Abstract
Conditional power combines the findings of a partially completed study with assumptions about the future. The goal is to estimate the probability that the eventual study result will be incompatible with a criterion value, such as acceptable risk or the null hypothesis. Some history and motivation for conditional power calculations are provided, with examples illustrating the application to drug safety studies. This is an expository article suggesting that conditional power, which is well-established in clinical trials research, also has application to observational studies. The utility may be highest in regulatory settings where resources are limited and interim decisions have to be made accurately in the shortest possible time.
Similar content being viewed by others
Notes
The “B” refers to statisticians’ use of the idea of Brownian motion, in which particles in a solution are continuously displaced from their previous position by random collisions. The statistical version is a number series generated by progressive summation, in which each new value to be added to the sum has distributional properties that are independent of the preceding increments.
Assuming a null-hypothesis slope of Θ0 = 0 for the remainder of the study, the expected value of B1 is 4.49, which exceeds the criterion value of 1.96 by 2.53. The exceedance probability is 99.4%.
The lower values with the exact conditional power reflect the fact that the size of the rejection region is typically smaller than its nominal value for an exact significance test with discrete data.
References
Coronary Drug Project Research Group. The Coronary Drug Project: design, methods, and baseline results. Circulation. 1973;47(suppl. 1):I1–79.
Coronary Drug Project Research Group. Clofibrate and niacin in coronary heart disease. JAMA. 1975;231(4):360–81.
Coronary Drug Project Research Group. The Coronary Drug Project: initial findings leading to modifications of its research protocol. JAMA. 1970;214:1303–13.
Coronary Drug Project Research Group. The Coronary Drug Project: findings leading to further modifications of its protocol with respect to dextrothyroxine. JAMA. 1972;220:996–1008.
Coronary Drug Project Research Group. The Coronary Drug Project: findings leading to discontinuation of the 2.5 mg/day estrogen group. JAMA. 1973;226:652–7.
Coronary Drug Project Research Group. Practical aspects of decision making in clinical trials: The Coronary Drug Project as a case study. Control Clin Trials. 1981;1:363–76.
Park JJH, Thorlund K, Mills EJ. Critical concepts in adaptive clinical trials. Clin Epidemiol. 2018;10:343–51.
Casali PG, Le Cesne A, Poveda Velasco A, Kotasek D, Rutkowski P, Hohenberger P, Fumagalli E, Judson IR, Italiano A, Gelderblom H, Adenis A, Hartmann JT, Duffaud F, Goldstein D, Broto JM, Gronchi A, Dei Tos AP, Marréaud S, van der Graaf WT, Zalcberg JR, Litière S, Blay JY. Time to definitive failure to the first tyrosine kinase inhibitor in localized GI stromal tumors treated with imatinib as an adjuvant: a European Organisation for Research and Treatment of Cancer Soft Tissue and Bone Sarcoma Group intergroup randomized trial in collaboration with the Australasian Gastro-Intestinal Trials Group, UNICANCER, French Sarcoma Group, Italian Sarcoma Group, and Spanish Group for Research on Sarcomas. J Clin Oncol. 2015;33(36):4276–83.
Woloshin S, Schwartz LM, White B, Moore TJ. The fate of FDA postapproval studies. N Engl J Med. 2017;377(12):1114–7.
Alling DW. Early decision in the Wilcoxon two-Sample test. J Am Stat Assoc. 1963;58(303):713–20.
Lan KKG, Simon R, Halperin M. Stochastically curtailed tests in long-term clinical trials. Commun Stat Part C Sequ Anal. 1982;1:207–19.
Halperin M, Ware J. Early decision in a censored Wilcoxon two-sample test for accumulating survival data. J Am Stat Assoc. 1974;69(414):422.
Lachin JM. A review of methods for futility stopping based on conditional power. Stat Med. 2005;24(18):2747–64.
Lan KK, Wittes J. The B-value: a tool for monitoring data. Biometrics. 1988;44(2):579–85.
Lin DY, Yao Q, Ying Z. A general theory on stochastic curtailment for censored survival data. J Am Stat Assoc. 1999;94:510–21.
Hunsberger S, Sorlie P, Geller NL. Stochastic curtailing and conditional power in matched case-control studies. Stat Med. 1994;13(5–7):663–70.
Siebert U, Rochau U, Claxton K. When is enough evidence enough? Using systematic decision analysis and value-of-information analysis to determine the need for further evidence. Z Evid Fortbild Qual Gesundhwes. 2013;107(7):575–84.
Ades AE, Lu G, Claxton K. Expected value of sample information calculations in medical decision modeling. Med Decis Making. 2004;24:207–27.
Heath A, Manolopoulou I, Baio G. A review of methods for analysis of the expected value of information. Med Decis Making. 2017;37(7):747–58.
Jennison C, Turnbull BW. Group sequential methods. Applications to clinical trials. Boca Raton: Chapman & Hall/CRC; 2000.
Chen J. Sequential testing for safety evaluation. In: Lawrence-Gould A, editor. Chapter 12: Statistical methods for evaluating safety in medical product development. Chichester: Wiley; 2015.
Martin D, Gagne JJ, Gruber S, Izem R, Nelson JS, Nguyen MD, Ouellet-Hellstrom R, Schneeweiss SS, Toh S, Walker AM. Sequential surveillance for drug safety in a regulatory environment. Pharmacoepidemiol Drug Saf. 2018. https://doi.org/10.1002/pds.4407.
Silva IR, Kulldorff M. Continuous versus group sequential analysis for post-market drug and vaccine safety surveillance. Biometrics. 2015;71:851–8.
Silva IR. Type I error probability spending for post-market drug and vaccine safety surveillance with binomial data. Stat Med. 2018;37(1):107–18.
Nelson JC, Cook AJ, Yu O, Dominguez C, Zhao S, Greene S, Fireman B, Jacobsen SJ, Weintraub ES, Jackson L. Challenges in the design and analysis of sequentially monitored postmarket safety surveillance evaluations using electronic observational health care data. Pharmacoepidemiol Drug Saf. 2012;21(S1):62–71.
Nelson J, Wellman R, Yu O, Cook A, Maro J, Ouellet-Hellstrom R, Boudreau D, Floyd J, Heckbert S, Pinheiro S, Reichman M, Shoaibi A. A synthesis of current surveillance planning methods for the sequential monitoring of drug and vaccine adverse effects using electronic health care data. eGEMS. 2016;4(1):1219.
Acknowledgements
Robert Glynn, Susan Gruber, Martin Kulldorff and Sebastian Schneeweiss, as well as members of the Division of Pharmacoepidemiology and Pharmacoeconomics at the Brigham and Women’s Hospital in Boston, provided important critiques of the material presented here. AMW is an employee of World Health Information Science Consultants, LLC.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
The author declares that he has no conflicts of interest.
Rights and permissions
About this article
Cite this article
Walker, A.M. Conditional power as an aid in making interim decisions in observational studies. Eur J Epidemiol 33, 777–784 (2018). https://doi.org/10.1007/s10654-018-0413-9
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10654-018-0413-9