had between 55.4% and 63.9% of cases waiting less than 18 weeks for elective care, with the remaining half of trusts spread more widely between 46.2% and 79.6%. There is no acute trust that has managed to defy the decline in performance that the pandemic precipitated: the acute trust with the smallest decline in performance since December 2019 is Barking, Havering and Redbridge University Hospitals Trust, where performance was only 4.1 percentage points lower in August 2025 than in December 2019 (falling from 75.7% to 71.6%). In December 2019, one in seven acute trusts met the 92% target. None has done so in any month since November 2020.

_{*    With more than 5,000 completed elective cases in 2024/25.}

There is also variation in elective performance between regions. The highest-performing region in March 2025 was the North East and Yorkshire, where 64.8% of cases had been waiting less than 18 weeks. In comparison, the East of England was the worst-performing region, with only 53.6% of cases waiting less than 18 weeks for care in that month.

There is also considerable variation within regions. For acute trusts that had more than 5,000 incomplete cases in March 2025, the region with the largest range of performance was the South East, with a 26.0 percentage-point difference between its worst-performing trust (University Hospitals Sussex NHS Foundation Trust, 48.9%) and the best-performing trust (Royal Berkshire NHS Foundation Trust, 74.9%).

In March 2025, 22 out of 117 acute NHS trusts (for which there is elective data for that month) were already meeting the 65% target for 2025/26. But this does not mean that they do not need to improve performance in 2025/26. The NHS requires all trusts to improve elective performance by 5 percentage points in 2025/26 compared to March 2025. ⁹² NHS England, 2025/26 Priorities and Operational Planning Guidance, 2025, p. 18, www.england.nhs.uk/wp-content/uploads/2023/04/PRN01625-25-26-priorities-and-operational-planning-guidance-january-2025.pdf

The highest-performing acute trust in March 2025 was Northumbria Healthcare NHS Foundation Trust, where 79.6% of cases had been on the elective waiting list for less than 18 weeks. Jim Mackey, the interim chief executive officer of NHS England, previously ran this trust.

By August 2025, some trusts were making progress towards their target of a 5 percentage point increase in elective performance. Of the 117 acute trusts for which there is data, 63 (53.8%) improved their elective performance between March 2025 and August 2025, and seven (6.0%) had already exceeded the 5 percentage point target (though performance could worsen again before March 2026). The remaining trusts (40.2% of the total) have seen their performance decline. The trust with the largest improvement between March and August 2025 was the Shrewsbury and Telford Hospital Trust, where performance went from 48.1% to 58.8%, a 10.7 percentage point increase, albeit from a very low base.

Elective activity continues to increase above pre-pandemic levels

A patient leaves the elective waiting list under one of the following scenarios:

•    Their treatment begins.

•    A decision is made not to treat the patient.

•    The patient declines treatment.

•    The patient misses an appointment when it had been clearly communicated to them. ⁹³ NHS England, ‘National elective access policy’, 5 February 2025, retrieved 1 November 2025, www.england.nhs.uk/long-read/national-elective-access-policy

At this point, the NHS deems the case to be ‘complete’ and removes that person from the waiting list for that referral. The number of completed^* elective cases is therefore an indicator of a trust’s elective activity (though it does not capture the number of appointments or diagnostic tests required to complete the case, which could range from zero to many).

_{*    A case does not count as ‘completed’ if the person is removed from the waiting list during a ‘validation’ process (in which a trust goes through its list to ensure that everyone still requires care) or if they do not attend an appointment after clear communication.}

Elective activity was slow to recover after the pandemic. But since late 2023, activity has been above pre-pandemic levels nationally and in most trusts. Hospitals completed 18.1 million elective cases in 2024/25 compared to 17.1 million in 2019 – an increase of 5.8% (Figure NHS 3.31 shows the average monthly completed cases over the previous three months). But there does seem to have been a slight slowdown in growth in 2025. In the eight months to August 2025, the NHS delivered 1.4% more completed cases compared to the same time in 2024. In comparison, the number of completed cases grew by 4.3% in the same months between 2023 and 2024. 1.4% is also almost half the average annual growth rate between 2010 and 2019 (2.7%).

For non-admitted activity, completed cases are still a reasonably long way behind the pre-pandemic trend increases. If non-admitted activity had grown at pre-pandemic rates from February 2020 onwards, we would expect hospitals to have completed 15.9 million cases in 2024/25 compared to the 14.4 million that they did deliver. This means activity was approximately 9.6% lower in 2024/25 than if it had risen in line with pre-pandemic trends. This is despite the growth in staff numbers being well above pre-pandemic trends, as discussed above.

In contrast, the amount of completed admitted activity has surpassed the pre-pandemic trend, though there was a slight downward trend in that metric between mid-2015 and the start of the pandemic.

