Propensity score methods to adjust for confounding in assessing treatment effects: bias and precision

Explore this paper's citation graph

Summary

Adjusted odds ratios using the PS stratification and the method using PS as a covariate were biased due to either residual confounding or over-adjustment, and matching on PS produced less biased average estimates than other methods but the precision of effect estimates was lower.

Type
article
Published
2008-12-31
Cited by
54
References
68
Access
Open access

Keywords

Propensity score matching, Confounding, Medicine, Statistics, Internal medicine

References

Cited by

Related papers