Premium
[P3–256]: THE PREVALENCE, CAUSES AND CONSEQUENCES OF UNDIAGNOSED DEMENTIA IN ENGLAND: A RECORD LINKAGE STUDY OF THE COGNITIVE FUNCTION AND AGEING STUDY II
Author(s) -
Aldus Clare,
Brayne Carol,
Matthews Fiona,
Arthur Antony,
Dening Tom,
Fox Chris,
Robinson Louise,
Stephan Blossom C.M.,
Savva George M.
Publication year - 2017
Publication title -
alzheimer's and dementia
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 6.713
H-Index - 118
eISSN - 1552-5279
pISSN - 1552-5260
DOI - 10.1016/j.jalz.2017.06.1470
Subject(s) - dementia , medical diagnosis , medicine , medical record , record linkage , cognition , gerontology , population , cognitive decline , cohort , cohort study , psychiatry , family medicine , disease , environmental health , pathology , radiology
dementia as having “diagnosed dementia” if the GH EMR contained a dementia diagnosis or dementia-related medication dispensing prior to the date of dementia detection in ACT (the index date) and as having “undiagnosed dementia” if none of these elements were present. Each “undiagnosed dementia” case was matched with three randomly selected ACT participants who did not develop dementia. Characteristics of “undiagnosed” and “no dementia” patients were compared using age-adjusted conditional logistic regression. Results:44% of dementia patients were undiagnosed. Undiagnosed dementia patients were more likely than no dementia patients to have dementia risk factors and comorbidities including cerebrovascular disease (22% vs. 10%, p<0.001), diabetes (20% vs. 13%, p<0.001), congestive heart failure (28% vs. 13%, p1⁄40.004), depression (25% vs. 14%, p<0.001) and traumatic brain injury (8% vs. 3%, p1⁄40.003). In addition, they were more likely to have had at least one emergency department visit (48% vs. 25%, p<0.001), hospitalization (31% vs. 21%, p1⁄40.024), or visit for an injury (14% vs. 6%, p1⁄40.006), and to “no-show” for a scheduled ambulatory visit (43% vs. 29%, p<0.001) in the previous two years. Conclusions: The EMR includes many elements that could be used to identify patients likely to have undiagnosed dementia. Additional studies are needed to develop an optimal predictive model and ultimately to investigate whether identifying and screening these patients could improve clinical outcomes and quality of life.