Response to Invited Commentary Stürmer et al. Respond to ‘‘Propensity Score Methods in Epidemiology’’

Abbreviation: PSC, propensity score calibration. We appreciate the thoughtful commentary of Oakes and Church (1) on our paper (2) and their conclusion that pro-pensity score calibration (PSC) may be helpful when some confounders are unmeasured. We agree that usual applica-tions of propensity score methods control only for confound-ing by ‘‘observable selection,’ ’ but we see much closer links between instrumental variables (3–5) and PSC than those de-scribed by Oakes and Church. Indeed, the gold-standard propensity score estimated in the validation study hopefully better approaches the true, but unknown, propensity of treat-ment than the error-prone propensity score and thus per-forms as an approximate instrument under assumptions similar to surrogacy (6, 7). PSC is no panacea for missing data on confounders