However, the level of completed elective case varies among trusts. Among the 98 acute trusts for which we have data, the trust at the 95th percentile of changes in elective activity between 2019 and the 12 months to August 2025 was Royal Devon University Healthcare NHS Foundation Trust, which increased activity by 49.5% in this time.^* More than a third (34.7%) of trusts completed fewer elective cases in the 12 months to August 2025 than in 2019.

_{*    This trust, as with others in this time series, is the result of a merger that happened during this period. We created a ‘synthetic’ trust for the analysis, which is the sum of the constituent trusts before the merger, to give a consistent time series for the newly formed trust.}

Additional activity is not translating into completed cases at the same rate as previously

One of the government’s key manifesto promises was to increase the number of appointments in the NHS by “40,000 per week” ⁹⁴ The Labour Party, ‘Build an NHS fit for the future’, 2024, retrieved 1 November 2025, https://labour.org.uk/change/build-an-nhs-fit-for-the-future  – or approximately two million per year. It was unclear at the time which activity would be included in the baseline for that target.

Since March 2025, the NHS has published a dataset called ‘additional elective appointments’, ⁹⁵ NHS England, ‘Recovery of elective activity’, NHS England, 2025, retrieved 1 November 2025, www.england.nhs.uk/statistics/statistical-work-areas/rtt-waiting-times/recovery-of-elective_activity-mi/  which details both the baseline of activity and the change since Labour came to power. This shows that the 40,000-per-week target includes operations, outpatient appointments and diagnostic tests that are all part of the elective pathway and contribute towards an elective case being completed.

This has allowed the Starmer government to claim that it increased appointments by 5.2 million between July 2024 and June 2025 – a 7.4% increase compared to the baseline.

It is difficult to know how much of this is attributable to actions taken since the general election. Activity tends to trend upwards in hospitals, as they hire more staff and treat a larger population. Full Fact has shown that the increase in activity since Labour came to power is slightly under the amount achieved in 2023/24 and in line with the increase in 2022/23 – though these were from low bases after a pandemic dip in activity. ⁹⁶ Benedictus L, ‘New data reveals rise in NHS appointments hailed by government is smaller than last year’s increase’, Full Fact, 23 May 2025, retrieved 27 October 2025, https://fullfact.org/health/labour-two-million-more-hospital-appointments

Potentially more worrying for the government is that its new dataset shows that additional activity is returning lower numbers of completed elective cases than it would have done in 2023/24. Between July 2023 and June 2024 – the current baseline for that activity – the NHS carried out 69.9 million elective appointments. In the same period, the NHS completed 17.7 million elective cases. In other words, it took 3.96 operations, outpatient appointments and diagnostic tests to complete a single elective case, on average. This number rose to 4.15 for the additional activity between July 2024 and June 2025, an increase of 4.9%.

This implies one or more of the following:

•    The NHS has become less efficient at completing elective cases.

•    Elective cases have become more complex (therefore requiring more care) over the past year.

•    As one interviewee suggested, trusts are reporting activity that they would have previously not recorded – with the government’s focus on this metric incentivising them to do so. ⁹⁷ Institute for Government interview.

If the first two reasons are behind the increase, then it means the NHS will have to carry out more activity to remove the same number of people from the elective waiting list.

There has been considerable policy focus on increasing staff numbers in trusts since 2019, as discussed above. But our analysis shows that there is almost no relationship between the level of activity or elective performance in a trust and the increases in staff, or the number of staff per patient in a trust’s catchment area. However, it could be that staff data is not granular enough to properly identify the effect of additional staff on elective pathways.

Surgical hubs and community diagnostic centres have proved to be an effective way to increase completed cases

Under the previous government, the NHS expanded the use of ‘surgical hubs’ to work through the backlog of ‘high-volume, low-complexity’ cases on the elective waiting list. ⁹⁸ NHS England, ‘Elective care improvement’, (no date), retrieved 1 November 2025, https://gettingitrightfirsttime.co.uk/hvlc/hvlc-programme  These cases include things like cataract removal and orthopaedic surgery.

Elective hubs are an innovation in which a trust dedicates resources exclusively to elective activity. This can be in the form of a separate physical location, a physically segregated section of an existing site or ring-fenced resources that sit within a site. ⁹⁹ Department of Health and Social Care, ‘Over 50 new surgical hubs set to open across England to help bust the COVID-19 backlogs’, press release, 26 August 2022, retrieved 1 November 2025, www.gov.uk/government/news/over-50-new-surgical-hubs-set-to-open-across-england-to-help-bust-the-covid-backlogs  The intention is to protect that capacity from being redirected towards emergency procedures, which would result in a hospital delaying the elective activity.

