RAPID RESPONSES FROM BMJ.COM: Drawbacks of primary prevention risk tables to assess  cardiovascular risk in type 2 diabetes

doi:10.1136/bmjusa.03040003

BMJ, doi: 10.1136/bmjusa.03040003, (Published 6 May 2003)

Letter

RAPID RESPONSES FROM BMJ.COM

Drawbacks of primary prevention risk tables to assess cardiovascular risk in type 2 diabetes

Following is an edited excerpt from one of the Rapid Responsesgenerated by this article, which can be read in their entiretyat http://bmj.com/cgi/eletters/326/7383/251 — Editor

From BMJ USA 2003;April:202

Editor — The proposal implied by Hall et al would leadto a rather rigid prescribing protocol whereby all patientswith scores above a threshold (eg, 15%) would receive additionaltherapy, and others would not. In addition, the overemphasison risk scores might ignore other key risk factors not representedby the Framingham risk equation, such as ethnicity, familyhistory, microalbuminuria, and triglyceride concentration.The low baseline prevalence of diabetes in the Framingham cohortleads to wide confidence intervals in the predicted risk (±5-10%).Thus, in a diabetic man with average risk factors, the upper95% confidence interval crosses the 15% 10-year threshold from age 40 onwards. It is quite plausible that individuals withrisk scores below a chosen cutoff have a higher true risk.The authors' proposal might lead to withholding treatment ina significant number of individuals.

It is conceivable that as new evidence emerges, a diagnosisof type 2 diabetes will be a prima facie reason to commencelipid-lowering therapy. Until such time, risk calculationsbased on Framingham data will continue to be incorporated intoguidelines. However, we suggest that an overemphasis on thisscore to determine therapeutic decisions is unhelpful and mightlead to the under-treatment of a significant number of patients.

Jamie C Smith, specialist registrar diabetes and endocrinology

Bristol Royal Infirmary, Bristol, UK jamie.smith{at}virgin.net