Impact of discontinuing disease-modifying therapies on health care utilization among midlife patients with multiple sclerosis in the United States.

backgroundMultiple sclerosis (MS) is a lifelong progressive neurological disease treated primarily with disease-modifying therapies (DMTs). Disease activity tends to decline as patients age. Midlife represents a crossroads where the risks of DMT may outweigh the benefits, prompting providers to consider DMT discontinuation to reduce treatment burden. However, real-world evidence on the impact of DMT discontinuation among midlife patients is lacking.

objectiveTo evaluate the association between DMT discontinuation and health care utilization among midlife patients with MS.

methodsMidlife patients with MS who received an injectable or oral DMT between 2001 and 2018 were identified from the MarketScan commercial claims database. DMT discontinuation, defined as a treatment gap exceeding 90 days in days supply, was the independent variable. Patients who discontinued DMTs had their index date set as the last gap day, whereas index dates for those who continued DMTs were matched based on the time distribution of index dates of discontinuers. Inpatient hospitalizations (all-cause, MS-related, and non-MS-related), emergency department (ED) visits (all-cause, MS-related, and non-MS-related), and relapse-related hospitalizations and outpatient visits were independently evaluated during the 365-day follow-up. Patients were observed until the occurrence of an event (depending on the model), deviation from the treatment group, disenrollment, death, end of follow-up, or data unavailability. Stabilized inverse probability of treatment weighting (sIPTW) was employed to balance the 2 groups. The associations between DMT discontinuation and each utilization outcome were estimated using Cox proportional hazard regression models with sIPTW.

resultsOf 149,721 midlife patients with MS, 22.8% discontinued DMTs and 77.2% continued DMTs. Patients who discontinued DMTs had a higher cumulative incidence for all utilization outcomes during the 365-day follow-up than those who continued DMTs. Cox regression showed that DMT discontinuation was associated with a 10.3% and 24.9% higher rate of all-cause and non-MS-related inpatient hospitalizations, respectively, with no significant association found for MS-related hospitalizations. Patients discontinuing DMTs exhibited higher utilization rates for ED visits, with an increase of 21.3% for all-cause, 23.0% for MS-related, and 20.9% for non-MS-related visits compared with those who continued DMTs. We also observed a 15.9% and 52.1% higher rate of relapse-related hospitalizations and outpatient visits associated with DMT discontinuation, respectively.

conclusionsThis study revealed that DMT discontinuation was associated with higher health care services utilization among midlife patients with MS, especially relapse-related outpatient visits. DMT discontinuation during midlife may be premature, and DMTs may still be necessary to reduce health care utilization.