Skip to main content
. 2012 Oct 17;94(20):1825–1832. doi: 10.2106/JBJS.K.00569

TABLE II.

Univariate and Multivariate Associations Between Baseline Variables and Risk of Revision Total Hip Replacement Among Medicare Beneficiaries Undergoing Primary Total Hip Replacement from July 1995 through June 1996

Bivariate*
Multivariate†
Variable Hazard Ratio 95% Confidence Interval Hazard Ratio 95% Confidence Interval
Sex
 Male 1.26 1.18, 1.34 1.23 1.15, 1.31
 Female 1.0
Age
 65-75 yr 1.49 1.40, 1.60 1.47 1.37, 1.58
 >75 yr 1.0
Race
 White 0.90 0.79, 1.03
 Nonwhite 1.0
Medicaid eligibility
 Yes 0.92 0.80, 1.06
 No 1.0
Comorbidity (Charlson score)
 >1 1.02 0.93, 1.12
 0–1 1.0
Hospital volume
 1–25 1.14 1.07, 1.23
 26–50 1.01 0.94, 1.10
 >50 1.0
Surgeon volume
 1–5 1.24 1.15, 1.34 1.21 1.12, 1.32
 6–12 1.09 1.02, 1.18 1.08 1.00, 1.17
 >12 1.0 1.0
*

Hazard ratios obtained from bivariate Cox proportional hazards models. †Hazard ratios obtained from multivariate Cox proportional hazards model. N = 49,136 for the multivariate model.

Listed volumes indicate only Medicare patients who had total hip replacement.