NHS0037

Written evidence submitted by Nuffield Trust

The Nuffield Trust is an independent health think tank. We aim to improve the quality of health care in the UK by providing evidence-based research and policy analysis and informing and generating debate.

Our researchers have been studying patterns and causes of NHS waiting times for several years, and extensively monitoring them as part of our QualityWatch initiative, funded by the Health Foundation. This submission draws on this expertise, along with work on finances, staffing, and health system response to Covid-19, to inform the Committee’s inquiry into the current backlog.

We conclude that:

• Covid-19 has accelerated a process of longer waits in the NHS stretching back many years, hitting the longest waiters hardest. The problem was getting worse even before the pandemic, suggesting a structural inability to keep up with the amount of care required. Reversing course and achieving shorter waits will be a major, complex task which will not occur quickly even if Covid-19 subsides.
• The direct impact of Covid-19 slowed down planned care because of the need to separate out patients and areas to prevent infection. There will be important choices about when and how to further loosen infection control provisions, with risks on both sides to be balanced.
• Experiences from when the UK previously drove down waiting times in the decade to 2009 shows that realistic targets with a genuine commitment from political leaders; tough decisions about prioritising patients; and prolonged growth in staff and money are needed. Progress took years, not months.
• Staff are the key resource that enables the NHS to take people off waiting lists by treating them. The workforce as a whole has been growing more slowly over the last decade, and this appears to be associated with lower growth in the number of procedures and growing waiting times. Medical trainees and consultants in key specialties for planned care have not been growing as quickly as they did in the period of falling waiting times.
• In addition to short term funding to help clear the backlog and cover infection control, NHS trusts in England are structurally spending at least £4bn a year more than was assumed under the long-term plan. This is both because initial assumptions were wrong and because the pandemic stopped efficiency improvements. A stable and realistic budget is needed to enable long-term improvements, including tackling the backlog.

1. The scale and origins of the backlog

The total number of people waiting to start consultant-led elective treatment in England increased to over 5.8 million in September 2021, the highest level since records of this measure began (Figure 1).

The increase follows a previous fall to 3.9 million in May 2020, as a result of the initial impact of Covid-19, which caused referrals to fall even faster than the amount of treatment provided.  However, the backlog was already increasing pre-Covid, reaching 4.6 million in February 2020. It has now resumed its growth, apparently accelerating.

Figure 1: The backlog for elective treatment

1.1 Trends in the longest waits

As well as more people waiting, those people are waiting longer than prior to Covid-19. 

In September 2021, 33.5% of people had been waiting over 18 weeks to start treatment, compared with 13% in June 2019 (Figure 2), and a target of only 8%.

The most dramatic change is seen for the longest waits. The number of people waiting over 52 weeks to start consultant-led elective treatment in September 2021 was 300,566, 230 times higher than in September 2019.  This figure peaked at 436,127 in March 2021.

12,491 patients were waiting more than two years at the end of September. The specialties with the highest number of patients waiting for the longest times were trauma and orthopaedics, general surgery, and ear, nose and throat.  Many of these patients will already have been waiting for treatment prior to the pandemic.

Figure 2: The 18 week waiting time standard has not been met for 5 years, and the 6 week wait for diagnostic tests has not been met for 7 ½ years

The number of people waiting over a year has dropped since April 2021.  Activity has been recovering towards pre-pandemic levels (Figure 3).  But the reduction in people waiting over 52 weeks may also reflect the big drop in referrals during the first Covid wave, meaning that fewer people are reaching a year of waiting than would have previously been the case. The most recent figures for the longest waiters are concerning: the number waiting over 52 weeks has begun to rise again from a low point in 2021 of 292,138.

Figure 3: referrals for elective care outpatient appointments and admissions

1.2. Diagnostic tests

Over one in four patients (26%) had been waiting over six weeks for a diagnostic test in September 2021. The diagnostic waiting time target has not been met for seven and a half years.  Long diagnostic waiting times are a particular concern while the backlog grows, because patients can often not be prioritised appropriately while their diagnosis is uncertain.

