Background: In the absence of a clear clinical history of reaction, diagnosis of cashew allergy using skin prick tests (SPT) or cashew-specific IgE requires a high number of oral food challenges (OFC). We recently showed that Ana o 3 sIgE alone, or a two-step diagnostic algorithm using cashew sIgE followed by Ana o 3 sIgE can reduce need for OFC. We aimed to determine if either of these approaches can provide a cost reduction to the health system compared to cashew SPT alone. Methods: Pooled individual level data from 6 studies was used to determine diagnostic accuracy and OFC rate. Two studies used cashew SPT (n=567, 198 allergic), with 95% positive and negative predictive values of ≥12mm and <3mm. Four studies were included in the pathways for Ana o 3 sIgE alone or a 2-step algorithm incorporating cashew and Ana o 3 sIgE (n=271, 156 allergic). Cut-offs used were ≥8.5kUA/L and ≤0.1kUA/L for cashew sIgE and ≥0.35kUA/L and ≤0.1kUA/L for Ana o 3 sIgE. Costs were constructed based on unit prices from hospital inpatient admissions, expenses incurred by families, individual patient data on allergic reaction types and rates and adrenaline autoinjector carriage, applying a health system perspective. Results: Modelled data through the Ana o 3 pathway resulted in a 46.43% cost reduction (\euro307,406/1000 patients) compared to using cashew SPT alone (\euro573,854/1000 patients). The 2-step algorithm resulted in a 44.94% cost reduction compared to SPT alone (\euro315,952.82/1000 patients). Both the Ana o 3 pathway and 2-step algorithm resulted in a 79-80% reduction in OFCs compared to SPT. Conclusions: Using Ana o 3 as a standalone test for cashew allergy diagnosis or a 2-step algorithm incorporating cashew sIgE and Ana o 3 sIgE is accurate and results in a large reduction in both OFCs and health system costs compared to cashew SPT alone.

Introduction: Children with peanut allergy are at increased risk of developing tree nut allergies, which can be severe and for most lifelong. Introduction of peanut in the first year of life can reduce the risk of peanut allergy, however, prevention strategies for tree nut allergies have not been established. We aimed to test the efficacy and safety of a novel strategy, a supervised multi-nut oral food challenge (OFC) compared to standard care for tree nut allergy prevention in infants at high risk of developing tree nut allergy, TreEAT. Methods and analysis: TreEAT is a 2-armed, open-label, randomised, controlled trial (RCT). Infants (n=212) aged 4-11months with peanut allergy will be randomised 1:1 at peanut allergy diagnosis to either a hospital-based multi-tree nut (almond, cashew, hazelnut and walnut) OFC using multi-nut butter or standard care (home introduction of individual tree nuts). All infants will be assessed at age 18months, with questionnaires and SPT to peanut and tree nuts. Peanut and tree nut OFCs will be performed as required to determine allergy status for each nut. The primary outcome is tree nut allergy at age 18 months. Secondary outcomes include peanut allergy resolution, proportion and severity of adverse events related to tree nut ingestion, number and frequency of tree nuts ingested, quality of life and parental anxiety and allergy related healthcare visits from randomisation to 18 months of age. Analyses will be performed on an intention-to-treat basis. Ethics and dissemination TreEAT was approved by the Royal Children’s Hospital Human Research Ethics Committee (#70489). Outcomes will be presented at scientific conferences and disseminated through publication. Trial registration number: ClinicalTrials.gov ID: NCT04801823

Background: In the absence of a clear clinical history of reaction, diagnosis of cashew allergy using skin prick tests (SPT) or cashew-specific IgE requires a high number of oral food challenges (OFC). By using Ana o 3 sIgE alone, or a two-step diagnostic algorithm using cashew sIgE followed by Ana o 3 sIgE, there is a reduced need for OFC. We aimed to perform a cost comparison for both of these approaches compared to cashew SPT alone. Methods: Pooled individual level data from 6 studies was used to determine diagnostic accuracy and OFC rate. Two studies used cashew SPT (n=567, 198 allergic), with 95% positive and negative predictive values of ≥12mm and <3mm. Four studies were included in the pathways for Ana o 3 sIgE alone or a 2-step algorithm incorporating cashew and Ana o 3 sIgE (n=271, 156 allergic). Cut-offs used were ≥8.5kUA/L and ≤0.1kUA/L for cashew sIgE and ≥0.35kUA/L and ≤0.1kUA/L for Ana o 3 sIgE. Costs were constructed based on unit prices from hospital inpatient admissions, expenses incurred by families, individual patient data on allergic reaction types and rates and adrenaline autoinjector carriage, applying a health system perspective. Results: Modelled data through the Ana o 3 pathway resulted in a 46.43% cost reduction (\euro307,406/1000 patients) compared to using cashew SPT alone (\euro573,854/1000 patients). The 2-step algorithm resulted in a 44.94% cost reduction compared to SPT alone (\euro315,952.82/1000 patients). Both the Ana o 3 pathway and 2-step algorithm resulted in a 79-80% reduction in OFCs compared to SPT. Conclusions: Using Ana o 3 as a standalone test for cashew allergy diagnosis or a 2-step algorithm incorporating cashew sIgE and Ana o 3 sIgE is accurate and results in a large reduction in both OFCs and health system costs compared to cashew SPT alone.

INTRODUCTION Home-based treatment of low-risk febrile neutropenia (FN) in children with cancer with oral or intravenous antibiotics is safe and effective. There are limited data on the economic impact of this model of care. We evaluated the cost-effectiveness of implementing a low-risk FN program, incorporating home-based intravenous antibiotics, in a tertiary pediatric hospital. METHODS A decision analytic model was constructed to compare costs and outcomes of the low-risk FN program, with usual in-hospital treatment with intravenous antibiotics. The program included a clinical decision rule to identify patients at low-risk for severe infection and home-based eligibility criteria using disease, chemotherapy and patient-level factors. Health outcomes (quality-of-life) and probabilities of FN risk classification and home-based eligibility were based on prospectively collected data. Patient-level costs were extracted from hospital records. Cost-effectiveness was expressed as the incremental cost per quality-adjusted life year (QALY). FINDINGS The mean healthcare cost of home-based FN treatment in low-risk patients was A$7,765 per patient compared to A$20,396 for in-hospital treatment (mean difference A$12,632 (95% CI,12,496-12,767)). Overall, the low-risk FN program was the dominant strategy, being more effective (0.0011 QALY (95% CI,0.0011-0.0012)) and less costly. Results of the model were most sensitive to proportion of children eligible for home-based care program. CONCLUSION Compared to in-hospital FN care, the low-risk FN program is cost-effective, with savings arising from cheaper cost of caring for children at home. These savings could increase as more patients eligible for home-based care are included in the program.