Long-term Socioeconomic Outcomes Associated With Pediatric-Onset Multiple Sclerosis

Kyla A McKay; Emilie Friberg; Neda Razaz; Kristina Alexanderson; Jan Hillert

doi:10.1001/jamaneurol.2020.5520

. 2021 Feb 22;78(4):1–5. doi: 10.1001/jamaneurol.2020.5520

Long-term Socioeconomic Outcomes Associated With Pediatric-Onset Multiple Sclerosis

Kyla A McKay ^1,^2,^✉, Emilie Friberg ³, Neda Razaz ⁴, Kristina Alexanderson ³, Jan Hillert ¹

¹Neuro Division, Department of Clinical Neuroscience, Karolinska Institutet, Stockholm, Sweden

²Centre for Molecular Medicine, Karolinska University Hospital, Stockholm, Sweden

³Division of Insurance Medicine, Department of Clinical Neuroscience, Karolinska Institutet, Stockholm, Sweden.

⁴Clinical Epidemiology Division, Department of Medicine, Solna, Karolinska Institutet, Stockholm, Sweden.

Accepted for Publication: December 18, 2020.

Published Online: February 22, 2021. doi:10.1001/jamaneurol.2020.5520

^✉

Corresponding Author: Kyla A. McKay, PhD, Neuro Division, Department of Clinical Neuroscience, Karolinska Institutet, SE-171 77 Stockholm, Sweden (kyla.mckay@ki.se).

Author Contributions: Dr McKay had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: McKay, Friberg, Alexanderson, Hillert.

Acquisition, analysis, or interpretation of data: McKay, Friberg, Razaz, Alexanderson.

Drafting of the manuscript: McKay.

Critical revision of the manuscript for important intellectual content: Friberg, Razaz, Alexanderson, Hillert.

Statistical analysis: McKay, Razaz.

Obtained funding: McKay, Alexanderson, Hillert.

Administrative, technical, or material support: Alexanderson.

Supervision: Hillert.

Conflict of Interest Disclosures: Dr McKay reported receiving research support from the Canadian Institutes of Health Research and the Swedish Research Council for Health, Working Life and Welfare. Dr Friberg reported being partly funded by Biogen and receiving unrestricted researcher-initiated grants from Celgene. Dr Razaz reported being funded by the Swedish Research Council for Health, Working Life and Welfare. Dr Alexanderson reported receiving unrestricted research grants from Biogen and the Swedish Research Council. Dr Hillert reported receiving honoraria for serving on advisory boards for Biogen, Celgene, Merck KGaA, Novartis, Sandoz, and Sanofi-Genzyme and speakers fees from Biogen, Celgene, Novartis, Merck KGaA, Teva, and Sanofi-Genzyme; serving as principal investigator for projects or receiving unrestricted research support from Biogen, Celgene, Merck KGaA, Novartis, Roche, and Sanofi-Genzyme; and receiving funding from the Swedish Research Council and the Swedish Brain Foundation for his multiple sclerosis research. No other disclosures were reported.

Funding/Support: This research was funded by postdoctoral fellowship awards to Dr McKay from the Canadian Institutes of Health Research, European Committee for Treatment and Research in Multiple Sclerosis, and The Swedish Research Council for Health, Working Life and Welfare, as well as an unrestricted grant from Biogen to Drs Alexanderson and Hillert.

Role of the Funder/Sponsor: The funding sources had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

^✉

Corresponding author.

PMCID: PMC7900935 PMID: 33616605

Key Points

Question

Is there an association between pediatric-onset multiple sclerosis (MS) and educational level, income, and use of disability benefits through adulthood?

Findings

This nationwide register-based cohort study included 485 patients with pediatric-onset MS and 4850 persons without MS, matched from the general population. Persons with pediatric-onset MS were less likely to attend university, had less earnings from work, and were more reliant on disability benefits than persons without MS.

Meaning

This study suggests that pediatric-onset MS may have lasting consequences that translate into lower educational achievements and earnings and a greater use of disability benefits in adulthood.

Abstract

Importance

Pediatric-onset multiple sclerosis (PoMS) is associated with significant cognitive and physical disability. Whether this disability translates into differences in educational achievements and earnings is unknown.

Objective

To evaluate the association between PoMS and educational level and income throughout adulthood.