While activity levels have recovered for many diagnostic tests, much higher throughput would be needed to catch-up on increasing demand and longer waits which have built up over many years.  For fifteen common diagnostic tests, including non-obstetric ultrasound, magnetic resonance imaging (MRI), and computed tomography (CT), there were over 1.9 million tests carried out in September 2021. This is slightly less than in September 2019, before the pandemic. It is more than twice as high as during May 2019, the peak of the Covid-19 slowdown.

1.3. Prognosis for the coming years

There is the potential for further growth as a “shadow waiting list” of people not referred during Covid-19 continues to join the queue. To what extent this will happen remains an open question.

In some cases, patients may receive treatment after an emergency attendance or admission, which may be reflected in the very high demand for emergency care in recent months. In other cases, after discussion with their GP, some patients may decide not to seek a specialist referral, but to manage their condition in other ways. This means that predicting how far the backlog will grow in coming months is difficult. There is no clear equivalent to Covid-19 since the founding of the NHS based on which we might extrapolate.

However, even if demand simply resumes its pre-pandemic trend, this would still result in a steady growth in the backlog, because it was already outstripping capacity.  Previous experience from the early 2000’s when waiting lists were at high levels indicates that it is likely to take years to address long waiting lists, and require a combination of increased investment and capacity, including more staff, along with rethinking some care pathways to reduce appointments or processes which provide limited benefit to patients.  The NHS long term plan had identified scope to reduce outpatient appointments and move some services to remote consultations.  There have already been big shifts in how out-patient care has been delivered during Covid, which may help, but will not be sufficient. In short, Covid-19 has accelerated an already worsening problem, and a quick resolution will be impossible given that all the drivers leading to falling performance before the pandemic continue to exist.

2. Understanding the impact of Covid-19

Why has the pandemic caused an apparent further worsening of waiting times, and how long will these factors remain? While some shifts reflect changes in patient and referral behaviour during the pandemic, we believe that the most important effects come on the supply side, through the impact in hospitals of the need to prevent and limit Covid-19 infection. Our report last year, Here to Stay, explores these dynamics in depth.[1]

A number of hospitals experienced high rates of transmission among patients and understandably, in response to this a cautious approach to infection prevention and control has been adopted. 

In our work with NHS services across the UK, we have heard how physical constraints and the design of hospitals mean that some have very great difficulties in separating planned care from emergency work. Where this is possible and patients and staff can be tested quickly and efficiently, planned treatment services can work at levels very close to normal (albeit with an enhanced level of PPE).  The most effective examples of this is where there is actual physical separation of these functions.

However, in many settings this will not be possible at all, it may raise staffing demands, and in areas such as theatres and imaging rooms the separation of work may be more difficult.  The requirement to recover medium and high risk patients in theatre rather than in post-anaesthetic recovery units has the potential to take theatres out of use for extended periods if it is not possible to test patients prior to surgery. 

There are also constraints imposed by the need to clean equipment between patients and to leave rooms empty where aerosol generating procedures – such as CPR, for example – have been performed. 

In some areas, it will be possible to overcome these issues by reconfiguring the physical layout of hospitals, or with more collaboration between hospitals – for example to make one site exclusively for planned work. This may be becoming more possible to a degree, as medical advances mean less intensive anaesthesia can be used for planned procedures, reducing the need to be near urgent care capabilities.

Policymakers also need to honestly consider at what point, with widespread testing available and if rates of infection fall, the guidance is overly cautious.  There are trade-offs between the impact on patients of delays to treatment, the additional cost of infection prevention and control, the costs of recovering the backlog of work and the risks and the impact of infection. It is possible that the balance of risk will shift so that an overly cautious approach to infection prevention and control will cause more harm than it prevents. The reduction through UKHSA guidance in September of social distancing to 1 metre, and the removal of PCR test requirement for some procedures, are an important decision point which should be scrutinised and learnt from.[2]

3. Staffing in historic context

In the longer term, understanding why the NHS before and after the pandemic struggles to deliver care at the rate GPs refer patients for it requires looking at its major resources: money and, above all, staff.

