Table 3.

Results of the multiple regression linear analyses for overall Cohen’s κ.

Predictors	INN vs BER		INN vs AUTO*		BER vs AUTO		CONS vs AUTO*		MAN vs AUTO*
	b	P	b	P	b	P	b	P	b	P
Intercept	−0.118	.003	−0.145	< .001	−0.188	< .001	−0.192	< .001	−0.208	< .001
Age	−0.106	< .001	−0.060	.053	−0.145	< .001	−0.097	.002	−0.132	< .001
Sex (F)	0.251	< .001	0.308	< .001	0.399	< .001	0.407	< .001	0.441	< .001
PLMS index	−0.023	.443	−0.047	.114	−0.033	.272	−0.057	.054	−0.042	.152
AHI	−0.227	< .001	−0.238	< .001	−0.175	<.001	−0.185	< .001	−0.200	< .001
BMI	−0.041	.189	−0.087	.005	−0.066	.031	−0.077	.012	−0.080	.008
Overall P	< .001		< .001		< .001		< .001		< .001

For each analysis, the overall Cohen’s κ was the outcome variable and age, sex (categorical), PLMS index, AHI, and BMI the predictors. Z-score transformations were applied to both outcome variable and predictors (except sex). For each model, the overall P value is reported, as well as the slope estimate (b) and the P value of each predictor. Statistical significance was set at the value of .05. INN vs BER: comparison of manual hypnograms scored in Innsbruck and Berlin; INN vs AUTO: comparison of manual hypnograms scored in Innsbruck to the automatic ones; BER vs AUTO: comparison of manual hypnograms scored in Berlin to the automatic ones; CONS vs AUTO: comparison of the epochs where manual scorers from Innsbruck and Berlin were in consensus to the respective epochs automatically scored; MAN vs AUTO: comparison of both manual hypnograms to the automatic one (in case of disagreement between manual scorers, an epoch was equally weighted between the 2 manually scored stages). AHI = apnea-hypopnea index, AUTO = automatic algorithm, BER = Berlin, BMI = body mass index, CONS = consensus, INN = Innsbruck. MAN = manual, PLMS = periodic limb movement during sleep. ^*Cubic transformation was applied to Cohen’s κ in the highlighted comparisons to meet the normality assumption of the model residuals.