Table 3.
Risk of non-persistence of biosimilar use versus originator use in etanercept-initiating patients, by pathology
| Pathology | Reference events | Biosimilars events | Raw HR | P value | IPTW aHR | P value |
| Rheumatoid arthritis | 48% (2731/5640) | 42% (2380/5625) | 0.84 (0.78 to 0.90) | <0.001 | 0.85 (0.78 to 0.93) | <0.001 |
| Ankylosing spondylitis | 55% (2977/5447) | 49% (2078/4255) | 0.86 (0.80 to 0.92) | <0.001 | 0.88 (0.80 to 0.96) | <0.001 |
| Psoriasic arthritis | 53% (484/905) | 49% (320/651) | 0.86 (0.72 to 1.03) | 0.033 | 0.82 (0.65 to 1.03) | 0.031 |
| Psoriasis | 57% (773/1362) | 51% (286/557) | 0.90 (0.76 to 1.07) | 0.123 | 0.80 (0.65 to 1.00) | 0.011 |
HRs are presented in point estimates (CI); bold font denotes significant differences. Bonferroni-corrected p value=0.0125.
IPTW computed through a multivariable logistic regression.
aHR, adjusted HR; IPTW, inverse probability of treatment weighting.