Association of incident dementia with hospitalizations.
Level 3 - non-randomized controlled study
Non-randomized longitudinal cohort follow-up study
PubMed 22235087 · doi:10.1001/jama.2011.1964
What was done
Retrospective analysis of hospitalizations among 3019 participants aged 65 years or older enrolled in the Adult Changes in Thought (ACT) longitudinal cohort study in an integrated health care system from 1994 to 2007. All participants had no dementia at baseline and were screened biennially. Automated data tracked hospital admissions from enrollment until death, disenrollment, or study end. Admission rates for all causes, body systems, and ambulatory care-sensitive conditions (ACSCs) were compared between participants who developed dementia and those who remained dementia-free.
What was found
Of 3019 participants, 494 developed dementia (86% admitted at least once) and 2525 remained dementia-free (59% admitted at least once). Unadjusted all-cause admission rates were 419 per 1000 person-years with dementia versus 200 per 1000 person-years without dementia. After multivariable adjustment, the admission rate ratio for all-cause hospitalizations was 1.41 (95% CI, 1.23-1.61; P < .001) and for ACSCs was 1.78 (95% CI, 1.38-2.31; P < .001). Adjusted rates were significantly higher in the dementia group for most body systems and for all individual ACSCs evaluated, including bacterial pneumonia, congestive heart failure, dehydration, duodenal ulcer, and urinary tract infection.
Why it matters
Incident dementia substantially increases the risk of hospitalization, particularly for conditions that might be managed in outpatient settings. This indicates that proactive ambulatory care strategies could target potentially preventable acute admissions in patients with dementia.
Limits
Observational cohort design cannot establish causality. The study was conducted within a single integrated health care system, potentially limiting generalizability to populations with different care models. Details on outpatient care quality and specific triggers for acute deterioration were not reported in the abstract.
Cited by
- supports The Adult Changes in Thought study and Rush Memory and Aging Project showed that longitudinal cognitive decline occurs in step-like decreases following significant acute illness or hospitalization.