TABLE 12.

Final multivariable exact logistic regression models for classic myositis-specific autoantibodies

Variable	Odds Ratio Estimate	95% Confidence Interval	p Value
Anti-Synthetase versus Anti-SRP Autoantibodies (n = 25)
Arthralgia	21.9	1.7-1000	0.01
Likelihood ratio chi square 9.36; c statistic 0.84			0.0002
Anti-Synthetase versus Anti-Mi-2 Autoantibodies (n = 30)
Arthralgia	8.5	1.3-77.1	0.02
Likelihood ratio chi square 7.19; c statistic 0.74			0.007
Anti-Synthetase Autoantibodies versus MSA/MAA negative (n = 117)
Interstitial lung disease	54.5	7.4-04.2	< 0.0001
“Mechanic’s” hands	53.4	5.0-953	0.0002
Age at diagnosis (≥ 14 years vs. < 14 years)	7.0	0.9–87.4	0.07
Likelihood ratio chi square 55.1; c statistic 0.94			< 0.0001
Anti-SRP Autoantibodies versus Anti-Mi-2 Autoantibodies (n = 17)
Wheelchair use	11.1	0.8-707	0.09
Likelihood ratio chi square 5.21; c statistic 0.78			0.02
Anti-SRP Autoantibodies versus MSA/MAA negative (n = 117)
Wheelchair use	11.1	0.8-707	0.09
Race (black vs. white/other)	22.8	1.4-1000	0.02
Arthralgia	0.052	0.001–0.95	0.04
Likelihood ratio chi square 28.99; c statistic 0.94			0.0001
Anti-Mi-2 Autoantibodies versus MSA/MAA negative (n = 127)
Race (non-white vs. white)	4.3	1.01-19.5	0.05
Likelihood ratio chi square 5.13; c statistic 0.67			0.02

Abbreviations: SRP = signal recognition particle; MSA/MAA = myositis-specific and myositis-associated autoantibody.