Healthcare resource use for pediatric low-grade glioma care: A cohort of linked electronic health records and claims data.

Abstract

10038 Background: Pediatric low-grade glioma (pLGG) is the most common brain tumor in children and is associated with persistent long-term tumor- and treatment-related morbidities. This study aims to characterize the costs associated with the treatment of pLGG. Methods: A retrospective study using the Optum de-identified Market Clarity Dataset linked claims (commercial, Medicare Advantage and Managed Medicaid beneficiaries) and electronic health records (EHRs) of cases ≤18 years of age, with an ICD-10 code for brain neoplasm and ≥1 physician notes between January 2017 and June 2018 was performed. Index date was the first claim or EHR with an ICD-10 code for brain neoplasm. Natural language processing was used to identify pLGG-relevant data from physician notes. The observation period included 3 months prior to index date and 6-month segments from index date for 36 months. Cases had either continuous insurance coverage or continuous EHR activity during this period. Results for procedures and medication use were reported as averages throughout the 36-month post-index follow-up period. Results: A total of 154 patients with pLGG were identified. Median age was 11 years (range 2–18). Among the cohort, 49% (n=76) were female; 75% (n=116) were non-Hispanic white, 13% (n=20) Hispanic, 5% (n=7) African American, 1% (n=1) Asian and 6% (n=10) other/unknown. Of benefits, 56% had commercial and 44% Medicaid. Brain imaging and neuropsychological evaluations were performed at least once in 85% and 57% of the cohort, respectively. Among pLGG relevant medications, 57% of the cohort used analgesics, 45% anti-emetics, 32% anxiolytics, and 29% anticonvulsants. A total of 100 patients received ≥1 pLGG disease control treatments: Of these, 53% underwent emergency brain surgery, 24% had planned brain surgery, 14% received chemotherapy, and 6% received targeted therapy. Of the 100 patients who received disease control treatments, 57% relapsed after first-line treatment; 37% of patients received 2 lines of therapy, and 20% of patients received ≥3 lines of therapy. Conclusions: Our findings show that patients with pLGG often require long-term medical care and high levels of healthcare resources to treat their disease and its sequalae. Additionally, a majority will relapse and face a higher burden of disease. Further studies using integrated data sources are warranted to help us understand the burden of pLGG and inform evidence-based health care planning for these patients.