Anaphylaxis in the community setting: determining risk factors for admission

Background: Although the identification and management of anaphylaxis in an emergency department setting has been well studied, our understanding of the risk factors for admission in a community-based hospital is lacking. Objective: To determine the demographics and the predictors of hospitalization, in patients presenting with anaphylaxis to a community-based emergency department (ED). Methods: We performed a five-year retrospective chart review of all patients seen in the ED of Winthrop University Hospital, a community-based institution, with an International Classification of Diseases, 9th Edition code related to anaphylaxis. Results: Fifty-eight visits met inclusion criteria, of which 34% resulted in hospital admission (95% CI: 22-48%). Univariate predictors for admission included (1) the involvement of 2, 3, and 4 organ systems (26%, 55%, and 75%, respectively; P < .02); (2) gastrointestinal symptoms vs no symptoms (59% vs 24%, P < .02); (3) non-sting (ingested and other allergens) vs insect sting allergen (50% vs 12.5%, P < .005); and (4) a history of an ED visit for anaphylaxis vs none (67% vs 30%, P < .05). Multivariate analysis (logistic regression) confirmed non-sting allergens (p < 0.02) and number of organ systems involved (P < .05) as independent predictors of hospitalization. Conclusion: In our study population, the involvement of multiple organ systems, particularly gastrointestinal involvement, a history of ED visits for anaphylaxis, and involvement of ingested or other allergens (non-sting) demonstrated higher admission rates. (C) 2012 Published by Elsevier Inc. on behalf of American Society for Histocompatibility and Immunogenetics.