3.1. Examining two eras: how staffing relates to waiting times

Staffing trends in the past decade, which has seen a steady worsening of waiting times, are quite different to those in the preceding decade, where waits generally fell. For instance, between 2000 and 2009 the number of hospital and community health doctors, nursing and midwifery staff and allied health professionals increased by 53%, 26% and 33% respectively. However, between 2010 and 2021, growth has been much slower for these groups, at 28%, 11% and 26% respectively.

The chart below shows the imperfect but clear relationship between increases in NHS output and labour force in the two periods. Labour inputs here, as designed by the ONS, represent the number of full time staff weighted by their salaries.

It is also worth noting that in the early years of the last decade, with austerity at its height, the health service managed to deliver increases in activity without more staff: this demonstrates how long and how vigorously options to do more with less or the same have been pursued.

Figure 1: Public service healthcare annual labour growth and annual output growth

Source: ONS (2021). Note: Data points are as at the end of each financial year. Data apply to England only.

Medical trainees and surgeons

In particular, there were large increases in the number of medical trainees between 1997 and 2008 – especially for A&E and anaesthetics. However, growth in medical trainees between 2009 and 2020 did not keep up with this increase seen previously across all specialties.

Figure 2: Percentage change in number of medical trainees by specialty, 1997-2008 and 2009-2020

Source: NHS Digital (2011); NHS Digital (2020). Note: Data show percentage changes of the headcount of medical trainees as at September of each year, with the exception of 2020 (March).

At consultant level, growth in the two periods is more even, perhaps reflecting that this group of doctors represents individuals starting their careers years or decades earlier. However, within several particular groups of surgeon which might be particularly associated with some types of elective care – general surgery and trauma and orthopaedics – growth has been significantly slower.

Figure 3: Percentage change in number of consultant surgeons by type, 1997-2008 and 2009-2021

Source: NHS Digital (2011); NHS Digital (2021). Note: Data show percentage changes of full-time equivalent consultant surgeons as at September each year, with the exception of 2021 (April).

Similarly, growth in the number of nurses working in the English NHS was considerable between 2000 and 2005, equating to an increase of 46,500 full-time equivalents over this five-year period. However, a sustained increase in the number of nurses at this scale has not been recorded since.

Based on the historic trend (see Figure 4), reaching the government’s ambition of 50,000 more nurses by 2024/25[3] will be challenging. We would expect an increase of, on average, 10,000 nurses a year for the government to achieve this. While there were nearly 13,000 more full-time equivalent nurses working in September 2020 compared to the previous year, this is likely to be partly due to atypical recruitment and retention practices during the Covid-19 pandemic. For example, recently retired nurses re-joined the NHS during the first wave of the pandemic and will have contributed to this large increase, but this trend will not continue and is likely to reverse.

Figure 4: Annual change in the number of nurses, 1960-2020

Source: NHS Digital (2020). Note: Gaps represent changes to the data collection. Includes hospital and community nurses and health visitors, but excludes nurses in general practice. Data are for full-time equivalents. All data points show staffing levels as at September each year.

Current situation

Increasing the rate at which staff become available is not a straightforward question. While EU recruitment has dropped off following Brexit, there is currently a very high rate of staff arriving from outside Europe – with registrations from overseas nurses and doctors each reaching around 10,000 in the last year of data.[4] [5] For nurses, this is actually above the rate we called for with the King’s Fund and Health Foundation in 2019 in order to close the staffing gap by 2023/24.[6] How sustainable this is will depend on political decisions about migration, the UK’s competitiveness and ethical concerns about draining staff from poorer countries.

Meanwhile, although medical training places have sharply increased during the pandemic, retention and whether staff choose to work in the NHS rather than other sectors or countries will remain crucial issues. This may be put at risk by pushing staff too hard after the pandemic or if the current disputes over pay present an unappealing prospect. It is concerning that the latest staff survey shows that the proportion of nurses considering retiring outright or leaving the health care sector has risen from 15.7% to 16.9%.[7]

4. Funding requirements

The other crucial resource for the NHS is its level of funding, which covers staff salaries as well as the equipment and medicines necessary for planned care to take place.

