Table 2.

Best-fitting Cox regression model of time to joint replacement surgery

Variable	Hazard ratio	95 % CI
Age (at cohort entry)	1.19	1.10, 1.28
Age (at cohort entry) squared	1.00	1.00, 1.00
Female sex	0.80	0.66, 0.96
Recipient of social assistance at cohort entry	0.89	0.59, 1.35
Baseline urban residence	1.03	0.82, 1.28
Baseline regional employment ratea	1.00	0.98, 1.01
Baseline regional family incomea	1.00	0.99, 1.01
Baseline regional percent university graduatesa	1.00	0.99, 1.01
Extra-articular RA manifestations (time-dependent)	1.08	0.91, 1.28
Rheumatology visits at baseline (log-transformed)	1.24	1.11, 1.38
Rheumatology visits in follow up, time dependent (log-transformed)	0.98	0.84, 1.15
High health care use at baseline	1.11	0.91, 1.35
Prior joint replacement	2.55	1.90, 3.43
Baseline comorbidities
Myocardial infarction	0.86	0.48, 1.56
Diabetes	0.99	0.76, 1.28
Osteoporosis	0.79	0.63, 0.99
Cerebrovascular disease	1.27	0.40, 4.02
Acute renal failure	1.16	0.68, 1.97
Chronic renal failure	1.60	0.90, 2.84
Coronary artery disease	1.01	0.81, 1.27
COPD-asthma	0.96	0.77, 1.18
Any type of cancer	1.01	0.72, 1.41
Osteoarthritis	2.15	1.81, 2.55
Charlson Comorbidity	0.92	0.85, 1.00
Drug use prior to cohort entry
Anti-TNF	1.00	0.35, 2.85
MTX	1.27	0.94, 1.70
Other DMARDs	1.13	0.88, 1.45
Systemic corticosteroids	1.15	0.97, 1.38
COXIBs	0.97	0.82, 1.15
NSAIDs	0.87	0.73, 1.03
Time-dependent cumulative drugs in follow up (years)
Anti-TNF	1.13	0.99, 1.29
Other biologic drugs	1.37	0.61, 3.08
Systemic corticosteroids	1.01	0.94, 1.08
COXIBs	1.20	1.11, 1.30
NSAIDs	1.31	1.20, 1.43
Cumulative MTX in first year (months)	0.94	0.92, 0.96
Cumulative other DMARDs in first year months)	0.97	0.95, 0.99

RA rheumatoid arthritis, MTX methotrexate, DMARD disease-modifying anti-rheumatic drug, COPD chronic obstructive pulmonary disease, TNF tumor necrosis factor, COXIB cyclooxygenase-2 inhibitor, NSAID nonsteroidal anti-inflammatory drug

^aDefined by census tract data