Abstract

MOLINA ARIAS, M.. Propensity score analysis: the desire to resemble clinical trials. Rev Pediatr Aten Primaria [online]. 2015, vol.17, n.65, pp.87-90. ISSN 1139-7632.  https://dx.doi.org/10.4321/S1139-76322015000100020.

Observational studies lack randomization. Therefore, there is no assurance that the different study groups have the same distribution of baseline covariates, which has the consequence that results cannot be compared directly without performing some form of adjustment for confounding or modifying effect. In this paper the use of propensity scores is described as an alternative method to the classic use of adjustment by logistic regression.

Keywords : Randomization; Propensity score; Logistic regression.

        · abstract in Spanish     · text in Spanish     · Spanish ( pdf )