Skip to main content
. 2018 Aug 24;18:192. doi: 10.1186/s12877-018-0880-6

Table 2.

Multiple Linear Regression Model Summary for TMT-A, TMT-B and Digit symbol

Dependent variables Independent variables
Model 1 Model 2
HG PPT HG PPT
β (95% CI) p-value β (95% CI) p-value β (95% CI) p-value β (95% CI) p-value
TMT-A −0.12 (− 0.55, 0.06) 0.114 −0.33 (−2.60, − 1.32) < 0.001 −0.33 (− 2.58, − 1.29) < 0.001 −0.09 (− 0.47, 0.10) 0.177
TMT-B −0.02 (−1.11, 0.84) 0.782 − 0.12 (− 4.38, − 0.06) 0.044 −0.12 (− 4.38, − 0.04) 0.046 −0.01 (− 1.04, 0.90) 0.888
Digit symbola 0.02 (− 0.39, 0.31) 0.827 0.30 (1.08, 2.56) < 0.001 0.30 (1.08, 2.56) < 0.001 0.03 (−0.39, 0.28) 0.751

Each of the regressions was performed separately for each independent variable (TMT-A, TMT-B, and digit symbol)

Model 1: includes two hand variables analyzed separately as independent variables and adjusted for age (a continuous variable), sex, years of education, TMIG-Index of Competence, GDS, hypertension, stroke, heart disease, diabetes and MMSE

Model 2: includes two hand variables analyzed together as independent variables and adjusted for the same covariates as Model 1

CI Confidence Interval, HG Handgrip strength, PPT Purdue Pegboard Test

aThe total number of subjects analyzed was 207