This study aim to investigate differences in rates of discontinuation and reasons for discontinuation between MS-clinics.

We included all patients previously treated with DMF in Denmark from national drug release in January 2014 to May 2020. Data were collected from the Danish Multiple Sclerosis Register. Small clinics with low discontinuation number (n<50) of patients were excluded to avoid bias. A total of 2571 patients were included. Descriptive statistics and hazard ratios (HR) were calculated using chi squared test and cox proportional hazards model. HR were calculated comparing sites with the lowest and highest rates of discontinuation.

Out of 2571 patients treated with DMF, 24.9 % discontinues due to adverse events, 12.0% due to disease activity or progression, 5.0% due to pregnancy or planned pregnancy and 6.1% due to other reasons.

Hazard ratios varied from 1.09-1.94 regarding differences in risk of discontinuation due to all reasons (significant in 4 sites, p<0.05), HR ratios varied from 1.20-2.16 due to adverse events (AE) (significant in 5 sites p<0.05) and disease activity or progression 1.01-2.02 (1 sites p<0.05). The distribution and frequency of AEs leading to discontinuation varied significantly from site to site with regards to flushing and lymphopenia. No difference was found in rates of discontinuation due to gastrointestinal AEs.

Our results demonstrate significant differences between MS clinics and rates of discontinuation and reason for discontinuation. Our results could possibly be due to differences in clinical management. Further studies designed to investigate impact from clinical practice are warranted.