Multivariable logistic regression of parameters potentially contributory to the classification of spondyloarthritis (SpA) in patients with peripheral arthritis, enthesitis, or dactylitis (n=266)
| All patients | ||
|---|---|---|
| OR (95% CI) | p Value | |
| Male gender | 1.09 (0.54 to 2.20) | 0.80 |
| Age (per year increase in age) | 0.96 (0.93 to 0.99) | 0.034 |
| Peripheral arthritis, ever | 1.04 (0.31 to 3.55) | 0.94 |
| Enthesitis at any site, ever | 5.61 (2.66 to 11.82) | <0.001 |
| Dactylitis, ever | 0.93 (0.39 to 2.23) | 0.87 |
| Uveitis, ever | 2.62 (0.17 to 39.90) | 0.49 |
| Psoriasis, ever | 12.20 (2.25 to 66.00) | 0.004 |
| Preceding infection | 3.86 (0.83 to 17.97) | 0.085 |
| IBP* in the past | 2.12 (0.85 to 5.35) | 0.11 |
| Family history for SpA (ankylosing spondylitis, reactive arthritis, uveitis, psoriasis, IBD) | 3.54 (1.24 to 10.12) | 0.018 |
| HLA-B27 | 13.72 (4.69 to 40.58) | <0.001 |
| Definite radiographic sacroiliitis (≥grade 2 bilateral or ≥grade 3 unilateral) | 13.10 (2.52 to 68.10) | 0.002 |
p Values of ≤ 0.05 are considered to be statistically significant. Sacroiliitis on MRI was not entered into the logistic regression analysis because of too few patients (n=60) with available information and because of infinite 95% CI. IBD was removed from the analysis because of infinite 95% CI. The R2 of this logistic regression analysis was 0.513 (Nagelkerke&#x0027;s).
↵* IBP according to the rheumatologist&#x0027;s judgement.
HLA-B27, human leucocyte antigen B27; IBD, inflammatory bowel disease; IBP, inflammatory back pain.