Multiple Sclerosis Journal - October 2017 - 1494

Multiple Sclerosis Journal 23(11)
Table 4. Hazard ratios and Kaplan-Meier estimates of median time to death from birth and disease onset.
Total number
(number died)
All
15,952 (1335)
Sex
Female
11,214 (842)
Male
4738 (493)
Initial phenotype
Bout onset
13,667 (975)
Progressive
1386 (265)
onset

Median age
(95% CI), years

Hazard ratio
(95% CI)

80.5 (79.9−81.1)

Median time from
disease onset
(95% CI), years

Hazard ratio
(95% CI)

51.9 (50.5−53.6)

81.4 (80.5−82.1)
78.9 (77.7−79.9)

−
1.45 (1.29−1.63)

53.6 (51.5−55.5)
48.3 (46.6−50.4)

−
1.44 (1.28−1.62)

80.5 (79.9−81.4)
79.9 (77.6−81.1)

−
1.12 (0.98−1.29)

53.0 (51.5−55.3)
37.9 (35.8−44.0)

−
2.96 (2.58−3.41)

CI: confidence interval.

in the overall estimations between the two populations can be attributed to difference in the percentage
of PPMS patients (8% in Sweden vs 12% in Canada).
The discrepancy seen between our estimations and
the other previous reports can be explained by differences in the population characteristics (percentage of
progressive patients) and the studies' methodological
approaches such as inclusion and exclusion criteria,
definition of sustained EDSS score, sampling strategies and more importantly the nature of the cohorts
(population-based vs clinic-based).
The main strength of our study is the use of a nationwide cohort. In comparison to regional and clinicbased studies, our data are more likely to capture MS
patients with a wider disability range. However, it is
still possible that some very severe patients have died
or have not been captured by the register. This may
result in overestimation of time to disability milestones in our historic cohorts (as more benign patients
were likely to be alive at the time of data collection)
but is unlikely to influence the overall estimations.
Majority of MS patients in Sweden are of White
European descent; hence, our results are more generalisable to this group of patients.
Here, we showed one of the highest survivals reported
in MS; however, this was still significantly shorter
than the survival of Swedish general population
(SMR: 2.02). Our reported overall and sex-specific
SMRs are in line with the previously reported SMRs
from other countries.22
Although the majority of the patients in our sample
had been exposed to at least a first-line DMT, the stratified estimations of time from disease onset to EDSS
score 6.0 in our sample was almost comparable with

1494

the estimation obtained from the mainly untreated
patient population of British Columbia (31.7 vs 30.3 in
bout-onset patients).9 On average, our treated patients
gained extra 1.6 years to EDSS score 6.0 due to exposure to first-line DMTs. This might explain the 1.4
years difference between the two populations.
However, time to conversion to SPMS was not influenced by exposure to treatments and the average time
gained to EDSS score 6.0 was only marginally significant. Overall, the evaluation of long-term effectiveness of DMTs merits further investigations (which is
out of the scope of this paper) and caution must be
taken when drug effectiveness is being explored in
lack of randomisation.
Our research highlights the importance of populationbased research for more consistent and unbiased
description of MS clinical course. Our work also
highlights the significant social and economic burden
of MS to the society, even two decades into the DMT
era. It is clear that the DMTs were unsatisfactory in
preventing eventual disability, due to either lack of
effectiveness or unfavourable timing.
Overall, age at major disability milestones in MS in
this nationwide cohort was significantly older than
previous clinical- and regional-based reports. Societal
and economic consequences of MS is immense as MS
patients not only die at younger age but also live
almost 20 years with moderate and 30 years with
severe impairment and disability.
Acknowledgements
The authors would like to thank all neurologists,
nurses and multiple sclerosis patients in Sweden and
the Swedish multiple sclerosis register for providing
data for this study.

journals.sagepub.com/home/msj
https://journals.sagepub.com/home/msj
Table of Contents for the Digital Edition of Multiple Sclerosis Journal - October 2017