The NHS has gradually expanded the number of surgical hubs since 2021. It is difficult to know how many are now operating in trusts, as the NHS does not publish consistent data. But in August 2022, the then government said that there were 91 hubs operating across England. ¹⁰⁰ Department of Health and Social Care, ‘Over 50 new surgical hubs set to open across England to help bust the COVID-19 backlogs’, press release, 26 August 2022, retrieved 1 November 2025, www.gov.uk/government/news/over-50-new-surgical-hubs-set-to-open-across-england-to-help-bust-the-covid-backlogs  In 2024, The Health Foundation said that “as of September 2024, there were 108 surgical hubs operating across the country with a further 26 due to open by the end of 2025”. ¹⁰¹ Clarke G, ‘Surgical hubs: key to tackling hospital waiting lists?’, blog, The Health Foundation, 5 September 2024, retrieved 27 October 2025, www.health.org.uk/features-and-opinion/blogs/surgical-hubs-key-to-tackling-hospital-waiting-lists  NHSE told us that there are 123 surgical hubs operating in England as of October 2025.

A recent Health Foundation evaluation found that surgical hubs led to a statistically significant increase in elective activity. On average, high-volume, low-complexity elective activity was 21.9% higher in trusts that had newly opened a hub compared to the counterfactual in which they did not. ¹⁰² Co M, Marks T, Tracey F, Conti S and Clarke G, ‘The impact of elective surgical hubs on elective activity in acute hospital trusts in England: a generalised synthetic control study’, Nature Communications, 2025, vol. 16, no. 1, p. 6192, retrieved 1 November 2025, https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4888136  It also found that elective activity recovered more quickly after the pandemic in trusts that had an established surgical hub. ¹⁰³ Co M, Marks T, Tracey F, Conti S and Clarke G, ‘The impact of elective surgical hubs on elective activity in acute hospital trusts in England: a generalised synthetic control study’, Nature Communications, 2025, vol. 16, no. 1, p. 6192, retrieved 1 November 2025, https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4888136  While hubs may increase activity, it may not be as much as the NHS has hoped for. The NAO found that trusts with surgical hubs reported roughly half as much (52%) additional activity as planned between April 2023 and September 2024. ¹⁰⁴ Comptroller and Auditor General, NHS England’s Management of Elective Care Transformation Programmes, Session 2024–25, HC 766, National Audit Office, 2025, p. 29, retrieved 1 November 2025, www.nao.org.uk/reports/nhs-englands-management-of-elective-care-transformation-programmes

The previous government also expanded the number of community diagnostic centres to increase diagnostic activity carried out in the NHS. Many cases on the elective waiting list require some form of diagnostic test before the NHS can remove them from the list: according to NHS management information, the NHS carried out 1.2 diagnostic tests for each completed elective case between July 2024 and June 2025. ¹⁰⁵ NHS England, ‘Recovery of elective activity’, 2025, retrieved 1 November 2025, www.england.nhs.uk/statistics/statistical-work-areas/rtt-waiting-times/recovery-of-elective_activity-mi  The NAO found that there were 167 community diagnostic centres operating in England in August 2024, with the NHS intending that a further three be operational by March 2025. ¹⁰⁶ Comptroller and Auditor General, NHS England’s Management of Elective Care Transformation Programmes, Session 2024–25, HC 766, National Audit Office, 2025, p. 8, retrieved 1 November 2025, www.nao.org.uk/reports/nhs-englands-management-of-elective-care-transformation-programmes  It also found that diagnostic activity in community diagnostic centres was 2% above the weekly target between July 2021 and November 2024. ¹⁰⁷ Comptroller and Auditor General, NHS England’s Management of Elective Care Transformation Programmes, Session 2024–25, HC 766, National Audit Office, 2025, p. 8, retrieved 1 November 2025, www.nao.org.uk/reports/nhs-englands-management-of-elective-care-transformation-programmes

The NHS aims for 99% of people to receive a test within six weeks of being added to the diagnostic waiting list. As with most NHS performance metrics, performance on this had declined gradually from November 2013 before tumbling during the pandemic. After the pandemic, performance reached a low of 68.7% in December 2022. There has been gradual improvement since then, likely because of the introduction of community diagnostic centres (though it is hard to model the counterfactual). In February 2025, 82.5% of cases had been waiting less than six weeks. There has been some backsliding since then. In August 2025, 76.0% had been waiting for less than six weeks.

Increased use of specialist advice is contributing to lower additions to the waiting list

It is not just the amount of activity that trusts carry out that affects elective performance, but also the number of new cases that are added to the waiting list. The number of additions to the waiting list dropped during the pandemic and has never fully recovered to pre-pandemic trends.

