Hospitals

Spending on hospitals has risen since 2009/10, as has the size of the workforce. But the amount of work that hospitals are doing has risen faster. Hospitals have made efficiencies – maintaining the clinical standard of care in the face of growing workloads – but these have not been enough to keep up with the growing cost of, and demand for, care. This has resulted in large financial deficits and longer waiting times for treatment.

There are several different types of hospitals that provide different types of care, such as community, mental health and acute hospitals. These are run by trusts and are paid to provide a certain type of treatment within a given geographical area. Most hospitals are defined as being ‘acute’ – meaning they provide active, but short-term and specific treatment. This involves emergency treatment for those with moderate to serious need, surgery, and specialist diagnostic and treatment services for a wide variety of conditions. It is these hospitals that this chapter focuses on. As health care is devolved to the individual nations, this chapter looks at England.

Spending on NHS providers has increased by 15% since 2009/10

Figure 1.7 Change in spending on NHS providers in England (real terms), since 2009/10

Spending on hospitals appears to have risen over the past eight years – although we cannot isolate spending on acute trusts across this period. In real terms, spending on NHS providers rose by 15% between 2009/10 and 2017/18 (reaching £82.4bn), although there was only 0.1% growth between 2016/17 and 2017/18. These figures include day-to-day spending on mental health, ambulance, community and specialist care NHS providers, as well as acute trusts.[1]

Unfortunately, Department of Health accounts do not separately identify what spending is specifically for hospitals. However, we do know that acute providers account for the majority of all NHS provider spending. NHS Improvement (which oversees trusts) has carried out some analysis, which shows that of the £72.7 million(m) income from patient care activities across all providers in 2017/18, £53.0m (about 73%) was income for acute services.[2] There is also evidence that spending on acute providers has been growing more than spending on other types of providers. The King’s Fund has estimated that the percentage increase in acute funding has been three times greater than the rate of growth of mental health funding since 2012/13.[3]

Despite spending increasing, the entire NHS is experiencing the slowest decade of spending growth of its existence. During the Coalition Government, the average annual growth rate of NHS spending was 1.1%; this increased to 2.3% between 2014/15 and 2016/17, but was still below the average of 3.7% throughout the history of the NHS.[4]

Demand for hospital services has continued to grow

An analysis of demographic change suggests that demand for hospital services is rising. The number of people living in England grew by 6.6% between 2009 and 2017.[5] But demand for hospital services is likely to have risen more than this because of the ageing population: the number of people over the age of 65 grew by over 19.4% in this same period.[6] The over-65s make up a greater proportion of health spending; for example, in 2016/17 more than 40% of finished consultant episodes were for patients aged over 65, despite representing 18% of the population.[7]

Meanwhile, long-term health conditions have remained prevalent, with 41% of the population estimated to have a longstanding illness in 2016.[8] Since 2009, life expectancy has grown more than healthy life expectancy and disability-free life expectancy, meaning that people are living longer in poor health.[9] Treating patients with multiple chronic conditions is estimated to account for over 70% of the health budget.[10]

Figure 1.8 Change in estimated demand for acute care, since 2009

We can model the changing demand for hospital services by applying an estimate of the need for acute services in different parts of the population, produced by the Department of Health in 2011, to the changing demographic profile of England.[11] This suggests that population-level demand for acute care grew by 9.5% between 2009 and 2017.

Input: the number of clinical staff working in hospitals is rising

Figure 1.9 Change in the total number of doctors and nurses (full-time equivalent) in NHS trusts and clinical commissioning groups, since September 2009

Staff costs make up the majority of NHS providers’ expenditure. In 2018, £52.0bn of the total £82.4bn of their spending was on staff.[12] Between September 2009 and May 2018, the size of the NHS workforce in hospital and community health services (measured as full-time equivalents) rose by 6.0%, comparable to England’s population increase. However, there was significant variation in the growth rate of different groups of workers over different time periods (for example, the number of managers and senior managers dropped by 13.0% over this period).