Design, Setting, and Participants

A prospective register-based cohort study of individuals with PoMS and a population-based matched reference cohort was conducted using nationwide microdata from linked registers in Sweden from January 1, 1990, to December 31, 2016; analyses were completed from May 1, 2019, to September 1, 2020. Of 772 persons with PoMS identified in the Swedish MS registry, 485 had an onset during the period from 1980 to 2014 and had socioeconomic data available. The general population reference cohort without multiple sclerosis (MS) (n = 4850) was randomly selected from the full Swedish population, matched 10:1 on age, sex, and country of birth.

Exposure

Pediatric-onset MS, diagnosed by a neurologist, with onset before 18 years of age.

Main Outcomes and Measures

Highest educational level (elementary school, high school, or university) was assessed using logistic regression. Income, measured as the mean annual earnings from paid work in US dollars, was compared using Tobit models, and net annual sickness absence and disability pension days were compared using zero-inflated negative binomial regression. Earnings and days receiving disability benefits were compared within 4 age periods (19-24, 25-34, 35-44, and 45-54 years).

Results

The median age of the cohort with PoMS (n = 485) and the matched reference cohort (n = 4850) in 2016 was 32 years (interquartile range, 26-40 years), and most participants were women (348 [71.8%] in the PoMS cohort and 3480 [71.8%] in the matched reference cohort). Persons with PoMS were less likely than persons in the matched reference cohort to attend university (odds ratio, 0.80 [95% CI, 0.66-0.97]) and had significantly lower annual earnings than the reference cohort, ranging from −$1618 (95% CI, −$2558 to −$678) in the youngest age period to −$10 683 (95% CI, −$18 187 to −$3178) in the eldest. Persons with PoMS received higher rates of disability benefits, as sickness absence days in the youngest age period (rate ratio, 3.06 [95% CI, 2.08-4.52]) and disability pension days in the oldest age period (rate ratio, 1.43 [95% CI, 1.11-1.85]).

Conclusions and Relevance

This study suggests that having PoMS is associated with less educational achievement, lower earnings, and greater use of disability benefits throughout the working-age life span. As adults, persons with PoMS never earned as much as their counterparts without MS, and they exhibited a heavier reliance on disability benefits.

This cohort study evaluates the association between pediatric-onset multiple sclerosis and education and income throughout adulthood.

Introduction

Between 2% and 10% of persons with multiple sclerosis (MS) experience their first neurologic symptom in childhood.¹ Patients with pediatric-onset MS (PoMS) may be more susceptible to inflammation early in disease² and impaired cognition³ relative to their counterparts with adult-onset MS. Individuals with PoMS may spend their entire adult lives with MS, potentially disrupting their ability to effectively study, work, and engage in society. Whether these factors translate into quantifiable differences in educational level or earnings is unknown. We evaluated the association between PoMS and educational level, earnings, and use of disability benefits.

Methods

Study Design and Data Sources

This Swedish nationwide register cohort study used linked individual-level data from January 1, 1990, to December 31, 2016, from the Swedish MS registry, the Total Population Registry, the Longitudinal Integration Database for Health Insurance and Labor Market studies (LISA), and the National Patient Register. Details of these registers are available in eTable 1 in the Supplement.

Study Population

Patients with PoMS included all persons with relapsing-remitting MS diagnosed by a neurologist with an onset at younger than 18 years, from 1980 to 2014. The general population–matched reference cohort was drawn from LISA and was individually matched 10:1 on sex, year of birth, country of birth (born in or outside of Sweden), and last year of available data in LISA. To ensure that the matched reference cohort did not include persons with MS, we excluded all persons with International Classification of Diseases, Ninth Revision and International Statistical Classification of Diseases and Related Health Problems, Tenth Revision codes for demyelinating disorders (eTable 1 in the Supplement).

Outcomes

The highest educational level was categorized as elementary school (≤9 years of school), high school (10-12 years), or university (>12 years). Individuals were linked to their parents to include the highest educational level of their parents as a covariate.

Total annual pretax earnings from paid work were recorded and converted to US dollars. Sickness absence (granted when work capacity is temporarily reduced owing to disease or injury) and disability pension (granted for long-term or permanent work incapacity due to disease or injury) are recorded as net days per year, annually. Details of the outcomes are available in the eAppendix in the Supplement. We explored each outcome separately within the following 4 age periods: 19 to 24, 25 to 34, 35 to 44, and 45 to 54 years.

