Prescriber feedback in England

Introduction
The provision of primary care in England is quite different from Australia. Virtually all patients are treated within the National Health Service (NHS) and allocated to individual general practitioners who usually work in groups. Each practitioner is responsible for the care of approximately 1900 individual patients. Almost all drugs that are prescribed are provided through the NHS and recorded in a large computer databank which was established by the Prescription Pricing Authority (PPA) for financial reimbursement of community pharmacists and dispensing doctors. For administrative purposes, England is divided into 100 health authorities, which have populations of about 0.25-1.0 million.

Feedback
In 1991, the Department of Health was concerned about rising prescribing costs. It started a program to give feedback to practitioners about their prescribing. The data source for this feedback was the PPA's reimbursement system. Each doctor was sent information in a quarterly paper, the prescription analysis and cost (PACT) report. Originally, the reports were of 3 types with differing 'levels' of detail. Now there are two: a standard report, and the detailed catalogue which is available on request.

The standard report
This contains details on the main sections of the British National Formulary (BNF), the doctor's 20 most commonly prescribed drugs, and their top 40 based on cost, together with local averages (Table 1). At present, the highest cost drugs in England are ulcer-healing drugs (H₂ antagonists and proton pump inhibitors). The unit used in the PACT report is the `item', i.e. each individual prescription of any medicinal product, no matter how large or small the quantity.

The detailed catalogue

This gives an analysis by drug group or therapeutic category, and time period. It provides information on all prescription items of the doctor and practice showing the actual product, the number of tablets prescribed on any prescription, the cost of each individual prescription item and the total cost for each product and drug.

Supervising prescribing
Each health authority has its own medical and pharmaceutical advisers whose role is to monitor prescribing, set prescribing budgets and to influence its practitioners' prescribing behaviours. The PACT analysis data set is a useful means of monitoring prescribing, but since individual prescription data are not linked to patient or diagnosis, the tool is rather blunt. Nevertheless, groups of practitioners have been able to develop quality markers based on PACT indicators. In addition, these data sets have been used to support financial incentive schemes, in which practitioners are rewarded for changing prescribing behaviour in a direction which is viewed as optimal. These schemes seem to have resulted in behaviour change. Generic prescribing, which has been encouraged by the Department of Health and health authorities as a means of reducing cost growth, has consistently increased. Currently, 60% of prescriptions are for generic drugs.

Other indicators that have been applied to PACT were produced by the U.K. Audit Commission. These were derived by examining prescribing behaviour in practices that were regarded by medical advisers as having 'good' behaviour and are therefore, perhaps, less objective than quality markers agreed by consensus groups.

Until 1991, most general practitioners in England were unaware of their prescribing patterns. PACT data not only allow them to look at their own behaviour, but provide a comparison with other practitioners in their health authority. Health authorities are similarly compared, using amalgamated data sets. In this way, prescribing behaviour can be compared across the country.

Conclusion
The PACT system's strength is that it provides information on all prescribing within the NHS. Its weakness is that it does not give details of the diagnosis, or indication for the prescription. Nevertheless, for drugs that have single indications (e.g. insulin) or where prescribing is undesirable (e.g. cerebral vasodilators or appetite suppressants), PACT data can be useful. The data also enable prescribing advisers and practitioners to plot trends in prescribing behaviour in response to interventions, and monitor the uptake of new drugs. When PACT was first introduced, prescribing costs in primary care were rising at around 13% annually. Since then, the rate of rise has steadily fallen to around 8%, where it seems to have levelled (10 year average of 10% per annum).

In common with many NHS strategies, there has been no controlled study to measure the effectiveness of PACT. It has generally been well received by general practitioners and prescribing advisers, and prescribing trends suggest an impact on some aspects of prescribing behaviour. An improvement would be to give information in the form of defined daily dose rather than items, since the latter can vary in quantity from practice to practice.

Further reading
Bateman DN, Campbell M, Donaldson LJ, Roberts SJ, Smith JM. A prescribing incentive scheme for non-fund holding general practices: an observational study. Br Med J 1996;313:535-8.

Bateman DN, Eccles M, Campbell M, Soutter J, Roberts SJ, Smith JM. Setting standards of prescribing performance in primary care: use of a consensus group of general practitioners and application of standards to practices in the north of England. Br J Gen Pract 1996;46:20-5.

A C K N O W L E D G E M E N T
I am grateful for comments from Dr John Ferguson, Medical Director of the Prescription Pricing Authority.