A retrospective claims analysis of fatigue in patients with multiple sclerosis on disease-modifying therapy

Thomas P Leist,Michele Cole,Sumit Verma,Alex Keenan,Hoa H Le

DOI: https://doi.org/10.1016/j.msard.2023.104917

Abstract:Background: Fatigue, one of the most common symptoms in patients with multiple sclerosis (MS), severely impairs quality of life and the ability to work or perform activities of daily living. Real-world data on fatigue in MS can help inform healthcare decisions and identify care gaps. We identified fatigue in patients with MS, using existing codes for fatigue and proxies of fatigue in healthcare claims database records and characterized cohorts with and without markers of fatigue who had been prescribed disease-modifying therapies for MS (MS-DMTs). Methods: In this cohort study, we retrospectively analyzed Optum's de-identified Clinformatics® Data Mart database from 1 January 2015 to 31 December 2019. The index date was defined as the first prescription record date for any MS-DMT during the study identification period. Included patient records were from adults (≥18 years) with ≥2 MS diagnosis claims listed within 12 months prior to the index date. Patients had ≥1 claim for any MS-DMT during the identification period (1 January 2016-31 December 2018), continuous enrollment in a health plan with medical and pharmacy benefits for 12 months before the index date (assessment one), and 12 months following the index date or to end of data availability (assessment two). After exploratory analyses, we applied the following definition to sort patient records into two cohorts according to presence or absence of markers of fatigue: ≥1 diagnosis (International Classification of Diseases, Ninth/Tenth Revisions code) claim for fatigue or ≥2 claims for stimulant drugs or ≥2 procedure claims for a sleep study or ≥2 pharmacy claims for sleep aid drugs; we used the broadest definition of fatigue so meeting any of these criteria qualified patients with MS as having fatigue. To minimize assessment one differences in selected patient characteristics between cohorts, we applied 1:1 propensity score matching with age, sex, US geographic region, and Charlson Comorbidity Index score as covariates. We analyzed demographic data, markers of fatigue, comorbidities at assessment one, and physical disabilities and neurologic impairment at assessment two. Results: Of 4077 patient records that met the eligibility criteria, 1976 had markers of fatigue. The propensity score-matched cohorts included 1519 patients each with and without fatigue. Assessment one comorbidities including anxiety (25.3% vs 10.5%; P<0.0001), arthritis (17.6% vs 12.9%; P = 0.0003), depression (15.0% vs 3.5%; P<0.0001), and gastrointestinal disorders (20.3% vs 14.2%; P<0.0001) were significantly more prevalent in the cohort with markers of fatigue at assessment one compared with those without fatigue. At assessment two, the cohort with baseline fatigue upon initial assessment was more likely to have indication of physical impairments (spasticity [63.5% vs 35.8%; P<0.0001], bladder dysfunction [37.8% vs 24.0%; P<0.0001], cognitive/behavioral dysfunction [27.0% vs 18.6%; P<0.0001]), neurologic impairments (pain [59.1% vs 44.0%; P<0.0001], depression [29.2% vs 9.9%; P<0.0001], and sensory disturbances [54.2% vs 36.7%; P<0.0001]), compared with the cohort without markers of fatigue at assessment one. Conclusions: In our analysis, patients with MS and fatigue were more likely to have comorbidities at assessment one and to develop physical disabilities and neurologic impairments at assessment two. Appropriate identification of patients with MS and fatigue may facilitate targeted care interventions to a group of patients at higher risk for disease progression and disability.