Must I endorse this 'Citizen Card'?
GPs and their practice teams contribute a staggering £5 billion to the British economy, a contribution that is likely to increase. But how is it measured, and do GPs get their just deserts? Philip Lee explains
The work done by GPs and practice staff makes a massive contribution to the economy of the UK. In fact they contribute around half a per cent of the UK national product that's a contribution of £5 billion!
They do this by keeping people in work, keeping them out of hospital, prescribing drugs effectively and so on. It can never be stressed enough that contributing to the economy is not just a question of 'producing goods'.
With increased emphasis on primary care within the NHS, more patients with more complex needs, and an ever-increasing range of drug and other treatments available in primary care, it seems likely that this output is growing and that GPs' contribution to the NHS has been underestimated.
So how is it measured? And are the measurements accurate?
General Household Survey
For general practice, the main source of information on activity is the General Household Survey (GHS). Questions are asked of people using GPs' services. One question asks respondents if they have made contact with their GP practice in the previous two weeks. This includes such contacts as a visit or a telephone consultation with a GP or practice nurse. Respondents are asked on how many occasions each of these occurred.
Minor operations commissioned by PCTs from GPs also get picked up in the NHS output measure, as part of the Reference Cost data set.
GP prescribing is also included in the NHS output measure. The Prescription Pricing Authority provides data on the number of prescriptions dispensed, broken down into almost 200 categories, each cost-weighted.
Calculations based on data from the GHS show the number of consultations with a GP went up from 217 million to 241 million between 2001/2 and 2002/3, an apparent increase of 11 per cent.
But the characteristics of the sampling scheme and the nature of the target population mean that there is always going to be a margin for error.
The GHS is widely acknowledged as a useful source for data, such as consultation rates for different age groups, and as a snapshot for any one year. But it is not a sufficiently accurate way of measuring growth in consultations with GP practices from one year to the next.
Joe Grice, executive director of the Atkinson Review and deputy head of the Government Economic Service, said: 'GPs do more things now than they did 10 or 15 years ago and we expect that trend to go on.
'If we just measure output by how often people see their GP and are failing to pick up the expanded range of things that used to be done in secondary care, we will be underestimating their contribution.'
The output measure for GPs also depends on the weight given to each type of consultation. The weight used is the average cost of each type of consultation. It is calculated for the Department of Health by the Personal Social Services Research Unit (PSSRU) at the University of Kent.
Things taken in to account include the average length of each type of consultation, including travel time, and the cost of the GP's time and immediate overheads. Clearly, a home visit is likely to cost more than the average consultation in the surgery and this is reflected in the weights currently used.
The calculations made by PSSRU make use of up-to-date information on GP earnings and other cost elements. However, the information on length of consultations of various types dates back to 1992, using a survey then commissioned by the Pay Review Body.
There have been many developments in general practice since 1992 and therefore consultation lengths may have changed. So there are weaknesses in both the volume and cost weights used for consultations in the current NHS output measure.
So improved methods for measuring GPs' output are being proposed. One priority is to consider whether a better data source could be found for estimating the number of consultations and their cost weights.
Work is currently under way by QResearch, based at the University of Nottingham, on their anonymised research database created from GPs' computer records. The database contains records of about 6 per cent of all patients registered with general practice. These come from nearly 500 practices widely spread through the UK. The sample data can be re-weighted to make it more representative of the UK population.
This source will provide data on a much larger number of consultations than currently obtained from the GHS, although its validity will naturally be checked carefully.
Other similar databases exist, or are in development, covering different groups of general practices (including ones for Scotland, Wales and Northern Ireland). Future development will examine valid ways of combining the different sources to give a larger overall sample.
In order to help ensure the quality of the NHS output figures in the National Accounts, it is also clearly very desirable to find a way of updating cost weights.
Philip Lee is in charge of health output measures at the Office for National Statistics
More accurate measure of GPs' work
When current developments are complete, the Office for National Statistics (ONS) will have a more accurate measure of the number of consultations by GPs and other practice staff, as well as the value of GP prescribing.
This is a big step forward. But is it the ideal way of measuring the way GPs contribute to the national income? Almost certainly not.
Areas for exploration must include:
·Better understanding of the effects of a typical consultation, including evidence linking length of consultation to benefits for patients;
·Better ways of measuring the general practice part in a complex care pathway for disease or co-morbidity, rather than counting 'health care episodes' separately;
·Better ways of taking into account the health gain to patients from what GPs do, including the preventive health measures recognised in the new GP contract; and
·More emphasis on patient perspective of the quality of care, for example by exploring use of patient survey information.
These are not easy subjects, but ONS is keen to work with the profession to see how progress can best be made, and comments are very welcome.
·Final Report of the Atkinson Review of the Measurement of Government Output and Productivity for the National Accounts http://www.statistics.gov.uk/atkinson
·Announcement by the National Statistician of the setting up of the new UK Centre for the Measurement of Government Activity http://www.statistics.gov.uk/pdfdir/lcatkinrep0105.pdf
·Note explaining the current methods for compiling NHS output http://www.statistics.gov.uk/cci/article.asp?id=988
·Please contact email@example.com for further information