Alan Milburn says he wants a new drive to boost bed numbers - but how easy will it be to achieve this at grassroots level, asks Thelma Agnew

Health secretary Alan Milburn has ordered the NHS to 'go for growth'and reverse the decline in bed numbers. Unsurprisingly, he appears to have no intention of leaving it up to local managers and clinicians to decide how many new beds they need. Hence, the promise of guidance next month which will, for the first time, offer a prescriptive 'model' to calculate the number of hospital and other beds needed in each area per head of population.

There is speculation about the shape, purpose and robustness of this model. HSJ understands that the Department of Health will issue a toolkit, including a national interactive database into which trusts and health authorities will input their own data on capacity for acute and community beds, 'forcing them to compare it with national standards'. An insider told HSJ: 'It will ask them why do you think you are different - and if you are, can you justify it?'

Work on developing the model is being led by John James, chief executive of Kensington, Chelsea and Westminster HA, and DoH officials, including chief economist Clive Smee. Regional offices will examine resulting data and negotiate with the DoH to agree what the distribution of 2,100 general and acute beds promised in the NHS plan will be.

But many question whether a blueprint handed down from the centre can possibly deliver his aim of 'increased numbers of the right beds in the right places'.

The current template has so far been tested on Kensington, Chelsea and Westminster HA and the DoH hopes to 'road test' it on some more HAs before the guidance is published.

But NHS Confederation policy director Nigel Edwards is so concerned that he is meeting DoH officials this week to try to get a sneak preview of the mysterious model. 'It could be anything, ' he says. 'If it is a developmental model it is a huge undertaking. I would be staggered if the DoH can come up with that in a month. '

Previous attempts to design models of services (by the former Institute of Health Services Management and, in the late 1990s, the British Medical Association) have not fared well, he points out. 'They all came a cropper on the issue of great variability. They were fine as ideals, but not so good when you tried to apply them locally. '

A centrally produced model will somehow have to accommodate not only variation in local NHS circumstances, but also the different priorities and resource allocations in social care.

Mr Edwards says: 'Say you have a model that decides bed use and doesn't take into account that a social services department is spending all its money on child protection - you would have old people on the street. It's a really complicated multi-agency problem that has some real subtleties of management. '

The NHS plan was based on reaching a national standard of no more than 82 per cent capacity.

HSJ understands that under the new beds model, trusts and HAs with higher capacity rates will be 'strongly encouraged' to improve community care and social services to bring numbers down.

But Mr Edwards points out: 'One of the problems with benchmarking information we have at the moment is that it tells you performance is better in one area than another, but it doesn't tell you why. '

Although Mr Milburn is adamant that the NHS culture of ever-decreasing bed numbers must be 'replaced', not everyone is convinced that he's got his sums right.

Mr Edwards agrees that the health service has had to cope with a level of bed occupancy that is 'unsustainable and leads to crisis'; but increasing bed numbers ('which would certainly help') may turn out to be a short-term solution. Relaxing the pressure on beds - or rather bed management - could backfire: 'The problem is that in the longer term, behaviour adapts to additional beds. If you just do more you don't do better. There is no easy fix. '

Charles Normand, professor of health economics at the London School of Hygiene and Tropical Medicine, suspects the NHS is not short of beds, though he admits he can't produce strong evidence to prove it.

Professor Normand knows just how difficult it is to calculate 'necessary' bed numbers: 'I did a survey for the European Union nine years ago trying to identify which parts of the then 12 member states had adequate or inadequate beds. [Conditions] turned out to be so varied that you couldn't make much sense. '

He insists, unfashionably, that there is still slack in the NHS system which could be pulled tight by better bed management. What worked at Central Middlesex trust in the 1990s, which reduced its bed stock by 10-20 per cent while he was a board member, could surely work throughout the service.

Measures such as ensuring emergency admissions were quickly seen by senior clinicians and 'not allowed to rot in beds', and full use of community facilities, kept patients moving. According to Professor Normand, it's the known benefits of bed management, rather than the imagined benefits of a beds model that the government should be pursuing: 'Really careful management of the bed stock and use of community step-down facilities are so important. I think a formulaic top-down approach, saying: 'You must use x number of beds' kind of misses the point. '

But a government-approved model for beds would at least give a much-needed 'national perspective', says Institute of Healthcare Management deputy chief executive Suzanne Tyler.

A 'phenomenal' number of beds were lost in the 1990s because decisions were being made 'without an overall picture', she says.

'Mr Milburn's model would at least give a starting point that you can then put a structure around. All of the deprivation indexes will have to be calculated in, and issues around bed location will depend on things like transport. I would have thought that head of population alone would be too simplistic. '

Whatever the shape of next month's model, nothing much will change unless clinicians can be persuaded to implement it.

Professor Normand knows just how tricky staff can be. He once kept an office in a hospital where the daily empty beds count remained stubbornly low: 'The nurses used to wheel empty beds on to the balconies until the count was finished. 'Beat that, Mr Milburn.