Statistical Analysis

Analyses were performed from May 1, 2019, to September 1, 2020. Demographic and clinical characteristics were summarized using frequency (percentage), mean (standard deviation), or median (interquartile range), based on the distribution of the data. Logistic regression models explored the association between PoMS and educational level, categorized as high school or more vs elementary school and as university level vs high school or elementary school. Analyses were adjusted for parental educational level, with results presented as odds ratios with 95% CIs. All P values were from 2-sided tests, and results were deemed statistically significant at P < .05.

Tobit regression models were used to estimate the differences in earnings between individuals with PoMS and individuals in the matched reference cohort, adjusted for educational level and year, with results reported as β coefficients in US dollars with 95% CIs. Last, zero-inflated negative binomial regression models were used to compare the mean net number of sickness absence and disability pension days within each age period between individuals with PoMS and individuals in the matched reference cohort. Findings were reported as rate ratios (RRs) with 95% CIs (crude RRs and adjusted for educational level).

In supplementary analyses, we dichotomized earnings (zero earnings vs earnings per age period) and disability benefits (≥1 day vs 0 days per age period). Second, we measured earnings among all persons with nonzero earnings by linear regression. Third, all analyses were stratified by childhood-onset MS (aged <13 years) and adolescent-onset MS (aged 13-17 years) and by era (MS onset during the period from 1980 to 1999 or during the period from 2000 to 2014).

Results

Of 772 persons with PoMS identified in the Swedish MS registry, 485 were eligible and matched to 4850 persons from the general population. Demographic and clinical characteristics are available in Table 1. There was no difference between groups in high school attendance; however, individuals in the PoMS group were less likely to attend university (odds ratio, 0.80 [95% CI, 0.66-0.97]) (eTable 2 in the Supplement).

Table 1. Demographic Features of the Full Cohort of Patients and the Individuals in the Matched Reference Cohort.

Characteristic	Participants, No. (%)		P value
Characteristic	Pediatric-onset MS (n = 485)	Matched reference cohort (n = 4850)	P value
Women	348 (71.8)	3480 (71.8)	>.99^a
Age in 2016, median (IQR), y	32 (26-40)	32 (26-40)	>.99^b
Born in Sweden	422 (87.0)	4220 (87.0)	>.99^a
Highest parental educational level
Elementary school	84 (17.3)	967 (19.9)	.35^a
High school	223 (46.0)	2117 (43.7)
University	178 (36.7)	1766 (36.4)
Highest educational level
Elementary school	52 (10.7)	458 (9.4)	.09^a
High school	235 (48.5)	2166 (44.7)
University	198 (40.8)	2226 (45.9)
Age, median (IQR), y
At MS onset	16.2 (14.4-17.3)	NA	NA
At MS diagnosis	18.6 (16.9-22.9)	NA	NA
DMT use during the study period
No therapy	26 (5.4)	NA	NA
First-line therapy only	114 (23.5)
Second-line therapy only	80 (16.5)
Switched from first- to second-line therapy	265 (54.6)

Open in a new tab

Abbreviations: DMT, disease-modifying therapy; IQR, interquartile range; MS, multiple sclerosis; NA, not applicable.

^{^a}

By χ² test.

^{^b}

By Wilcoxon rank sum test.

In all 4 age periods, the individuals in the PoMS cohort had lower annual earnings than the individuals in the matched reference cohort, ranging from −$1618 (95% CI, −$2558 to −$678) among those aged 19 to 24 years to −$10 683 (95% CI, −$18 187 to −$3178) among those aged 45 to 54 years (Table 2; eTable 3 in the Supplement). Patients with PoMS were more likely to have zero earnings, 1 or more sickness absences, or 1 or more disability pension days in all 4 age periods (Figure; eTable 4 in the Supplement) and more net sickness absence days per year in the first 3 age periods (19-24 years: RR, 3.06 [95% CI, 2.08-4.52]; 25-34 years: RR, 3.16 [95% CI, 2.37-3.96]; and 35-44 years: RR, 1.77 [95% CI, 1.14-2.74]) (Table 2). Among those aged 19 to 24 years and those aged 25 to 34 years, the adjusted rate ratio of disability pension days was not significantly higher for the PoMS cohort compared with the matched reference cohort. However, in the last 2 age periods, those in the PoMS cohort had significantly higher rates (35-44 years: RR, 1.41 [95% CI, 1.17-1.71]; 45-54 years: RR, 1.43 [95% CI, 1.11-1.85]) (Table 2).

