Gustaf Wellhagen1, Dr Sebastian Ueckert1, Professor Elisabet Nielsen1, Carl Smith1, Xinyi Li1, Associate Professor Joachim Burman2, Professor Mats O. Karlsson1
1Department of Pharmacy, Uppsala University, 2Department of Medical Sciences, Uppsala University
Introduction Multiple sclerosis (MS) is a chronic, progressive, disease which impacts the ability to work. It is a leading cause of disability among young adults (1). Losing the ability to work has high personal and societal costs. On the positive side, new therapies have evolved in the past decades (2), and several disease-modifying treatments are available (3–5). As some of these are relatively expensive biological agents there is a need to evaluate their cost-effectiveness, which is a complicated task that requires real-world data. Combined registry data can provide insights about the relation between disease severity and days on sickness-related benefits. The most common way to assess MS disease severity is the Expanded Disability Status Scale (EDSS), which is physician-reported and ranges from 0 (normal neurological exam) to 10 (death due to MS) in steps of 0.5, but patient-reported outcomes like the Multiple Sclerosis Impact Scale (MSIS-29) or Fatigue Scale for Motor and Cognitive Functions (FSMC) are sometimes also used. The MSIS-29 consists of 29 statements about the disease impact on daily life, all rated from 1 (not at all) to 5 (extremely), which can also be split into two subscales: physical (phy) and psychological (psy). The FSMC similarly has 20 questions about cognitive (cog) and motor (mot) fatigue, rated from 1 (does not apply at all) to 5 (applies completely), and can also be split in two subscales. Objectives To investigate the link between disease severity, as measured by different instruments, and number of days on sickness-related benefits in Swedish registry data. Methods EDSS, MSIS-29 and FSMC, patient and clinical characteristics were collected from the Swedish MS registry (SMSreg), pertaining to individuals (18-65 years old) with at least one record in the registry in the 5-year period starting from the beginning of 2014 to the end of 2018. Benefits for these individuals were retrieved from the Swedish Social Insurance Agency. For each individual and day, they were assigned one of 5 states: 0%, 25%, 50%, 75% or 100% sickness benefit level related to MS. An item response theory (IRT) model was developed for the SMSreg data, linking the EDSS, MSIS-29 and FSMC to the same disease severity construct via five correlated latent variables (LV). Then, a Markov model for the MS sickness benefit levels was constructed. The model assumed that each day, a subject belonged to one of the states. The next day, any state was possible. The estimated sickness propensity decided the transition probabilities, where age and the IRT LVs were used as covariates. Simulations to predict the number of sickness benefit days for scenarios across age and score ranges were performed from a model including all LVs. Results The final IRT model had 5 correlated LVs, describing EDSS, MSIS-29 psy, MSIS-29 psy, FSMC mot and FSMC cog, respectively. In addition to the subscales within MSIS-29 and FSMC, high correlations (>0.7) were found between EDSS and MSIS-29 phy and FSMC mot and MSIS-29 phy. The lowest correlations among LVs (=0.3) were between EDSS and MSIS-29 psy and EDSS and FSMC cog. Simulations revealed a strong trend with EDSS, where the mild (score 0-2.5), moderate (3-5.5) and severe (6-9.5) groups were expected to have 21, 127 and 262 days of sickness benefit, respectively (Table 1). When stratifying on EDSS group, age group and subscale score groups, other trends became apparent: notably a high MSIS-29 phy was predictive of a higher number of expected days with sickness benefit when EDSS was low. Table 1. Expected number of days with MS-related sickness benefits Stratification | Days FSMC cog 26-50: 88 FSMC cog 10-25: 26 FSMC mot 26-50: 95 FSMC mot 10-25: 16 MSIS-29 psy 2.55-5: 106 MSIS-29 psy 1-2.5:43 MSIS-29 phy 2.55-5: 172 MSIS-29 phy 1-2.55: 32 EDSS 6-9.5: 262 EDSS 3-5.5: 127 EDSS 0-2.5: 21 Age 46-62: 144 Age 31-45: 43 Age 18-30: 9 Conclusions The number of expected days on sickness benefit varied considerably with disease severity. EDSS, which is physician-reported, was the strongest predictor of the number of MS-related sickness benefit days. The self-reported MSIS-29 phy particularly had an impact when EDSS was low. This work is an important step towards quantifying the societal and individual cost for different MS treatments, as it shows who is at risk of work disability.
1. Mark J. Tullman MD. Overview of the Epidemiology, Diagnosis, and Disease Progression Associated With Multiple Sclerosis. 2013 Feb 25 [cited 2023 Nov 22];19. Available from: https://www.ajmc.com/view/ace008_13feb_ms_tullmans15tos20 2. Rieckmann P, Boyko A, Centonze D, Coles A, Elovaara I, Havrdová E, et al. Future MS care: a consensus statement of the MS in the 21st Century Steering Group. J Neurol. 2013 Feb;260(2):462–9. 3. Wiendl H, Gold R, Berger T, Derfuss T, Linker R, Mäurer M, et al. Multiple Sclerosis Therapy Consensus Group (MSTCG): position statement on disease-modifying therapies for multiple sclerosis (white paper). Ther Adv Neurol Disord. 2021;14:17562864211039648. 4. Faissner S, Gold R. Efficacy and Safety of Multiple Sclerosis Drugs Approved Since 2018 and Future Developments. CNS Drugs. 2022 Aug;36(8):803–17. 5. Radick L, Mehr SR. The Latest Innovations in the Drug Pipeline for Multiple Sclerosis. Am Health Drug Benefits. 2015 Nov;8(8):448–53.
Reference: PAGE 33 (2025) Abstr 11746 [www.page-meeting.org/?abstract=11746]
Poster: Drug/Disease Modelling - Other Topics