Table 5.
Multivariable logistic regression model for HSC-cupula subtype classification.
| Variable | β | OR [95% CI] | p value |
|---|---|---|---|
| Age, years | −0.21 | 0.979 [0.932–1.029] | 0.410 |
| Sex (male) | 0.628 | 1.874 [0.741–4.738] | 0.185 |
| Presence of SSNHL history | −21.004 | 0 [0-] | >0.999a |
| Presence of stroke history | 1.369 | 3.930 [0.405–38.111] | 0.238 |
| Presence of NP1 | −0.468 | 0.626 [0.235–1.669] | 0.349 |
| Presence of NP2 | 21.173 | 1.56 × 109 [0-] | >0.999a |
| Presence of NP3 | −0.008 | 0.992 [0.338–2.916] | 0.989 |
| CP value, percentage | 0.010 | 1.010 [0.987–1.034] | 0.384 |
Reference group: HSC-Hcu (coded as 0). aComplete separation occurred for SSNHL (exclusively in HSC-Lcu group) and NP2 (all present in HSC-Hcu group), resulting in non-interpretable estimates. β, regression coefficient; CI, confidence interval; HSC, horizontal semicircular canal; Lcu, light cupula; Hcu, heavy cupula; SSNHL, sudden sensorineural hearing loss; OR, odds ratio; NP, null point; CP, canal paresis.