Last updated:

Beds in the NHS

NHS trusts forced to open thousands of extra beds to cope with year-round pressures

Even well beyond the pressures of the winter, hospitals in England are having to deploy escalation measures to cope with levels of demand. Nine in 10 NHS trusts were still using escalation beds – extra hospital beds brought in to cope with demand - on 1 May 2019, with little sign of this practice ending.

This is yet more evidence that so-called 'winter pressures' and the associated escalation measures are now a year-round reality for the NHS. The pressure on hospitals and doctors that forces them to use escalation beds – traditionally rolled out primarily during the winter months - remains long after winter is over.

This research shows that hospitals' 'core bed stock' is no longer able to deal with the level of year-round demand on the NHS. Doctors on the frontline know this, with 92 per cent now agreeing that the NHS is 'in a state of year-round crisis' (BMA quarterly survey).

The government must ensure the 'core bed stock' grows by at least 3,000 beds to reach a level that can cope with year-round demand, reserving escalation beds for responses to peaks in demand and addressing dangerous bed occupancy levels.


What do we know about escalation beds?

Hospitals experience (often seasonal) fluctuations in demand. NHS trusts produce escalation plans designed to cope with these increases in pressure, including opening and staffing escalation beds. These beds are additional to the permanent bed stock and are brought in to service for limited periods of time, either in temporary or re-purposed wards, or added to existing wards, to provide extra capacity.

Almost all NHS trusts use escalation beds during winter, and data about their use is published from December to March. For the rest of the year, there is no public data on the use of escalation beds.

On 3 March 2019 (the final day of data publication), trusts across England were still using thousands of escalation beds. To better understand what has happened since then, the BMA sent a freedom of information (FOI) to NHS trusts asking them about their use of escalation beds during the five-week period following this (4 March to 7 April), and then later for 1 May.

This new data uncovers the stark reality that almost all trusts are having to continue to use escalation beds well into spring, and that escalation beds were being used at similar levels in May to the height of winter.


Escalation beds are no longer just for winter

Escalation beds were required daily beyond winter and into the spring. Furthermore, there is little evidence to suggest that trusts will be able to close all their escalation beds before summer or even autumn. This is particularly concerning given that BMA projections show that demand for beds will rise again this summer.

On 4 March, the day after NHS England reporting ended, just nine trusts (out of 105) were not using escalation beds. Five weeks later on 7 April, well in to spring, still only 18 were managing without escalation beds. This left 87 trusts (83 per cent) still using escalation beds. Although some trusts did manage to close all their escalation beds over this period, many of them were forced to reopen escalation beds, and by 1 May, a large majority were still using escalation beds.

Alarmingly, one trust was still using 147 escalation beds, while in another trust, escalation beds accounted for over one in every five beds (465 beds). Having ended winter, on 4 March, using an average of 25 escalation beds per trust, trusts were still using an average of 21 escalation beds each five weeks later, on 7 April. At the current rate, it could take until late summer for trusts to close their escalation beds.

Figure 1 - Escalation beds per NHS trust (rolling 7-day average)

ARM content images 
  • Escalation bed numbers in the first week of April were comparable to those in early January, and only marginally lower than during the height of winter.
  • Of the 104 Trusts who reported data for the full five weeks, 23 had more escalation beds open by the end of the period than at the start, and 23 had the same number.

Figure 2 - Trusts using escalation beds

ARM content images 

Figure 3 - Average number of escalation beds per NHS trust on the 1st of each month

ARM content images 
  • There were more escalation beds open on 1 May (20.5 beds per trust) than on 1 January, March, or April.
  • Since March, doctors experienced the pressure on beds (April to June) even worse than last year, with over four in every 10 respondents reporting that there are fewer available beds than last year. Only 8 per cent say there are more beds available than last spring (BMA quarterly survey).


What is the picture across England?

While there was some regional variation, hospital trusts in all regions used significant numbers of escalation beds beyond winter. No region came close to closing all their escalation beds in the five weeks following 3 March.

