Modeling of atrophy size trajectories: variable transformation, prediction and age-of-onset estimation

Table 3 Analysis of risk factors in the analysis data set

Variable	Estimate	95% CI	p-value
time [in years]	0.42	(0.36, 0.50)	<0.0001
time x (hyperchol.= no)	0.11	(0.06, 0.17)	<0.0001
time x (hypertension = no)	-0.09	(-0.17,-0.03)	0.0004
Variance Term	Estimate
Eye:Patient γ_i	1.83²
Patient ζ_j	4.03²
Residuals ε	0.42²

The table presents the coefficient estimates and bootstrap 95% confidence intervals that were obtained from fitting Model (6) with transformed response (λ=0.45) to an imputed version of the analysis data set. The model parameter μ_Δ, which reflects the mean disease age at study entry, was estimated to be \(\hat {\mu }_{\Delta }=4.74\) (95% CI [3.41, 4.83]). P-values were obtained using the R package lmerTest [25]

ISSN: 1471-2288