A population-based epidemiologic study of emergency department visits for anaphylaxis in Florida

Abstract

Background: Previous population-based analyses of emergency department (ED) visits for anaphylaxis have been limited to small populations in limited geographic areas and focused on children or have included patients who had allergic conditions other than anaphylaxis.

Objective: We sought to describe the epidemiology and risk factors among patients with anaphylaxis presenting to Florida EDs.

Methods: Two thousand seven hundred fifty-one patients with anaphylaxis were identified for 2005-2006 within ED records by using the International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM), and a validated ICD-9-CM-based algorithm. Age- and sex-specific rates were calculated. Regression analyses were used to determine relative risks for anaphylaxis caused by various triggers (food, venom, and medication) and risk factors (age, sex, race, and ethnicity).

Results: The highest observed rates were among the youngest male subjects (8.2/100,000 Floridians aged 0-4 years) and among adult female subjects (15-54 years) grouped in 10-year age categories (9.9-10.9/100,000 Floridians). Male and black subjects were 20% and 25%, respectively, more likely to have a food trigger than female and white subjects. White, male, and older subjects were more likely to have an anaphylaxis-related ED visit caused by insect stings. Venom-induced anaphylaxis was more likely in August through October. Children were less likely than those older than 70 years (referent) to have medication-induced anaphylaxis (P < .03).

Conclusion: This is the only ED-based population study in a US lower-latitude state. The overall rate is considerably lower than other US ED-based population studies. The rates of anaphylaxis by age group differed by sex. Male and black subjects were more likely to have a food trigger.

Figure 1

Proportion of anaphylaxis ED visits by month and type of trigger for the years 2005-2006 (n=2751). Caption: Compared to the overall monthly distribution of ED visits, a statistically significant difference was found only for venom-induced anaphylaxis (p<0.001).