Figure 4 - Escalation beds by region (percentage of total beds)

ARM content images 
  • With an average of 37, the East Midlands had the highest number of escalation beds per trust of the 10 regions, although at 3 per cent of total beds, the region was middle ranking.
  • The North East used the fewest escalation beds as a percentage of total beds of any region throughout the five weeks, consistently below 2.5 per cent. This region averaged 12 escalation beds per trust.
  • South Central increased their use of escalation beds over the course of the five weeks, from 3 per cent to 4 per cent of total beds.
  • The South East was the most successful region in reducing their use of escalation beds over the five weeks, falling to 16.5 beds per trust, from 35. This meant escalation beds fell from 4.7 per cent to 2.7 per cent of total beds.
  • Although London began and ended the five week period using the most escalation beds as a percentage of total beds, there was a one percentage point reduction in their use, from 5 per cent to just under 4 per cent of total beds.


How can the government ensure the 'core bed stock' is sufficient to cope with year-round pressures?

Figures show that the UK has far fewer hospital beds per 1,000 people than other OECD countries, so it is no wonder that so many trusts are having to use escalation beds throughout the year.

Figure 5 - Hospital beds per 1,000 people (2017) (source)

ARM content slides 

Given limited bed capacity, escalation beds can be a useful, flexible tool for trusts to cope with fluctuations in demand. Over the winter, when demand goes up, trusts can temporarily open additional beds, at a lower cost than staffing extra permanent beds year-round.

However, it is a sign of a broken system, and of intense all year-round pressures, if trusts have to use them beyond winter and throughout the spring. Our analysis shows that large numbers of escalation beds are consistently being used for months. These beds can be very expensive to staff, often requiring locum or agency workers. Trust board papers note the staffing costs associated with the continued use of escalation beds, for example:

"Expenditure increased by £682K overall this month due the use of locum and bank staff to cover escalation wards." (source)

"The overspend was due to a combination of locum and agency expenditure, due to recruitment challenges and covering for increased activity and escalation beds." (source)

"Ward staffing … during March continued to be challenging. The causative factors remain vacancies and the number of escalation beds in use." (source)

Furthermore, with trusts using so many escalation beds routinely, their capacity to be able to open further beds to meet peaks in demand is seriously limited. That is why we believe that the level and prevalence of escalation bed use that we have uncovered is not sustainable.

The 'core bed stock' must be at a level at which escalation beds are not routinely needed beyond winter.

The BMA has estimated that a minimum of 3,000 additional core beds are needed. This is the number of extra beds that would have enabled trusts to avoid the overuse of escalation beds outside of winter. The North West and London need over 400 additional core beds each, with the East Midlands, the East of England, and Yorkshire and the Humber needing over 300 more beds. The North East only requires an additional 123 beds (see calculation below).

Figure 6 - Beds required by region

ARM content images 


Is this a big ask?

NHS leaders reportedly asked for 4,000 extra beds ahead of the winter last year. Unfortunately, these additional beds were not provided. In fact, hospital bed numbers over the winter were consistently down on last year, perhaps resulting in the widespread use of escalation beds described above.

3,000 more beds are an additional three beds for every existing 100 beds across England. For a trust with 500 beds, this would mean an extra 15 beds. Additionally, funding for the adequate staffing of these beds must be provided by the government. While this would require upfront government funding, we believe this is likely to save costs in the long run – through the reduction in locum costs.

Increasing the core bed stock to the point where trusts will not be forced to regularly use escalation beds beyond winter is a sensible and achievable first step towards addressing the bed shortage in our NHS.


Even this won’t completely solve bed pressures

Although these additional beds are needed to address the year-round use of escalation beds, even 3,000 additional beds would not address the dangerously high bed occupancy levels seen across the NHS. Tackling the problematic use of escalation beds is necessary but addressing high bed occupancy is an additional challenge.

