Background:
Atypical disease presentations, such as delirium, are associated with adverse health outcomes. They are also markers of frailty in elderly people, which is itself associated with adverse hospital outcomes. We investigated the relationship between frailty and atypical disease presentation in predicting adverse hospital outcomes and complications of the hospital course of elderly patients admitted to general medical services.
Methods:
We conducted a cohort study in a large (800 beds) tertiary care university hospital. The prevalence of atypical disease presentations and the incidence of adverse hospital outcomes (death, nursing home admission, prolonged hospital stay, and failure to regain premorbid functional status) were studied in previously well and previously frail elderly patients.
Results:
Patients were classified as being well or frail on the basis of the premorbid Barthel Index (well, score of ≥95 [n=76]; frail, score of <95 [n=117]). Frail elderly were older (80 vs 76 years), more often female (62% vs 46%), and less likely to be community dwelling (89% vs 99%). Atypical disease presentation was more common in the frail elderly (59% vs 25%; P<.001). Of those who presented atypically, the frail most often presented with delirium (61%) and the well presented with falls (37%) and delirium (32%). Of the frail elderly with atypical symptoms, 60% had adverse hospital outcomes compared with 32% of the well elderly who presented typically (P<.05). Logistic regression analysis showed that premorbid functional dependence (odds ratio, 2.48; 95% confidence interval, 1.17 to 5.22), atypical disease presentation (odds ratio, 2.37; 95% confidence interval, 1.20 to 4.67), and functional decline at admission (odds ratio, 5.64; 95% confidence interval, 2.37 to 13.44) were all independently predictive of poor hospital outcomes. By contrast, severity of disease, age, and sex did not confer an increased risk of adverse events.
Conclusions:
Premorbid functional dependency, atypical disease presentation, and functional decline on admission have independent impacts on adverse hospital outcomes. Assessment of each should be incorporated into the routine care of elderly patients.(Arch Intern Med. 1995;155:1060-1064)