Instrumental variable method for time‐to‐event data using a pseudo‐observation approach

Observational studies are often in peril of unmeasured confounding. Instrumental variable analysis is a method for controlling for unmeasured confounding. As yet, theory on instrumental variable analysis of censored time-to-event data is scarce. We propose a pseudo-observation approach to instrumental variable analysis of the survival function, the restricted mean, and the cumulative incidence function in competing risks with right-censored data using generalized method of moments estimation. For the purpose of illustrating our proposed method, we study antidepressant exposure in pregnancy and risk of autism spectrum disorder in offspring, and the performance of the method is assessed through simulation studies.

[1]  T. Tosteson,et al.  Using instrumental variables to estimate a Cox’s proportional hazards regression subject to additive confounding , 2014, Health Services and Outcomes Research Methodology.

[2]  J. Fine,et al.  Instrumental variable additive hazards models , 2015, Biometrics.

[3]  J. Robins,et al.  Correcting for non-compliance in randomized trials using rank preserving structural failure time models , 1991 .

[4]  An Introduction to Instrumental Variables , 2008 .

[5]  J. Klein,et al.  Survival Analysis: Techniques for Censored and Truncated Data , 1997 .

[6]  Maja Pohar Perme,et al.  Pseudo-observations in survival analysis , 2010, Statistical methods in medical research.

[7]  M. Schumacher,et al.  On pseudo-values for regression analysis in competing risks models , 2009, Lifetime data analysis.

[8]  D. Schendel,et al.  Antidepressant exposure in pregnancy and risk of autism spectrum disorders , 2013, Clinical epidemiology.

[9]  Paul J Rathouz,et al.  Two-stage residual inclusion estimation: addressing endogeneity in health econometric modeling. , 2008, Journal of health economics.

[10]  M. Glymour,et al.  Instrumental variable estimation in a survival context. , 2015, Epidemiology.

[11]  E. Kaplan,et al.  Nonparametric Estimation from Incomplete Observations , 1958 .

[12]  T. Martinussen,et al.  A Note on the Large Sample Properties of Estimators Based on Generalized Linear Models for Correlated Pseudo‐observations , 2016 .

[13]  John P Klein,et al.  Regression Modeling of Competing Risks Data Based on Pseudovalues of the Cumulative Incidence Function , 2005, Biometrics.

[14]  John P. Klein,et al.  SAS and R functions to compute pseudo-values for censored data regression , 2008, Comput. Methods Programs Biomed..

[15]  Erik T. Parner,et al.  Regression Analysis of Censored Data Using Pseudo-observations , 2010 .

[16]  E. Goetghebeur,et al.  A Causal Proportional Hazards Estimator for the Effect of Treatment Actually Received in a Randomized Trial with All‐or‐Nothing Compliance , 2003, Biometrics.

[17]  P Gustafson,et al.  Use of instrumental variables in the analysis of generalized linear models in the presence of unmeasured confounding with applications to epidemiological research , 2008, Statistics in medicine.

[18]  B. Briesacher,et al.  Use of instrumental variable in prescription drug research with observational data: a systematic review. , 2011, Journal of clinical epidemiology.

[19]  G. Bijwaard.  Instrumental Variable Estimation for Duration Data , 2008 .

[20]  J. Robins,et al.  Instruments for Causal Inference: An Epidemiologist's Dream? , 2006, Epidemiology.

[21]  Kurt Brännäs,et al.  Estimation in a Duration Model for Evaluating Educational Programs , 2000 .

[22]  Sander Greenland,et al.  An introduction to instrumental variables for epidemiologists. , 2018, International journal of epidemiology.