Disabilities and Health-Related Quality of Life in Multiple Sclerosis: Prospective 10-year Web-Based Study of a Population-Based Real-Life Cohort in the Netherlands

Abstract

Purpose

To assess at population level long-term changes in disabilities and health-related quality of life (HRQoL) in persons with multiple sclerosis (PwMS) in the Netherlands.

Patients and Methods

A prospective web-based study in a real-world population-based self-enrolled cohort; 6-month assessments by means of the Multiple Sclerosis Impact Profile (MSIP) and Multiple Sclerosis Quality of Life-54 (MSQoL-54) questionnaires; Generalized Linear Mixed Models (GLMM) analysis; predetermined anchor points 5, 8 and 10 years follow-up.

Results

Nation-wide 391 patients started (2011–2012), 272 participants were included in the MSIP and 267 in the MSQoL-54 analyses. Female 74%, mean (standard deviation) age 45 (11) years. Baseline Expanded Disability Status Scale scores: 58.9% 0.0–4.0, 35.5% 4.5–6.5, 5.6% >6.5. Baseline demographic and disease characteristics did not differ between participants with and without follow-up data. GLMM indicated that after 10 years disability related to Activities of Daily Living and Basic Movement Activities had increased. Conversely, at 5, 8 and 10 years an improvement was reported in disability related to Lack of Support from Environmental Factors, more specifically Professional Care and Assistance, and Social Security Services. The other MSIP disabilities and symptoms remained stable over time, as did the physical and mental MSQoL-54 scores.

Conclusion

Our findings indicate that in the Dutch MS population in the period 2011–2021 the limitations in activities of daily living and in basic movement activities worsened over time. The improved support from professional care, professional assistance and social security services may be related to the implementation of nation-wide decentralized legislation to support people with disabilities, and the emergence of MS-specific neurological and nursing care. These findings at population level regarding the long-term course of disabilities and HRQoL in PwMS in the Netherlands in the era of (high) efficacy disease-modifying drugs may be useful for policymakers, healthcare providers, and patient organizations.

Introduction

Multiple sclerosis (MS) is a chronic immune-mediated inflammatory and degenerative disease of the central nervous system with a wide range of symptoms and disabilities that may develop over time. The availability since the mid-1990s of low-efficacy disease-modifying drugs (DMDs) resulted in a new course of MS.¹ In the past two decades, more efficacious DMDs have become available. DMDs inhibit the immune-mediated inflammatory aspects of the disease, that largely determine the initial relapsing-remitting phase (relapsing-remitting MS [RRMS]),¹ but do not affect the neurodegeneration, that prevails in the primary and secondary progressive phases.¹ In recent years various studies have assessed long-term clinical changes in patients with MS (PwMS) in the era of high-efficacy DMDs.^2,3

However, the results of these studies cannot be easily generalized. First, findings in a given country are not per se applicable to other countries given the differences in access to DMDs, organization of healthcare systems, and cultural barriers.⁴ Second, some studies mainly involved patients treated in MS-dedicated academic centers, the findings of which do not necessarily apply to patients treated in general neurological practices or to the entire MS population. To the best of our knowledge, no population-based data are available on the long-term course of MS in the Netherlands.

Against the background of limited financial resources and the complexity of MS care, healthcare professionals, patient organizations, and health authorities may benefit from detailed information on long-term changes in MS-related disabilities and health status at the population level. A differentiated knowledge of the occurrence of and changes in the various disabilities and health aspects in PwMS could contribute to a better assessment of treatments and healthcare needs as well as the risk that patients have to reach major levels of handicaps and dependencies. In fact, the availability of prospectively collected real-world data may enable a more realistic evaluation of the long-term prospects for both PwMS and society as a whole and may help increase the effectiveness and efficiency of MS-related policies.

Patient-reported outcome measures (PROMs) assess the subjective elements of the patients’ conditions.⁵ The US Food and Drug Administration and the European Medicines Agency define patient-reported outcomes (PROs) as any outcomes related to the patient’s health or treatment that is evaluated directly by the patient, without any interpretation by a doctor or anyone else.^6,7 Thus, PROs are characterized by intrinsic clinical relevance as they reflect the condition as reported by patients themselves.