4.1. Chronic funding deficits

The backlog is rooted in permanent capacity shortfalls, not just the pandemic. Before considering temporary additional funding to address the post-Covid backlog we need to consider whether NHS funding is sufficient to fund the service on an ongoing basis.

In the financial year 2019-20, the first covered by the Long-Term Plan (LTP), the NHS providers which together account for over two thirds of NHS England’s budget spent £2bn more than planned for the year. This continued similar overspends in earlier years. It suggests that the problem is the planned budget, which has assumed unrealistically low rates of activity growth on one side of the equation, and unrealistically high rates of efficiency savings on the other. Both have been systematically missed.

Over the course of 2020-21 we estimate that the gap between the original provider spending plan implicit in the LTP and providers’ actual recurrent costs, removing temporary spending to cover Covid-19 costs, grew to at least £4bn. This was predominantly due to the inability of providers to make efficiency savings during the pandemic.

Even if providers manage to make the headline 1.1% efficiency savings pledged by NHS England in SR 2018, the gap will grow again to at least £5bn by the end of March 2022.  These calculations are explained at greater length on our website.[8]

The higher costs than anticipated are currently being funded by NHS England diverting spending originally ear marked for LTP objectives, such as investing in primary care and reducing waiting times.

The recent spending review increased NHS England’s revenue budget for 2022-23 and 2023-4 by £6.6bn and £3.6bn respectively in cash terms against the original LTP. However this additional funding is non recurrent and by April 2024 the NHS is essentially being asked to return to a revenue budget that will be just £1bn higher in cash terms than planned for before the pandemic. That entails that yet again NHS providers will be asked to make unrealistically high efficiency savings to claw back efficiencies “missed” during the pandemic – and indeed systematically missed in the years before it. This will undermine attempts to increase provider capacity to address elective waiting times as it means they will be at best left again with at least £2bn of their annual income coming only as a temporary bailout when the efficiency targets are inevitably missed. That would deter trusts from recruiting and retaining permanent staff, or investing in expanded physical capacity. It will focus managers’ attentions into one off, non-recurrent savings measures. 

Our view is that, as a minimum starting point, the recurrent £2bn provider “overspend” against the planned budget from 2015/16 to 2019/20 should be funded with a reset to NHS England’s baseline budget – before considering additional temporary funding for both Covid 19 and the elective backlog. Beyond that, ambitions to claw back the additional £3bn or more excess recurrent cost accrued by the end of 2021-22 through extra efficiency savings need to be tempered by a sober analysis of what is genuinely possible.

4.2. Temporary funding

The NHS will also require further temporary funding: for Covid-19 itself, and for short-term additional costs to address the backlog. The IFS have published a possible estimate of £3.5bn for the latter, though they note that it is subject to the deep uncertainty about how many of those patients not referred during the pandemic will re-enter the system.[9]

However, we should remain cautious about how much can be achieved by temporary funding alone. There are limited reservoirs of extra staff to be pulled, and because the private sector often employs the same doctors as the NHS, the amount of genuine added capacity there is easily overestimated – especially if efforts to reduce the waiting list backlog involve increased use of private sector capacity.

At least in the short term, it may therefore be useful when estimating the likely cost of expanding activity to address waiting lists to consider that during 2020-21 the additional incremental provider pay costs for dealing with Covid-19 totalled £2.7bn. Although further work is needed to explore the staff groups and additional working hours and shifts purchased for this sum, it is possible that it represents a ceiling on the total additional workforce resource available to the healthcare provider sector in England as a whole.

5. Solutions and approaches: looking to history and abroad

A large number of operational, clinical and policy changes are available which may help deliver the capacity and prioritisation needed to get waiting times falling once more. We are confident the Committee will be provided with many examples and cases by NHS stakeholders. In this section we will look to provide wider context by looking at what other countries across Europe are doing to address similar problems; and by looking at the policies which secured national falls in waiting times in the NHS in the recent past.