Since 2009 there have been greater increases in the clinical workforce than the overall workforce. Generally these increases did not take place before 2012, but rising demand and new guidance from the National Institute for Health and Care Excellence (after the Francis Review into Mid Staffordshire NHS Foundation Trust found that inadequate staffing compromised patient safety)[13] encouraged hospitals to maintain a steady pace of clinical recruitment from that time. Between September 2009 and May 2018, the number of doctors and nurses increased by 13.2% and 10.6% respectively, well ahead of our calculated rate of growth of demand (although this growth rate appears to have slowed in recent years).*

Efforts to recruit enough staff quickly to meet rising demand safely also led to the increased use of expensive agency workers, the cost of which reached a peak of £3.6bn in 2015/16. Following a concerted effort by NHS Improvement, this spending has begun to decline: between 2016/17 and 2017/18, spending on agency staff fell from £2.9bn to £2.4bn (a real-terms fall of nearly 20% and a return to 2009/10 spending levels).[14]

However, there are signs that hospitals are struggling to maintain the size of workforce that they need. According to NHS Digital, full-time equivalent vacancies within acute trusts increased from 53,236 between July and September 2015, to 59,609 between July and September 2017, a 12.0% increase. Vacancies for medical and dental staff rose by 16.3%, and those for nursing and midwifery staff by 23.4%.[15]**

Much attention has been given to issues with recruiting nurses, after nurse training bursaries were scrapped in 2017 and replaced with loans (with the Government arguing that this would allow it to save around £800m a year, and increase the number of training places).[16] Between 2016 and 2017, the number of applicants for UK nursing courses dropped significantly, by 19.6%. However, the number of applicants accepted on courses fell far less: acceptances were 3.8% lower 15 days after A-Level results were announced, and just 0.9% lower by the end of the recruitment cycles. A similar pattern has occurred in 2018, where the number of applicants was 27.6% lower than in 2016, but the number accepted 15 days after A-Level results was just 5.8% lower.[17] The decline in nurse trainees is therefore not as dramatic as one might expect, but it is still the case that the Government is failing to increase the number of student nurses – one of the stated purposes of scrapping the bursaries.

There are also concerns about the implications of new language requirements and Brexit for international recruitment. The number of workers registering with the Nursing & Midwifery Council for the first time decreased by 12.3% between 2016/17 and 2017/18. While there was a 13.4% increase in those registering from overseas countries that are not part of the European Union (EU), there was an 87.4% decrease in those registering from the EU/European Economic Area.[18]

Staff are also leaving the NHS in increasing numbers: voluntary resignations in the NHS rose from 74,285 in 2011/12 to 114,870 in 2017/18. Workload appears to have been a key driver of this. The number of people resigning and citing work-life balance as the reason more than doubled in this period, from 6,699 to 18,013.[19] The NHS Staff Survey has also found that the percentage of staff in acute trusts reporting feeling unwell because of work-related stress rose from 27.9% to 37.0% between 2010 and 2017.[20] At the same time, throughout this period, hospitals have been able to hold down spending on staff through the public sector pay cap.

The rest of the money that hospitals spend goes on items like medicines, equipment, catering and maintaining buildings. It is less clear what has happened to these other individual items – whether the cost of buying them has risen or fallen, placing or relieving pressure on hospital budgets.

The NHS produces an estimate of the overall spending on medicines in hospitals, which increased from £4.17bn in 2010/11 to £8.3bn in 2016/17 (a real-terms increase of 80%).[21] However, it is not possible to determine what has caused this increase – whether more or different medicines have been prescribed, or whether the cost of each product is rising – as data on volume, product mix and prices is not publicly available. Indeed, this change in spending on medicines is an estimation based on list prices, and may not reflect the prices that the NHS actually paid: the NHS receives discounts but keeps the terms confidential.[22] If we were able to take these discounts into account, we might observe a different rate of spending change.

However, there is evidence that the overall cost of providing care is increasing. Health inflation has increased at a greater speed than consumer inflation, with the Department of Health estimating that the price of health inputs rose by 10.5% between 2009/10 and 2015/16 (compared with about 9.5% economy-wide inflation).[23] The Personal Social Services Research Unit at the London School of Economics has estimated that, between 2009/10 and 2016/17, national average costs per care episode increased by 27% for elective inpatient stays, by 20% for non-elective long stays and by 7% for non-elective short stays.[24] Reference costs (which provide a measure of the average unit cost to the NHS of providing a particular service) also suggest that the cost of care has increased, particularly for Accident & Emergency (A&E) care (with an estimated 22.1% increase in real terms between 2012/13 and 2016/17).[25]

*    We have tried to isolate staff working in acute settings from the data, but these figures will contain some staff working in other types of hospital. The ‘doctors’ category consists of consultants (including directors of public health), associate specialists, specialty doctors, staff grades, specialty registrars, foundation doctors and hospital practitioners/clinical assistants. The ‘nurses’ category consists of nurses (adults) and nurses (children).