In the last decades, the use of PROMs has increased considerably and currently those outcomes are used to assess the effects of treatments and the quality of care, to evaluate policies and to inform health economics.^5,8,9 One main advantage of PROMs is that they may provide indicators of treatment effectiveness that are not captured by objective markers or clinical assessments.^5,7 Moreover, PROMS are not time-consuming for healthcare organizations, as data are generated by patients’ self-reports, while web-based programs enhance the ease of use – and thus utilization – for patients and healthcare providers alike.

PROMS that assess the health status are to be distinguished from those that measure symptoms and disabilities. For, organizational changes in health services may further improve the health status in patients who do receive optimal medical treatment. A general health-related PRO is health-related quality of life (HRQoL), which can be defined as the functional effect of an illness, its therapy and the received care on a patient, as perceived by the patient.^10–12 HRQoL is multidimensional, encompassing physical and occupational function, psychological state, social interaction, and somatic sensation.^10–12

PROMs are increasingly being used in research and in daily practice,^8,9 also given the fact that they can be prognostic of hard clinical outcomes.⁸ The Multiple Sclerosis Impact Profile (MSIP) is an MS-specific questionnaire that provides detailed information about various MS-related disabilities that cannot be derived from physician-based or technical measures.^13–15 The MSIP is a validated instrument that has been shown to detect clinically relevant changes over time in various disability domains.¹⁵ With respect to the health status of PwMS, the MSQoL-54 is a validated and frequently applied questionnaire to assess the MS-specific HRQoL.¹⁶ Notably, a low HRQoL in PwMS has been found to be an independent predictor of long-term disability.¹⁷

In view of these considerations, a prospective long-term PROM-based study was conducted in a real-life open cohort of PwMS in the Netherlands. The aim of the study was 1) to evaluate changes over time in MS-related disability in the DMD-era by measuring patient-reported disabilities in various domains, 2) to evaluate changes over time in HRQoL, as this outcome relates to treatment, health care and societal circumstances.

Given that DMDs mainly target immune-mediated inflammation but not neurodegenerative processes, and that DMD treatments are only partially effective, we hypothesized that patient-reported disabilities and HRQoL would deteriorate over time. We further hypothesized that worsening over time would be more pronounced among patients with progressive disease, higher baseline EDSS scores, longer disease duration, and older age, as these characteristics are generally associated with more advanced or progressive MS.

Patients and Methods

This is a prospective, web-based, patient-centred study of long-term changes in disabilities, perception of disabilities, and HRQoL in individuals with MS or clinically isolated syndrome (CIS) in the Netherlands (“Dutch Multiple Sclerosis Study”). The study protocol was published¹⁸ and registered in the Netherlands Trial Register (ID number: NL9520).

Study Objectives and Design

The primary objective of this study was to assess long-term changes in patient-reported disabilities and perceptions of disabilities.¹⁸ The secondary objective was to study long-term changes in HRQoL.¹⁸

This study had a prospective, web-based, patient-centred, longitudinal, observational design with voluntary patient-driven inclusion. The first patient was enrolled in May, 2011. Initially, a follow-up period of 2 years was planned to assess the feasibility of the project. Since then, the study has been extended annually. Participants were asked to complete questionnaires about disabilities, disability perception, and HRQoL at baseline and every six months thereafter.

Study Population and Patient Recruitment

The study population included individuals in the Netherlands diagnosed with MS or CIS. The criteria for being able to participate were 1) MS or CIS diagnosis, 2) willingness and ability to comply with the assessments as per the protocol, 3) access to the Internet, and 4) having provided informed consent. Exclusion criteria were not defined.

Patients were informed about the option to participate via the websites of patient organizations (Dutch National MS Foundation [DNMSF], Multiple Sclerose Vereniging Nederland [MSVN]) and the MS4 Research Institute (www.ms4ri.nl), and the study website (www.msstudie.nl). Neurologists and MS nurses in the Netherlands were sent an information letter by regular mail with patient brochures, which they were asked to hand over to their patients. In addition, the brochure was sent to the donors of the DNMSF as an attachment for its journal and mail. In the DNMSF journal, information was provided by the principal investigator (PJJ). On two occasions, study information was published in national and regional newspapers.

