Linear mixed model (LMM) has the advantage of modeling the corelated data. Alzheimer’s disease (AD) is a chronic neurogenerative disease that affects the brain of the subject. No study was found to study the longitudinal effect of apolipoprotein E epsilon 4 (APOE4) genotype on the logical memory test total score in AD. A longitudinal data of 844 with AD, 2167 with cognitive normal (CN), and 4472 with mild cognitive impairment (MCI) participants who underwent logical memory examination test in the Alzheimer's Disease Neuroimaging Initiative (ADNI) were investigated. Episodic memory of the study participants was monitored based on a short story told to the participants and then participants asked to recall what was told. The multivariate LMM was used to determine the longitudinal changes in the logical memory test total score adjusting for age and sex. The Akaike information criterion (AIC) statistic and the Bayesian information criterion (BIC) statistic were used to select the best covariance structure. The repeated measures longitudinal analysis was performed using PROC MIXED in SAS 9.4. Both AIC and BIC statistics favor the unstructured correlated structure (UN). Using a UN model in the LMM, the APOE gene was is significantly associated with logical memory test total score (pUN covariance structure is the best. This study provided the first evidence of the effect of APOE4 genotype on the logical memory related to AD.
| Identifer | oai:union.ndltd.org:ETSU/oai:dc.etsu.edu:asrf-1239 |
| Date | 12 April 2019 |
| Creators | Fokuoh, Evelyn, Wang, Kesheng |
| Publisher | Digital Commons @ East Tennessee State University |
| Source Sets | East Tennessee State University |
| Detected Language | English |
| Type | text |
| Format | application/pdf |
| Source | Appalachian Student Research Forum |
| Rights | http://creativecommons.org/licenses/by/4.0/ |
Page generated in 0.0021 seconds