Methods: There were 18468 ED visits involving individuals 65+. Factor analysis was used to identify clusters of ED diagnoses which factored into medical diagnoses and symptoms. Discriminant analysis was used to evaluate whether diagnoses and/or symptoms predicted group membership.
Results: Demographically, significant differences between the groups included: more diagnosed conditions, longer length of hospital stay, and more African American or individuals with Hispanic ethnicity in the clustered repeater group compared to the other two groups. Groups were similar with regard to medical diagnoses. Discriminant analysis showed that older adults with syncope/dizziness, vomiting or an accident/fall were more likely to be in the single user group; whereas persons with generalized symptoms such as abdominal pain, shortness of breath, malaise were more likely to be in the clustered repeater group.
Conclusion: Different symptoms predicted group membership, suggesting that geriatric syndromes may explain some of the differences in patterns of use. Symptom management is a major nursing role and attention to the fit of the elder’s needs with the ED model of care may be important in reducing particular visit clusters, reducing costs and improving the outcomes of care for older ED users.