Technical Aspects

The Dutch Multiple Sclerosis Study is a modular application of the Curavista e-Health Platform built on an Oracle database with Java-scripting, XML-applets, and AJAX protocols. The data processing was 256-bit encrypted with VPN tunnelling. The physical presence and software of the databases were secured in a dedicated data centre in the Netherlands. The database used in this study is compliant with EU regulations on data storage and activation for medical purposes. There were four separate databases: one with personal identifiers (name, address, and identification number), one with medical records (answers to the questions and identification number), one with the social security number, and one with the key. Only after the login are the data presented on the screen (encrypted key).

Data Acquisition

After providing consent, patients received a personal code and logged on to the study website (www.msstudie.nl) to choose their username and password. When online, they had to scroll through webpages containing electronic case record forms (e-CRFs) with questionnaires (demographics, disease characteristics, disabilities, perception of disability, HRQoL, one-time questionnaires). Patients were informed by e-mail that an assessment was due and that the corresponding e-CRFs had been made available and were to be completed anytime within one week. Participants were allowed to attend as many sessions as needed, as the answers were automatically saved. The questionnaire items were fixed, and the responses were captured automatically. Automated checks for completeness were performed before questionnaires were submitted. The patients saw an overview of all questions and answers before submission and were allowed to change their answers before submission. After confirmation, the e-CRF was automatically sent to the study centre, and changes were no longer possible. If an e-CRF was not completed within one week, the help desk sent a reminder via e-mail. Data extraction for the current analysis was performed in September 2021.

In 2021, all participants, including those who discontinued the study, were asked to complete the MSIP and MSQoL-54 questionnaires. In 2016 and 2021, additional questionnaires inquired about demographics, socioeconomic aspects, disease characteristics, medicine use, comorbidities, and lifestyle. To maximize the number of respondents, all participants were contacted via a three-fold stepwise approach: e-mail, phone, and personal letters. In addition, calls were posted on the websites of patient organizations.

Outcome Measures

Multiple Sclerosis Impact Profile (MSIP)

Disabilities and disability perceptions were assessed using the MSIP. The MSIP is based on the International Classification of Functioning, Disability, and Health (ICF), which is a framework for describing and organizing information on functioning and disability. The ICF provides a standard language and conceptual basis for the definition and measurement of health and disability and is structured around the following components: body functions and structure, activities (related to tasks and actions by an individual), participation (involvement in a life situation), and additional information on severity and environmental factors. Together with the International Classification of Diseases (ICD), ICF constitutes the core classification of the World Health Organization Family of International Classifications (WHO-FIC).

The MSIP measures MS-related disabilities and disability perceptions and has established psychometric properties.^13–15 It reflects an objective view of the severity of the impact of MS (disabilities) as well as the subjective impact (disability perception). The MSIP comprises 36 questions assessing disabilities (Questions Q1a-Q36a) and disability perception (Q1b-Q36b) in seven domains and four symptoms. Domain and symptom scores were obtained by adding the scores for each item. The domains and symptoms included a) Impairments in Muscle and Movement Functions (minimum score 0, maximum score, 16), b) Impairments in Excretion and Reproductive Functions (0–12), c) Impairments in Mental Functions (0–12), d) Limitations in Basic Movement Activities (0–15), e) Limitations in Activities of Daily Living (0–24), f) Restrictions in Participation in Life Situations (0–26), g) Lack of Support from Environmental Factors (0–20), h) Fatigue (0–4), i) Pain (0–4), i) Impairment in Speech Functions (0–4) and j) Impairment in Seeing Functions (0–4). A higher score indicates a worse condition.

Multiple Sclerosis Quality of Life-54 Questionnaire (MSQoL-54)

HRQoL was assessed using the MSQoL-54. The MSQoL-54 is a psychometrically validated MS-specific multidimensional inventory of patient-centred health status.¹⁶ It consists of the 36-item Short Form Health Survey as a generic core measure to enable comparisons with other patient populations and the general population, supplemented by 18 additional questions exploring items relevant to PwMS in the areas of health distress (four items), sexual function (four items), satisfaction with sexual function (one item), overall quality of life (two items), cognitive function (four items), energy (one item), pain (one item), and social function (one item). Physical and mental dimensions underlie the MSQoL-54: physical and mental domains. Composite scores for each domain range from 0 to 100, with higher values indicating better HRQoL.