There were 20.8 million additions to the waiting list in 2019 and 20.9 million in 2024, an increase of only 0.5% in total or 0.1% per year on average. In comparison, between 2016 (the first complete year for which we have data) and 2019, additions grew at a rate of 2.0% per year on average. If additions had carried on growing at pre-pandemic rates from 2019, there would have been 22.8 million additions in 2024/25 – 1.8 million or 8.6% more than there were.

One contributor to lower-than-expected additions is increased use of what is known as ‘specialist advice’. This is a mechanism that allows a professional (mostly, but not always, a GP) to consult a specialist doctor in secondary care before referring their patient for care in hospital. If the specialist does not believe that the referral should be made, the professional may then choose not to make the referral. This is known as a ‘diverted referral’. Diverted referrals reduce the number of additions to the waiting list. The amount of specialist advice given has increased rapidly in recent years, as discussed in more depth in the chapter on general practice.

In total, there were 2.6 million diverted referrals from specialist advice in 2024/25. As previously shown, there were 1.8 million fewer additions to the waiting list than there would have been if additions had followed pre-pandemic trends. Adding on these diverted referrals would therefore take the number of additions in 2024/25 above the level expected if they had followed pre-pandemic trends. But this is an imperfect comparison. There was still specialist advice before the pandemic that likely resulted in diverted referrals, though there is no data on this. It is also unlikely that every diverted referral would result in an addition to the elective waiting list.

Despite the imperfect comparison, it is clear that expanded use of specialist advice is holding down additions to the elective waiting list.
 

The NHS intends to expand specialist advice to further reduce additions to the elective waiting list

The government is trying to grow the use of specialist advice further. In 2025/26, the NHS will for the first time pay GPs for each request for pre-referral specialist advice. The stated aim is to “incentivise even closer working between general practice and secondary care and support the government’s commitment to move more care from secondary care into community settings… whilst also supporting elective recovery”. ¹⁰⁸ NHS England, ‘Changes to the GP contract in 2025/26’, 28 February 2025, retrieved 1 November 2025, www.england.nhs.uk/long-read/changes-to-the-gp-contract-in-2025-26  The NHS’s Medium Term Planning Framework lays out an ambition for integrated care boards and primary care to plan for how “greater access to specialist advice… could support GPs to manage more patients without the need for referral”. ¹⁰⁹ NHS England, Medium Term Planning Framework – Delivering Change Together 2026/27 to 2028/29, 2025, p. 26, retrieved 30 October 2025, www.england.nhs.uk/wp-content/uploads/2025/10/medium-term-planning-framework-delivering-change-together-2026-27-to-2028-29.pdf

As a result, the NHS hopes to increase the number of pre-referral requests from 2.9 million in 2024/25 to 4 million in 2025/26, a rise of 36% or an average of 2.6% per month – faster growth than at any point since 2019.

If the NHS achieves the target of 4 million requests for pre-referral specialist advice in 2025/26 and the same proportion result in diverted referrals as in the previous three years, it will mean 1.8 million fewer additions to the waiting list, compared to 1.3 million fewer in 2024/25.^*

Data for the first five months of 2025/26 shows that there were 1.4 million requests for pre-referral specialist advice during that time. This was a 19.9% increase on the same period in 2024/25 – but still around 220,000 requests (13.1%) beneath where it should be if all 4 million requests were spread evenly through the year.

_{*  This differs to the amount quoted above because this only refers to pre-referral specialist advice, rather than all specialist advice, as above. This is because the government has only set a target for pre-referral specialist advice.}

Far more patients are added to the waiting list in some integrated care boards than others

The amount of demand for elective care varies across England. The NHS added 0.33 cases to the elective waiting list per patient in the country in 2024/25. Of these, 0.30 cases were added to the waiting list of NHS providers and 0.03 were added to the waiting list of independent service providers (ISPs).^* But there was substantial variation around the country. At the upper end of the scale, 0.48 cases were added to the waiting list, per weighted patient,^** in North Central London Integrated Care Board. At the lower end, 0.19 cases per weighted patient joined the waiting list in Lincolnshire Integrated Care Board.

_{*    These are private providers of health care that the NHS contracts to provide care to patients. Patients themselves do not pay any money at the point of delivery.}

_{**   A ‘weighted patient’ is a metric the NHS uses to adjust the patient population for the complexity of care required to treat them.}

Use of the independent sector has declined from a post-pandemic high

As part of its elective reform programme (published in January 2025), the government said that it wanted to “strengthen the relationship with the independent sector” ¹¹⁰ Department of Health and Social Care, Reforming Elective Care for Patients, 2025, p. 22, www.england.nhs.uk/wp-content/uploads/2023/04/reforming-elective-care-for-patients.pdf  as a way of improving elective performance. This includes working with ISPs to expand independent sector capacity for both diagnostics and electives. ¹¹¹ NHS England, ‘Elective recovery: a partnership agreement between the NHS and the independent sector’, 6 January 2025, retrieved 1 November 2025, www.england.nhs.uk/long-read/elective-recovery-a-partnership-agreement-between-the-nhs-and-the-independent-sector

In many ways, this was a continuation of existing policy. The previous government also attempted to increase the use of ISPs to reduce the elective waiting list.

