Table 6.

Performance of models (N=1273, cases=634, controls=639)

	WAIC	Weight	AUC	Sens	Spec	NPV	PPV	F1
Animal Fluency
SD timings	1586.29	0.955	0.625	0.631	0.585	0.656	0.558	0.592
SD timings, ASR	1592.42	0.045	0.621	0.644	0.571	0.659	0.555	0.596
Letter F Fluency
Mayr scores	1757.24	0.991	0.591	0.552	0.579	0.566	0.565	0.559
Mayr scores, ASR	1766.71	0.009	0.580	0.539	0.606	0.570	0.576	0.557
Combined Model
SD timings + Mayr	1584.08	0.931	0.632	0.599	0.637	0.656	0.578	0.588
SD timings + Mayr, ASR	1589.29	0.069	0.630	0.606	0.621	0.655	0.571	0.588

ASR = scores estimated from transcriptions derived directly from automatic speech recognition; AUC = area under the ROC curve; F1 = F1 measure (harmonic mean of PPV and Sens); NPV = negative predictive value; PPV = positive predictive value; SD = slope difference; Sens = sensitivity; Spec = specificity; WAIC = Widely applicable information criterion.