We thank Dr. Alter, Dr. Mascitelli and Mr. Vos, and Dr. Goldstein for their careful reviews of our paper. Treatment effect was qualitatively similar in participants age 35 to 64 years and age 65 to 75 years, providing some reassurance that age does not importantly modify treatment effect. Because treatment duration was 4.9 years, some follow-up is derived from participants older than 75 years. Nonetheless, we agree that the TNT study results do not provide definitive efficacy or safety information for the 5% of the U.S. population older than 75 years. A manuscript comparing outcomes by sex is being published (1).

The primary study outcome for TNT overall, and this subanalysis in particular, was time to first occurrence of a major cardiovascular event, such as coronary heart disease death; nonfatal, non–procedure-related myocardial infarction; resuscitated cardiac arrest; and fatal or nonfatal stroke (2). Angina was not included in the primary TNT end point, and neither the overall trial nor the subanalysis was powered to determine whether small differences in total mortality were real or were simply due to chance. Apparent differences in total and noncardiovascular mortality between subgroups should be interpreted with great caution, especially in an older population in which competing causes of death (such as cancer) closer to the end of the normal life span are likely to play a larger role. Analyses by type of cancer in this and other statin trials do not show any organ specificity (3, 4). Ancillary analysis of noncardiovascular mortality in TNT (4) showed no relationship between cancer mortality and achieved low-density lipoprotein cholesterol level; participants in the lowest quintile of achieved low-density lipoprotein cholesterol (the majority taking 80 mg of atorvastatin) had the lowest cancer mortality rate (4). Data on incidence of nonfatal cancer by age subgroup are not available. Given study design and power, such analyses are unlikely to yield definitive results. We agree that our study does not support intensive lipid-lowering therapy among older or younger patients to reduce total mortality; but mortality should not be the sole measure of treatment benefit. Nonfatal cardiovascular events, such as myocardial infarction and stroke, significantly affect functional status and quality of life, worsen prognosis, and result in substantial health care expenditures. Nonetheless, a 2008 hierarchical meta-analysis of statin therapy versus placebo in patients age 65 years or older showed a reduction in all-cause mortality (relative risk reduction of 22% over 5 years) (5). We affirm our conclusion that intensive lipid-lowering therapy (atorvastatin, 80 mg daily vs. 10 mg daily) in patients age 65 years or older with established coronary heart disease prevents potentially disabling cardiovascular events, with an absolute risk reduction similar to that in younger individuals.