A health economic analysis of noninjectable epinephrine compared with intramuscular epinephrine

Noninjectable epinephrine to treat allergic reactions addresses an unmet need. Intranasal epinephrine is approved and a sublingual form is under development. Inhaled epinephrine is poorly studied for anaphylaxis. These forms have unknown cost-effectiveness.

To evaluate cost-effectiveness of commercially available noninjectable epinephrine compared with intramuscular epinephrine for treatment of anaphylaxis.

Markov cohort analyses evaluated the cost-effectiveness of noninjectable epinephrine forms. The base-case assumed exaggerated anaphylaxis fatality rates (50-fold increase) for using inhaled epinephrine given low certainty evidence in anaphylaxis and deliberately reduced fatality risk for nasal or sublingual forms (10-fold reduction) theorizing higher adherence and early use during an allergic reaction.

In the base-case scenario, assuming a 10-fold decreased risk in peanut allergy fatality associated with intranasal or sublingual epinephrine treatment for a severe allergic reaction (net monetary benefit [NMB], $2,189,134) vs intramuscular epinephrine use (NMB, $2,189,114), intranasal or sublingual epinephrine was the most cost-effective option (incremental cost-effectiveness ratio [ICER], $83,748/quality-adjusted life-year [QALY]), but only at a marginal annual cost of $4. Intramuscular epinephrine was cost-effective (ICER, $17,900/QALY) vs inhaled epinephrine (NMB, $2,183,531), although inhaled epinephrine reached cost-effectiveness (willingness to pay [$100,000/QALY]) if associated fatality risk fell below 2.5-fold. Substituting a single noninjectable form of epinephrine for a second injectable device (in patients prescribed 2 autoinjectors already) would be cost-effective; however, adding a supplemental noninjectable device was not cost-effective, even assuming a 10-fold risk reduction with multiple device carriage (ICER, $858,462).

Noninjectable routes of epinephrine can be cost-effective options provided fatality risk is not significantly elevated. Carriage of redundant epinephrine autoinjectors with noninjectable forms is not cost-effective if associated with excess cost of redundant device packs.