Univariate Multivariate model 1b Multivariate model 2c
  OR (95% CI)a OR (95% CI) OR (95% CI)
Sex, age, anthropometric factors, and medical history
Sex      
female (vs. male) 0.98 (0.65-1.47) 1.49 (0.93-2.39) 2.19 (1.15-4.15)
Age      
40.0 - 56.7 1 1 1
56.8 - 65.1 4.16 (2.41-7.16) 3.54 (2.01-6.23) 3.75 (2.10-6.72)
65.2 - 83.7 13.3 (7.74-22.9) 12.4 (6.96-21.9) 14.9 (8.07-27.5)
p for trend       <0.0001       <0.0001       <0.0001
Height      
136 - 151 1   1
152 - 156 0.75 (0.49-1.15)   1.21 (0.71-2.05)
157 - 183 0.61 (0.40-0.93)   1.03 (0.53-2.00)
p for trend 0.020   0.864
Weight at 40 years of age      
35.0 - 49.0 1   1
50.0 - 55.0 1.13 (0.73-1.75)   1.27 (0.76-2.14)
56.0 - 95.0 1.45 (0.94-2.24)   2.50 (1.41-4.43)
p for trend 0.093   0.001
BMI at 40 years of age      
15.81 - 21.04 1 1  
21.05 - 23.18 1.55 (1.00-2.42) 1.27 (0.78-2.07)  
23.19 - 35.37 1.95 (1.26-3.01) 1.76 (1.10-2.84)  
p for trend 0.003 0.019  
Weight      
33.0 - 51.0 1    
52.0 - 58.0 1.30 (0.84-2.02)    
59.0 - 96.0 1.46 (0.94-2.27)    
p for trend 0.095    
BMI      
13.84 - 21.874 1    
21.875 - 24.44 1.30 (0.83-2.02)    
24.45 - 38.79 1.98 (1.28-3.05)    
p for trend 0.002    
History of treatment for knee injury (≥2 weeks)
yes (vs. no) 2.73 (1.75-4.26) 2.07 (1.29-3.33) 2.01 (1.24-3.24)
Occupation and lifestyle
Occupation      
physical worker (vs. non-physical worker) 1.79 (1.25-2.54) 1.93 (1.31-2.85) 2.09 (1.41-3.11)
Often good posture while sitting before age 40
yes (vs. no) 1.73 (1.21-2.48) 1.19 (0.78-1.81) 1.23 (0.80-1.87)
a OR = odds ratio, 95% CI = 95% confidence interval
b Multivariate model included sex, age, BMI at 40 years of age, history of treatment for knee injury, occupation, and often good posture while sitting before age 40. Analysis based on a sample size of 493.
c Multivariate model included sex, age, height, weight at 40 years of age, history of treatment for knee injury, occupation, and often good posture while sitting before age 40. Analysis based on a sample size of 493.
Table 2: Association between selected characteristics and knee OA among Japanese.