TABLE 2.

Estimates for Risk of 2 Outcomes in the Example Study

Outcome	Within-Center Analyses				Combined Analysis
	British Columbia (n = 19,979)	Pennsylvania (n = 4176)	New Jersey (n = 3998)	Horizon (n = 3451)	Pooled* (n = 31,604)
Myocardial infarction hospitalization
Unadjusted	1.68 (1.39, 2.03)	1.19 (0.83, 1.70)	1.35 (0.91, 2.02)	~1.47 (0.87, 2.48)	1.52 (1.30, 1.76)
Adjusted by individual covariates	1.23 (1.00, 1.52)	1.18 (0.80, 1.72)	~1.21 (0.79, 1.86)	~1.05 (0.60, 1.86)	1.20 (1.03, 1.41)
Adjusted by universal propensity score	1.23 (1.01, 1.50)	1.19 (0.82, 1.74)	1.14 (0.75, 1.71)	~1.11 (0.65, 1.91)	1.16 (1.00, 1.36)
Adjusted by local propensity score	1.15 (0.94, 1.41)	1.03 (0.69, 1.56)	1.27 (0.82, 1.98)	~0.99 (0.55, 1.78)	1.11 (0.95, 1.31)
Second hospitalization for revascularization
Unadjusted	1.30 (1.06, 1.61)	1.05 (0.81, 1.35)	0.92 (0.73, 1.15)	~1.38 (1.08, 1.75)	1.15 (1.02, 1.29)
Adjusted by individual covariates	1.23 (0.99, 1.53)	1.02 (0.78, 1.34)	0.96 (0.75, 1.22)	~1.39 (1.09, 1.79)	1.15 (1.02, 1.30)
Adjusted by universal propensity score	1.27 (1.02, 1.57)	1.03 (0.79, 1.35)	0.94 (0.74, 1.20)	~1.38 (1.07, 1.77)	1.13 (1.00, 1.28)
Adjusted by local propensity score	1.21 (0.97, 1.50)	1.07 (0.81, 1.43)	0.88 (0.68, 1.14)	~1.31 (1.00, 1.71)	1.07 (0.94, 1.21)

Figures indicated are odds ratios and their 95% confidence intervals, estimated with logistic regression models adjusted as indicated. The models adjusted by individual covariates and universal propensity score include the same variables and should therefore be substantially similar. Adapted from Pharmacoepidemiol Drug Saf. 2010 Feb 16, Epub ahead of print.

Universal PS included all measured covariates. Local PS included all measured covariates plus the best information available at the center.

PS-adjusted models are adjusted by decile of PS. Deciles are computed within each center.

^~indicates model failed to converge and value is approximate.

^*Pooled odds ratios are estimated with conditional logistic regression.