Instrumental variables I: instrumental variables exploit natural variation in nonexperimental data to estimate causal relationships.

The gold standard of study design for treatment evaluation is widely acknowledged to be the randomized controlled trial (RCT). Trials allow for the estimation of causal effect by randomly assigning participants either to an intervention or comparison group; through the assumption of "exchangeability" between groups, comparing the outcomes will yield an estimate of causal effect. In the many cases where RCTs are impractical or unethical, instrumental variable (IV) analysis offers a nonexperimental alternative based on many of the same principles. IV analysis relies on finding a naturally varying phenomenon, related to treatment but not to outcome except through the effect of treatment itself, and then using this phenomenon as a proxy for the confounded treatment variable. This article demonstrates how IV analysis arises from an analogous but potentially impossible RCT design, and outlines the assumptions necessary for valid estimation. It gives examples of instruments used in clinical epidemiology and concludes with an outline on estimation of effects.

[1]  A. Walker,et al.  Drug Copayment and Adherence in Chronic Heart Failure: Effect on Cost and Outcomes , 2006, Pharmacotherapy.

[2]  David A. Jaeger,et al.  Problems with Instrumental Variables Estimation when the Correlation between the Instruments and the Endogenous Explanatory Variable is Weak , 1995 .

[3]  J. Angrist,et al.  Digitized by the Internet Archive in 2011 with Funding from Estimation of Limited-dependent Variable Models with Dummy Endogenous Regressors: Simple Strategies for Empirical Practice , 2011 .

[4]  J. Avorn.  In defense of pharmacoepidemiology--embracing the yin and yang of drug research. , 2007, The New England journal of medicine.

[5]  M Alan Brookhart,et al.  Instrumental variables II: instrumental variable application-in 25 variations, the physician prescribing preference generally was strong and reduced covariate imbalance. , 2009, Journal of clinical epidemiology.

[6]  Elliott S Fisher,et al.  Analysis of observational studies in the presence of treatment selection bias: effects of invasive cardiac management on AMI survival using propensity score and instrumental variable methods. , 2007, JAMA.

[7]  M. Kramer,et al.  Uses of ecologic studies in the assessment of intended treatment effects. , 1999, Journal of clinical epidemiology.

[8]  J. Robins.  Correcting for non-compliance in randomized trials using structural nested mean models , 1994 .

[9]  A. Walker.  Confounding by indication. , 1996, Epidemiology.

[10]  Jennifer G. Robinson,et al.  Conjugated equine estrogens and peripheral arterial disease risk: the Women's Health Initiative. , 2006, American heart journal.

[11]  John M Brooks,et al.  Was breast conserving surgery underutilized for early stage breast cancer? Instrumental variables evidence for stage II patients from Iowa. , 2003, Health services research.

[12]  M Alan Brookhart,et al.  Evaluating Short-Term Drug Effects Using a Physician-Specific Prescribing Preference as an Instrumental Variable , 2006, Epidemiology.

[13]  J. Robins,et al.  Marginal structural models to estimate the causal effect of zidovudine on the survival of HIV-positive men. , 2000, Epidemiology.

[14]  Jeffrey M. Wooldridge,et al.  Introductory Econometrics: A Modern Approach , 1999 .

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

[16]  E. Rimm,et al.  Vitamin E consumption and the risk of coronary heart disease in men. , 1993, The New England journal of medicine.

[17]  M Alan Brookhart,et al.  Simultaneous assessment of short-term gastrointestinal benefits and cardiovascular risks of selective cyclooxygenase 2 inhibitors and nonselective nonsteroidal antiinflammatory drugs: an instrumental variable analysis. , 2006, Arthritis and rheumatism.

[18]  J. Angrist,et al.  Estimation of Limited Dependent Variable Models With Dummy Endogenous Regressors , 2001 .

[19]  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.

[20]  J. Avorn,et al.  A review of uses of health care utilization databases for epidemiologic research on therapeutics. , 2005, Journal of clinical epidemiology.

[21]  Sebastian Schneeweiss,et al.  Selective prescribing led to overestimation of the benefits of lipid-lowering drugs. , 2006, Journal of clinical epidemiology.

[22]  B J McNeil,et al.  Does more intensive treatment of acute myocardial infarction in the elderly reduce mortality? Analysis using instrumental variables. , 1994, JAMA.

[23]  J. Stock,et al.  Instrumental Variables Regression with Weak Instruments , 1994 .

[24]  S. Johnston,et al.  Combining ecological and individual variables to reduce confounding by indication: case study--subarachnoid hemorrhage treatment. , 2000, Journal of clinical epidemiology.

[25]  E. Chrischilles,et al.  Heterogeneity and the Interpretation of Treatment Effect Estimates From Risk Adjustment and Instrumental Variable Methods , 2007, Medical care.

[26]  J. Newhouse,et al.  Econometrics in outcomes research: the use of instrumental variables. , 1998, Annual review of public health.

[27]  S. Ebrahim,et al.  What can mendelian randomisation tell us about modifiable behavioural and environmental exposures? , 2005, BMJ : British Medical Journal.

[28]  J. Avorn,et al.  Determinants of selective cyclooxygenase-2 inhibitor prescribing: are patient or physician characteristics more important? , 2003, The American journal of medicine.

[29]  S. Greenland.  An introduction To instrumental variables for epidemiologists , 2000, International journal of epidemiology.

[30]  K. Harris,et al.  Who is the marginal patient? Understanding instrumental variables estimates of treatment effects. , 1998, Health services research.

[31]  J. Avorn,et al.  Risk of death in elderly users of conventional vs. atypical antipsychotic medications. , 2005, The New England journal of medicine.

[32]  Sebastian Schneeweiss,et al.  A Medicare database review found that physician preferences increasingly outweighed patient characteristics as determinants of first-time prescriptions for COX-2 inhibitors. , 2005, Journal of clinical epidemiology.

[33]  J. Rassen,et al.  Evaluating the Validity of an Instrumental Variable Study of Neuroleptics: Can Between-Physician Differences in Prescribing Patterns Be Used to Estimate Treatment Effects? , 2007, Medical care.

[34]  J. Manson,et al.  Vitamin E in the Primary Prevention of Cardiovascular Disease and Cancer: The Women’s Health Study: A Randomized Controlled Trial , 2005, JAMA.