Background: Purchasers may now define how many clinical procedures they wish to buy. In the investigation of chest pain, coronary angiography provides a definitive diagnosis. In clinical practice, the use of coronary angiography varies, so purchasers may choose the sort of service they wish to buy. We reviewed the extent of variation in the investigation of patients in the Trent Region and present a system for establishing criteria for the purchase of coronary angiography.
Methods: Three studies are presented, based on patients referred to the three referral centres in the Trent Region. First, the appropriateness for investigation, defined as benefit outweighing risk, is determined by expert panel; second, an audit of patients referred for coronary angiography is presented; and third, the likelihood or probability of significant coronary disease is assessed by application of a logistic regression model.
Results: According to the panel criteria, 27 per cent of patients in one centre were judged inappropriate for coronary angiography, 28 per cent in the second centre and 10 per cent in the third. When referred for angiography, extent of anti-anginal medication, symptom severity, use of exercise testing, and result of and planned management after angiography varied widely among centres. The likelihood of significant coronary artery disease could be predicted in 86 per cent of patients.
Conclusions: Variation in clinical practice does not make the purchasers' task easy. Based upon these studies and clinical trials of coronary artery bypass surgery and coronary angioplasty, we present potential performance indicators which may form the basis of contracts or be used as audit measures.