Expanded Disability Status Scale (EDSS)

The baseline assessment included the Expanded Disability Status Scale (EDSS). The EDSS is an investigator-based standard tool that quantifies MS-related disabilities in eight functional systems: pyramidal, cerebellar, brainstem, sensory, bowel and bladder, visual, cerebral and other.¹⁹ Scores range from 0.0 to 10.0, with higher values indicating greater disability. A version of the EDSS that can be used by phone is available and was used in this study.²⁰ To prevent inter-rater variation, all EDSS scoring was performed by the same experienced MS nurse (MH). The EDSS scores were categorized as low (<4.5), medium (4.5–6.5) or high (>6.5).

Data Analysis

The original data analysis plan was conceived by IK, LB, GvM, and PJJ, and has been published at https://onderzoekmetmensen.nl/nl/trial/22564. Statistical analyses were performed at the Department of Health Services Research of Maastricht University, Maastricht, Netherlands. Baseline characteristics and observed outcome scores were summarized using descriptive statistics. Continuous variables were presented as means with standard deviations (SDs) or with 95% Confidence Intervals (95% CIs). Categorical variables were presented as frequencies and percentages.

Longitudinal changes in MSIP and MSQoL-54 scores were analysed using Generalized Linear Mixed Models (GLMM) to account for repeated measurements within participants.^21,22 The primary GLMM analysis focused on longitudinal changes in MSIP domain scores, because the main objective of the study was to assess changes in disability and disability perception over time. MSIP scores are bounded scores with a fixed minimum and maximum value for each domain. Therefore, MSIP domain scores were analysed using beta-binomial GLMM. Time was included as a categorical fixed effect, with baseline as the reference category.^21,22 Participant was included as a random intercept to account for repeated measurements within the same individual.

The secondary GLMM analysis focused on longitudinal changes in MSQoL-54 physical and mental health composite scores, because HRQoL was the secondary objective of the study. MSQoL-54 composite scores are continuous scores ranging from 0 to 100.²³ Therefore, these scores were analysed using Gaussian GLMM with identity link. As in the MSIP models, time was included as a categorical fixed effect, with baseline as the reference category, and participant was included as a random intercept.

The GLMMs were unadjusted because the aim was to estimate changes over time within the cohort, rather than to identify baseline predictors of change. Adjusted sensitivity analyses were performed including baseline age, sex, disease course, disease duration, and baseline EDSS category. Subgroup analyses were performed by adding time-by-subgroup interaction terms for sex, age category, disease course, disease duration, and baseline EDSS category. Adjusted and subgroup results are presented in the Appendix. Pairwise comparisons between follow-up timepoints and baseline were performed for 5, 8, and 10 years of follow-up. We looked at whether the patients changed over time compared with the start of the study. Baseline was the starting point. We then compared the results at 5, 8, and 10 years with baseline. Because we made several comparisons with the same baseline, we used Dunnett adjustment to make the p-values more cautious.^24,25 We considered a result statistically significant only if the adjusted p-value was below 0.001. Statistically significant changes from the baseline of approximately 15% or 0.5 SD were considered clinically relevant. Baseline adjusted results are presented in the Appendix.

Paired t-tests were used for MSQoL-54 physical and mental composite scores when paired differences were approximately normally distributed. Wilcoxon signed-rank tests were used for MSIP item-level analyses because these scores were ordinal, bounded, and showed non-normal paired differences.

All analyses were conducted using R version 4.5.3.

Ethics Approval and Informed Consent

The study protocol was submitted to the Medisch Ethische Toetsing Onderzoek Patiënten en Proefpersonen, an ethical committee established in Tilburg, the Netherlands (M379). This committee concluded that a review was not necessary, since the study did not qualify for examination under the Dutch Medical Research Involving Human Subjects Act of 1999 (http://wetten.overheid.nl/BWBR0009408).

The study was conducted in accordance with the Declaration of Helsinki (Ethical Principles for Medical Research Involving Human Subjects, version 2013; 64th World Medical Association General Assembly, Fortaleza, Brazil, October 2013) (www.wma.net) and Dutch Medical Research Involving Human Subjects Act http://wetten.overheid.nl/BWBR0009408).

The study was registered in the Netherlands Trial Register (ID number NL9520) on May 27 2021. https://www.onderzoekmetmensen.nl/en/node/22564/pdf https://trialsearch.who.int/Trial2.aspx?TrialID=NL9520.

