Abstract
The estimation of treatment effects is one of the primary goals of statistics in medicine. Estimation based on observational studies is subject to confounding. Statistical methods for controlling bias due to confounding include regression adjustment, propensity scores and inverse probability weighted estimators. These methods require that all confounders are recorded in the data. The method of instrumental variables (IVs) can eliminate bias in observational studies even in the absence of information on confounders. We propose a method for integrating IVs within the framework of Cox’s proportional hazards model and demonstrate the conditions under which it recovers the causal effect of treatment. The methodology is based on the approximate orthogonality of an instrument with unobserved confounders among those at risk. We derive an estimator as the solution to an estimating equation that resembles the score equation of the partial likelihood in much the same way as the traditional IV estimator resembles the normal equations. To justify this IV estimator for a Cox model we perform simulations to evaluate its operating characteristics. Finally, we apply the estimator to an observational study of the effect of coronary catheterization on survival.
Similar content being viewed by others
References
Baker, S.: Analysis of survival data from a randomized trial with all-or-none compliance: estimating the cost-effectiveness of a cancer screening program. J. Am. Stat. Assoc. 93, 929–934 (1998)
Bijwaard, G.E.: IV Estimation for Duration Data. Tinbergen Institute Discussion Paper TI 2008–032/4. Tinbergen Institute, Amsterdam (2008)
Bijwaard, Govert E., Ridder, Geert: Correcting for selective compliance in a re-employment bonus experiment. J. Econom. 125(1–2), 77–111 (2005)
Brannas, K.: Estimation in a duration model for evaluating educational programs. IZA Discussion Papers, Institute for the Study of Labor (2001)
Buckley, J., James, I.: Linear regression with censored data. Biometrika 66, 429–436 (1979)
Buzas, J.S., Stefanski, L.A.: IV estimation in generalized linear measurement error models. J. Am. Stat. Assoc. 91, 999–1006 (1996)
Buzas, J., Stefanski, L., Tosteson, T.: Measurement Error. In: Ahrens, W., Pigeot, I. (eds.) Handbook of Epidemiology, pp 729–765. Springer-Verlag, Berlin (2005)
Carslake, D., Fraser, A., et al.: Associations of mortality with own height using son’s height as an IV. Econ. Human Biol. 11(3), 351–359 (2013)
Cole, S.R., Platt, R.W., Schisterman, E.F., Chu, H., Westreich, D., Richardson, D., Poole, C.: Illustrating bias due to conditioning on a collider. Int. J. Epidemiol. 39(2), 417–420 (2010)
Cox, D.R.: Regression models and life-tables. J. R Stat. Soc. Ser. B 34(2), 187–220 (1972)
Curtis, L.H., Hammill, B.G., Eisenstein, E.L., Kramer, J.M., Anstrom, K.J.: Using inverse probability-weighted estimators in comparative effectiveness analyses with observational databases. Med. Care 45(10), S96–S102 (2007)
Foster, E.M.: IVs for logistic regression: an illustration. Soc. Sci. Res. 26, 287–504 (1997)
Greenland, S.: An introduction to IVs for epidemiologists. Int. J. Epidemiol. 29(4), 722–729 (2000)
Hernan, M.A., Hernandez-Diaz, S., Robins, J.M.: A structural approach to selection bias. Epidemiology 15, 615–625 (2004)
Hernan, M.A., Robins, J.M.: An epidemiologist’s dream. Epidemiology 17, 360–372 (2006)
Hirano, K., Imbens, G.W.: Estimation of causal effects using propensity score weighting: an application to data on right heart catherization. Health Serv. Outcomes Res. Methodol. 2, 259–278 (2001)
Huang, Y., Wang, C.Y.: Cox regression with accurate covariates unascertainable: a nonparametric-correction approach. J. Am. Stat. Assoc. 95(452), 1213–1219 (2000)
Johnston, K., Gustafson, P., Levy, A.R., Grootendorst, P.: Use of IVs in the analysis of generalized linear models in the presence of omitted confounding with applications to epidemiological research. Stat. Med. 27, 1539–1556 (2008)
