Covariate adjustment in randomized trials with binary outcomes: Targeted maximum likelihood estimation

Covariate adjustment using linear models for continuous outcomes in randomized trials has been shown to increase efficiency and power over the unadjusted method in estimating the marginal effect of treatment. However, for binary outcomes, investigators generally rely on the unadjusted estimate as the literature indicates that covariate-adjusted estimates based on the logistic regression models are less efficient. The crucial step that has been missing when adjusting for covariates is that one must integrate/average the adjusted estimate over those covariates in order to obtain the marginal effect. We apply the method of targeted maximum likelihood estimation (tMLE) to obtain estimators for the marginal effect using covariate adjustment for binary outcomes. We show that the covariate adjustment in randomized trials using the logistic regression models can be mapped, by averaging over the covariate(s), to obtain a fully robust and efficient estimator of the marginal effect, which equals a targeted maximum likelihood estimator. This tMLE is obtained by simply adding a clever covariate to a fixed initial regression. We present simulation studies that demonstrate that this tMLE increases efficiency and power over the unadjusted method, particularly for smaller sample sizes, even when the regression model is mis-specified.

[1]  Marie Davidian,et al.  Improving Efficiency of Inferences in Randomized Clinical Trials Using Auxiliary Covariates , 2008, Biometrics.

[2]  M. Saag,et al.  Treatment of cryptococcal meningitis associated with the acquired immunodeficiency syndrome. National Institute of Allergy and Infectious Diseases Mycoses Study Group and AIDS Clinical Trials Group. , 1997, The New England journal of medicine.

[3]  N. Jewell,et al.  Some surprising results about covariate adjustment in logistic regression models , 1991 .

[4]  Ewout W Steyerberg,et al.  Covariate adjustment in randomized controlled trials with dichotomous outcomes increases statistical power and reduces sample size requirements. , 2004, Journal of clinical epidemiology.

[5]  Mark J. van der Laan,et al.  Why prefer double robust estimators in causal inference , 2005 .

[6]  J. Robins,et al.  Doubly Robust Estimation in Missing Data and Causal Inference Models , 2005, Biometrics.

[7]  James M. Robins,et al.  Unified Methods for Censored Longitudinal Data and Causality , 2003 .

[8]  J. Stockman.  Supplemental Perioperative Oxygen and the Risk of Surgical Wound Infection: A Randomized Controlled Trial , 2007 .

[9]  Peter J. Bickel,et al.  INFERENCE FOR SEMIPARAMETRIC MODELS: SOME QUESTIONS AND AN ANSWER , 2001 .

[10]  J. Robins,et al.  Adjusting for Nonignorable Drop-Out Using Semiparametric Nonresponse Models , 1999 .

[11]  Aad van der Vaart,et al.  Estimating a survival distribution with current status data and high-dimensional covariates , 2006 .

[12]  J. Robins.  A graphical approach to the identification and estimation of causal parameters in mortality studies with sustained exposure periods. , 1987, Journal of chronic diseases.

[13]  D. Sessler,et al.  Antenatal Betamethasone and Incidence of Neonatal Respiratory Distress After Elective Caesarean Section: Pragmatic Randomized Trial , 2005, JAMA.

[14]  S. Assmann,et al.  Subgroup analysis and other (mis)uses of baseline data in clinical trials , 2000, The Lancet.

[15]  Mark J van der Laan,et al.  Deletion/Substitution/Addition Algorithm in Learning with Applications in Genomics , 2004, Statistical applications in genetics and molecular biology.

[16]  C. Klaassen,et al.  Discussion to "Inference for semiparametric models: some questions and an answer" by Peter J. Bickel and Jaimyoung Kwon , 2001 .

[17]  S. Pocock,et al.  Subgroup analysis, covariate adjustment and baseline comparisons in clinical trial reporting: current practiceand problems , 2002, Statistics in medicine.

[18]  Christina Wolfson,et al.  Randomised trial of home-based psychosocial nursing intervention for patients recovering from myocardial infarction , 1997, The Lancet.

[19]  M. J. van der Laan,et al.  The International Journal of Biostatistics Targeted Maximum Likelihood Learning , 2011 .

[20]  J. Robins.  A new approach to causal inference in mortality studies with a sustained exposure period—application to control of the healthy worker survivor effect , 1986 .

[21]  Why Prefer Double Robust Estimates? Illustration with Causal Point Treatment Studies , 2002 .

[22]  M. Davidian,et al.  Covariate adjustment for two‐sample treatment comparisons in randomized clinical trials: A principled yet flexible approach , 2008, Statistics in medicine.

[23]  Randall Wetzel,et al.  Effect of mechanical ventilator weaning protocols on respiratory outcomes in infants and children: a randomized controlled trial. , 2002, JAMA.