In our analysis of bed occupancy rates over the winter, we found that, even while using thousands of escalation beds, only 35 (out of 134) trusts kept their average bed occupancy below the NHS’s own recommended maximum occupancy of 92 per cent. Previous analysis by the BMA showed that, on top of the number of escalation beds already used, 5,000 additional core beds are needed to bring occupancy down to safe levels. It is clear that reducing bed occupancy should be a priority, with 59 per cent of doctors believing that high bed occupancy has compromised patient safety this spring (April to June). Concerningly, 58 per cent believe this will continue to compromise patient safety over the summer (BMA quarterly survey).

As shown below, escalation bed use and high bed occupancy go hand in hand. Investment to increase the core bed stock is likely to reduce bed occupancy and, at the same time, reduce the number of escalation beds that trusts are having to regularly open. Importantly, this will also empower trusts to use their escalation beds to respond to genuine peaks in demand, rather than just to cope with all year pressure, keeping occupancy down even when pressures are high.

Figure 7 - Escalation beds and bed occupancy

ARM content images 

Notes on methodology

How did we get to 3,000 beds?

This figure has been calculated by summing the average number of escalation beds used by each trust over the five-week period between 4 March and 7 April 2019 (outside the NHS winter reporting period). We did not use the 1 May figures as this is only a snapshot, not a full picture of use over time.

These numbers were then aggregated into regions and scaled up to account for trusts that did not provide data. This figure would therefore provide the number of beds needed across England to ensure that trusts would only be using escalation beds to respond to the peaks in demand over this period.

Using the maximum number of escalation beds used by each trust over the same time gives a figure of 4,400 beds. Taking the minimum number of escalations beds used gives an equivalent figure of 2,000 beds.

We believe the average number is the most relevant as this is the best estimate of the number of beds that are needed to ensure that trusts do not have to routinely use escalation beds outside the NHS winter reporting period, and can instead use them flexibly to respond when demand is at its highest.

More information: NHS bed facts and figures

  • Addressing dangerous levels of bed occupancy

    winter pressures

    The reasons behind the problems at trusts in winter are complex, overlapping and numerous. One issue that often lies at the heart of the problem is the increasingly high proportion of beds that are occupied during the colder months.

    As a general guideline, evidence suggests that to ensure safe patient care occupancy should ideally not exceed 85%1, while NHS Improvement has said that above 92%, the deterioration in A&E performance begins to accelerate.2

    Furthermore, high bed occupancy leads to issues across trusts, and is often a decisive factor in the decision to cancel operations (for example last winter, which saw tens of thousands of operations cancelled with little warning).

    In last winter’s situational reports (an NHS England dataset which indicates the level of pressures on secondary care), between the last week of November and the first week of March, 80 trusts (of 137) spent at least half of that period at 92% occupancy or more. Only five never exceed 92%.

    It is clear that the NHS no longer has the bed capacity to cope with high levels of demand in winter.


    How many beds are needed in England?

    Using bed occupancy and availability data from previous winters, the BMA has calculated how many additional general and acute beds each trust would have needed to meet certain bed occupancy thresholds.

    For example, on 20 February 2018 (the day that saw the highest occupancy), trusts would have required an additional 4,809 beds to have achieved under 92% bed occupancy across England. In order to have achieved 85%, an additional 13,066 beds would have been needed.

    The more achievable goal of 92% would still have required almost 5,000 more beds, and in the last two winters, the addition of 5,000 beds would have reduced occupancy by 4.6% in 2017/18 and 5% in 2016/174,5, in both cases. It is also worth noting that 5,000 is also the number of extra beds endorsed by the Royal College of Emergency Medicine.6

    Whilst 85% bed occupancy remains the long-term goal, in the short-term the BMA believes the NHS needs 5,000 more beds than last winter if the more realistic goal of 92% is to be achieved, and if patient safety and A&E performance are not to be severely compromised in the 2018/19 winter. This is in addition to the 2,500 – 5,000 escalation beds normally opened during winter, so could represent as many as 10,000 additional beds on top of the NHS’ core bed stock.

