Skip to main content

Table 5 Factors associated jointly with change in WOMAC at 10 years exceeding minimal clinically important difference: multivariable analysis

From: Long-term health related quality of life in total knee arthroplasty

Factors

Change in WOMAC exceeding MCID*

 

Pain

Stiffness

Function

 

Odds ratio**

(95% CI)

p

value

Odds ratio

(95% CI)

p

value

Odds ratio

(95% CI)

p

value

 

Baseline WOMAC

1.06 (1.04–1.08)

< 0.001

1.07 (1.06–1.09)

< 0.001

1.06 (1.04–1.08)

< 0.001

 

Age (continuous)

    

0.94 (0.90–0.97)

0.001

 

BMI categorized

       

 BMI < 25

  

-ref-

-ref-

-ref-

-ref-

 

 25 ≤ BMI < 30

  

0.19 (0.05–0.66)

0.009

0.52 (0.21–1.29)

0.159

 

 30 ≤ BMI < 35

  

0.18 (0.05–0.66)

0.009

0.35 (0.14–0.88)

0.026

 

 BMI ≥ 35

  

0.20 (0.05–0.79)

0.022

0.25 (0.08–0.74)

0.012

 

Comorbidity

       

 Peripheral vascular disease

    

0.49 (0.24–0.99)

0.048

 

 Depression

0.42 (0.18–0.94)

0.036

0.26 (0.09–0.77)

0.015

0.19 (0.06–0.56)

0.003

 

 Rheumatologic disease

0.37 (0.17–0.80)

0.011

     

Readmissions (6 months)

    

0.41 (0.18–0.94)

0.035

 

Complications from hospital discharge

  

0.31 (0.11–0.91)

0.032

   

Rehabilitation from hospital discharge

  

2.12 (1.18–3.80)

0.012

   

AUC (95%CI)

0.74 (0.69–0.79)

0.85 (0.81–0.89)

0.78 (0.73–0.82)

  1. Abreviations: BMI, Body Mass Index; WOMAC, Westen Ontario and McMaster Universities Osteoarthritis Index (0, best; 100, worst); CI, Confidence interval; MCID, Minimal clinically important difference; AUC: Area under the ROC curve; ref: reference group
  2. *Changes in WOMAC exceeding the cut-offs points of the MCID (Pain = 29.88; Stiffness = 26.76; Function = 32.72)
  3. **An Odds ratio < 1 indicates a lower likelihood of exceeding minimal clinically important difference; an Odds ratio > 1 indicates a higher likelihood of exceeding minimal clinically important difference
  4. These models have been adjusted for age and gender and they only appear in the table if some of them were statistically significant