**    NHS Digital has added a caveat that this information is provisional and experimental.

 

Output: hospitals are doing much more work

Figure 1.10 Number of emergency admissions via type 1 Accident & Emergency, since 2009/10

Hospital activity – the things that hospitals do, such as performing treatments and tests – has risen rapidly since 2009/10. For example, the number of people admitted to hospital via A&E for type 1 emergency admissions* rose from 3.36m in 2009/10 to 4.4m in 2017/18 (a 30.9% increase).

Significant increases in activity have taken place across all hospital services, continuing a long-running trend that began before 2009.[26] Between 2009/10 and 2017/18, elective admissions and non-elective admissions both increased by nearly 20% (to 8.2m and 6.2m respectively). Meanwhile, the number of outpatient appointments rose by 41.8% (to 119.4m) and the number of diagnostic tests rose by over 50% (to 21.9m).[27]

It is notable that across these measures, the number of people receiving treatment in hospital (including the number turning up at A&E in need of treatment) is rising by more than the demographic pressures predicted by our model of population-level demand. There are several possible explanations for this. The first is the rise in the prevalence of chronic medical conditions, which increase the number of repeat attendances. The Health Foundation (an independent health think tank) has found that, in 2015/16, one in three patients admitted as an emergency had five or more health conditions, a significant increase from one in ten in 2006/07.[28] Similarly, in 2017 it was estimated that 77% of inpatient bed days were for people with long-term conditions.[29]

Another possibility is that there has been an increase in people accessing hospital services when another care setting would be more appropriate. For example, the rate of emergency admissions for acute conditions that should not usually require hospital admission increased from 1,083 per 100,000 in 2009/10 to 1,359 per 100,000 in 2016/17.[30] Finally, in some instances it is possible that medical improvements and changing patient expectations have driven unnecessary uses of treatment, such as unnecessary prescriptions or diagnostic tests.[31]

*    Referring to a major A&E department, with a consultant-led, 24-hour service. This does not include walk-in centres.

 

Output: clinical quality and patient satisfaction have been maintained

Figure 1.11 Overall patient experience scores for adults treated as inpatients, since 2009/10

The data suggests that the standard of hospitals’ work has remained good. Patient satisfaction scores are high, fluctuating between 75.6% and 77.3% in the period from 2009/10 to 2016/17. In 2017/18 the introduction of new methodology recorded a score of 78.4%.* The June 2018 Friends and Family Test (a survey of how likely patients are to recommend the health service they have used) also found that 95.9% of inpatients would recommend the hospital service and 87.4% of patients would recommend the A&E service (comparable to scores of 94.3% and 88.2% respectively in June 2013).[32]

In contrast, satisfaction scores in the British Social Attitudes (BSA) survey have remained high, but have declined slightly. Satisfaction with the NHS was at 57% in 2017 (compared with 64% in 2009), with satisfaction for inpatient appointments at 55% (from 59%), outpatient appointments at 65% (from 67%) and A&E attendances at 52% (from 59%). This can perhaps be explained as the difference between perceptions and experience of care.[33]

Clinical indicators suggest that the safety of care has been maintained or is improving. Across all care settings, the number of health care associated infections, such as MRSA and C. difficile fell between 2009/10 and 2017/18 (from 1,898 to 846, and from 25,604 to 13,286, respectively), despite an increase in the number of patients treated.[34] In 2017, 13 of 134 trusts reported a mortality indicator higher than expected (Summary Hospital-level Mortality Indicator – SHMI), only a slight increase from 10 of 143 in

2011.[35] And finally, between August 2012 and June 2018, the proportion of patients deemed ‘harm free’ in acute hospital point-of-care surveys increased from 91.6% to 94.2% (meanwhile there have been declines in rates of new pressure ulcers, new venous thromboembolism and patient falls).[36]

Despite these improvements, hospitals do not perform as well as other health settings (such as general practice) during Care Quality Commission (CQC) inspections. Following its inspections of 265 hospitals between 2014 and 2016, the CQC found that 52% ‘required improvement’ and a further 6% were ‘inadequate’.[37]

*    It should be noted that these scores refer to those already receiving care. The data does not include the experiences of those queuing for services and still waiting to be admitted.