Informed consent was obtained online via www.msstudie.nl and www.ms4ri.nl. For additional information and eventual questions, the principal investigator (PJJ) was contacted via webmail. Patients willing to participate confirmed that they had taken notice of the information and agreed to participate by clicking a button on a specific page on the study website. This consent form also covered the storage of personal and study-related data in the database. The patient could begin with the baseline assessment as soon as informed consent was provided. Patients were informed that they could discontinue participation or withdraw consent at any time without having to provide a reason. The completion of the web-based questionnaires took approximately 30–45 minutes every six months; two additional questionnaires (Questionnaire 2016, Questionnaire 2021) took approximately 30 minutes to complete. The assessment of the EDSS score by phone took approximately 10–20 min.

Results

Patients

391 patients enrolled in the study, and 341 completed the baseline assessment, 88% of them between April 2011 and September 2012 (Figure 1). The data analysis group consisted of 272 patients who completed the baseline questionnaires and at least one follow-up MSIP (n=272) and MSQoL-54 (n=267) questionnaire. The demographic and disease characteristics are presented in Table 1.

Figure 1.

Study flow chart. *of the participants who had reached this time point.

Table 1.

Demographic and Disease Characteristics of the Data Analysis Group

		MSIP (n=272)		MSQoL-54 (n=267)
Age	Years (SD)	45.2 (10.8)		45.4 (10.8)
		N	%	N	%
Sex	Female	201	74	201	75
	Male	71	26	66	25
Diagnosis	MS	259	95	254	95
	CIS	7	3	7	3
	Unknown	6	2	6	2
Disease course	RRMS	177	65	173	65
	SPMS	35	13	34	13
	PPMS	32	12	32	12
	Unknown	28	10	28	10
EDSS (n=197)	Low (<4.5)	116	59	116	59
	Medium (4.5–6.5)	70	36	70	36
	High (>6.5)	11	6	11	6
Disease duration	<5 years	72	26	71	27
	5–15 years	108	40	106	40
	≥15 years	92	34	90	34

Abbreviations: MSIP, Multiple Sclerosis Impact Profile; MSQoL-54, Multiple Sclerosis Quality of Life-54 item; SD, standard deviation; MS, multiple sclerosis; CIS, clinically isolated syndrome; RRMS, relapsing-remitting multiple sclerosis; SPMS, secondary progressive multiple sclerosis; PPMS, primary progressive multiple sclerosis; EDSS, Expanded Disability Status Scale.

For the MSIP the numbers of completed questionnaires at baseline, month (M) 6, M12, M18, M24, M30, M36, M42, M48, M54, M60, M66, M72, M78, M84, M90, M96, M102, M108, M114 and M120 are 273, 269, 235, 223, 203, 193, 167, 160, 154, 148, 135, 123, 116, 109, 110, 103, 94, 85, 88, 80, 66, and for the MSQoL-54 267, 263, 227, 216, 194, 181, 159, 154, 146, 148, 131, 118, 116, 108, 109, 99, 91, 88, 85, 56 and 61. Notably, in view of the length of the inclusion period, 6.3% of the patients in the data analysis group had not yet reached M96 (8 years) at the time of analysis and 30.5% of the patients had not reached M120 (10 years) at the time of analysis.

MSIP and MSQoL-54 data were obtained from 135 patients for the 5-year analysis, from 91 patients for the 8-year analysis, and from 61 for the 10-year analysis, being 49.6%, 35.7%, and 32.3% of the participants who had reached the respective endpoints.

MSIP

The mean (SD) MSIP disabilities and disability perception scores at the baseline are shown in Table 2. The mean scores (95% CI) at follow-up are presented in Figure 2.

Table 2.

Multiple Sclerosis Impact Profile (MSIP) Disabilities and Disabilities Perception Scores (Minimum-Maximum) at Baseline (n=272)

MSIP Scores (Min.-Max).	Disabilities		Disabilities Perception
	Mean	SD	Mean	SD
Muscle and Movement Functions (0–16)	3.7	2.46	1.7	2.24
Excretion and Reproductive Functions (0–12)	2.7	2.17	2.3	1.95
Mental Functions (0–12)	2.5	1.73	2.3	1.88
Basic Movement Activities (0–15)	3.2	3.01	2.2	2.76
Activities of Daily Living (0–24)	5.8	5.23	4.7	4.40
Participation in Life Situations (0–26)	3.6	3.80	2.5	2.73
Environmental Factors (0–20)	5.3	3.89	1.0	1.49
Fatigue (0–4)	1.8	0.98	1.6	0.93
Pain (0–4)	0.8	0.85	0.7	0.84
Speech functions (0–4)	0.4	0.52	0.3	0.58
Seeing functions (0–4)	0.7	0.85	0.6	0.77

Abbreviations: Min., minimum; Max., maximum; SD, standard deviation.

