All over the developed world, public and private health insurance organisations are attempting to reduce the cost of prescribing in general practice. Many use financial incentives as carrots and/or sticks (U.K., Germany, Netherlands, Republic of Ireland). Some use obligatory prescribing guidelines with financial rewards for those who observe them and penalties for those who repeatedly ignore them (France). Others use strictly limited lists of subsidised drugs (Canada, Australia, parts of the diverse U.S.A. primary care system).

The educational approach
Beginning in Northern Ireland in 19761, continued in the U.S.A. in 19862, taken up briefly in England in 19923 and more enthusiastically in Sweden in 19914 and in Germany in 19935, educational approaches were introduced. Each country used different feedback methods and all were successful for the duration of the intervention, but the effect decayed once feedback ceased. They were better than financial incentives because general practitioners tend to adopt anything which they believe will improve the quality of their diagnosis and treatment. Doctors will discard therapy which they are persuaded is outdated, illogical (not evidence-based) or economically irrational. The educational approach is powerful and results in permanent change for the best of reasons - that the doctor is better informed. This is not the case for financial incentives - when these cease, the doctor is none the wiser.

Northern Ireland
Since 1976, general practitioner prescribing education has involved practice visits by an experienced former general practitioner or pharmacist. The visitor discusses the practice's own historical prescribing patterns. This began as a simple review of drug groups, but gradually became more and more complex, more and more educational and less and less economically orientated.

In 1992, the Drug Utilization Research Unit in Belfast developed a fully computerised prescribing feedback system called 'COMPASS'. This takes every general practitioner's prescribing, every month, and compares it, in all therapeutic groups, against an ideal template. Deviations from the prescribing `ideal' are noted and the general practitioners receive an attractive, easily-read report, giving the clinical, pharmacological or economic reasons why they might consider modifying their prescribing.6

COMPASS has been adopted by Cape Province in South Africa and by the Icelandic National Health Service. Although this system was developed by an academic unit with quality and rationality as the primary aim, full adoption of its recommendations would allow the average general practitioner to achieve savings 5 times greater than generic substitution alone. In Northern Ireland, this is 15% of the total general practitioner drug budget.

COMPASS works better if it is taken to a practice for discussion, by an experienced prescribing visitor, and best of all if it is accompanied by a positive financial incentive. For example, in the U.K. and the Republic of Ireland, general practitioners were allowed to use a percentage of prescribing savings to improve any aspect of their practices approved by Health Boards.

The ideal prescribing visitor is a former general practitioner who has made pharmacology and therapeutics a second career. Pharmacists can be almost as effective once they have established good relationships with the doctors in their districts. This mirrors the new discipline of clinical pharmacy in hospitals. An area pharmaceutical adviser in Northern Ireland achieved savings of £1.6 million and £2.2 million in her first two years of prescribing liaison, in only 55 practices (108 general practitioners).

The major benefit of COMPASS, and of all academically-orientated feedback programs, is that it empowers doctors to select the most economically and scientifically rational prescription in therapeutic areas where several drugs might be effective, but where only one fulfils the criterion of cost-effectiveness. The other benefit is that such feedback is a powerful educational tool, making doctors aware of current evidence and, if used regularly, keeping them up to date throughout their professional lives.


  1. McGavock H. Rational discussion as a guide to more cost-effective and scientific prescribing in general practice:15 years of experience in Northern Ireland. In: Kochen MM, editor. Rational pharmacotherapy in general practice. Berlin/Heidelberg: Springer-Verlag, 1991:186-200.
  2. Soumerai SB, Avorn J. Economic and policy analysis of university-based drug 'detailing'. Med Care 1986;24:313-31.
  3. Newton-Syms FA, Dawson PH, Cooke J, Feely M, Booth TG, Jerwood D, et al. The influence of an academic representative on prescribing by general practitioners. Br J Clin Pharmacol 1992;33:69-73.
  4. Ekedahl A, Hovelins B, Molstad S. Regional 2-day seminars in primary health care - influence on prescribing habits. Apoteksbolaget Report, Stockholm, 1991:1-6.
  5. Von Ferber L. Auditing drug therapy by peer review - the German experience. Pharmacoepidemiol Drug Safety 1993;2:195-200.
  6. McGavock H, Wilson-Davis K, Rafferty T. A 'compass' for general practitioner prescribers. Health Trends 1994;26:28-30.