Abstract
In randomized clinical trials, substantial imbalance in the baseline outcome variable may occur by chance. Inference on treatment effect may be confounded by such imbalance if not properly accounted for. The usual unadjusted analysis may be conditionally biased with an inflated Type I error rate or reduced power conditioning on the baseline imbalance in the observed sample. This paper reviews methods for baseline adjustment with emphasis on the analysis of co-variance (ANCOVA) model with the baseline outcome as a covariate. Many issues on the ANCOVA model, including the nature of the adjustment, unconditional and conditional properties of treatment effect estimate from the ANCOVA, and baseline with measurement errors are discussed. In summary, the ANCOVA model with the baseline outcome as a covariate is more efficient unconditionally, and has better statistical properties conditionally, than the usual analysis of variance (ANOVA) analysis based on either posttreat-ment outcome or change from baseline and, therefore, is recommended. A real-life data set is analyzed and used to illustrate the unconditional and conditional properties.
Similar content being viewed by others
References
Altman DG. Comparability of randomized groups. Statistician. 1985;34:125–136.
Chuang-Stein C, Tong DM. The impact and implication of regression to the mean on the design and analysis of medical investigations. Stat Methods Med Research. 1997;6:115–128.
Overall JE, Magee KN. Directional baseline differences and type I error probabilities in randomized clinical trials. J Biopharm Stat. 1992;2:189–203.
Senn SJ. Covariate imbalance and random allocation in clinical trials. Stat Med. 1989;8:467–475.
Rothman KJ. Epidemiologic methods in clinical trials. Cancer. 1977;39:1771–1775.
Chambless LE, Roeback JR. Methods for assessing difference between groups in change when initial measurement is subject to intra-individual variation. Stat Med. 1993;12:1213–1237.
Begg CB. Significance tests of covariate imbalance in clinical trials. Control Clin Trials. 1990;11:223–225.
Permutt T. Testing for imbalance of covariate in controlled experiments. Stat Med. 1990;9:1455–1462.
Crager MR. Analysis of covariance in parallel-group clinical trials with pretreatment baselines. Biometrics. 1987;43:895–901.
Carroll RJ, Gallo P, Gleser LJ. Comparison of least squares and error-in-variable regression, with special reference to randomized analysis of covariance. J Am Stat Assoc. 1985;80(392):929–932.
Cochran WG. Errors of measurement in statistics. Technometrics. 1968;10(4):637–666.
Carroll RJ. Covariance analysis in generalized linear measurement error models. Stat Med. 1989;8:1075–1093.
Senn S. Letters to Editor: method for assessing dif ference between groups in change when initial mea surement is subject to intra-individual variation. Stat Med. 1994;13:2280–2283.
Senn S. Letters to Editor: In defense of analysis of covariance: a reply to Chambless and Roeback. Stat Med. 1995;14:2283–2288.
Reichardt CS, Trochim WMK, Cappelleri JC. Report of the death of regression discontinuity analysis are greatly exaggerated. Evaluation Review. 1995;19: 39–63.
Senn S. Testing for baseline balance in clinical trials. Stat Med. 1994;13:1715–1726.
Weigel RM, Narvaez M. Multiple regression analysis of differential response to treatment in randomized controlled clinical trials. Control Clin Trials. 1991;12:378–394.
Knoke JD. Nonparametric analysis of covariance for comparing change in randomized studies with baseline values subject to error. Biometrics. 1991; 47:523–533.
McKean JW, Vidmar TJ. A comparison of two rank-based methods for the analysis of linear models. Am Statistician. 1994;48:(3):220–229.
Maxwell SE. Covariate imbalance and conditional size: dependence on model-based adjustments. Stat Med. 1993;12:101–109.
Tukey JW. Tightening the clinical trial. Control Clin Trials. 1993;14:266–285.
Atiqullah M. The robustness of the covariance analysis of a one-way classification. Biometrika. 1964; 51:365–372.
Maxwell SE, Delaney HD, Dill CA. Another look at ANCOVA versus blocking. Psychological Bull. 1984;95(1):136–147.
Lair N. Further comparative analysis of pre-test post-test research design. Am Statistician. 1983(Novem-ber);37(4):329–330.
Chesher A. Non-normal variation and regression to the mean. Stat Methods Med Research. 1997;6:147–166.
Appleton DR, French JM, Vanderpump MPJ. Ignoring a covariate: an example of Simpson’s paradox. Am Statistician. 1996;50:340–341.
Beach ML, Meier P. Choosing covariates in the analysis of clinical trials. Control Clin Trials. 1989;10: 161S–175S.
Meier P. Stratification in the design of a clinical trial. Control Clin Trials. 1981;1:355–361.
Senn SJ. Regression to the mean. Stat Methods Med Research. 1997;6:99–102.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Wei, L., Zhang, J. Analysis of Data with Imbalance in the Baseline Outcome Variable for Randomized Clinical Trials. Ther Innov Regul Sci 35, 1201–1214 (2001). https://doi.org/10.1177/009286150103500417
Published:
Issue Date:
DOI: https://doi.org/10.1177/009286150103500417