Diagnostic utility and psychometric properties of the Technology Activities of Daily Living Questionnaire (T-ADLQ) in people with non-formally educated with Alzheimer's disease in Lima, Peru

Rosa Montesinos¹Loreto Olavarria²Fernando Henríquez²Diego Chambergo-Michilot¹Belen Custodio¹CAROLINA ANDREA DELGADO³Marco Malaga¹Nilton Custodio^1*Andrea Slachevsky^2*

• ¹Peruvian Institute of Neurosciences (IPN), Lince, Peru
• ²Geroscience Center for Brain Health and Metabolism (GERO), Santiago, Chile
• ³Hospital Clinico UC, Santiago, Chile

ABSTRACT Objective: To evaluate the discriminative capacity and psychometric properties of the Technology–Activities of Daily Living Questionnaire (T-ADLQ) in distinguishing cognitively unimpaired individuals from those with amnestic mild cognitive impairment (aMCI) and Alzheimer's disease dementia (ADD) in a population with no formal education. Methods: This cross-sectional study included individuals with no formal education over 50 years of age residing in Callao, Peru. Participants were classified into three cognitive groups: cognitively unimpaired (CU; n = 64), aMCI (n = 60), and early ADD (n = 63). Functional decline was assessed with the T-ADLQ. Group comparisons were conducted using the chi-square or ANOVA tests, as appropriate. Pearson correlations were used to assess concurrent validity. The reliability of the T-ADLQ was assessed using Cronbach's alpha. Receiver operating characteristic (ROC) curve analyses and area under the curve (AUC) metrics were used to assess the discriminative validity of the measures across the three cognitive groups. Results: In a sample of 187 illiterate older adults, the T-ADLQ demonstrated excellent internal consistency (Cronbach's α = 0.966) and strong inverse correlations with global and executive cognitive measures (MMSE, RUDAS, IFS). It also showed a moderate positive correlation with PFAQ. ROC analyses revealed excellent discriminative performance of the T-ADLQ. The total score and the instrumental (IADL) and advanced (AADL) domains achieved perfect accuracy in differentiating cognitively unimpaired individuals from those with aMCI or ADD. The basic activities of daily living (BADL) domain also showed high accuracy, particularly in distinguishing aMCI from ADD. Conclusions: The T-ADLQ and its subdomains exhibit strong psychometric properties and high discriminative capacity in detecting functional decline in individuals with aMCI and