'Most of our customers survive the NHS even though we do not measure whether they feel better as a consequence'

NHS management is moving increasingly quickly from old-style crisis firefighting to strategic evaluation of performance using a combination of financial, activity and outcome data. This process may be uncomfortable for many.

Payment by results is obliging hospitals to count what they do in order to claim income. The Department of Health and the Audit Commission are shocked at how poor these processes are and will be rattling cages in an effort to get improved levels of billing to primary care trusts. In time, this problem will convert from under-claiming to over-claiming. This is what the Americans call 'DRG [diagnostic resource group] creep', where finance directors 'warp' the rules to over-bill and increase revenue.

Playing these financial games will focus finance directors' attention on activity. The DoH has published productivity matrices (www.productivity.nhs.uk) which are derived from routine NHS data, in particular hospital episode statistics. This material has been collected for nearly 20 years and largely ignored by NHS managers. With PbR, managers are realising HES nicely describes their consultants' activity and that this material is an essential ingredient of their job planning, appraisal and the allocation on consultant excellence awards.

PbR has already required hospitals to invest in better coding of activity and improve the training of their workforce. However, the collection of better data is only part of the reforms. The next problems to be resolved are the analysis of the data and its use in decision-making by managers and boards. Most hospitals and PCTs have little or no analytical capacity. Consequently what is emerging is central capacity to provide analysis with the NHS funding private entrepreneurs such as CHKS and Dr Foster Intelligence to manipulate and repackage publicly provided data such as HES.

This is a nice product of the NHS lacking the skills to do this work and the DoH being slimmed down to levels of 'efficiency' that result in its not having the personnel to analyse and disseminate quantitative information.

There is an urgent issue about quality control of such streams of information to the NHS. For instance, with such data as the productivity matrices there is an issue of whether to include hospitals with high levels of specialisation as they may bias measures of dispersion and estimates of potential activity and resource savings. Hopefully peer review of this work will inform its fine tuning by public commissioners.

Doctors are obsessed with measures of failure, such as mortality and complications. However, most of our customers survive their encounters with the NHS even though we do not measure whether they feel better as a consequence.

But times are changing. CHKS is working with four NHS trusts introducing patient-reported outcome measures that will report changes in physical and mental well-being of patients before and after their care. They are using a reduced version of SF36 (www.sf36.org) and the EQ5D (www.euroqol.org). And BUPA, which has been investing in outcome measurement since 1998, has taken over Outcome Technologies, which is beginning to market outcome measurement to public and private providers. Hopefully it too will use SF36 and EQ5D so their comparative merits can be identified.

NHS managers have to understand HES and the activity data it yields and face the issues around the collection and management of data from outcome measurement. This is an ambitious agenda.

Transforming NHS managers in this way requires major changes in recruitment, education, training and lifetime skills acquisition. Ministers are impatient for change and the DoH's advocacy needs to be heeded as it considers how to develop swiftly quantitative management in routine NHS practice with new incentives and performance mechanisms.

Professor Alan Maynard is director of the health policy group at York University.