Further Comments on the Alpha Spending Function



The use of alpha spending functions has become a powerful tool in designing group sequential trials since 1990. In this manuscript, we present practical problems we encountered and suggest possible solutions. Since group sequential methods deal with only one endpoint, they have certain intrinsic limitations in medical research. As a result, we often modify the design in conducting clinical trials since the sequential statistical framework is not sophisticated enough to handle many practical problems related to multiple endpoints in clinical trials. Topics discussed include the choice of a primary endpoint, the use of symmetric and asymmetric boundaries in interim analysis, data-driven interim analyses, overruling of a group sequential boundary, comparison of two means and the use of logrank tests in survival trials.


Two one-sided hypotheses testing Alpha and beta spending Sequential boundaries and stopping guidelines Proportional hazards assumption Information scale 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Armitage P, McPherson CK, Rowe BC (1969) Repeated significance tests on accumulating data. J R Stat Soc, Ser A 132:235–244 CrossRefMathSciNetGoogle Scholar
  2. 2.
    Chen JYH, DeMets DL, Lan KKG (2003) Monitoring mortality at interim analyses while testing a composite endpoint at the final analysis. Control Clin Trials 24:16–27 CrossRefGoogle Scholar
  3. 3.
    Cook T, DeMets DL (2008) Introduction to statistical methods for clinical trials. Chapman & Hall/CRC, Taylor & Francis Group, LLC, Boca Raton MATHGoogle Scholar
  4. 4.
    DeMets DL, Ware JH (1980) Asymmetric group sequential boundaries for monitoring clinical trials. Biometrika 67:651–660 CrossRefMathSciNetGoogle Scholar
  5. 5.
    Ellenberg S, Fleming T, DeMets D (2002) Data monitoring committees in (clinical trials: a practical perspective). John Wiley & Sons, West Sussex Google Scholar
  6. 6.
    East5 Manual (2007) Cytel Inc., Cambridge, MA Google Scholar
  7. 7.
    Hwang IK, Shih WJ, DeCani JS (1990) Group sequential designs using a family of type I error probability spending functions. Stat Med 9:1439–1445 CrossRefGoogle Scholar
  8. 8.
    Friedman LM, Furberg CD, DeMets DL (1998) Fundamentals of clinical trials, 3rd edn. Springer, New York Google Scholar
  9. 9.
    Kim K, DeMets DL (1987) Design and analysis of group sequential tests based on the type I error spending rate function. Biometrika 74:149–154 MATHCrossRefMathSciNetGoogle Scholar
  10. 10.
    Lan KKG, Hu P, Proschan MA (2009) A conditional power approach to the evaluation of predictive power. Stat Biopharm Res 1(2):131–136. doi:10.1198/sbr.2009.0035 CrossRefGoogle Scholar
  11. 11.
    Lan KKG, Lachin JM (1990) Implementation of group sequential logrank tests in a maximum duration trial. Biometrics 46:759–770 CrossRefGoogle Scholar
  12. 12.
    Lan KKG, Lachin JM, Bautista O (2003) Overruling of a group sequential boundary—a stopping rule versus a guideline. Stat Med 22:3347–3355 CrossRefGoogle Scholar
  13. 13.
    Lan KKG, DeMets DL (1983) Discrete sequential boundaries for clinical trials. Biometrika 70(3):659–663 MATHCrossRefMathSciNetGoogle Scholar
  14. 14.
    Lan KKG, DeMets DL (1989) Group sequential procedures: calendar versus information time. Stat Med 8:1191–1198 CrossRefGoogle Scholar
  15. 15.
    Lan KKG, DeMets DL (1989) Changing frequency of interim analysis in sequential monitoring. Biometrics 45:1017–1020 CrossRefGoogle Scholar
  16. 16.
    O’Brien PC, Fleming TR (1979) A multiple testing procedure for clinical trials. Biometrics 35:549–556 CrossRefGoogle Scholar
  17. 17.
    Pampallona S, Tsiatis AA, Kim K (2001) Interim monitoring of group sequential trials using spending functions for the type I and type II error probabilities. Drug Inf J 35:1113–1121 Google Scholar
  18. 18.
    Pocock SJ (1977) Group sequential methods in the design and analysis of clinical trials. Biometrika 64:191–199 CrossRefGoogle Scholar
  19. 19.
    Proschan MA (1999) Properties of spending function boundaries. Biometrika 86:466–473 MATHCrossRefMathSciNetGoogle Scholar
  20. 20.
    Proschan MA, Follmann DA, Waclawiw MA (1992) Effects of assumption violations on type I error rate in group sequential monitoring. Biometrics 48:1131–1133 CrossRefGoogle Scholar
  21. 21.
    Proschan MA, Lan KKG, Wittes JT (2006) Statistical monitoring of clinical trials: a unified approach. Springer Science+Business Media, LLC, New York MATHGoogle Scholar
  22. 22.
    Reboussin DM, DeMets DL, Kim KM, Lan KKG (2000) Computations for group sequential boundaries using the Lan–DeMets spending function method. Control Clin Trials 21(3):190–207 CrossRefGoogle Scholar
  23. 23.
    Reboussin DM, DeMets DL, Kim KM, Lan KKG (1995) Programs for computing group sequential bounds using the Lan–DeMets method, version 2.1. UW Department of Biostatistics, Technical Report No. 95, October 1995 Google Scholar
  24. 24.
    Spiegelhalter DJ, Freedman LS, Blackburn PR (1986) Monitoring clinical trials: conditional or predictive power? Control Clin Trials 7:8–17 CrossRefGoogle Scholar
  25. 25.
    Taylor AL, Ziesche S, Yancy C, Carson P, D’Agostino R Jr, Ferdinand K, Taylor M, Adams K, Sabolinski M, Worcel M, Cohn JN for the African–American Heart Failure Trial Investigators (2004) Combination of isosorbide dinitrate and hydralazine in blacks with heart failure. NEJM 351(20):2049–2057 CrossRefGoogle Scholar

Copyright information

© International Chinese Statistical Association 2009

Authors and Affiliations

  1. 1.Johnson & Johnson, Pharmaceutical Research & Development, L.L.C.RaritanUSA
  2. 2.Department of Biostatistics & Medical InformaticsThe University of Wisconsin–MadisonMadisonUSA

Personalised recommendations