. 2022 Sep 18;41(28):5517–5536. doi: 10.1002/sim.9580

TABLE 4.

For Scenario 3 with continuous outcome, interim analysis performance using each estimator with O'Brien‐Fleming and Pocock stopping boundaries under (a) the null hypothesis $β = 0$ and (b) the alternative $β = 6.24$ , with maximum sample size $n_{\max} = 300$ and $t_{e n d} = 208$ days

P(reject)

MC E(SS)

MC E(Stop)

P(reject)

MC E(SS)

MC E(Stop)

(a) Null hypothesis

O'Brien‐Fleming

Pocock

{\hat{β}}_{𝒯_{F}} (t)

0.026

299.9 (3.1)

207.4 (5.5)

0.025

298.6 (11.3)

206.2 (12.7)

{\hat{β}}_{I P W} (t)

0.026

299.8 (3.4)

207.4 (5.8)

0.025

298.6 (11.5)

206.1 (12.9)

{\hat{β}}_{A I P W 1} (t)

0.025

299.9 (2.6)

207.5 (5.0)

0.026

298.6 (11.2)

206.2 (12.7)

{\hat{β}}_{A I P W 2} (t)

0.026

299.8 (4.0)

207.0 (7.4)

0.029

298.1 (13.2)

205.7 (14.5)

(b) Alternative hypothesis

O'Brien‐Fleming

Pocock

{\hat{β}}_{𝒯_{F}} (t)

0.875

286.3 (26.0)

167.8 (30.6)

0.823

259.9 (45.2)

151.9 (41.9)

{\hat{β}}_{I P W} (t)

0.876

286.0 (26.4)

167.4 (30.7)

0.826

259.1 (45.4)

151.1 (41.9)

{\hat{β}}_{A I P W 1} (t)

0.930

286.5 (25.5)

165.5 (28.9)

0.896

255.9 (45.2)

146.0 (39.6)

{\hat{β}}_{A I P W 2} (t)

0.930

265.9 (37.3)

146.7 (31.2)

0.892

239.8 (45.3)

133.5 (37.8)

Note: Entries are as in Table 2. The standard error for entries for P(reject) in (a) is $\approx 0.0016$ .