Skip to main content
. Author manuscript; available in PMC: 2017 Oct 1.
Published in final edited form as: J Am Geriatr Soc. 2016 Aug 22;64(10):2065–2073. doi: 10.1111/jgs.14253

Figure 2. Hypothetical Distribution of Propensity Scores Between Treated and Untreated Patients.

Figure 2

Abbreviation: PS, propensity score.

Figure 2A depicts a situation with high C statistic, in which the propensity score clearly distinguishes treated patients from untreated patients and there is a small overlap in propensity score. This may reflect a consistent clinical practice regarding treatment use or inclusion of strong predictors of the treatment, such as instrumental variables. Figure 2B depicts a situation with low C statistic, in which the propensity score modestly distinguishes treated patients from untreated patients and there is a large overlap in propensity score. This may indicate that a majority of patients have a chance of being treated or untreated, or that important confounders may have been omitted.