    In the longer term, even with a greater emphasis on primary and community care, a year-round 85% occupancy figure (as recommended by the BMA and a number of royal colleges) will only be achievable with a substantial increase to the number of beds in England. Therefore, a long-term bed strategy should be implemented with a view to reintroducing a sufficient number of beds across England to the point where 85% bed occupancy can be reached.


    How many beds are needed in different parts of England

    Using trust-level data, the BMA has also calculated approximate numbers of beds required by different regions of England.7

    East Midlands 385 South Coast 430
    East of England 220 South East 285
    London 915 South West 345
    North East 195 West Midlands 560
    North West 720 Yorkshire and The Humber 520
    South Central 495    


    Are trusts attempting to address the possible shortfall?

    Earlier this year the BMA sent an FOI (freedom of information) request to 135 trusts asking for clarification on their winter planning, including how many additional beds they believed they would need this winter, and whether they would be increasing their bed stock accordingly. Of the 135 trusts we contacted, just under two thirds responded.

    The responses varied significantly, with some trusts suggesting that they could need as many as 150 beds, while others said that they did not expect to need any additional beds at all. A small number of trusts (7) were explicit about their plans to make no additions to their bed stock – citing reasons including staff shortages, space constraints and lack of funding.

    However, it was clear from the majority of other responses that where trusts are planning winter bed increases these are fairly modest in scale and are unlikely to meet the levels of demand this winter if the BMA’s projections are accurate.

    Of the trusts which replied to say that they would not be increasing the number of beds to match the projected shortfall, almost all suggested that other measures (e.g. length of stay reduction, improved patient flow) would be sufficient to reduce demand to manageable levels.

    Though some hospitals are increasing their bed stock, the results of the FOI would indicate that a bed strategy is only being implemented where sufficient funding has been made available, and that reductions in admissions and length of stay (which are often both contingent on uncontrollable factors such as weather and flu) form a much larger part of winter pressures contingency planning.



    Short term:

    • With winter fast approaching, it seems unlikely that any trust will be able to substantially increase its number of available beds at this stage. However, those trusts with the greatest need should be supported to address their shortfall as much as possible in the time that remains before winter pressures become more substantial.
    • Greater transparency about winter pressures is needed. Data publication often changes year-to-year which severely affects comparability. NHS England should resume publishing the OPEL (operating pressure escalation levels framework) data as part of the Winter Situational Reports dataset (which was available in 2017 but not 2018) and should ensure that data gathering and publication is consistent every year.

    Long term:

    • NHS England must set out clear plans of how occupancy levels will reach acceptable levels over the coming years in its forthcoming Long Term Plan, and the government must back NHS England with sufficient resources to achieve this.


    1. A safe bed occupancy figure is contingent on the clinical setting (eg an urgent care setting will require a lower occupancy figure than a ward catering for patients undergoing planned procedures) but broadly speaking, 85% represents the benchmark for patient safety, and it is inadvisable for bed occupancy to regularly exceed that figure.

    2. NHS Improvement, review of winter 2017/18

    4. Percentage points

    5. Winter in this context refers to the period covered by the relevant data, which traditionally starts publication in late November and ends in early March

    6. NHS Improvement, review of winter 2017/18

    7. Please note, these have been calculated differently to the national figures and so the total number of beds is slightly higher than the national ask. The pan-England ask is partially based on the bed shortfall on the busiest day in February, but some trusts had much higher occupancy on other days that week. In these cases, we have used the figures from the busier days. The regions have been devised using STP boundaries, with STPs combined to form larger regions. The figures are rounded.


  • Read 2017/18 report

    Download the full report for the whole of the UK published in February 2017.

    State of the health system - Beds in the NHS (PDF)

  • Winter pressures

    As part of our ongoing work evaluating pressures in the NHS, the BMA is publishing monthly analysis of a number of key indicators, including those directly related to winter pressures.

    Read our analysis of pressure points in the NHS

    As part of our analysis of the winter pressures we have calculated the future trends and projections for 2018/19