Figure 2.

Multiple Sclerosis Impact Profile (MSIP) mean domain scores with 95% confidence intervals (CI) over time.

Notes: M, month. (a) Impairments in Muscle and Movement Functions (minimum score 0, maximum score, 16), (b) Impairments in Excretion and Reproductive Functions (0–12), (c) Impairments in Mental Functions (0–12), (d) Limitations in Basic Movement Activities (0–15), (e) Limitations in Activities of Daily Living (0–24), (f) Restrictions in Participation in Life Situations (0–26), (g) Lack of Support from Environmental Factors (0–20), (h) Fatigue (0–4), (i) Pain (0–4), (j) Impairment in Speech Functions (0–4) and (k) Impairment in Seeing Functions (0–4). A higher score indicates a worse condition. Bars indicate 95% CI values.

Table 3 presents the significant longitudinal changes in MSIP domains based on mixed-model analysis. Significant changes occurred in three MSIP domains. Activities of Daily Living disability scores had increased from 3.90 at baseline to 4.93 at 8 years and 6.10 at 10 years, indicating a worsening of disability. Disability regarding Basic Movement Activities had also worsened at 10 years, with scores rising from 2.05 to 2.98. Perceived disability related to Basic Movement Activities increased from 1.09 to 1.75 at 10 years.

Table 3.

The Significant Longitudinal Changes in MSIP Domains Based on Mixed-Model Analysis

MSIP Domain	Outcome	Score Range	Timepoint	Baseline Predicted Score	Follow-Up Predicted Score (95% CI)	Change from Baseline	Change from Baseline (%)	OR vs Baseline (95% CI)	Adjusted p-value	Direction
Activities of daily living	Disabilities	0–24	8 years (n= 94)	3.90	4.93 (4.09 to 5.88)	+1.02	+26.2%	1.33 (1.09 to 1.63)	<0.001	Worse
Activities of daily living	Disabilities	0–24	10 years (n=63)	3.90	6.10 (5.07 to 7.26)	+2.20	+56.3%	1.76 (1.39 to 2.21)	<0.001	Worse
Basic movement activities	Disabilities	0–15	10 years (n-63)	2.05	2.98 (2.41 to 3.65)	+0.93	+45.2%	1.56 (1.16 to 2.10)	<0.001	Worse
Basic movement activities	Disability perception	0–15	10 years (n=63)	1.09	1.75 (1.32 to 2.28)	+0.66	+60.9%	1.69 (1.17 to 2.44)	<0.001	Worse
Environmental factors lack of support	Disabilities	0–20	5 years (n=136)	4.31	2.92 (2.47 to 3.45)	−1.39	−32.2%	0.62 (0.48 to 0.81)	<0.001	Better
Environmental factors lack of support	Disabilities	0–20	8 years (n=94)	4.31	2.62 (2.16 to 3.16)	−1.69	−39.2%	0.55 (0.41 to 0.74)	<0.001	Better
Environmental factors lack of support	Disabilities	0–20	10 years (n=63)	4.31	2.41 (1.92 to 2.99)	−1.91	−44.2%	0.50 (0.35 to 0.71)	<0.001	Better

Abbreviations: MSIP, Multiple Sclerosis Impact Profile; 95% CI, 95% Confidence Interval; OR, Odds Ratio.

Conversely, disability scores related to Lack of Support from Environmental Factors had improved at 5, 8, and 10 years, with predicted scores declining from 4.31 at baseline to 2.92 at 5 years, 2.62 at 8 years, and 2.41 at 10 years. This domain evaluates support from immediate family, professional care and assistance, social security services, and healthcare services. Exploratory item-level paired analyses using Wilcoxon signed-rank tests indicated that improvements were primarily attributable to enhanced support from professional care and assistance and social security services. Descriptive analysis of item-level scores over time indicated that these improvements occurred mainly during the first two years of the study, corresponding to 2011–2014 for most participants (Figure 3). Perceived disability related to Lack of Support from Environmental Factors did not change significantly.

