Format

Send to

Choose Destination
See comment in PubMed Commons below
J Alzheimers Dis. 2014;41(3):937-45. doi: 10.3233/JAD-140297.

Further validation of the Internet-based Dementia Risk Assessment.

Author information

1
School of Medicine, The Johns Hopkins University, Baltimore, MD, USA Bloomberg School of Public Health, The Johns Hopkins University, Baltimore, MD, USA The Copper Ridge Institute, Sykesville, MD, USA.
2
The Copper Ridge Institute, Sykesville, MD, USA.
3
School of Medicine, The Johns Hopkins University, Baltimore, MD, USA The Copper Ridge Institute, Sykesville, MD, USA.
4
Bloomberg School of Public Health, The Johns Hopkins University, Baltimore, MD, USA.

Abstract

Most approaches to the detection of presymptomatic or prodromal Alzheimer's disease require the costly collection and analysis of biological samples or neuroimaging measurements. The Dementia Risk Assessment (DRA) was developed to facilitate this detection by collecting self-report and proxy-report of dementia risk variables and episodic memory performance on a free Internet website. We now report two validation studies. In Study 1, 130 community-residing older adults seeking memory screening at senior health fairs were tested using the Mini-Cog, and were then observed while taking the DRA. They were compared to a demographically-matched subsample from our anonymous Internet sample. Participants seeking memory screening had more dementia risk factors and obtained lower scores on the DRA's recognition memory test (RMT) than their Internet controls. In addition, those who failed the Mini-Cog obtained much lower scores on the RMT than those who passed the Mini-Cog. In Study 2, 160 older adults seeking evaluation of cognitive difficulties took the DRA prior to diagnostic evaluations at outpatient dementia clinics. Patients who ultimately received the diagnosis of a dementia syndrome scored significantly lower on the RMT than those diagnosed with other conditions or deemed normal. Lower education, family history of dementia, presence of hypercholesterolemia and diabetes, and memory test score distinguished the dementia and no-dementia groups with around 82% accuracy. In addition, score on the RMT correlated highly with scores on other instruments widely used to detect cognitive decline. These findings support the concurrent validity of the DRA for detecting prevalent cognitive impairment. Prospective studies of cognitively normal persons who subsequently develop dementia will be necessary to establish its predictive validity.

KEYWORDS:

Memory testing; neuropsychology; risk factors; screening

PMID:
24705550
DOI:
10.3233/JAD-140297
[Indexed for MEDLINE]
PubMed Commons home

PubMed Commons

0 comments
How to join PubMed Commons

    Supplemental Content

    Full text links

    Icon for IOS Press
    Loading ...
    Support Center