Early stopping to accept H(o) based on conditional power: approximations and comparisons.

It is intuitively appealing to clinicians to stop a trial early to accept the null hypothesis Ho if it appears that this will be the likely outcome at the planned end of the trial. We consider procedures that calculate at each time point the conditional probability of rejecting Ho at the end of the trial given the current data and some value of the parameter of interest. Lan, Simon, and Halperin (1982, Communications in Statistics C1, 207-219) calculate this probability under the design alternative, and Pepe and Anderson (1992, Applied Statistics 41, 181-190) use an alternative based solely on the current data. We investigate a modification to Pepe and Anderson's (1992) procedure that has a more satisfying interpretation. We define all of these procedures as formal sequential tests with lower stopping boundaries and study them in this context. This facilitates an improved understanding of the interplay of parameters by introducing visual displays, and it leads to an approximation for power by treating it as a boundary crossing probability. We use these tools to compare the performances of the different designs under a variety of parameter configurations.

[1]  K. K. Lan,et al.  Stochastically curtailed tests in long–term clinical trials , 1982 .

[2]  Choi Sc,et al.  Monitoring clinical trials based on predictive probability of significance. , 1989 .

[3]  D. Siegmund.  Sequential Analysis: Tests and Confidence Intervals , 1985 .

[4]  J. O'fallon,et al.  Stopping when the experimental regimen does not appear to help. , 1994, Statistics in medicine.

[5]  Anastasios A. Tsiatis,et al.  Group sequential designs for one-sided and two-sided hypothesis testing with provision for early stopping in favor of the null hypothesis , 1994 .

[6]  D. Siegmund.  Large Deviations for Boundary Crossing Probabilities. , 1982 .

[7]  S S Ellenberg,et al.  An efficient design for phase III studies of combination chemotherapies. , 1985, Cancer treatment reports.

[8]  S. Pocock.  Group sequential methods in the design and analysis of clinical trials , 1977 .

[9]  Margaret S. Pepe,et al.  Two-stage experimental designs : early stopping with a negative result , 1992 .

[10]  D. Siegmund.  Boundary Crossing Probabilities and Statistical Applications , 1986 .

[11]  Barry R. Davis,et al.  Upper bounds for type I and type II error rates in conditional power calculations , 1990 .

[12]  D. Spiegelhalter,et al.  Monitoring clinical trials: conditional or predictive power? , 1986, Controlled clinical trials.

[13]  A. Tsiatis,et al.  Approximately optimal one-parameter boundaries for group sequential trials. , 1987, Biometrics.

[14]  J. Herson,et al.  Predictive probability early termination plans for phase II clinical trials. , 1979, Biometrics.

[15]  Christopher Jennison,et al.  Statistical Approaches to Interim Monitoring of Medical Trials: A Review and Commentary , 1990 .

[16]  P. Armitage,et al.  Repeated Significance Tests on Accumulating Data , 1969 .

[17]  R A Betensky,et al.  Conditional power calculations for early acceptance of H0 embedded in sequential tests. , 1997, Statistics in medicine.

[18]  Xiaoping Xiong,et al.  A Class of Sequential Conditional Probability Ratio Tests , 1995 .

[19]  David L. DeMets,et al.  Group sequential methods for clinical trials with a one-sided hypothesis , 1980 .

[20]  P. O'Brien,et al.  A multiple testing procedure for clinical trials. , 1979, Biometrics.

[21]  David L. DeMets,et al.  Asymmetric group sequential boundaries for monitoring clinical trials , 1982 .