Healthcare Resource Utilization and Costs Among Patients with Essential Tremor

Abstract

Background: Essential tremor (ET) is a chronic neurologic movement disorder primarily characterized by anaction tremor, most commonly involving the upper extremities. ET can interfere with activities of daily living, social functioning, work productivity, and health-related quality of life. Treatment options remain limited, with propranolol and primidone generally considered first-line pharmacologic therapies, though many patients do not respond to or tolerate available medications. Prior literature suggests that ET is associated with substantial clinical and economic burden; however, additional real-world evidence is needed to characterize healthcare resource utilization (HCRU) and costs attributable to ET. Objective: To describe HCRU and quantify healthcare costs among commercially insured, Medicare and Medicaid beneficiaries with ET and compare against a cohort made up of the general population using U.S. administrative claims data. Methods: We conducted a retrospective matched cohort study using Merative MarketScan databases fromOctober 15th, 2015, through December 31st, 2024. Adults with ET were identified using ICD-10-CM diagnosis code G25.0 during the identification period. The index date was defined as the first qualifying ET diagnosis date. General population controls were sampled from the same payer databases and were assigned a random index date from a range based on the earliest and latest index dates of the ET cohort. Patients were required to have 12 months of continuous enrollment before index and were followed for 24 months after index. Patients with ET were matched on propensity scores using these variables: baseline demographics, payer type, comorbidities, pre index all-cause healthcare costs, and all-cause HCRU. All-cause costs and HCRU were examined during the 24 months following index date, including inpatient, outpatient, emergency department, and pharmacy costs. Between group mean differences were assessed using Welch Two Sample t-tests. Costs were also evaluated using a two-part model, with a logistic regression to estimate the probability of incurring a cost, and a generalized linear model with gamma distribution and log link to estimate costs among patients with positive costs. Marginal mean differences were calculated from the two-part models, with nonparametric bootstrapping to construct 95% confidence intervals. Post-index Healthcare utilization across the different categories were evaluated using a negative binomial model and results were reported as incidence rate ratios (IRR) between ET patients and matched controls. Both regression analyses were adjusted for age, sex, insurance type, index year, and CCI. Results: The final matched cohort included 72,343 ET patients and 72,343 matched controls from the generalpopulation. Baseline characteristics were all balanced after matching, with standardized mean differences below the prespecified threshold of 0.05. Results showed significantly higher total costs between the essential tremor and matched general population cohorts with a mean difference of $5,747.22 (95% CI: 4,610 – 6,875, p<0.001). A two-part model with additional regression components also showed significantly higher total costs between the essential tremor and general population cohorts with a mean difference of $9,927.92 (95% CI: 8797.56 – 11,019.98, p<0.001). Due to the difference between these two results, additional models were run as a sensitivity analysis and showed a range of mean differences from $5,995.32 to $9,373. All models were directionally consistent and showed ET being associated with higher total costs compared to the general population. Costs were primarily driven by outpatient services, and secondarily by pharmacy costs. During the post-index follow-up period, ET patients were found to have higher (incidence rate ratio [IRR] = 1.32, p < 0.001) overall utilization compared with matched controls. Rates of inpatient encounters (IRR = 1.42, p < 0.001) and emergency department encounters (IRR = 1.47, p = < 0.001) were also both found to be significantly higher among ET patients. Outpatient encounter rates were also found to be higher (IRR = 1.31, p = < 0.001) among ET patients. Discussion: Adults with ET experienced greater HCRU and costs compared with the general population, asshown by increased healthcare costs and higher rates of total encounters in every setting. These findings highlight the real-world economic burden associated with ET and call for the need for future evaluations of disease management strategies and emerging therapies.