Skip to main content
. 2022 Sep 29;62(5):1877–1886. doi: 10.1093/rheumatology/keac557

Table 3.

Agreement of classification by discipline, along dimensions of severity, risk of progression and progression

A. Determined by Kappa statistic
Kappa calculation n (pair)a Average n (profile)b Mean Bootstrapped mean (95% CI)c
Severity
 Between rheumatologists and pulmonologists 66 7.6 0.13 0.13 (0.00, 0.25)
 Among rheumatologists 44 8.7 0.17 0.17 (−0.01, 0.45)
 Among pulmonologists 17 6.6 0.20 0.18 (0, 0.25)
Risk of progression
 Between rheumatologists and pulmonologists 66 6.6 0.61 0.59 (0.49, 0.69)
 Among rheumatologists 44 8.3 0.70 0.66 (0.51, 0.86)
 Among pulmonologists 17 5.9 0.48 0.4618 (0.26, 0.66)
Progression
 Between rheumatologists and pulmonologists 66 3.1 0.56 0.51 (0.18, 0.70)
 Among rheumatologists 44 3.5 0.78 0.70 (0.36, 0.95)
 Among pulmonologists 17 3.1 0.29 0.24 (−0.00, 0.50)
B. Determined by χ2 analysis
χ2 calculation Rheumatology Pulmonology P-value
Severityd
 Clinical ILD 205 (93.2%) 114 (89.8%) 0.26
 Subclinical ILD 15 (6.8%) 13 (10.2%)
Risk of progression
 High risk 97 (45.3%) 55 (46.2%) 0.88
 Low risk 117 (54.7%) 64 (53.8%)
Progression
 Progressive 17 (18.9%) 11 (20.0%) 0.20
 Stable 57 (63.3%) 40 (72.7%)
 Improved 16 (17.8%) 4 (7.3%)
a

Number of paired used to calculate kappa statistics.

b

Average number of profile in each pair.

c

100 bootstrap datasets, randomly selecting based on profile with replacement.

d

‘Cannot tell’ was removed from this calculation.