Importance: Biphasic anaphylaxis may occur in up to 20% of patients with anaphylaxis; however, the optimal observation time of patients with resolved anaphylaxis is unknown.
Objective: To characterize the cost-effectiveness of short vs prolonged medical observation times after resolved anaphylaxis.
Design, setting, and participants: An economic evaluation was performed of computer-simulated adult patients observed in outpatient allergy clinics and emergency departments, with rates of biphasic anaphylaxis derived from a 2019 meta-analysis.
Exposures: Computer-simulated patients (10 000 per strategy) were randomized to undergo 1 hour of medical observation (associated with 95% negative predictive value of biphasic anaphylaxis) or 6 or more hours of observation (associated with a 97.3% negative predictive value of biphasic anaphylaxis).
Main outcomes and measures: Cost-effectiveness of 6- to 24-hour medical observation of resolved anaphylaxis evaluated at willingness-to-pay thresholds of $10 000 per case of biphasic anaphylaxis observed and $10 million per death prevented, assuming that observation is associated with a 10- to 1000-fold reduction in the risk of death due to biphasic anaphylaxis.
Results: Biphasic anaphylaxis occurred after hospital discharge in 365 patients observed for 1 hour and in 213 patients undergoing prolonged observation. From a health care sector perspective, with medical observation costs of $286.92 per hour, the incremental cost of extended medical observation of resolved anaphylaxis (1 hour vs 6 hours) was $62 374 per case of biphasic anaphylaxis identified ($68 411 from the societal perspective). In Monte Carlo simulations, with hourly costs ranging from $100 to $500 and extended observation ranging from 6 to 24 hours (health care sector perspective), the mean (SD) costs were $295.36 ($81.22) for 1 hour of observation vs $3540.42 ($1626.67) for extended observation. The incremental cost-effectiveness ratio was $213 439 per biphasic anaphylaxis observed ($230 202 from the societal perspective). A 6-hour observation could be cost-effective if the risk of biphasic anaphylaxis after 1-hour observation of resolved anaphylaxis was 17% or if hourly observation costs were less than $46 in the base case. Cost-effectiveness could also be achieved (willingness-to-pay of $10 million per death prevented, health care sector perspective) when a baseline fatality rate of 0.33% per biphasic anaphylactic event was assumed, with a no greater than 24% relative risk of fatality associated with 6-hour observation.
Conclusions and relevance: This study indicates that prolonged medical observation (6-24 hours) for resolved anaphylaxis may not be cost-effective for patients at low risk for biphasic anaphylaxis; however, in particular clinical circumstances of low observation costs, high postdischarge risk of biphasic anaphylaxis, or large incremental fatality risk reduction associated with extended observation, longer medical observation can be justified.