Skip to main content
. Author manuscript; available in PMC: 2023 Feb 1.
Published in final edited form as: Neurobiol Aging. 2021 Oct 14;110:88–95. doi: 10.1016/j.neurobiolaging.2021.10.005

Table 4.

Moderating effect of each CR proxy on the relationship between AD-CT and cognition

Model term B SE β p
Model 1
 Education 0.98 0.14 0.30 <.001b
 AD-CT 39.74 3.41 0.51 <.001b
 Education × AD-CT −0.61 0.63 −0.04 .33
Model 2
 pIQ 0.55 0.05 0.40 <.001b
 AD-CT 37.27 3.24 0.48 <.001b
 pIQ × AD-CT −0.18 0.24 −0.03 .45
Model 3
 OC 1.93 0.49 0.17 <.001b
 AD-CT 39.18 3.56 0.50 <.001b
 OC × AD-CT −6.28 2.19 −0.11 .004a
Model 4
 LCA 5.81 0.89 0.26 <.001b
 AD-CT 40.03 3.50 0.51 <.001b
 LCA × AD-CT 0.59 4.20 0.01 .89

Note: Age, sex, and APOE4 carrier status were treated as covariates. B: unstandardized coefficient, β: standardized coefficient.

Key: CR, cognitive reserve; AD-CT, AD-signature region cortical thickness; pIQ, premorbid Intelligence Quotient; OC, occupational complexity; LCA, lifetime cognitive activity; SE, standard error; APOE4, apolipoprotein E ε4.

a

p < 0.01.

b

p < 0.001.