| Popis: |
This study aimed to establish criterion validity of normative regression GAITRite values for discriminating between amnestic Mild Cognitive Impairment (a-MCI) and cognitively intact controls using a reference sample (N=312, age 69 - 95) of healthy participants from the Victoria Longitudinal Study. Regression norming was used to derive weighted equations which were deployed in a target sample of 34 older adults (a-MCI=12, controls=22, 70–85 years of age). The difference between observed and predicted scores was derived, standardized as T-scores, and then used in logistic regression analyses to classify aMCI and healthy controls. The standardized difference scores for velocity, stride time variability and support base variability while under cognitive load (counting backward by 7’s) yielded a classification accuracy of 93.2% with specificity of 96.9% and sensitivity of 83.3%. This is the first evidence of clinically meaningful classification of aMCI using regression norming values and variability indicators derived from the GAITRite system. |