Elective activity ground to a halt in both NHS trusts and ISPs during the pandemic, but the decline in activity was larger in the latter than the former. However, activity also recovered far more quickly in ISPs: in the year to July 2024, they completed 35.2% more elective cases than in 2019. In comparison, NHS providers only increased their activity by 5.2% during the same period.

Despite the current government’s intention to make greater use of the independent sector, there has been a decline in the number of cases that ISPs have completed compared to the high of the year before the general election. In the 12 months to August 2025, ISPs completed 3.4% fewer elective cases than at the peak in the 12 months to July 2024. But that still means that they completed 30.6% more cases in that 12-month period than in 2019.

When looking at other ISP elective activity that does not result in the completion of an elective case, activity continued to rise throughout 2024 and 2025, while completed cases fell, as shown in the chart below.

An interviewee offered a few explanations for this change in ISP activity, all of which could be happening simultaneously. ¹¹² Institute for Government interview.

First, patients may be starting treatment in ISPs earlier than in other years. For example, someone might have previously had a test and an appointment with a consultant in an NHS trust, before they were then referred to an ISP who completed the pathway and they potentially had one follow-up appointment. Instead, it could be that in 2024/25, the patient was instead referred to the ISP who then carried out all the activity.

Second, there could be an increase in follow-up activity in ISPs. Follow-up appointments take place after a case is completed and would therefore not contribute to more completed elective cases. However, why this would happen is unclear.

Third, it could be that ISPs are improving the quality of the data they collect and therefore recording more activity that was previously unrecorded.

A fourth alternative explanation is that ISPs’ caseload is becoming more complex, therefore requiring more treatment for every completed case. This could either reflect increasing complexity of cases across the entire elective waiting list – something interviewees often reported, but which is difficult to prove – or that ISPs are diversifying away from their usual case mix of relatively straightforward cases.

Despite the large increase in ISP activity since 2019, independent activity remains a relatively small proportion of completed elective cases. In the year ending August 2025, ISPs carried out 9.0% of completed cases, up from 7.7% in 2019.

Use of ISPs has increased the most in London

There has been considerable variation in the use of ISPs around England since 2019. ISP activity grew fastest in London and the East of England between 2019 and the year ending August 2024, by 97.4% and 50.2% respectively. Activity also grew in all other regions, but with the slowest growth happening in the South East, where activity grew by 2.5% compared to the national average of 30.6%. That could be because the South East was the region that conducted the highest proportion of elective activity in ISPs in 2019 (12.6% compared to the England average of 7.7%, as shown in the chart below), meaning there was less capacity in the region to expand independent sector activity.

The overall 3.4% decline in ISP elective activity between the year ending July 2024 and the year ending August 2025 was similarly variable across England. The number of elective cases that ISPs completed rose in London and the South West in that time, by 31.0% and 1.1% respectively. It fell in every other region, with the largest fall happening in the South East: 9.6%.

When looking further into the data, a handful of ISPs are driving the activity increase in the sector in London. The first is Oviva Ltd. In the 12 months to August 2025, Oviva delivered 20,185 completed elective cases, even though it only operated in the final five months of that period. Oviva grew the number of completed cases month on month, reaching 11,283 in August 2025. This accounted for 8.0% of elective cases that ISPs in the whole of England completed and 0.8% of all completed elective cases including from NHS providers in that month. Oviva is a company that specialises in ‘weight loss medication and expert care’, ¹¹³ Oviva, ‘Weight loss medication and expert care with Oviva’, 2025, retrieved 27 October 2025, https://oviva.com/uk/en  likely meaning that it is a provider of ‘weight-loss jabs’ that the NHS started providing for the first time in 2025. ¹¹⁴ Gregory A, ‘NHS begins mass rollout of weight-loss jabs to patients in England’, The Guardian, 23 June 2025, retrieved 27 October 2025, www.theguardian.com/society/2025/jun/23/nhs-begins-mass-rollout-of-weight-loss-jabs-to-patients-in-england

The other major source of growth for London is a provider called Medefer. The number of completed elective cases that this provider delivered grew from 495 in 2023/24 to 14,948 in the year ending August 2025. But the attribution of those cases to London may be a reporting quirk. Medefer is a parent company, which operates 19 sites across England, but records all its elective activity through the parent company, which is registered in London.