 

Have hospitals become more efficient and can that be maintained?

Although spending in hospitals has increased, it appears that outputs have risen by more. Taking this and the rising costs of providing health care into account, we can likely conclude that efficiencies have taken place in hospitals since 2009/10. Most recently, NHS Improvement has estimated that NHS providers on average used their resources 3.7% more efficiently in 2017/18 than in 2016/17, having made efficiency savings of the same magnitude the year before as well. However, this was below the target of 4.3% efficiency savings a year and a quarter of the savings were non-recurrent (meaning they cannot be expected to produce further efficiencies).[38]

In NHS England’s Five Year Forward View in 2014, ambitious targets were set for productivity improvements of between 2% and 3% a year, against the long-run average of 0.8%.[39] Our analysis does not allow us to put a precise number on hospital productivity improvements, but the Office for National Statistics has estimated that the total productivity of public service health care as a whole increased by 8.6% between 2009 and 2015. This suggests an annual average growth rate of 1.4%, which is above the long-run average, but below the targets set out in the Five Year Forward View. The rate of growth may be slowing (with only 0.1% growth in 2015 according to the Office for National Statistics).[40] It is also likely that many of the gains have been made outside of hospitals and in non-acute care (which have seen greater reductions in inputs, such as staff).

Economy in staff costs has played a big part in achieving efficiencies. Although health inflation overall rose by 10.5% between 2009/10 and 2015/16, this was largely driven by rising prices. Pay cost inflation only increased by 6.3% (of which 3.1% occurred between 2009/10 and 2010/11), below the 9.5% increase in the economy-wide inflation.[41] The Nuffield Trust has estimated that by letting inflation overtake wage growth, in 2016/17 the NHS made £2.6bn in savings (more than the national spending on ambulance services).[42] We can also assume that labour productivity has increased, as staff numbers have risen at a slower rate than activity.

However, these gains may not be possible to sustain. The NHS Staff Survey has recorded an increase in work-related stress (from 27.9% to 37.0% between 2010 and 2017), as well as a decline in pay satisfaction. In 2010, 2.9% more acute staff reported being happy with their pay than unhappy. By 2017, the balance had flipped, with 14.0% more respondents saying they were unhappy than happy.[43] In the face of significant pressure, the Government lifted the longstanding pay cap in March 2018 for all health care staff (excluding doctors) and agreed a fully funded pay rise from 2018/19.[44] Subsequent announcements in July 2018 mean that doctors will now also receive a pay rise.[45]

The length of time that patients spend in hospital has fallen (continuing a decades-long trend). Medical advances have enabled the average stay of overnight patients to decrease from 5.6 days in 2009/10 to 4.9 days in 2015/16.[46] This is a productivity improvement, which helps to reduce the need for beds. Reflecting this change, the number of acute and general beds declined by 4.7% between 2010/11 and 2017/18 (with a 6.3% reduction in overnight beds, but an 11.0% increase in day-only beds).[47]

However, there is evidence that this approach cannot produce further efficiencies. Between 2010/11 and 2017/18, bed occupancy rates for general and acute care rose from 82.4% to 86.0% for day beds, and from 87.1% to 90.4% for overnight beds.[48] Occupancy rates over 85% are associated with regular bed shortages and queuing (as the number of beds available at any given time to admit a new patient is reduced), and increased risks of infections.[49] Increased occupancy is usually most prevalent during the winter, as cold weather and viruses both create new conditions and exacerbate existing ones. However, all quarters in 2017/18 had a higher percentage of occupied beds than the winter of 2010/11, suggesting that a previously seasonal pressure has become the norm.[50]

