Objective: To determine the effects of current age and disease duration on excess mortality in multiple sclerosis, we described the dynamics of excess deaths rates over these two time scales and studied the impact of age at multiple sclerosis clinical onset on these dynamics, separately in each initial phenotype.

Methods: We used data from 18 French multiple sclerosis expert centers participating in the Observatoire Français de la Sclérose en Plaques. Patients with multiple sclerosis living in metropolitan France and having a clinical onset between 1960 and 2014 were included. Vital status was updated on January 1st, 2016. For each multiple sclerosis phenotype separately (relapsing onset (R-MS) or primary progressive (PPMS)), we used an innovative statistical method to model the logarithm of excess death rates by a multidimensional penalized spline of age and disease duration.

Results: Among 37524 patients (71% women, mean age at multiple sclerosis onset ± standard deviation 33.0 ± 10.6 years), 2883 (7.7%) deaths were observed and 7.8% of patients were lost-to-follow-up. For R-MS patients, there was no excess mortality during the first 10 years after disease onset; afterwards, whatever age at onset, excess death rates increased with current age. From current age 70, the excess death rates values converged and became identical whatever the age at disease onset, which means that disease duration had no more impact. Excess death rates were higher in men with an excess hazard ratio of 1.46 (95% confidence interval 1.25-1.70). In contrast, in PPMS patients, excess death rates rapidly increased from disease onset, and were associated with age at onset, but not with sex.

Conclusions: In R-MS, current age has a stronger impact on multiple sclerosis mortality than disease duration while their respective effects are not so clear in PPMS. 

It deals with the observed and expected survival by the methods of Kaplan-Meier and Ederer 1 in multiple sclerosis patients.