Skip to main content
. 2023 Nov 15;15:200. doi: 10.1186/s13195-023-01344-0

Table 4.

Linear mixed effect models to identify predictors of HR-QoL

Variable EQ-5D utility VAS
Estimates (SE) P value Estimates (SE) P value
Clinical stage  < 0.05  < 0.001
 SCD Reference Reference
 MCI  − 0.012 (0.004)a  − 1.554 (0.286)a
 Mild  − 0.017 (0.015)  − 2.063 (0.777)a
 Moderate 0.03 (0.021)  − 1.308 (0.995)
 Severe 0.202 (0.155)  − 8.674 (3.27)a
 Time (year)  − 0.006 (0.001)  < 0.001  − 0.504 (0.065)  < 0.001
Clinical stage × time interaction  < 0.001
 SCD × time Reference
 MCI × time 0.001 (0.002)
 Mild × time 0.02 (0.005)a
 Moderate × time 0.009 (0.006)
 Severe × time  − 0.057 (0.039)
Baseline age  − 0.001 (0.001)  < 0.05  − 0.016 (0.032) 0.061
Sex female  − 0.067 (0.006)  < 0.001  − 1.743 (0.562)  < 0.05
Education (year) 0.001 (0.001) 0.23 0.172 (0.09) 0.06
BMI  − 0.005 (0.001)  < 0.001  − 0.122 (0.063) 0.05
Diabetes  − 0.032 (0.011)  < 0.05  − 3.076 (0.972)  < 0.05
Cardiovascular history  − 0.04 (0.009)  < 0.001  − 2.784 (0.791)  < 0.001
Institutionalisation  − 0.05 (0.017)  < 0.05 1.633 (1.542) 0.29
Depression  − 0.014 (0.001)  < 0.001  − 1.063 (0.089)  < 0.001
IADL  − 0.125 (0.005)  < 0.001  − 3.009 (0.459)  < 0.001

Models included diabetes, cardiovascular history, institutionalisation, depression, and IADL in addition to the baseline demographics. The purpose of the models was to identify predictors for HR-QoL. We did not interpret the estimates for the clinical stage since we included institutionalisation, depression, and IADL which are the consequences of cognitive decline (clinical stage in the model)

SCD Subjective cognitive decline, MCI Mild cognitive impairment, BMI Body mass index; Depression, measured by neuropsychological inventory clinician-rated version (higher scores indicate more symptoms), IADL Instrumental activities of daily living (higher scores indicate more functional impairment)

aStatistically significant