The number of NHS hospital beds has fallen

“NHS cuts 15,000 beds in six years”

Daily Mail, 20 February 2017

“Number of NHS hospital beds has plunged by 20% in ten years, BMA warns”

Daily Mirror, 20 February 2017

“Between 2006/7 and 2015/16, the number of overnight hospital beds has decreased by a fifth.”

British Medical Association, 20 February 2017

“This analysis is inaccurate — the figures come from two different time periods when the way of counting beds was different, and so they aren't comparable.‎”

Department of Health, 20 February 2017

The claims making the headlines aren’t fully accurate.

But they are telling one part of a wider story.

It’s true that the number of hospital beds in England is falling. They’re more and more likely to be occupied overnight, leaving less slack in the system to cope with surges in demand.

The significance of the fall isn’t clear-cut. Per person, we probably don’t need quite as many beds we used to: the NHS is getting more efficient at seeing patients in hospital; more of us are being seen on the same day rather than staying overnight; and when we do stay overnight, on average we’re spending slightly less time in hospital.

Finally, these figures don’t tell us everything. For example, they don’t tell us about beds in private hospitals, paid for by the NHS.

Official figures show at least 12,000 fewer overnight hospital beds over six years

Whatever periods you compare between now and six years ago, the number of overnight beds on the NHS has fallen.

On an average night from July to September in 2010, there were about 141,500 beds across the English NHS. During the same period last year – the latest figures – there were 129,500.

So that’s a fall of about 12,000.

You can get slightly different figures by comparing different points in the year. For example, if you compared April to June of both years you’d see a fall of 14,000.

The headlines get 15,000 by comparing the spring months in 2010 with the summer months in 2016. This isn’t a fair comparison because the number of beds varies depending on the time of year. Typically the number of beds increases at the start of the year, during winter, before falling back in the later months.

Figures before 2010 aren’t comparable with today’s figures

This was the government’s line of argument in response to the headlines this week.

Before 2010, the NHS’ figures included more types of bed. It counted beds for residential care or what’s called ‘non-consultant-led’ treatment (beds for care managed by a nurse or GP).

Since 2010, only beds managed by consultant doctors in hospitals are part of the figures.

That was seen as a “major factor” behind the sharp fall in the number of overnight beds counted in these figures.

That means claims that simply compare the number of beds before and after 2010 will overstate the fall.

We asked the BMA – which made claims about the change over the past decade – about these concerns. They pointed us to part of their report which warned that “care should be taken … when comparing figures from before and after 2009/10” due to the change in methodology.

Whatever rough estimates you care to make, it’s still clear the number of overnight beds has been falling. The difficulty is knowing by how much.

We don’t necessarily need as many beds per patient as we used to

It’s tempting to conclude that any number of fewer beds is bad news for the NHS, especially considering hospitals are busier than ever.

It’s not quite as simple as that.

On one hand, the number of overnight beds has been falling for years – it fell by about 9% over the last six.

But on the other, the proportion of beds which are occupied has been rising slowly by comparison. Six years ago about 85% of overnight beds were occupied. Last year it was more like 87%.

A few other trends help to explain why this might have happened.

First, a growing proportion of hospital visits involving a consultant have been day-only over the last decade, as opposed to patients staying overnight. 29% of cases were day-only in 2005/6 while 36% were in 2015/16. This is also reflected in a slight rise in the number of day-only beds over the same period.

The average length of stay in hospital is also falling consistently over time. This is because patients who are in hospital for longer periods are leaving sooner. It’s been the case since 2007 that half of overnight patients only stay a day.

But there are still fewer than the recommended number of beds available

Hospitals across England are getting more and more likely to reach capacity. A rise in occupancy from 85% to 87% might not sound like much, but the upward trend has been consistent for the last six years

From July to September last year, bed occupancy hit 87.5% - the highest recorded level for that time of year. Across hospital trusts, almost two thirds were using over 85% of their overnight beds on average. 10% of trusts had 95% occupancy rates or higher.

This is dangerous for patients, according to the trade association, NHS Providers. It and the Royal College of Psychiatrists (RCP) say 85% is the standard rate. The RCP goes on:

“[85%] means that patients can be admitted in a timely fashion to a local bed, retain the connections with their social support network and take leave without the risk that they cannot return to their ward should they need a longer period of in-patient care. It allows functioning space to accommodate those newly admitted and to provide proper treatments to current patients.”

The problem is particularly acute over winter, when ‘unplanned’ care demand is highest so the system needs slack. Health think tank the King’s Fund warned last month that bed occupancy rates had routinely exceeded 85% over the past winter and on some days often passed even 95%.