NUMBER-NEEDED-TO-TREAT AND CARDIAC REHABILITATION

Abstract

The number-needed-to-treat [NNT], the reciprocal of the absolute risk reduction, is used to estimate the benefits and risks associated with treatment so providing an additional measure of treatment effectiveness in this era of cost-containment. If the minimal important difference for an outcome is known, then the proportions of patients improving, stable, or deteriorating can be determined. Improved health-related quality of life [HRQL] is an important, clinically meaningful, and attainable outcome of cardiac rehabilitation. Data were derived from two randomized controlled trials of rehabilitation after myocardial infarction, the Cardiac Rehabilitation in Advanced Age trial [CR-AGE; n = 270] with generic HRQL and the McMaster Early Rehabilitation Study [MERS; n = 201] with specific HRQL as the outcomes. We estimated the NNT [95% confidence limits, CI] with each HRQL outcome. In CR-AGE, the generic global SIP score increased significantly with rehabilitation [p = 0.010] but with no differences in the proportions of rehabilitation and usual care patients improving [0.40 vs. 0.32], stable [0.58 vs 0.63], or deteriorating [0.02 vs 0.05]. The marginal proportion improving with rehabilitation was 0.089 and the estimated NNT = 11.2 [95% CI, 4.6, −25.6]. In MERS, the specific global QLMI score increased significantly with rehabilitation [p = 0.005] and a significant difference [p = 0.036] in the proportions of rehabilitation and usual care patients improving [0.63 vs. 0.44], stable [0.35 vs 0.51], or deteriorating [0.02 vs 0.05]. The marginal proportion improving with rehabilitation was 0.19 and the estimated NNT = 5.4 [95% CI, 3.0, 21.0]. In conclusion, and as hypothesized, the NNT estimated from specific HRQL as the outcome measure was smaller than the NNT with generic HRQL as the outcome measure.