Sensitivity analysis and power for instrumental variable studies

In observational studies to estimate treatment effects, unmeasured confounding is often a concern. The instrumental variable (IV) method can control for unmeasured confounding when there is a valid IV. To be a valid IV, a variable needs to be independent of unmeasured confounders and only affect the outcome through affecting the treatment. When applying the IV method, there is often concern that a putative IV is invalid to some degree. We present an approach to sensitivity analysis for the IV method which examines the sensitivity of inferences to violations of IV validity. Specifically, we consider sensitivity when the magnitude of association between the putative IV and the unmeasured confounders and the direct effect of the IV on the outcome are limited in magnitude by a sensitivity parameter. Our approach is based on extending the Anderson-Rubin test and is valid regardless of the strength of the instrument. A power formula for this sensitivity analysis is presented. We illustrate its usage via examples about Mendelian randomization studies and its implications via a comparison of using rare versus common genetic variants as instruments.

[1]  M. Daly,et al.  Searching for missing heritability: Designing rare variant association studies , 2014, Proceedings of the National Academy of Sciences.

[2]  Jerry A. Hausman,et al.  Specification and estimation of simultaneous equation models , 1983 .

[3]  Paul R. Rosenbaum,et al.  Design sensitivity in observational studies , 2004 .

[4]  T. W. Anderson,et al.  Estimation of the Parameters of a Single Equation in a Complete System of Stochastic Equations , 1949 .

[5]  Christopher Winship,et al.  Counterfactuals and Causal Inference: Methods and Principles for Social Research , 2007 .

[6]  S. Ebrahim,et al.  'Mendelian randomization': can genetic epidemiology contribute to understanding environmental determinants of disease? , 2003, International journal of epidemiology.

[7]  Sebastian Schneeweiss,et al.  Preference-Based Instrumental Variable Methods for the Estimation of Treatment Effects: Assessing Validity and Interpreting Results , 2007, The international journal of biostatistics.

[8]  Peter E. Rossi,et al.  Plausibly Exogenous , 2008, Review of Economics and Statistics.

[9]  J. MacKinnon,et al.  Estimation and inference in econometrics , 1994 .

[10]  S. Ebrahim,et al.  Mendelian randomization: prospects, potentials, and limitations. , 2004, International journal of epidemiology.

[11]  Dylan S. Small,et al.  War and Wages : The Strength of Instrumental Variables and Their Sensitivity to Unobserved Biases , 2007 .

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

[13]  Thomas J. Rothenberg,et al.  Approximating the distributions of econometric estimators and test statistics , 1984 .

[14]  C. Nelson,et al.  The Distribution of the Instrumental Variables Estimator and its T-Ratiowhen the Instrument is a Poor One , 1988 .

[15]  N. Sheehan,et al.  Mendelian randomization as an instrumental variable approach to causal inference , 2007, Statistical methods in medical research.

[16]  J. Robins,et al.  Credible Mendelian randomization studies: approaches for evaluating the instrumental variable assumptions. , 2012, American journal of epidemiology.

[17]  Jonathan H. Wright,et al.  A Survey of Weak Instruments and Weak Identification in Generalized Method of Moments , 2002 .

[18]  B. Cowling,et al.  Power and sample size calculations for Mendelian randomization studies using one genetic instrument. , 2013, International journal of epidemiology.

[19]  Dylan S. Small,et al.  War and Wages , 2008 .

[20]  Eric Zivot,et al.  Inference on Structural Parameters In Instrumental Variables Regression With Weak Instruments , 1998 .

[21]  Dylan S. Small,et al.  Calibrating Sensitivity Analyses to Observed Covariates in Observational Studies , 2013, Biometrics.

[22]  G. Imbens.  Instrumental Variables: An Econometrician's Perspective , 2014, SSRN Electronic Journal.

[23]  Raj Chetty,et al.  Identification and Inference With Many Invalid Instruments , 2011 .

[24]  Greg Gibson,et al.  Rare and common variants: twenty arguments , 2012, Nature Reviews Genetics.

[25]  Dylan S. Small,et al.  Instrumental variable methods for causal inference , 2014, Statistics in medicine.

[26]  L. Smeeth,et al.  Limits to causal inference based on Mendelian randomization: a comparison with randomized controlled trials. , 2006, American journal of epidemiology.

[27]  Donald W. K. Andrews,et al.  Optimal Two‐Sided Invariant Similar Tests for Instrumental Variables Regression , 2006 .

[28]  P. Holland.  CAUSAL INFERENCE, PATH ANALYSIS AND RECURSIVE STRUCTURAL EQUATIONS MODELS , 1988 .

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

[30]  George Davey Smith,et al.  Mendelian randomization: Using genes as instruments for making causal inferences in epidemiology , 2008, Statistics in medicine.

[31]  S. Kruger.  Design Of Observational Studies , 2016 .

[32]  S. Norby.  [Mendelian randomization]. , 2005, Ugeskrift for laeger.

[33]  Marcelo J. Moreira.  Tests with correct size when instruments can be arbitrarily weak , 2009 .

[34]  T. DiPrete,et al.  7. Assessing Bias in the Estimation of Causal Effects: Rosenbaum Bounds on Matching Estimators and Instrumental Variables Estimation with Imperfect Instruments , 2004 .

[35]  Geert Molenberghs,et al.  Ignorance and uncertainty regions as inferential tools in a sensitivity analysis , 2006 .

[36]  Dylan S. Small,et al.  Sensitivity Analysis for Instrumental Variables Regression With Overidentifying Restrictions , 2007 .

[37]  Paul R. Rosenbaum,et al.  Robust, accurate confidence intervals with a weak instrument: quarter of birth and education , 2005 .

[38]  S. Ebrahim,et al.  Mendelian randomization: can genetic epidemiology help redress the failures of observational epidemiology? , 2008, Human Genetics.