Table 3.

Estimates of predictive performance

	GLMM (LS)				GLMM (TS)
	Sens	Spec	PPV	NPV	Sens	Spec	PPV	NPV
K=200:
LS FP< 0.05	0.42	0.95	0.98	0.25	0.66	0.96	0.99	0.31
LS FP< 0.10	0.61	0.91	0.97	0.32	0.77	0.96	0.99	0.39
K=350:
LS FP< 0.05	0.50	0.95	0.93	0.60	0.61	0.95	0.95	0.59
LS FP< 0.10	0.66	0.90	0.90	0.67	0.79	0.90	0.93	0.71
(a) GLMM approach

	LMM (LS)				LMM (TS)
	Sens	Spec	PPV	NPV	Sens	Spec	PPV	NPV
K=200:
LS FP< 0.05	0.66 (0.60, 0.75)	0.95	0.99 (0.98, 0.99)	0.36 (0.31, 0.46)	0.77	0.96	0.99	0.39
LS FP< 0.10	0.79 (0.74, 0.83)	0.90	0.98 (0.97, 0.98)	0.47 (0.37, 0.54)	0.84	0.96	0.99	0.49
K=350:
LS FP< 0.05	0.57 (0.44, 0.67)	0.95	0.94 (0.92, 0.95)	0.63 (0.56, 0.70)	0.73	0.93	0.95	0.67
LS FP< 0.10	0.71 (0.65, 0.79)	0.90	0.90 (0.88, 0.92)	0.70 (0.66, 0.78)	0.84	0.90	0.94	0.77
(b) LMM approach