Skip to main content

Advertisement

Table 3 Estimated parameters of variance (Var) and covariance (Cov) from the primary analysis and the three sensitivity analyses using bivariate linear mixed models

From: Estimating correlation between multivariate longitudinal data in the presence of heterogeneity

Parameters for variance-covariance& Primary analysis Sensitivity analysis #1 Sensitivity analysis #2 Sensitivity analysis #3
 Var(E1) 2.02 ± 0.06# 1.99 ± 0.06# 1.44 ± 0.05# 2.03 ± 0.06#
 Var(E2) 19.19 ± 0.96# 19.21 ± 0.96# 17.03 ± 0.86# 19.30 ± 0.96#
 Var(I1) 5.09 ± 0.51# 2.98 ± 0.34# 2.69 ± 0.28# 1.65 ± 0.20#
 Var(I2) 38.77 ± 4.63# 39.26 ± 4.78# 36.12 ± 4.33# 36.50 ± 4.41#
 Var(S1) 0.29 ± 0.04# 0.19 ± 0.03# 0.12 ± 0.02# 0.09 ± 0.02#
 Var(S2) 1.35 ± 0.24# 1.31 ± 0.24# 0.73 ± 0.16# 0.81 ± 0.18#
 Cov(I1, S1) 0.63 ± 0.10# 0.14 ± 0.07* 0.02 ± 0.05 −0.11 ± 0.04#
 Cov(I2, S2) −0.73 ± 0.79 −1.11 ± 0.81 −1.11 ± 0.63 −1.13 ± 0.69
 Cov(I1, I2) 3.88 ± 1.17# 3.01 ± 0.98# 2.14 ± 0.79# 1.13 ± 0.69
 Cov(S1, S2) 0.37 ± 0.08# 0.26 ± 0.06# 0.09 ± 0.04* 0.03 ± 0.04
 Cov(I1, S2) 1.16 ± 0.27# 0.53 ± 0.21* 0.13 ± 0.15 0.13 ± 0.14
 Cov(I2, S1) 0.67 ± 0.31* 0.44 ± 0.27 0.28 ± 0.19 0.40 ± 0.19*
  1. &Ek, Ik, Sk: error term, random intercept, and random slope for MD (k = 1) and VA (k = 2)
  2. * p < 0.05, # p < 0.01