Skip to main content
. 2013 Nov 13;8(11):e78991. doi: 10.1371/journal.pone.0078991

Table 4. Sensitivity analyses of multivariable-adjusted association of diabetes with moderate-severe ADL limitation at 2- and 5-years.

2-years 5-years
OR (95% CI) p-value OR (95% CI) p-value
Model 2 (Model 1 +anxiety + depression)a
Control 1.0 (ref) 1.0 (ref)
Diabetes without complications 1.45 (1.04, 2.01) 0.03 1.59 (1.04, 2.41) 0.03
Diabetes with complications 2.49 (1.59, 5.47) 0.001 3.00 (1.28, 7.06) 0.01
Model 3 (Model 1 +anxiety + depression + ipsilateral hip involvement)b
Control 1.0 (ref) 1.0 (ref)
Diabetes without complications 1.37 (0.92, 2.06) 0.12 1.46 (0.87, 2.45) 0.15
Diabetes with complications 2.51 (1.23, 5.09) 0.01 3.13 (1.29, 7.61) 0.01
Model 4 (Model 1 minus Deyo-Charlson index)c
Control 1.0 (ref) 1.0 (ref)
Diabetes without complications 1.56 (1.14, 2.17) 0.006 1.60 (1.51, 2.43) 0.03
Diabetes with complications 3.86 (2.13, 6.99) <0.001 3.34 (1.51, 7.37) 0.003
a

Adjusted the main multivariable model for anxiety and depression, i.e. the model is adjusted for a total of 11 additional covariates/confounders: Age, gender, BMI, Deyo-Charlson comorbidity score, ASA class, distance from the medical center, income, underlying diagnosis, preoperative moderate-severe ADL limitation, anxiety and depression

b

Adjusted the main multivariable model for anxiety, depression and ipsilateral hip pain, i.e. the model is adjusted for a total of 12 additional covariates/confounders: Age, gender, BMI, Deyo-Charlson comorbidity score, ASA class, distance from the medical center, income, underlying diagnosis, preoperative moderate-severe ADL limitation, anxiety, depression and ipsilateral hip involvement

c

Adjusted the main multivariable model for anxiety and depression, but not for Deyo-Charlson, i.e. the model is adjusted for a total of 8 additional covariates/confounders: Age, gender, BMI, ASA class, distance from the medical center, income, underlying diagnosis, preoperative moderate-severe ADL limitation