5.2. Learning from Europe

A number of themes have emerged in the way other countries are reforming services in the short, medium and long-term to address care bottlenecks and proactively manage the demand for care that has built up throughout the pandemic. The table below illustrates a selection of these which were submitted to the European Commission in recovery plans. This draws on our analysis, soon to be published, of national health system approaches.

As in the NHS, the pandemic has been a catalyst for innovation in many countries, and many of the interventions described below are beneficial changes unlocked to respond to the pandemic that are now being sustained to support recovery. The pandemic has also served as an impetus in several countries to pass or agree reforms that have been long-standing policy objectives but have otherwise lacked funding or political consensus to implement.

While it is still too early to be clear on all best practices, and how different contexts will contribute to different results, there is value in understanding the solutions other health systems are turning to. Significant points may include:

• Expanding hospital and diagnostic capacity is a priority for many countries, illustrating similar issues as are seen in the NHS.
• However, there is considerable emphasis on improving preventive and out-of-hospital services, rather than focusing solely on planned treatment. Focusing on waiting times for recovery is a choice, and the many countries are taking quite a wide view.
• As in the NHS, opportunities through the recovery to digitalise care and introduce new technologies are being widely emphasised.

• Notes: 1) Some of the solutions and approaches described above are happening within regions or hospital networks but are not necessarily being applied nationally or come from a central directive. 2) Some of the initiatives above describe efforts that predated the pandemic, but have been accelerated or expanded as part of recovery. 3) The absence of a country does not mean that the initiative or intervention is not taking place in that country.

5.2. Lessons from history

The NHS itself also presents a good source of lessons in reversing long waiting times: as discussed above, around the turn of the millennium it successfully reversed a prolonged period of worsening waiting times. This is visible in the graph below, which combines different measures used as the NHS tried to understand and control waiting times better: beginning by only capturing the numbers of ordinary (ie not day case) patients waiting annually, then all inpatients at increasingly frequent intervals, and now all patients between GP referral and treatment along with many other indicators.

The different measures show the waiting list mostly increasing from the 1970s up to 1997, following which it declines across all the different data series, before starting again to rise around 2012.

While sporadic, ad hoc and locally-driven ‘waiting list blitzes’ had been a feature of the NHS in the 1980s and 90s, it was only when inpatient lists reached a post war high of 1.3 million in 1997 that more systematic policy action started to evolve following a pledge by the then incoming Labour government to reduce waiting list numbers by 100,000 (achieved over two years).

From 2000 to 2004 policy focussed on achieving, among other things, national targets for having maximum waits for outpatients to three months, abolishing waiting lists and replacing them with booking systems for patients and reducing the maximum wait for inpatient care from eighteen months to six.

In 2004, targets were tightened and the scope of what was considered waiting widened with the aim of a maximum wait of eighteen weeks from referral by a GP to treatment in hospital. Following a large improvement by 2008, the target was broadly achieved by around 2011.

Some key lessons from this period include:

• Targets can work, but they must be carefully designed and implemented. In this period the NHS raised standards over time: currently we begin with unobtainable standards which are decreasingly relevant.
• Resources are needed. In addition to staff, as described above, this period saw very rapid increases in spending.
• Reducing waiting lists and times takes years, not months, even with rapid growth in spending far beyond anything likely today.
• Political and managerial commitment are vital.
• Prioritisation was necessary, at least early on, ranking specialties, operations, and patient groups most likely to benefit (and conversely, most likely to suffer if they waited too long).
• Decisions are needed about what constitutes a reasonable wait: should recovery aim for no more than 8% of patients waiting longer than 18 weeks for example? Or, immediate pre-pandemic levels of performance of 17%?

While there are lessons from the past, the NHS and the waiting times situation now is not the same as it was a quarter of a century ago. The way the NHS has performed and the way it has changed how it delivers services, with for example more care provided remotely, may help reduce waiting faster than before. On the other hand, the NHS is in a more difficult position, with larger lists, and some major reforms, such as centralising specialty lists, have already been done and cannot be done again. Many procedures have already been shifted to day surgery, and while more efficiency should be demanded from hospitals each year, the health service has already run into financial difficulties through the temptation to overestimate how fast this can be done.

December 2021