Table 2.

Gain in efficiency for total slope and chronic slope compared with the clinical outcome when the long-term treatment effect is intermediate between uniform and proportional

Mean Baseline GFR (ml/min per 1.73 m2)Mean Slope (ml/min per 1.73 m2 per year)Total SlopeTotal SlopeChronic Slope
No Acute EffectAcute Effect =−1.25 ml/min per 1.73 m2Acute Effect =−1.25 ml/min per 1.73 m2
Relative EfficiencyRequired N for Clinical Outcome in a 4–6 yr RCTaRelative EfficiencyRequired N for Clinical Outcome in a 4–6 yr RCTRelative EfficiencyRequired N for Clinical Outcome in a 4–6 yr RCTa
Total Slope in 2 yr RCT versus Clinical Outcome in 4–6 yr RCTTotal Slope in 4–6 yr RCT versus Clinical Outcome in 4–6 yr RCTTotal Slope in 2 yr RCT versus Clinical Outcome in 4–6 yr RCTTotal Slope in 4–6 yr RCT versus Clinical Outcome in 4–6 yr RCTChronic Slope in 2 yr RCT versus Clinical Outcome in 4–6 yr RCTChronic Slope in 4–6 yr RCT versus Clinical Outcome in 4–6 yr RCT
27.5−1.51.141.0749800.370.4171401.270.767140
−3.251.511.5821700.821.4921901.291.812190
−5.01.241.408701.131.539601.321.77960
42.5−1.50.711.1141300.280.3950100.711.685010
−3.251.411.7117500.261.2319401.132.321940
−5.01.171.618300.641.599301.262.33930
67.5−1.51.061.8360900.460.4682401.182.838240
−3.251.282.1624800.160.6929401.423.652940
−5.01.632.4213100.091.2512601.363.291260
  • All calculations assume a 25% intermediate long-term effect. Relative efficiencies are given by the ratio of sample size (N) for the clinical end point over 4–6 yr versus the slope analysis over the indicated follow-up period. Relative efficiencies greater than one indicate that a smaller sample size is required to achieve the same statistical power with the slope outcome over the indicated follow-up period compared with the clinical end point over 4–6 yr. See Supplemental Figure 6 for additional detail including simulation SEM.

  • a Increases in required N for the clinical end point with baseline GFR=27.5 compared with 42.5 ml/min per 1.73 m2 are due to assumptions that the acute effect is smaller at lower GFR levels, a higher proportion of slow progressors reaching clinical events with smaller effect sizes under the intermediate long-term treatment effect model, and larger effects of random error in GFR measurements on end points requiring smaller GFR change.