Norrie disease

Authors did not use latest version of Sheffield table

EDITOR--Isles et al have compared the Sheffield, New Zealand, and joint British methods for estimating risk of coronary heart disease.[1] But they used an earlier version of the Sheffield table and not the current table, published in the stone issue of the journal.[2]

Earlier versions of the Sheffield table included left ventricular hypertrophy on electrocardiography as a principle risk factor, but this proved too complex for many general practitioners (and, presumably, practice nurses). Could the authors confirm that the Sheffield table they studied included left ventricular hypertrophy and that the other methods compared with it did not, and would they comment on how this may have influenced the preferences expressed?

Their conclusions on the accuracy of the three methods differ from other data available. In our study (in preparation) the Sheffield and joint British methods proved similarly accurate in their measurement of coronary risk, but the New Zealand chart was much less accurate (table). The Sheffield table was even slightly more accurate than the New Zealand chart for estimating cardiovascular rather than coronary risk (table). We are aware of two unpublished, large, independent studies in different patient populations that confirm the lesser accuracy of the New Zealand chart but similar accuracy for the Sheffield and joint British methods.

Given the lower accuracy of the New Zealand chart and the fact that it estimates five year cardiovascular risk whereas British and European guidelines are expressed in terms of 10 year coronary risk, the New Zealand chart is no longer suitable for use in Britain or the rest of Europe. More suitable paper based options for implementing recent British guidelines are the new Sheffield table[2] and the joint British chart,[3] which have similar accuracy. When choosing between these it is important to remember that the joint British chart is a risk assessment method and nothing more. The new Sheffield table is a risk assessment method and an accurate screening tool and provides a summary of current guidelines on a single page.

Erica J Wallis research assistant ej.wallis@sheffield.ac.uk

Lawrence E Ramsay professor of clinical pharmacology and therapeutics

Joseph I N M Yikona research fellow Peter R Jackson reader in clinical pharmacology and therapeutics

Clinical Pharmacology and Therapeutics, Royal Hallamshire Hospital, Sheffield S10 2JF

Competing interests: None declared.

[1] Isles CG, Ritchie LD, Murchie P, Norrie J. Risk assessment in primary prevention of coronary heart disease: randomised comparison of three scoring methods. BMJ 2000;320:690-1. (11 March.)

[2] Wallis EJ, Ramsay LE, Haq IU, Ghahramani P, Jackson PR, Rowland-Yeo K, et al. Coronary and cardiovascular risk estimation for primary prevention: validation of a new Sheffield table in the 1995 Scottish health survey population. BMJ 2000;320:671-6. (11 March.)

[3] Wood D, Durrington PN, Poulter N, McInnes G, Rees A, Wray R on behalf of the British Cardiac Society, British Hyperlipidaemia Association, British Hypertension Society and endorsed by the British Diabetic Association. Joint British recommendations on prevention of coronary heart disease in clinical practice. Heart 1998;80(suppl 2):S1-29.

Authors' reply

EDITOR--We used the penultimate version of the Sheffield table mainly because the latest version was not available to us at the time of our study. The penultimate Sheffield table does not include left ventricular hypertrophy on electrocardiography as a risk factor, and so this cannot have been the reason why general practitioners and nurses expressed a preference for the New Zealand and joint British charts.

Our study was not a test of the accuracy of the three risk assessment methods but was designed to test how well general practitioners and nurses interpreted the three methods and to determine which they preferred. Our main findings were that some nurses had difficulty interpreting the Sheffield table and that both general practitioners and nurses preferred the New Zealand and joint British charts.

We disagree with Wallis et al's statement that the joint British chart is a risk assessment method and nothing more. Like the Sheffield table and New Zealand guideline it can be used as a screening tool: it gives clinicians an opportunity not to measure serum lipid concentrations if it is clear from the patients' age, sex, smoking habit, blood pressure, and glucose tolerance that their risk of coronary heart disease is unlikely to exceed the threshold for drug intervention.

Christopher Isles consultant physician Medical Unit, Dumfries and Galloway Royal Infirmary, Dumfries DG1 4AP C.Isles@dgri.scot.nhs.uk

Lewis Ritchie Mackenzie professor of general practice Department of General Practice and Primary Care, University of Aberdeen, Aberdeen AB42 2AY

Competing interests: None declared.

Accuracy of Sheffield table, joint British chart, and New Zealand chart compared with full Framingham function in estimating 15% risk of coronary heart disease over 10 years and 20% risk of cardiovascular disease over 10 years

COPYRIGHT 2000 British Medical Association
COPYRIGHT 2000 Gale Group