Table 2. β Coefficient of Earnings and Rate Ratios of Annual Net Sickness Absence and Disability Days Among Individuals With PoMS Compared With Individuals in the Matched Reference Cohort.

Age period, y	Adjusted (95% CI)^a
	β Coefficient of earnings, $	Rate ratio of net annual days
	β Coefficient of earnings, $	Sickness absence	Disability pension
19-24	−1618 (−2558 to −678)	3.06 (2.08 to 4.52)	0.96 (0.77 to 1.19)
25-34	−4226 (−5861 to −2592)	3.16 (2.37 to 3.96)	0.95 (0.78 to 1.14)
35-44	−8654 (−12 583 to −4724)	1.77 (1.14 to 2.74)	1.41 (1.17 to 1.71)
45-54	−10 683 (−18 187 to −3178)	1.05 (0.51 to 2.14)	1.43 (1.11 to 1.85)

Open in a new tab

Abbreviation: PoMS, pediatric-onset multiple sclerosis.

^{^a}

Adjusted for educational level and matched on age, sex, country of birth, and year.

Figure. — Vertical lines indicate 95% CIs. PoMS indicates pediatric-onset multiple sclerosis.

When stratified by childhood-onset MS and adolescent-onset MS, the estimates were largely in the same direction (eTables 5-7 in the Supplement). The childhood-onset group did not always reach significance, possibly owing to small numbers (n = 55). Similarly, the risk estimates in the 2 eras were similar, although the effect size for earnings and sickness absence in the PoMS cohort compared with the matched reference cohort was larger for the earlier era (eTables 8-10 in the Supplement).

Discussion

In this nationwide register study, we found differences between individuals with PoMS and individuals in a matched reference cohort without MS in socioeconomic outcomes throughout adulthood. Those with PoMS were less likely to attend university and never attained the same level of earnings from work as their counterparts without MS. Before the age of 25 years, individuals with PoMS made, on average, more than $1600 less than their matched references, but by the ages of 45 to 54 years, this gap had widened to more than $10 600 per year. At the same time, they had 3 times the mean number of sickness absence days per year before the age of 34 years, and the number of disability pension days used in later life was higher than that of the reference cohort.

Unique to these young patients is that the disease imparts damage to the central nervous system while it is still in development.⁴ Although enhanced compensatory mechanisms can mitigate some of this damage, there appear to be lasting effects through adulthood.^2,3 Our results highlight how this damage can translate into real-world consequences.

To our knowledge, this study is the first to explore the association between childhood MS and educational level at the national level. We found that persons with PoMS were less likely to attend university relative to persons without MS in a matched reference cohort. The available data on educational level capture only the initiation of schooling, not whether these persons completed schooling or their academic standing, but prior research has suggested that these patients face distinctive scholastic challenges.^5,6,7

The median earnings of the PoMS cohort never reached those of the matched reference cohort, and the PoMS cohort exhibited an atypical trajectory. Earnings plateaued in early adulthood among individuals with PoMS, whereas they continued to increase among individuals in the matched reference cohort. The lower earnings of those in the PoMS cohort may indicate an inability to attain or remain in paid work as we saw a higher proportion of those in the PoMS cohort with zero earnings. However, the supplementary analysis that included only persons with nonzero earnings also revealed significantly lower earnings among the PoMS cohort. Sweden has a social welfare system to help defray the economic cost of chronic disease. Similar to persons with adult-onset MS, we found that persons with PoMS used substantially more of such benefits than those in their matched reference cohort.⁸ When stratified by era, the older cohort (MS onset during the period from 1980 to 1999) appeared to be most responsible for these associations, suggesting that patients who received a diagnosis more recently may be performing better, perhaps owing to improved treatment strategies.

Limitations and Strengths

This study has some limitations, including the lack of details on educational achievements and the potentially limited generalizability outside of Sweden. It also has some strengths, including the use of large population-based cohorts followed up prospectively over decades. The matched reference cohort was randomly selected at the national level, limiting any potential selection bias. Information on exposures and outcomes was obtained from quality administrative registers and recorded in real-time, thereby avoiding recall bias.