As with the change in completed elective cases, there is variation in the proportion of cases completed in independent sector (as opposed to NHS sector) providers.

Even after rapid growth between 2019 and 2025, London still has the lowest proportion of elective activity carried out in ISPs at 5.2%, compared to the England average of 9.1%. This could explain why London has grown independent activity fastest since 2019: there was more room for NHS trusts to shift activity into the independent capacity in that region, combined with the fact that there has been a rapid increase in the use of a few providers that are registered in the capital. The South East is the only region where the proportion of elective activity carried out by the independent sector fell in this time.

The independent sector is more appropriate for some specialties than others 

The independent sector is typically more amenable to elective cases that are known as ‘high-volume, low-complexity’. These are things like cataract removals and hip and knee replacements, where treatment is relatively straightforward. This work is well suited to ISPs, because they can create a standardised process for these procedures, allowing them to complete a lot of work for as low a cost as possible, making them more profitable than complex activity, which is harder to standardise.

Before the pandemic, in 2019, the specialty that made up the greatest proportion of ISPs’ elective work was trauma and orthopaedics. This is where joint replacement procedures are recorded. This accounted for almost a third (29.9%) of ISPs’ elective work in 2019. In comparison, that specialty made up 9.6% of NHS providers’ elective work. Apart from ‘other’, ophthalmology (where cataract removals are recorded) accounted for the next highest proportion of ISPs’ elective work in 2019.

Since 2019, independent sector activity has not increased equally across all specialties. An increase in two specialties in particular – ophthalmology and dermatology – has driven the growth.

ISPs completed 187.2% more ophthalmology cases in the 12 months to August 2025 compared to 2019 and 115.6% more dermatology cases. Without those two specialties, ISPs completed 4.0% fewer elective cases in the 12 months to August 2025 than in 2019. Much of the increase in ophthalmology activity comes from rising numbers of cataract removal surgeries. There is no publicly available data splitting out ophthalmology activity, but the NHS estimates that the proportion of ophthalmology procedures that were cataract removals rose from 58% in 2017/18 to 65% 2023/24, ¹¹⁵ NHS England, ‘Proposed amendment to the 2025/26 NHS Payment Scheme – consultation notice’, 8 July 2025, retrieved 1 November 2025, www.england.nhs.uk/long-read/2526-nhsps-amendment-consultation  indicating an ever-greater focus on the more straightforward elective procedures.

After the rapid increase in NHS-funded ophthalmology activity taking place in the independent sector, ophthalmology became the specialty with the highest proportion of activity taking place in the independent sector, overtaking trauma and orthopaedics. In the 12 months to August 2025, ISPs completed 23.4% of NHS-funded elective cases in that time compared to 21.7% of trauma and orthopaedic cases. In comparison, the same levels were 9.4% and 20.8% in 2019.

When looking at the elective performance of those specialties that have become more reliant on the independent sector, there is a range of outcomes. Ophthalmology – the specialty that has shifted the highest proportion of work into the independent sector since 2019 – is also the specialty that had one of the smallest declines in elective performance between December 2019 and August 2025. In the former, 83.8% of ophthalmology cases were waiting less than 18 weeks, in the latter it was 69.8%.^* That decline is second only to elderly medicine, which was the highest-performing specialty in both December 2019 and August 2025 (94.6% and 83.6% respectively). Ophthalmology rose from being the 10th best performing specialty (out of 19) to the second best.

Dermatology (83.7% in December 2019 and 64.1% in August 2025) also rose up the performance rankings between 2019 and 2025, but by only three places. That means it was the eighth best performing specialty in August 2025.

From that mixed performance, it is difficult to conclude that high use of the independent sector alone leads to unusually good performance. Even if independent sector use was a guarantee of substantially improved performance, it will only ever remain a viable option for a small subset of specialities that are low complexity and profitable enough for those providers to take on. There is therefore only so far the independent sector can support the NHS in its bid to return elective performance to target by the end of this parliament.

_{*    This is the performance for all providers, in both the independent and NHS sectors.}

Urgent and emergency care

Under the previous government, the NHS experienced arguably the worst winter on record – the winter of 2022/23 – for urgent and emergency care. In the face of a pandemic productivity hangover, hospital flow slowed to a crawl. The NHS recorded its worst ever A&E waiting times in December 2022, when just 43.4% of people attending a major^* A&E were seen within four hours. Ambulances queued for hours outside hospitals and response times reached previously unknown heights.

That is not to say that pre-pandemic performance was strong. As with elective care, urgent and emergency care performance had been declining for years before Covid ripped through hospitals. The last time major A&E departments hit the target for 95% of patients to be seen within four hours was in June 2013.

