INTRODUCTION: Mild traumatic brain injury (mTBI) is a frequent cause of emergency department (ED) visits in France, with a substantial impact on healthcare utilization and ED costs. Blood-based, brain biomarker tests such as S100B and a combination of glial fibrillary acidic protein (GFAP) and ubiquitin C-terminal hydrolase-L1 (UCH-L1) have been shown to accurately identify mTBI patients who do not have intracranial lesions (ICL), potentially reducing the need for computed tomography (CT) scanning.
OBJECTIVES: To assess the cost-effectiveness of the GFAP + UCH-L1 compared with S100B for screening adults with suspected mTBI presenting to an ED.
METHODS: We developed a decision tree model to estimate the costs and health outcomes of using two biomarker tests in patients presenting to the ED with suspected mTBI (Glasgow Coma Scale of 13–15). Model parameters were extracted from peer-reviewed papers, clinical guidelines, and expert opinion. The diagnostic test accuracy measures were extracted from the pivotal trial (GFAP + UCH-L1) and a meta-analysis (S100B). Lifelong costs (2020 €) and health outcomes were calculated from a French healthcare system perspective and discounted over time at 2.5%. Cost for the GFAP + UCH-L1 test was assumed to be the same as the S100B test. In this model, patients with a positive biomarker receive a CT scan to confirm the presence of ICLs. Depending on the biomarker test outcome and the CT results (where relevant), patients may be discharged immediately from the ED, kept for <24-hour observation, admitted to a short stay general ward, or admitted to a neurosurgery ward. Patients with ICLs not identified by the biomarker (false negatives) could suffer from delayed treatment and poorer health outcomes. Clinical outcomes, including Glasgow Outcome Score and radiation-induced cancer were used to calculate quality-adjusted life years (QALYs). One-way and probabilistic sensitivity analyses were conducted to manage uncertainty.
RESULTS: In a hypothetical cohort of 1,000 patients, the GFAP + UCHL1 biomarker was associated with a decrease of 48 CT scans compared to S100B (769 versus 817, respectively). In the same cohort, the GFAP + UCH-L1 biomarker resulted in €4,785 overall savings and a minor improvement of 0.08 QALYs. In probabilistic sensitivity analysis, at the willingness to pay threshold of €50,000 per QALY, there is 63% probprobability of the GFAP + UCH-L1 test to be cost-saving and a 73% probability to be cost-effective when compared to S100B.
CONCLUSION: The findings of this cost-effectiveness model show that for patients with suspected mTBI who present to an ED, compared to S100B, screening with the GFAP + UCH-L1 test is associated with cost savings and a reduction in CT scans, while resulting in similar health outcomes.
