Table 2.

Cause-specific Cox regression models for predictors of discontinuation in CORRONA and NinJa (2 separate models).

Variable	CORRONA		NinJa
Variable	HR (95% CI)	p	HR (95% CI)	p
Female	1.06 (0.86–1.30)	0.613	2.19 (0.98–4.90)	0.055
Race
White	Reference
Black	0.96 (0.65–1.43)	0.838
Other	1.35 (0.85–2.14)	0.202
Age, each decade	1.04 (0.97–1.12)	0.246	1.06 (0.87–1.28)	0.568
RA duration, each decade bDMARD class	0.88 (0.79–0.99)	0.031	0.53 (0.33–0.85)	0.009
TNF inhibitors	Reference		Reference
Non-TNF inhibitors	0.93 (0.63–1.38)	0.731	0.80 (0.38–1.69)	0.552
CDAI	0.89 (0.80–1.00)	0.048	0.78 (0.56–1.07)	0.118
MTX use	1.56 (1.28–1.90)	< 0.001	0.80 (0.43–1.49)	0.483
Glucocorticoid use	1.24 (0.99–1.56)	0.064	0.94 (0.53–1.66)	0.820
Time since bDMARD, yrs	0.94 (0.83–1.05)	0.255	1.25 (0.95–1.64)	0.106
Index yr	1.01 (0.97–1.05)	0.688	1.10 (0.90–1.35)	0.359
C statistics	0.60		0.66
R²	0.01		0.04

CORRONA: COnsortium of Rheumatology Researchers Of North America; NinJa: National Database of Rheumatic Diseases by iR-net in Japan; RA: rheumatoid arthritis; bDMARD: biological disease-modifying antirheumatic drugs; TNF inhibitors: tumor necrosis factor inhibitors (adalimumab, certolizumab pegol, etanercept, golimumab, and infliximab); Non-TNF inhibitors: non-tumor necrosis factor inhibitor biologic agents (abatacept, anakinra, and tocilizumab); CDAI: Clinical Disease Activity Index; MTX: methotrexate.