_{*  We refer to type 1 A&E departments as ‘major’ in this report to avoid too much jargon. Type 1 departments are located in large hospitals, have a 24-hour service led by consultants and have full resuscitation facilities. They deal with the most serious/complex cases.}

A&E performance improved slightly in 2024/25

In 2023/24 – the last financial year before Labour won the general election – 56.9% of people attending a major A&E waited less than four hours for care, up from 49.6% in 2022/23. This rose slightly again in 2024/25, to 59.3%, though is still a long way off from the national standard of 95% and well below the rate of 70.4% in 2019.

Performance in major A&E departments has improved marginally in the 16 months since Labour came to power. In the 12 months to June 2024, 58.1% of people waited for less than four hours in major A&E departments. This rose to 59.7% in the 12 months to September 2025. In July 2025, the NHS also recorded its best four-hour performance since the height of the pandemic: 63.1%.

In recent years, the NHS has set targets below the national standard as intermediary targets. But the NHS was also a long way off achieving its reduced target (set by the Conservative government) in 2024/25 that 78% of patients should wait less than four hours in March 2025. The actual number was 60.9%. The current government has kept the same target for March 2026.

Performance on longer waits is also poor. Before the pandemic there was no month in which more than 5% of people attending A&E waited for more than 12 hours. Since July 2021, there has been no month in which less than 5% of attendances resulted in a 12-hour wait.

Performance on 12-hour waits has remained broadly stable since Labour came to power. In the 12 months before the July 2024 general election, 10.3% of attendees waited more than 12 hours in A&E. This remained steady in the 12 months to September 2025, with 10.3% of attendees waiting more than 12 hours.

A&E waiting times are much better in some hospitals than others

In 2024/25, trusts in the South East of England saw the greatest proportion of patients seen within four hours of their arrival at a major A&E: 63.1%. The North West was the worst-performing region, where 56.4% of people were seen within four hours.

However, there was wide variation in all regions. No region had less than a 19.7 percentage-point gap between its worst-performing and best-performing trust. In the North East and Yorkshire – the region with the widest range – there was a 47.4 percentage-point difference between the worst- and best-performing trusts.

Only five trusts had four-hour performance above 75% in 2024/25: Birmingham Women’s and Children’s NHS Foundation Trust, Sheffield Children’s NHS Foundation Trust, Alder Hey Children’s NHS Foundation Trust, Homerton Healthcare NHS Foundation Trust and Maidstone and Tunbridge Wells NHS Trust. Three of those five are the three specialist children trusts for which there is data in 2024/25.

Poor performance is not down to excessively high demand

The NHS often argues that high demand drives poor A&E performance. The logic is appealing: more people attending A&E means that the queue of people waiting grows, and people therefore wait longer. But there is little evidence for this at the national or trust level.

Growth in national attendances is below the pre-pandemic trend. There were 16.9 million attendances at major A&E departments in the 12 months to September 2025 compared to 17.5 million if attendances had grown in line with pre-pandemic trends. Attendances are therefore 3.4% lower than they would be if they had continued growing at the pre-pandemic rate.

While there is a moderate relationship between the number of attendances per person in a trust’s catchment area and a trust’s performance on the four-hour target, as the chart above shows, that relationship is positive, that is, the trusts that had more attendances per person had fewer people who waited more than four hours for care on average.

Instead, slow patient flow through hospitals is driving poor performance. ¹¹⁶ Freedman S and Wolf R, The NHS Productivity Puzzle: Why has hospital activity not increased in line with funding and staffing?, Institute for Government, 2023, p. 15, retrieved 1 November 2025, www.instituteforgovernment.org.uk/publication/nhs-productivity  As previously shown, bed occupancy remains so consistently high in hospitals that it makes it hard for A&E departments to admit patients or for ambulances to hand patients over to hospitals.

Ambulance performance is improving but remains poor

Most regions are missing the targets for ambulance response times

In relation to both category 1 and category 2 ambulance response times – for cardiac arrest and for serious conditions such as strokes, respectively – there was a large spike in ambulance response times in the winter of 2022/23.

Since then, response times have improved somewhat but are still on average above the target for both categories, which is seven minutes for category 1 and 18 minutes for category 2. In the 12 months before the 2024 general election, category 1 response times were eight minutes and 25 seconds on average and category 2 response times were 36 minutes and 21 seconds. These have both improved slightly since Labour came to power, to eight minutes and eight seconds and 33 minutes and 26 seconds respectively, but are still 16.1% and 85.8% higher than national targets respectively.

This is true nationally, but also for most regions. Between April 2021 and September 2025, no region met its target for category 2 response times in any month.

