Table 4.
Logistic regression analysis of the presence of recurrence
| B | SE | Wald | df | p | Exp(B) | PPV, % | NPV, % | Accuracy, % | |
|---|---|---|---|---|---|---|---|---|---|
| Constant | –2.949 | 0.948 | 9.680 | 1 | 0.002* | 0.052 | 75.0 | 85.6 | 85.2 |
| Births, n | 4.714 | 2 | 0.095 | ||||||
| None | –1.926 | 1.007 | 3.656 | 1 | 0.056 | 0.146 | |||
| Single | –0.442 | 0.793 | 0.311 | 1 | 0.577 | 0.643 | |||
| Vitamin B12 | 2.112 | 0.734 | 3.638 | 1 | 0.002 | 2.783 | |||
| Rheumatological disease | 2.312 | 0.783 | 5.734 | 1 | 0.004 | 4.536 | |||
| Complaint – fistula | 1.189 | 0.563 | 4.456 | 1 | 0.035 | 3.285 | |||
| Number of complaints ≥3 | 2.117 | 0.934 | 8.729 | 1 | 0.012 | 2.423 | |||
| Multicentricity | 2.745 | 0.884 | 9.651 | 1 | 0.002 | 2.064 | |||
| Erythema nodosum | 1.726 | 0.876 | 4.645 | 1 | 0.003 | 2.332 | |||
| Medical treatment (alone) | 1.517 | 0.480 | 9.974 | 1 | 0.002 | 4.557 |
Dependent: recurrence; independent: investigated variables. Bold values are statistically significant at p < 0.05. Backward method.