Abstract

AbstractBackgroundWorldwide prevalence of dementia in older adults ranges from 5‐7%, with higher rates observed in Latin America (8.5%). However, underdiagnosis remains frequent. Our aim is to identify the profile of individuals with probable dementia that were underidentified using an epidemiological cohort.MethodAn epidemiological study named ELSI‐Brazil was evaluated for potential underidentification of dementia. Individuals at or above 60 years of age with probable dementia were selected using semantic fluency test scores (adjusted for age and education) and loss of any basic activity of daily living (Katz scale<6). Groups were divided into reported and unreported diagnosis of dementia (UpD), according to previous history of Alzheimer’s disease. UpD group compounds individuals potentially living with probable dementia without a proper medical evaluation. Individuals were evaluated for self‐reported risk factors, previous medical history, frailty scores, cognitive tests and neuropsychiatric symptoms. A multivariate logistic regression was performed to demonstrate predictors of UpD.ResultAmong 5221 older adults of the ELSI‐Brazil sample, 474 individuals (mean age 73.30±8.65, 67% females) were classified as probable dementia (pD), and 4747 individuals (mean age 69.6±7.42, 59% females) as probable not dementia (pND). Among probable dementia, unreported dementia (UpD) comprehended 459 older adults (96.4%), with significantly more clinical comorbidities, such as hypertension, diabetes, ischemic events (stroke or myocardial infarction), hearing loss and malignancy (p < 0.001). They also presented higher neuropsychiatric symptoms, such as poor quality of sleep, frequent usage of sleep drugs, and depression (p<0.001). Strong predictors of UpD included frailty score (OR: 2.06 [1.58 ‐ 2.68]), poor sleep quality (OR: 1.46 [1.13 ‐ 1.90]), frequent usage of sleep drugs (OR: 1.54 [1.15 ‐ 2.05]), and recent hospitalization (OR: 1.67 [1.20 ‐ 2.30]).ConclusionA specific profile of vulnerability was demonstrated in older adults with unidentified probable dementia in this cohort. Frailty, poor sleep quality, frequent usage of sleep drugs, and recent hospitalization were important predictors of UpD. This study provides insights to guide public health policies and target individuals more vulnerable to underidentified dementia.

Full Text
Paper version not known

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call