Exploiting multiple outcomes in Bayesian principal stratification analysis with application to the evaluation of a job training program

The causal effect of a randomized job training program, the JOBS II study, on trainees' depression is evaluated. Principal stratification is used to deal with noncompliance to the assigned treatment. Due to the latent nature of the principal strata, strong structural assumptions are often invoked to identify principal causal effects. Alternatively, distributional assumptions may be invoked using a model-based approach. These often lead to weakly identified models with substantial regions of flatness in the posterior distribution of the causal effects. Information on multiple outcomes is routinely collected in practice, but is rarely used to improve inference. This article develops a Bayesian approach to exploit multivariate outcomes to sharpen inferences in weakly identified principal stratification models. We show that inference for the causal effect on depression is significantly improved by using the re-employment status as a secondary outcome in the JOBS II study. Simulation studies are also performed to illustrate the potential gains in the estimation of principal causal effects from jointly modeling more than one outcome. This approach can also be used to assess plausibility of structural assumptions and sensitivity to deviations from these structural assumptions. Two model checking procedures via posterior predictive checks are also discussed.

[1]  D. Rubin,et al.  Assessing the effect of an influenza vaccine in an encouragement design. , 2000, Biostatistics.

[2]  Fabrizia Mealli,et al.  Comparing principal stratification and selection models in parametric causal inference with nonignorable missingness , 2008, Comput. Stat. Data Anal..

[3]  D. Rubin Bayesianly Justifiable and Relevant Frequency Calculations for the Applied Statistician , 1984 .

[4]  Fan Li,et al.  A Bayesian Semiparametric Approach to Intermediate Variables in Causal Inference , 2011 .

[5]  D. Rubin,et al.  Inference from Iterative Simulation Using Multiple Sequences , 1992 .

[6]  Michael G Hudgens,et al.  Evaluating Candidate Principal Surrogate Endpoints , 2008, Biometrics.

[7]  M. J. Bayarri,et al.  P Values for Composite Null Models , 2000 .

[8]  V. Johnson Bayesian Model Assessment Using Pivotal Quantities , 2007 .

[9]  Avi Feller,et al.  Principal Stratification , 2015 .

[10]  Fan Li,et al.  Improving inference of Gaussian mixtures using auxiliary variables , 2014, Stat. Anal. Data Min..

[11]  Bengt Muthén,et al.  Modeling of Intervention Effects With Noncompliance: A Latent Variable Approach for Randomized Trials , 2001 .

[12]  Donald B. Rubin,et al.  Estimation of Causal Effects via Principal Stratification When Some Outcomes are Truncated by “Death” , 2003 .

[13]  D. Basu Randomization Analysis of Experimental Data: The Fisher Randomization Test , 1980 .

[14]  Michael R Elliott,et al.  A Bayesian Approach to Surrogacy Assessment Using Principal Stratification in Clinical Trials , 2010, Biometrics.

[15]  Michael R Elliott,et al.  Bayesian inference for causal mediation effects using principal stratification with dichotomous mediators and outcomes. , 2010, Biostatistics.

[16]  Jeremy MG Taylor,et al.  Causal assessment of surrogacy in a meta-analysis of colorectal cancer trials. , 2011, Biostatistics.

[17]  Joshua D. Angrist,et al.  Identification of Causal Effects Using Instrumental Variables , 1993 .

[18]  P. Rosenbaum The Consequences of Adjustment for a Concomitant Variable that Has Been Affected by the Treatment , 1984 .

[19]  Juni Palmgren,et al.  Sensitivity Analysis for Principal Stratum Direct Effects, with an Application to a Study of Physical Activity and Coronary Heart Disease , 2009, Biometrics.

[20]  S. Zeger,et al.  On estimating efficacy from clinical trials. , 1991, Statistics in medicine.

[21]  F. Mealli,et al.  Using secondary outcomes and covariates to sharpen inference in randomized experiments with noncompliance , 2012 .

[22]  F. Gosselin,et al.  A New Calibrated Bayesian Internal Goodness-of-Fit Method: Sampled Posterior p-Values as Simple and General p-Values That Allow Double Use of the Data , 2011, PloS one.

[23]  S. Chib Marginal Likelihood from the Gibbs Output , 1995 .

[24]  Central role , 2005, Veterinary Record.

[25]  R. Little,et al.  Statistical Techniques for Analyzing Data from Prevention Trials: Treatment of No-Shows Using Rubin's Causal Model , 1998 .

[26]  Dylan S Small,et al.  Discussion of "Identifiability and estimation of causal effects in randomized trials with noncompliance and completely nonignorable missing data". , 2009, Biometrics.

[27]  Amiram D. Vinokur,et al.  Effects of Recent and Past Stress on Mental Health: Coping with Unemployment Among Vietnam Veterans and Nonveterans1 , 1987 .

[28]  S. Chib,et al.  Bayesian analysis of cross-section and clustered data treatment models , 2000 .

[29]  D. Rubin Estimating causal effects of treatments in randomized and nonrandomized studies. , 1974 .

[30]  A. Vinokur,et al.  Impact of the JOBS intervention on unemployed workers varying in risk for depression , 1995, American journal of community psychology.

[31]  Donald B. Rubin,et al.  Principal Stratification Approach to Broken Randomized Experiments , 2003 .

[32]  Paul Gustafson,et al.  What Are the Limits of Posterior Distributions Arising From Nonidentified Models, and Why Should We Care? , 2009 .

[33]  Xiao-Li Meng,et al.  POSTERIOR PREDICTIVE ASSESSMENT OF MODEL FITNESS VIA REALIZED DISCREPANCIES , 1996 .

[34]  F Mealli,et al.  Application of the Principal Stratification Approach to the Faenza Randomized Experiment on Breast Self‐Examination , 2007, Biometrics.

[35]  D. Rubin,et al.  The central role of the propensity score in observational studies for causal effects , 1983 .

[36]  N. Hjort,et al.  Post-Processing Posterior Predictive p Values , 2006 .

[37]  Valen E. Johnson A Bayesian x2 test for goodness-of-fit , 2004 .

[38]  D. Rubin,et al.  Principal Stratification in Causal Inference , 2002, Biometrics.

[39]  Donald B Rubin,et al.  Principal Stratification for Causal Inference With Extended Partial Compliance , 2008 .

[40]  Donald B. Rubin,et al.  Bayesian Inference for Causal Effects: The Role of Randomization , 1978 .

[41]  Jerome P. Reiter,et al.  Sensitivity analysis for unmeasured confounding in principal stratification settings with binary variables , 2012, Statistics in medicine.

[42]  D. Rubin,et al.  Bayesian inference for causal effects in randomized experiments with noncompliance , 1997 .

[43]  C. Manski Nonparametric Bounds on Treatment Effects , 1989 .

[44]  R. Gallop,et al.  Mediation analysis with principal stratification , 2009, Statistics in medicine.