Winter continues to be challenging for hospitals. There were 1,100 fewer beds available each day in the winter of 2017 than in the winter of 2016, contributing to trolley queuing, and norovirus bed closures in 2017 were at the highest level for three years.[51] In January 2018, in the face of great demand, the NHS National Emergency Pressures Panel pre-emptively encouraged hospitals to delay non-urgent operations. The Royal College of Surgeons has estimated that the number of operations taking place between November 2017 and January 2018 was 7% lower than between November 2016 and January 2017.[52] NHS England has also estimated a 3% decrease in elective admissions to hospital between January 2017 and January 2018, which freed up the equivalent of 1,400 beds in the latter month.[53]

Have efficiencies been enough to meet demand?

The provider sector (including acute trusts) has increased its productivity more quickly than the rest of the economy over the past eight years. But it is still underperforming against the ambitious targets that have been set for the service.[54],[55]

That shortfall in efficiencies has had two major consequences. The first is that, for the past five years, hospitals have routinely overspent their budgets. NHS providers have been in deficit since 2013/14, reaching a peak of £2.4bn in 2015/16.[56] This was reduced to £0.79bn in 2016/17 after the Government introduced an additional £1.8bn ‘Sustainable Transformation Fund’.[57] This was intended to support transformation but has largely been used to plug gaps in hospital finances. There are also concerns that the NHS’s capital budget has been squeezed in order to reduce this level of deficit, and that high-risk, backlog maintenance reached £0.95bn in 2016/17 (nearly tripling in real terms from 2010/11).[58]

However, the reduction in deficits has not lasted, and in 2017/18 NHS Improvement recorded the provider deficit rising again to £0.96bn. This overall deficit has been driven by overspending by the acute sector, which reached £1.7bn in 2017/18.[59] The emergence of hospital deficits suggests that service costs or demand have risen more than expected, and/or that efficiencies have not taken place at the anticipated level.

The other consequence is that people are having to wait longer for treatment. This is in spite of the NHS Constitution, which contains a series of pledges about maximum waiting times that the Government remains committed to.[60]

Figure 1.12 Percentage of emergency attendances (type 1) admitted, transferred or discharged within four hours, August 2010 to August 2018

In August 2010, 97.1% of type 1 emergency attendances were discharged, admitted or transferred within four hours of arrival, meeting the 95% target for the service. However, type 1 attendances have not met this standard since July 2013. By August 2018, only 83.9% of attendances met the target (following a record low of 76.4% in March 2018). Queuing and missed targets can be seen across hospital care (see Box 1.1).[61]

Box 1.1: Queuing in hospitals

  • In August 2018, 89.7% of all emergency attendances (not just type 1) were discharged, admitted or transferred within the four-hour window (below the 95% target and the 98.1% rate in August 2010).
  • The number of patients not treated within 28 days of a cancelled elective operation tripled between 2009/10 and 2017/18 (from 2,258 to 7,158).
  • In June 2018, only 87.8% of people began treatment for non-urgent conditions within 18 weeks of referral, against a 92% target. At 4.1 million people, the waiting list is the highest it has been for a decade; 66% higher than in June 2009.
  • Performance against cancer waiting targets is more mixed:
    • Some targets are still being met, such as the target that 96% of first treatments should begin within one month from the decision to treat.
    • The target that 93% of consultations should begin within two weeks of a GP referral held up until 2017/18, but was not met in the first quarter of 2018/19.
    • The target that 85% of people should begin treatment for cancer within 62 days of an urgent GP referral has not been met for several years and was 82.2% in 2017/18.

Before the announcement of a new five-year funding settlement in June 2018,[62] NHS leaders said in a 2017 planning paper for 2018/19 that – without trade-offs in care – it was likely that waiting times would lengthen in 2018/19. They stated that “our current forecast is that – without offsetting reductions in other areas of care – NHS Constitution waiting times standards… will not be fully funded and met next year”.[63]

Part of the stated purpose of the new funding settlement is to ensure “that over the medium term no NHS organisation is in financial deficit” and to deliver “core performance standards so patients are never left waiting when they most depend on the NHS” (although a review of targets has also been announced).[64] Now that Simon Stevens (Chief Executive of NHS England) has publicly accepted the new settlement, the pressure will be back on hospitals to meet these expectations.[65]