Published: 29/07/2004, Volume II4, No. 5916 Page 23

A fresh attempt to measure NHS outputs should help us see if we are getting value for money, says John Appleby

For years the key summary measure of NHS outputs has been the cost-weighted activity index (CWAI).

And for as long as it has existed, it has been subject to criticism for failing to reflect properly what the NHS actually produces. But now economists at the Department of Health and Office for National Statistics are inching towards what will hopefully turn out to be a truer measure of what the NHS does.

It would appear that with a few strokes of their calculator keys, economists have managed to increase NHS outputs. In so doing they have also, albeit very slightly, increased total gross domestic product - ironically making it slightly harder for the government to reach its spending target as a percentage of GDP.

Despite some criticisms prompted by the timing of the new output figures, these revisions are long overdue. The old output measure added together very broad categories of NHS activity - inpatients and day cases, ambulance journeys and so on; 16 categories in all. In order to be able to add together such disparate types of activity, each was weighted by their share of total expenditure. So, a unit of inpatient contributed much more to overall outputs than, say, an ambulance journey because it took a much bigger share of spending. And within each category all activities were equally weighted.

The coarseness of the old categories and the fact that many NHS activities were simply ignored meant that the CWAI has misrepresented the level of outputs. But until recently it has not been possible to say how inaccurate the CWAI has been. Now, the new measure uses over 1,700 categories of activity which are then bundled up into one grand output using unit costs reported in the DoH's national reference costs database as weights.

The result of all this has been a reported increase in NHS outputs from 1996-2003 of 28.5 per cent compared with an increase of 19 per cent under the old measure.

These revisions are only an interim step, however. An implicit assumption retained in the new measure of output is that the quality of that output remains unchanged over time: a knee replacement in 1996 is of the same quality as one carried out in 2003.

But technology moves on, techniques improve, more effective drugs are substituted for less effective ones and so on. This means that unless it is thought that the changes in unit costs used to weight changes in knee replacement activity reflect changes in quality (unlikely), the value of NHS output has changed, but has not been adequately captured by the unit cost weights. It is this issue of quality changes (and also, whether to focus on NHS outcomes, rather than outputs) which is the subject of current research commissioned by the DoH.

Despite the rather technical, if not arcane, nature of this work (base-weighted Laspeyres index, or Paasche index anyone? ), getting the NHS output measure right is essential in answering basic questions such as what we are getting for our money. Of course, whether what we get is produced efficiently or does us any good is another matter.

Professor John Appleby is chief economist at the King's Fund.