Skip to main content
. Author manuscript; available in PMC: 2014 Jun 1.
Published in final edited form as: Arthritis Care Res (Hoboken). 2013 Jun;65(6):969–976. doi: 10.1002/acr.21926

Table 3.

Relationships between patient race/gender and willingness to undergo arthroplasty

Willing, n Unwilling, n OR (95% CI) a
Model 1 b,f
White Male (Referent) 177 35
White Female 232 67 0.68 (0.44, 1.08)
AA Female 131 75 0.35 (0.22, 0.55)
AA Male 45 31 0.29 (0.16, 0.51)
Model 2 c,f
White Male (Referent) 163 28
White Female 205 57 0.57 (0.34, 0.96)
AA Female 119 64 0.33 (0.18, 0.60)
AA Male 44 31 0.26 (0.13, 0.52)
Model 3 d,f
White Male (Referent) 163 28
White Female 205 57 0.59 (0.35, 1.01)
AA Female 119 64 0.36 (0.20, 0.65)
AA Male 44 31 0.28 (0.14, 0.55)
Model 4 e,f
White Male (Referent) 163 28
White Female 205 57 0.60 (0.35, 1.02)
AA Female 119 64 0.35 (0.19, 0.64)
AA Male 44 31 0.28 (0.14, 0.54)
a

Odds Ratio (95% Confidence Interval); 95% CIs are calculated using robust variance estimates

b

Model 1 = Race by sex only

c

Model 2 = Model 1 + site (VA or non-VA), age, income, and WOMAC total score

d

Model 3 = Model 2 + marital status + #close friends or family

e

Model 4 = Model 3 + MOS Social Support

f

Likelihood Ratio Test for Heterogeneity, p ≤0.001

AA=African-American; MOS=Medical Outcomes Study; WOMAC=Western Ontario McMaster Instrument