Conclusions

In this study, we found that PoMS has profound, lasting consequences that translate into lower educational achievements and earnings and a greater use of disability benefits in adulthood. Assessing whether specific clinical or demographic factors are associated with these socioeconomic outcomes will be the next step in understanding, and potentially preventing, these outcomes.

Supplement.

eTable 1. Details of the National Registers in Sweden and How They Were Used in this Study

eTable 2. Educational Achievement in PoMS vs the General Population Reference Cohort

eTable 3. Earnings Among the PoMS and Matched Reference Cohort Over the Four Age Periods

eTable 4. Number and Percentage of Persons in the PoMS Group and the Reference Cohort Who Had Zero Earnings, ≥1 Sickness Absence Day, or ≥1 Disability Pension Day in Each of the Four Age Periods

eTable 5. Educational Level Among PoMS Compared to the Matched Reference Cohort in Childhood-Onset and Adolescent-Onset Cases

eTable 6. Earnings Among the PoMS Compared to the Matched Reference Cohort in Childhood-Onset and Adolescent-Onset Cases

eTable 7. Sickness Absence and Disability Pension Days Among the PoMS Compared to the Matched Reference Cohort in Childhood- and Adolescent-Onset MS

eTable 8. Educational Level Among PoMS Compared to the Matched Reference Cohort in the Two Eras

eTable 9. Earnings Among PoMS Compared to the Matched Reference Cohort in the Two Eras

eTable 10. Sickness Absence and Disability Net Days Among the PoMS Compared to the Matched Reference Cohort in Two Eras

eAppendix. Details of Study Outcomes and Analyses

Click here for additional data file.^{(241.2KB, pdf)}

References

1.Waldman A, Ness J, Pohl D, et al. Pediatric multiple sclerosis: clinical features and outcome. Neurology. 2016;87(9)(suppl 2):S74-S81. doi: 10.1212/WNL.0000000000003028 [DOI] [PMC free article] [PubMed] [Google Scholar]
2.McKay KA, Hillert J, Manouchehrinia A. Long-term disability progression of pediatric-onset multiple sclerosis. Neurology. 2019;92(24):e2764-e2773. doi: 10.1212/WNL.0000000000007647 [DOI] [PMC free article] [PubMed] [Google Scholar]
3.McKay KA, Manouchehrinia A, Berrigan L, Fisk JD, Olsson T, Hillert J. Long-term cognitive outcomes in patients with pediatric-onset vs adult-onset multiple sclerosis. JAMA Neurol. 2019;76:1028-1034. doi: 10.1001/jamaneurol.2019.1546 [DOI] [PMC free article] [PubMed] [Google Scholar]
4.Gogtay N, Giedd JN, Lusk L, et al. Dynamic mapping of human cortical development during childhood through early adulthood. Proc Natl Acad Sci U S A. 2004;101(21):8174-8179. doi: 10.1073/pnas.0402680101 [DOI] [PMC free article] [PubMed] [Google Scholar]
5.Mikaeloff Y, Caridade G, Billard C, Bouyer J, Tardieu M. School performance in a cohort of children with CNS inflammatory demyelination. Eur J Paediatr Neurol. 2010;14(5):418-424. doi: 10.1016/j.ejpn.2010.02.003 [DOI] [PubMed] [Google Scholar]
6.Amato MP, Goretti B, Ghezzi A, et al. ; Multiple Sclerosis Study Group of the Italian Neurological Society . Cognitive and psychosocial features of childhood and juvenile MS. Neurology. 2008;70(20):1891-1897. doi: 10.1212/01.wnl.0000312276.23177.fa [DOI] [PubMed] [Google Scholar]
7.Montiel-Nava C, Peña JA, González-Pernía S, Mora-La Cruz E. Cognitive functioning in children with multiple sclerosis. Mult Scler. 2009;15(2):266-268. doi: 10.1177/1352458508097471 [DOI] [PubMed] [Google Scholar]
8.Wiberg M, Murley C, Tinghög P, et al. Earnings among people with multiple sclerosis compared to references, in total and by educational level and type of occupation: a population-based cohort study at different points in time. BMJ Open. 2019;9(7):e024836. doi: 10.1136/bmjopen-2018-024836 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials