Table 1. Association with change in global cognition from the models without mixture (K=1).

	Model 1 Estimate (SE, p)	Model 2 Estimate (SE, p)	Model 3 Estimate (SE, p)
Fixed Effects on baseline level (intercept)
Age at baseline (α1)	-	-.024 (.002, <.001)	-.022 (.002, <.001)
Male sex (α2)	-	-.054 (.035, .120)	-.074 (.034, .030)
Education (α3)	-	.034 (.005, <.001)	.033 (.004, <.001)
τ tangle density	-	-	-.016 (.002, <.001)
Macroscopic infarcts	-	-	-.067 (.033, .042)
Cortical Lewy bodies	-	-	-.027 (.049, .586)
Hippocampal sclerosis	-	-	-.099 (.057, .083)
Fixed Effects on decline (slope)
Age at baseline (β1)	-	-.002 (.0001, .006)	-.002 (.0008, .027)
Male sex (β2)	-	.006 (.013, .639)	-.004 (.012, .728)
Education (β3)	-	.003 (.002, .077)	.003 (.002, .040)
τ tangle density	-	-	-.006 (.001, <.001)
Macroscopic infarcts	-	-	.005 (.011, .664)
Cortical Lewy bodies	-	-	-.024 (.016, .138)
Hippocampal sclerosis	-	-	-.007 (.018, .716)
Fixed Effects on decline (quadratic)
Age at baseline (β4)	-	-.0003 (.0001, <.001)	-.0002 (.0001, <.001)
Male sex (β5)	-	.0005 (.0007, .492)	.0005 (.0007, .476)
Education (β6)	-	-.0002 (.0001, .032)	-.0002 (.0001, .039)
τ tangle density	-	-	-.0004 (.0001, <.001)
Macroscopic infarcts	-	-	-.003 (.001, <.001)
Cortical Lewy bodies	-	-	-.006 (.001, <.001)
Hippocampal sclerosis	-	-	-.003 (.001, .004)
Random Effects
Mean Intercept (μ0)	-.003 (.018, .863)	.001 (.020, .944)	.134 (.027, <.001)
Variance ( ${\sigma }_1^2$)	.219 (.012, <.001)	.173 (.010, <.001)	.163 (.010, <.001)
Mean Linear (μ1)	-.034 (.006, <.001)	-.032 (.007, <.001)	.006 (.009, .497)
Variance ( ${\sigma }_2^2$)	.018 (.001, <.001)	.017 (.001, <.001)	.012 (.001, <.0001)
Covariance (σ12)	.017 (.003, <.001)	.010 (.002, <.001)	.006 (.002, .004)
Mean Quadratic (μ2)	-.010 (.0003, <.001)	-.011 (.0005, <.001)	-.006 (.0007, <.001)
Residuals Variance	.088 (.002, <.001)	.087 (.002, <.001)	.086 (.002, <.001)

SE: Standard error.