Background: Fatal anaphylaxis is very rare, with an incidence ranging from 0.5 to 1 deaths per million person-years. Objective: Based on a systematic review, we aimed to explain differences in the reported incidence of fatal anaphylaxis based on the methodological and demographic factors addressed in the various studies. Methods: We searched PubMed/MEDLINE, EMBASE, and the Web of Science for relevant retrospective and prospective cohort studies and registry studies that had assessed the anaphylaxis death rate for the population of a country or for an administrative region. The research strategy was based on combining “anaphylaxis” with “death”, “study design”, and “main outcomes” (incidence). Results: A total of 46 studies met the study criteria and included 16,541 deaths. The range of the anaphylaxis mortality rate for all causes of anaphylaxis was 0.002-2.51 deaths per million person-years. Fatal anaphylaxis due to food (range 0.002-0.29) was rarer than deaths due to drugs (range 0.004-0.56) or Hymenoptera venom (range 0.02-0.61). The frequency of deaths due to anaphylaxis by drugs increased during the study period (IRR per year, 1.02, 95%CI 1.00-1.04). We detected considerable heterogeneity in almost all of the meta-analyses carried out. Conclusion: The incidence of fatal anaphylaxis is very low and differs according to the various subgroups analyzed. The studies were very heterogeneous. Fatal anaphylaxis due to food seems to be less common than fatal anaphylaxis due to drugs or Hymenoptera venom.

This article has been accepted for publication and undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process, which may lead to differences between this version and the Version of Record. Please cite this article as

<b><i>Introduction:</i></b> The drug provocation test (DPT) is the gold standard for the drug allergy workup; however, it is not free from severe adverse reactions. Our aim was to obtain robust data that predict a reaction during or after the DPT at the first contact with the patient in the allergy outpatient clinic. <b><i>Methods:</i></b> The population of this cross-sectional study comprised all patients undergoing a drug allergy workup (clinical assessment, specific IgE, or skin tests, or DPT) at University Hospital Fundacion Alcorcon in 2016. DPTs were performed until therapeutic doses were reached, and late reactions were checked. The clinical disorders assessed in our study were classified mainly as absence of allergic reactions, morbilliform rash, urticaria, anaphylaxis, and other cutaneous disorders. <b><i>Results:</i></b> Physicians from the Allergy Unit programmed drug allergy workups in 977 patients (median age, 52 years; women, 64.54%). DPTs were not performed for 165 drugs involved in the reactions. Patients who did not undergo DPT were older than patients who did (positive or negative) (<i>p</i> = 0.0001). Positive DPT results were detected in 6.00% of DPTs performed, and most were for amoxicillin and metamizole (15–25% each). Multinomial logistic regression showed that positive reactions were more probable after DPT if the same clinical disorder was diagnosed at the first visit, including the episodes not considered allergic episodes (OR = 0.2, &#x3c;0.01), except for anaphylaxis, which favored not performing DPTs (OR = 11, <i>p</i> &#x3c; 0.001). <b><i>Conclusion:</i></b> We conclude that clinical practice in the diagnosis of drug allergy in our Allergy Department is safe, without over-diagnosis.