The author responds:
Shahar and Shahar1 make the following points:
- The built-in selection bias in hazard ratios can be motivated by references to susceptibility or by causal diagrams.
- Predicting the direction of the bias requires information on the nature of the interaction between exposure and other causes of disease.
- Selection on prebaseline variables may introduce bias in an observational study.
These 3 points have been previously made in the epidemiologic literature. See, for example, the articles cited by Shahar and Shahar. Two clarifications: (a) the concept of susceptibility (point 1) needs not be restricted to deterministic models as Shahar and Shahar seem to imply, and (b) the selection bias described in Shahar and Shahar's causal diagram (point 3) can be eliminated by restriction or stratification on prebaseline exposure. For some recent discussions on adjustment for prebaseline exposure see the references below.2–4
Miguel A. Hernán
Department of Epidemiology
Harvard School of Public Health
1.Shahar E, Shahar DJ. More on selection bias [letter]. Epidemiology
2.Robins JM, Hern´n MA, Rotnitzky A. Invited commentary: Effect modification by time-varying covariates. Am J Epidemiol
3.Ray WA. Evaluating medication effects outside of clinical trials: new user designs. Am J Epidemiol
© 2010 Lippincott Williams & Wilkins, Inc.
4.Hernán MA, Alonso A, Logan R, et al. Observational studies analyzed like randomized experiments: an application to postmenopausal hormone therapy and coronary heart disease [with discussion]. Epidemiology