Table 2.
Logistic regression analyses predicting persistent vs. non-persistent to first TNF inhibitor
| Regression coefficient | Standard error | Chi-square | Odds ratio (95% CI) | p value | |
|---|---|---|---|---|---|
| Intercept | − 0.9700 | 0.2470 | 15.4200 | < 0.0001 | |
| Gender (female vs. male) | − 0.4563 | 0.1311 | 12.1186 | 0.634 (0.49–0.81) | 0.0005 |
| Age | 0.00974 | 0.0053 | 3.3297 | 1.010 (0.99–1.02) | 0.0680 |
| Medicaid | − 0.7560 | 0.4170 | 3.2873 | 0.470 (0.20–1.06) | 0.0698 |
| Medicare | − 0.2249 | 0.3030 | 0.5510 | 0.799 (0.44–1.44) | 0.4579 |
| Use of cDMARD | 0.2633 | 0.1333 | 3.9043 | 1.301 (1.00–1.69) | 0.0482 |
| Use of opioid | − 0.4637 | 0.1226 | 14.3172 | 0.629 (0.49–0.80) | 0.0002 |