Table 5.

Posterior distribution [mean (SD) and 95 % equal-tailed credibility interval (CI)] of the parameters of the Bayesian non-linear mixed model for interobserver agreement

	Piecemeal deglutition			Postswallow vallecular pooling			Postswallow pyriform sinus pooling			Laryngeal penetration/tracheal aspiration
	Mean (SD)	95 % CI		Mean (SD)	95 % CI		Mean (SD)	95 % CI		Mean (SD)	95 % CI
Intercept	1.45 (0.21)	1.08	1.88	0.70 (0.15)	0.41	1.01	0.43 (0.19)	0.072	0.80	0.36 (0.16)	0.056	0.67
Consistencya	−0.51 (0.17)	−0.86	−0.19	−0.86 (0.20)	−1.27	−0.51	−0.28 (0.18)	−0.66	0.081	0.64 (0.17)	0.33	0.98
Groupb	−0.14 (0.20)	−0.53	0.25	−0.0011 (0.19)	−0.37	0.36	−0.049 (0.21)	−0.48	0.34	0.012 (0.17)	−0.33	0.32
σ 2	0.18 (0.13)	0.0045	0.48	0.051 (0.060)	0.00	0.21	0.13 (0.086)	0.0012	0.32	0.022 (0.017)	0.00	0.063

To facilitate interpretation of the table, a more detailed description is given. Mean, SD, and, 95 % CI are presented separately per FEES variable. When ‘0’ is not entailed in the 95 % CI, the difference between the predictors (thin and thick liquid consistency, or neurological and oncological group) is statistically significant. A positive mean indicates that the agreement of the predictor used as reference is lower. For instance, in the line consistency, the agreement level between thin and thick liquid consistencies is compared. ‘0’ is not entailed in the 95 % CI of all FEES variables, except for postswallow pyriform sinus pooling (−0.66, 0.081). It means that there is a statistically significant difference on the agreement level depending on the consistency scored. A negative mean for piecemeal deglutition (−0.51) and postswallow vallecular pooling (−0.86) indicates that the interobserver agreement for thick was higher than that for thin liquid

SD standard deviation

The groups used as reference are:

^aThick liquid for bolus consistency

^bNeurological patients for the etiological group