to the editor: In this “Tips from Other Journals,” Dr. Crawford-Faucher reviewed the meta-analysis by Brugts and colleagues that concluded that statins are beneficial for primary prevention of cardiac disease.1 I agree with this conclusion, but with some reservations. The review stated that the relative risk reduction for all-cause mortality was 12 percent, which sounds very good; however, I think it is important to look at the absolute risk reduction. All-cause mortality after a mean follow-up of 4.1 years was 5.1percent in the group treated with statins and 5.7 percent in the control group. That translates into an absolute risk reduction of 0.6 percent and a number needed to treat (NNT) of 167. This means that 167 patients would need to be treated with a statin for 4.1 years to prevent one death. Additionally, based on the study data, the NNT to prevent one major coronary event is 77, and the NNT to prevent one major cerebrovascular event is 250.
The studies that were used in the meta-analysis included persons with diabetes mellitus. I would disagree with the authors of the meta-analysis who suggested that there is little reason to suspect different treatment effects between persons with diabetes and those without diabetes on a pathophysiological basis. This statement seems to contradict the concepts that persons with diabetes have different low-density lipoprotein treatment goals than those without diabetes, and that with regard to lipid-lowering therapy, diabetes is considered a coronary disease equivalent.2
Finally, statins are not completely benign drugs and can be expensive. Except for noting that there does not seem to be an increased risk of cancer secondary to statins, no other harms were mentioned. What are the harms of statins and what is the number needed to harm? With a NNT above 75, a number needed to harm would be helpful in determining if we should prescribe these medications to our patients.
It would be interesting to see if a patient's decision whether or not to take the drug was influenced by using one of the following statements: “This drug will decrease your chance of dying by 12 percent.”; or “Out of the 167 patients that take this pill for the next four years, one patient will not die who would have if he or she had not taken the drug, but 166 will take the medicine and get no benefit, and may be harmed.”
There is a big difference between relative and absolute risk reductions. Conveying the NNT to our colleagues and patients provides a more accurate assessment of the effectiveness of a drug, and might help physicians and patients make more informed decisions.