Landrum, M.B., Ayanian, J.Z.: Causal effect of ambulatory specialty care on mortality following myocardial infarction: a comparison of propensity score and iv analyses. Health Serv. Outcomes Res. Methodol. 2, 221–245 (2001)
Newhouse, J.P., McClellan, M.: Econometrics in outcomes research: the use of IVs. Annu. Rev. Public Health 19, 17–34 (1998)
McClellan, M., McNeil, B.J., Newhouse, J.P.: Does more intensive treatment of acute myocardial infarction in the elderly reduce mortality? Analysis using instrumental variables. JAMA 272, 859–866 (1994)
Rassen, J.A., Schneeweiss, S., Glynn, R.J., Mittleman, M.A., Brookhart, M.A.: IV analysis for estimation of treatment effects with dichotomous outcomes. Am. J. Epidemiol. 169(3), 273–284 (2009)
Robins, J.M., Tsiatis, A.A.: Correcting for non-compliance in randomized trials using rank preserving structural failure time models. Commun. Stat. Theory Methods 20(8), 2609–2631 (1991)
Staiger, D., Stock, J.H.: Instrumental variables regression with weak instruments. Econometrica 65(3), 557–586 (1997)
Stukel, T.A., Fisher, E.S., Wennberg, D.E., Alter, D.A., Gottlieb, D.J., Vermeulen, M.J.: Analysis of observational studies in the presence of treatment selection bias: effects of invasive cardiac management on AMI survival using propensity score and IV methods. J. Am. Med. Assoc. 297(3), 278–285 (2007)
Suissa, S.: Immortal time bias in pharmacoepidemiology. Am. J. Epidemiol. 167(4), 492–499 (2008)
Tan, Z.: Marginal and nested structural models using IVs. JASA 105(489), 156–169 (2010)
Thanassoulis, G., O’Donnell, C.J.: Mendelian randomization. J. Am. Med. Assoc. 301(22), 2386–2388 (2009)
Vansteelandt, S., Bowden, J., Babanezhad, M., Goetghebeur, E.: On IVs estimation of causal odds ratios. Stat. Sci. 26(3), 403–422 (2011)
Wooldridge, J.M.: Econometic analysis of cross section and panel data. The MIT Press, Cambridge (2002)
Acknowledgments
A. James O’Malley’s research that contributed to this paper was supported by NIH Grant 1RC4MH092717-01. The authors have no conflicts of interest to report.
Author information
Authors and Affiliations
Corresponding author
Appendices
Appendix 1: The marginal is Cox proportional hazards model
Here we show that if the treatment \(X\) is applied independently of \(U\) in model (1) that the marginal distribution of the time-to-event with respect to \(X\) is proportional hazards. This is relevant because it shows that parameter \(\beta\) of that model has an intepretation as the logarithm of the hazard ratio comparing the potential outcome if a subject recieves \(X=1\) to the potential outcome if a subject recieves \(X=0\) and \(U\) is unknown (i.e. not conditioned on).
The survival curve corresponding to model (1), \(S(t;x|U)={\mathrm{Pr}}[\widetilde{T}(x)\ge t|U]\) is
where \(H(u,t)=\int _0^t h(u,t^{\prime })dt^{\prime }\). If \(X\) is applied independently of \(U\) then the marginal causal model for the survival curve, obtained by integrating over \(U\) equals
Suppose \(H(u,t)=A(t)u + B(t)\). Then (7) becomes
Therefore if we take \(B(t)=\ln {\mathrm{mgf}}_U[-A(t)]\) the marginal causal survival curve is \(S(t;x)=\exp \biggl [-\Lambda _{01}(t)\exp (\beta ^{\prime }x)\biggr ]\) which is a Cox proportional hazards model.
This particular causal conditional model has the property that E\([h(u,t)|\widetilde{T}(x) \ge t]\) equals zero as demonstrated below:
when \(B(t)=\ln {\mathrm{mgf}}_U[-A(t)]\).
Appendix 2: R code for implementing the estimating equation
Rights and permissions
About this article
Cite this article
MacKenzie, T.A., Tosteson, T.D., Morden, N.E. et al. Using instrumental variables to estimate a Cox’s proportional hazards regression subject to additive confounding. Health Serv Outcomes Res Method 14, 54–68 (2014). https://doi.org/10.1007/s10742-014-0117-x
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10742-014-0117-x