Using a Computer Vision-Based System for Reading Skin Prick Test Results: A Cost-Minimisation Analysis

OBJECTIVES: Skin prick tests (SPTs) are the first diagnostic approach for people with a suspected allergy. Together with the clinical history, SPTs allow us to draw conclusions on allergies based on the sensitization pattern. This study investigates whether reading skin prick allergy results using a computer vision-based system – Nexkin – is cost-saving compared to standard current practice where SPTs are conducted and read manually.

METHODS: A cost-minimization analysis was conducted to evaluate the cost-cutting effect of reading skin prick test results using the computer vision-based system, Nexkin. The patient population consists of individuals who were referred to a the allergology department at one of the largest University hospitals in Switzerland, Inselspital Bern. Which averages 100 tests a week, i.e., 5000 tests annually. Our model parameters - test costs, sensitivity, specificity, prevalence - are sourced from literature and expert input by allergy clinicians.

RESULTS: From the perspective of a university hospital’s allergology department, the average expected value of costs saved by reading SPT results using computer vision-based Nexkin was approximately CHF 7 per test. Monte Carlo simulation which accounts for parameter value uncertainty also indicated that SPTs conducted using Nexkin saved approximately CHF 7 in majority of the simulations. Our sensitivity analyses demonstrate that parameters representing the costs associated with manual and semi-automated procedures have the largest influence on the incremental cost. However, a higher allergy prevalence favored standard manual SPTs.

CONCLUSIONS: While this study is based on early modelling and requires additional evidence for more robust conclusions, it demonstrates the cost-saving nature of computer aided SPTs. Against the backdrop of rising healthcare costs especially in Switzerland, using Nexkin could lead to considerable cost savings, thus contributing to cost containment efforts. Results should be taken with caution since our study is based on hay fever allergy prevalence in Switzerland.