For category 1 response times, only two regions – London and the North West – have met the target for a mean response time of seven minutes since April 2021. In London, this has happened in six of the 14 months after the 2024 general election. For the North West, it only happened in one month (August 2025).

Despite being the best-performing region for category 1 response times in 2024/25 (an average of seven minutes and 23 seconds compared to a target of seven minutes), London was the third worst-performing region for category 2 response times (an average of 37 minutes and 41 seconds compared to a target of 18 minutes).

There were record ambulance handover delays in the winter of 2024/25

Handover delays have worsened since Labour won the general election in July 2024. The number of ambulance handovers that resulted in a delay of more than 30 minutes increased in 2024/25 compared to 2023/24, and was the highest on record: between the end of November 2024 and the end of March 2025, 31.9% of handovers resulted in a delay of more than 30 minutes. This compares to 28.7% in 2023/24 and 28.1% in 2022/23.

Throughout the winter of 2024/25, there was wide variation in the performance
of different regions when it came to ambulance handovers. The South East was consistently the best-performing region: only 14.7% of ambulance arrivals resulted in a delay of more than 30 minutes. In contrast, almost half (49.7%) of ambulance arrivals in the South West resulted in a delay of more than 30 minutes. That reflects the South West’s performance more widely. That region is the worse performing for handover delays, and category 1 and 2 response times.

Cancer care

Waits for treatment are improving

There has been an improvement in the proportion of people starting treatment within two months of an urgent cancer referral: in 2024/25, 68.4% of patients started treatment within two months, compared to 64.7% in 2023/24, an increase of 3.7 percentage points. However, this is still some way off from the national target of 85%.

Interviewees attributed improved performance to a few factors. First, they argued that the national focus on two-month waits between referral and treatment has meant that the system has dedicated time and resources to meeting the target. Partly, this focus could have been the result of the previous government’s decision to cut the number of cancer targets. ¹¹⁷ Bloch B, ‘Ministers confirm plans to scrap two-thirds of cancer targets by autumn – but new Faster Diagnosis Standard has never been met’, Sky News, 17 August 2023, retrieved 27 October 2025, https://news.sky.com/story/ministers-confirm-plans-to-scrap-two-thirds-of-cancer-targets-by-autumn-but-new-faster-diagnosis-standard-has-never-been-m…  Second, interviewees thought that falling demand, as discussed below, contributed to improved waits. Third, they pointed to implementation of the ‘faster diagnosis standard’ – a target introduced under the Johnson government in 2021, which aims for 75% of patients to be diagnosed with cancer or have cancer ruled out within 28 days of an urgent referral ¹¹⁸ NHS England, NHS Cancer Programme: Faster Diagnosis Framework, 2022, p. 4, www.england.nhs.uk/wp-content/uploads/2019/07/B1332-NHS-Cancer-Programme-Faster-Diagnosis-Framework-v5.pdf  – as contributing to improved two-month waiting times. This makes sense. If the NHS is diagnosing cancers more quickly, then hospitals will be able to start treatment earlier.

Since the NHS introduced the faster diagnosis standard, hospitals have improved performance against that target. In 2021/22 (the first year the NHS introduced the standard), 72.0% of patients were given a definitive answer within 28 days. This dipped in 2022/23, to 70.2%, and rose to 76.4% in 2024/25, meaning the NHS hit its target in that year. Performance dipped slightly in the first five months of 2025/26, to 75.9%, but remained above target.

Falling waiting times are helped by flattening numbers of urgent referrals for cancer care

Cancer care was one of the stand-out areas of NHS performance after the pandemic, with activity rapidly rebounding to pre-pandemic trends. In the 12 months to August 2025, hospitals carried out 3.1 million urgent cancer appointments (Figure NHS 3.53 shows average monthly amount over the previous three months). That was almost a third (32.2%) higher than in 2019. But the growth between 2023/24 and 2024/25 (3.7%) was below the average annual growth rate of 10.4% between 2010 and 2019. Also, if increases in urgent cancer appointments in the 12 months to August 2025 had followed pre-pandemic trends, the NHS would have carried out 3.8 million appointments in that time, not 3.1 million. This means that appointments are 15.0% lower than expected.

It is difficult to tell why this flattening has happened or whether it is a positive or negative development. It could be that cancer detection has declined, while there is still a need for more appointments. It could also be that there are lengthening waits for cancer appointments. Or it could be that it reflects a genuine flattening of demand for cancer care. But interviewees reported that it reflects a reduction in referrals for cancer care more than a lengthening of the wait for an appointment. The improvement in the two-month target, whereby people should start cancer treatment within two months of their referral, described above, supports this: if people were waiting longer for a first cancer appointment after their referral, it would be less likely that they would be diagnosed within four weeks or be able to start treatment within two months of their referral. It also seems unlikely that detection rates have declined.