A QI Initiative: Assessing Predictive Value of Jewelry Sensitivity for Patch Testing Outcomes in a Medium Sized Hospital System

Abstract

RATIONALE: True prevalence and patient morbidity of allergic contact dermatitis (ACD) due to metals remains obscure in part due to vague practice parameters creating variability in history collection.1-5 The practice parameters only acknowledge the importance of a patient history for diagnosis with “moderate” strength, and give no definite guidelines for which historical indicators best signify positive patch test results.1 The Deaconess Clinical Research Institute aims to assess the prevalence of jewelry sensitivity and if such sensitivity impacts the days when tests turn positive. We hypothesize that by identifying if jewelry sensitivity correlates with test results, physicians can prioritize specific questions to optimize visit efficiency. OBJECTIVE: Identify the difference in prevalence of metal allergies between patients with and without history of jewelry sensitivity, if history of jewelry sensitivity changed what days patients turned positive, which metals were most common with this history and reveal if the “jewelry sensitivity” historical cue predicts positive results. METHODS: A blinded retrospective chart review of 157 patients in a single outpatient allergy clinic from July 2020 to July 2023 referred for the evaluation of ACD. RESULTS: There was no significant difference in days tests turned positive between patients with or without history of jewelry sensitivity (p=0.3383). 75.8% of allergic patients were jewelry sensitive, and 42.0% of patients with a jewelry sensitivity were positive. Patients with jewelry sensitivity were 25.5% more likely to test positive than those without jewelry sensitivity but were not significantly associated with a positive test result for all or specific metals (p>0.05). 85.1% of positive results came from nickel, palladium, and cobalt. CONCLUSION: History of jewelry sensitivity cannot reliably predict the days when tests turn positive or likelihood of positive results, although low sample size likely caused the latter observation.