Author Affiliations: Department of Epidemiology, Tulane University School of Public Health and Tropical Medicine, New Orleans, Louisiana (email@example.com).
In Reply: The letters from Drs Taranikanti and Banerjee and Drs LaRiccia and Morgan raise the issue of the ARR and NNT. Calculating NNT as the inverse of the pooled ARR yielded the following NNTs in our meta-analysis: 130 for stroke, 23 for CHF, 37 for composite CVD outcomes, and 73 for all-cause mortality. Applying the median duration of studies included for each outcome, the corresponding 5-year NNTs would be 94 for stroke, 8 for CHF, 25 for composite CVD outcomes, and 30 for all-cause mortality. These numbers compare favorably with the 5-year NNTs reported in other studies of antihypertensive medication treatment and with those for other CVD prevention measures.1- 3 For instance, among patients with hypertension, the 5-year NNT to avoid a CVD event is 86 for treatment with diuretics3 and 140 for treatment with β-blockers.4 For this same outcome, 5-year NNTs range from 20 to 63 for treatment with statins among patients with hyperlipidemia and from 346 to 426 for men and women for treatment with aspirin.2
Thompson AM, He J, Bazzano LA. Antihypertensive Treatment of Patients With Cardiovascular Disease but Without Hypertension—Reply. JAMA. 2011;305(21):2170-2171. doi:10.1001/jama.2011.706