Table 5.
Linear regression models of phantosmias/phosphenes intensity and hedonic ratings
| B | SE | t | p | 95% CI | ||
|---|---|---|---|---|---|---|
| LB | UB | |||||
| Model 1: Phantosmia intensity | ||||||
| History of dust/chemical exposure = Yes | −1.52 | 0.61 | −2.50 | 0.02 | −2.75 | −0.29 |
| Model 2: Phantosmia hedonic | ||||||
| History of dust/chemical exposure = Yes | 0.49 | 0.21 | 2.30 | 0.03 | 0.06 | 0.93 |
| Model 3: Phosphene intensity | ||||||
| Tumor duration [months] | 0.11 | 0.04 | 3.15 | < 0.01 | 0.04 | 0.18 |
| Food allergy = Yes | 2.77 | 0.86 | 3.21 | < 0.01 | 1.04 | 4.50 |
| Epilepsy = Yes | −1.50 | 0.63 | −2.39 | 0.02 | −2.76 | −0.24 |
| Model 4: Phosphene hedonic | ||||||
| Analgesics = Yes | 0.47 | 0.17 | −2.84 | < 0.01 | 0.14 | 0.81 |
B beta coefficient, SE standard error, t statistical test used to determine the significance of the B coefficient, p level of significance, 95% CI 95% confidence interval, LB lower bound, UB upper bound
Stepwise method was used to select the most important predictors of phosphene or/and phantosmia during the radiation therapy