Table S2. Univariate series analysis for predicting percent sleep stage by mouth guard setting.
| Variable | Mean (%) | Wake
estimate bc (SE) |
p-value | Mean (%) | Light
estimate bc (SE) |
p-value |
|---|---|---|---|---|---|---|
| μ 0 | 15.6 | 0.52 (0.04) | < 1.9e-14 | 33.3 | 0.19 (0.01) | < 2e-16 |
| β MG0 x | 7.4 | -0.12 (0.04) | 0.0021 | — | — | — |
| β MG4 x | 4.6 | -0.18 (0.04) | 0.0002 | 39.3 | 0.05 (0.02) | 0.0089 |
| β MG6 x | 2.7 | -0.24 (0.06) | 0.0004 | 39.0 | 0.05 (0.02) | 0.0185 |
| βNight | 16.0 | 0.01 (0.00) | 0.0146 | — | — | — |
| bc: exponent = 0.355 | bc: exponent = 1.5 | |||||
| Adjusted R2: 0.3310 | Adjusted R2: 0.1875 | |||||
| p-value = 0.0022 | p-value = 0.0123 | |||||
Adjusted mean in percent, transformed mean estimate bc, standard error (SE) and p-value (Pr > |t-value|). R2: R-squared; bc: Box-Cox transformed variable raised to exponent given in final model. Back-transformation to original units was performed (after adjustments relative to intercept) by taking the nth (exponent) root of estimate.