© 1984 by Biometrika Trust
Biased estimates of treatment effect in randomized experiments with nonlinear regressions and omitted covariates
Biometry Branch of the Division of Cancer Prevention and Control, National Cancer Institute Bethesda, Maryland, U.S.A.
Department of Mathematics and Statistics, University of Pittsburgh Pittsburgh, Pennsylvania, U.S.A.
Biometry Branch of the Division of Cancer Prevention and Control, National Cancer Institute Bethesda, Maryland, U.S.A.
Certain important nonlinear regression models lead to biased estimates of treatment effect, even in randomized experiments, if needed covariates are omitted. The asymptotic bias is determined both for estimates based on the method of moments and for maximum likelihood estimates. The asymptotic bias from omitting covariates is shown to be zero if the regression of the response variable on treatment and covariates is linear or exponential, and, in regular cases, this is a necessary condition for zero bias. Many commonly used models do have such exponential regressions; thus randomization ensures unbiased treatment estimates in a large number of important nonlinear models. For moderately censored exponential survival data, analysis with the exponential survival model yields less biased estimates of treatment effect than analysis with the proportional hazards model of Cox, if needed covariates are omitted. Simulations confirm that calculations of asymptotic bias are in excellent agreement with the bias observed in experiments of modest size.
Key Words: Bias • Clinical trial • Nonlinear model • Omitted covariate • Randomization
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  S. G Baker and B. S Kramer Randomized trials for the real world: making as few and as reasonable assumptions as possible Statistical Methods in Medical Research, June 1, 2008; 17(3): 243 - 252. [Abstract] [PDF]
|
|
|
|
 |
|
 E. P Martens, W. R Pestman, A. d. Boer, S. V Belitser, and O. H Klungel Systematic differences in treatment effect estimates between propensity score methods and logistic regression Int. J. Epidemiol., May 3, 2008; (2008) dyn079v1. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Cai, L. Tian, S. D. Solomon, and L.J. Wei Predicting future responses based on possibly mis-specified working models Biometrika, March 1, 2008; 95(1): 75 - 92. [Abstract] [PDF]
|
|
|
|
 |
|
 G. Verbeke and S. Fieuws The effect of miss-specified baseline characteristics on inference for longitudinal trends in linear mixed models Biostat., October 1, 2007; 8(4): 772 - 783. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Sturmer, S. Schneeweiss, K. J. Rothman, J. Avorn, and R. J. Glynn Performance of Propensity Score Calibration--A Simulation Study Am. J. Epidemiol., May 15, 2007; 165(10): 1110 - 1118. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Shen and Y. Cheng Adaptive design: estimation and inference with censored data in a semiparametric model Biostat., April 1, 2007; 8(2): 306 - 322. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. R. Rao and D. A. Schoenfeld Survival Methods Circulation, January 2, 2007; 115(1): 109 - 113. [Full Text] [PDF]
|
|
|
|
|
|
L. S Freedman, N. Potischman, V. Kipnis, D. Midthune, A. Schatzkin, F. E Thompson, R. P Troiano, R. Prentice, R. Patterson, R. Carroll, et al. A comparison of two dietary instruments for evaluating the fat-breast cancer relationship Int. J. Epidemiol., August 1, 2006; 35(4): 1011 - 1021. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. A. Brookhart, S. Schneeweiss, K. J. Rothman, R. J. Glynn, J. Avorn, and T. Sturmer Variable Selection for Propensity Score Models Am. J. Epidemiol., June 15, 2006; 163(12): 1149 - 1156. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. A. Knaus Designing Trials and Deciding on Therapy in Complex Diseases: What Can We Do to Improve? Med Decis Making, July 1, 2005; 25(4): 366 - 369. [PDF]
|
|
|
|
|
|
L. M LaVange, T. A Durham, and G. G Koch Randomization-based nonparametric methods for the analysis of multicentre trials Statistical Methods in Medical Research, June 1, 2005; 14(3): 281 - 301. [Abstract] [PDF]
|
|
|
|
|
|
M. Abrahamowicz, R. du Berger, D. Krewski, R. Burnett, G. Bartlett, R. M. Tamblyn, and K. Leffondre Bias due to Aggregation of Individual Covariates in the Cox Regression Model Am. J. Epidemiol., October 1, 2004; 160(7): 696 - 706. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Ritz and D. Spiegelman Equivalence of conditional and marginal regression models for clustered and longitudinal data Statistical Methods in Medical Research, August 1, 2004; 13(4): 309 - 323. [Abstract] [PDF]
|
|
|
|
|
|
G. Zou A Modified Poisson Regression Approach to Prospective Studies with Binary Data Am. J. Epidemiol., April 1, 2004; 159(7): 702 - 706. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. C. Johnston, A. F. Connors Jr, D. P. Wagner, and E. C. Haley Jr Risk Adjustment Effect on Stroke Clinical Trials Stroke, February 1, 2004; 35 (2): e43 - e45. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. W. Steyerberg and M. J. C. Eijkemans Heterogeneity Bias: The Difference between Adjusted and Unadjusted Effects Med Decis Making, January 1, 2004; 24(1): 102 - 104. [PDF]
|
|
|
|
|
|
M. S. Cepeda, R. Boston, J. T. Farrar, and B. L. Strom Comparison of Logistic Regression versus Propensity Score When the Number of Events Is Low and There Are Multiple Confounders Am. J. Epidemiol., August 1, 2003; 158(3): 280 - 287. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. A. Hanley, A. Negassa, M. D. deB. Edwardes, and J. E. Forrester Statistical Analysis of Correlated Data Using Generalized Estimating Equations: An Orientation Am. J. Epidemiol., February 15, 2003; 157(4): 364 - 375. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. C. Johnston, A. F. Connors Jr, D. P. Wagner, and E. C. Haley Jr Predicting Outcome in Ischemic Stroke: External Validation of Predictive Risk Models Stroke, January 1, 2003; 34(1): 200 - 202. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. B. Begg On the Use of Familial Aggregation in Population-Based Case Probands for Calculating Penetrance J Natl Cancer Inst, August 21, 2002; 94(16): 1221 - 1226. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. A. Betensky, D. N. Louis, and J. Gregory Cairncross Influence of Unrecognized Molecular Heterogeneity on Randomized Clinical Trials J. Clin. Oncol., May 15, 2002; 20(10): 2495 - 2499. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. D. ALLISON Comparing Logit and Probit Coefficients Across Groups Sociological Methods Research, November 1, 1999; 28(2): 186 - 208. [Abstract] [PDF]
|
|
|
|
|
|
G. Howard, G. B. Russell, R. Anderson, G. W. Evans, T. Morgan, V. J. Howard, and G. L. Burke Role of Social Class in Excess Black Stroke Mortality Stroke, October 1, 1995; 26(10): 1759 - 1763. [Abstract] [Full Text]
|
|