Figure 3.

Disability (mean and standard deviation) due to lack of support from the four environmental factors: immediate family, professional care and assistance, social security services, and healthcare services.

Notes: Bars indicate standard deviation. M, month. Number of patients per time point: 272 (baseline), 269 (M6), 235 (M12), 223 (M18), 203 (M24), 193 (M30), 167 (M36), 160 (M42), 154 (M48), 148 (M54), 135 (M60), 123 (M66), 116 (M72), 109 (M80), 110 (M86), 103 (M92), 94 (M98), 85 (M104), 88 (M110), 80 (M116) and 66 (M120).

No statistically significant changes were observed in the remaining interpretable MSIP models at the predefined significance threshold of p <0.001. Models that exhibited convergence problems were excluded from interpretation.

The results of adjusted sensitivity analyses were generally consistent with those of the primary MSIP models. After adjustment for baseline age, sex, disease course, and disease duration, the main 10-year findings for Activities of Daily Living, Basic Movement Activities, and Lack of Support from Environmental Factors remained statistically significant. In the model that additionally included baseline EDSS category, improvement in Lack of Support from Environmental Factors remained significant at 5, 8, and 10 years, while worsening in Activities of Daily Living disability remained significant at 10 years. In contrast, the findings for Basic Movement Activities were attenuated and did not meet the predefined significance threshold of p <0.001 following EDSS adjustment. Full adjusted results are provided in the Appendix.

Subgroup analyses were performed to explore whether longitudinal MSIP changes differed by sex, age category, disease course, disease duration, and baseline EDSS category. No statistically significant subgroup-by-time interactions were observed at the predefined threshold of p <0.001. Full subgroup results are presented in the Appendix.

MSQoL-54

The mean (SD) baseline physical and mental health composite scores were 56.38 (16.56) and 70.15 (18.11), respectively.

GLMM analyses indicated no statistically significant longitudinal changes in either score at 5, 8, or 10 years compared with baseline (all Dunnett-adjusted p-values >0.001) (Figure 4) (Table 4). Subgroup analyses revealed no statistically significant within-subgroup changes for either the physical or mental health composite scores at the predefined threshold of p <0.001. No significant subgroup-by-time interactions were identified for the mental health composite score. In contrast, significant subgroup-by-time interactions for the physical health composite score were observed for age category, disease duration, and baseline EDSS category, indicating that physical HRQoL trajectories varied across these subgroups. However, as no individual subgroup demonstrated a statistically significant change from baseline, these results should be interpreted with caution. Full subgroup results are provided in the Appendix.

Figure 4.

Mean MSQoL-54 mental and physical composite scores with 95% confidence intervals (CI) on 10 year follow-up. Bars indicate 95% CI values.

Table 4.

Longitudinal Changes in MSQoL-54 Physical and Mental Health Composite Scores Based on Mixed-Model Analysis

MSQoL-54 Composite Score	Timepoint	N	Baseline Predicted Score	Follow-Up Predicted Score, Mean (95% CI)	Mean Difference vs Baseline (95% CI)	% Change from Baseline	p-value	Interpretation
Physical health composite	5 years	131	56.38	57.11 (54.60 to 59.61)	+0.72 (−2.24 to 3.69)	+1.3%	0.991	No significant change
Physical health composite	8 years	91	56.38	56.07 (53.34 to 58.79)	−0.32 (−3.69 to 3.06)	−0.6%	1.000	No significant change
Physical health composite	10 years	61	56.38	57.22 (54.17 to 60.27)	+0.84 (−3.11 to 4.78)	+1.5%	0.996	No significant change
Mental health composite	5 years	131	70.15	71.64 (68.95 to 74.33)	+1.48 (−2.19 to 5.16)	+2.1%	0.897	No significant change
Mental health composite	8 years	91	70.15	70.89 (67.89 to 73.89)	+0.74 (−3.45 to 4.92)	+1.1%	0.998	No significant change
Mental health composite	10 years	61	70.15	70.63 (67.19 to 74.07)	+0.48 (−4.41 to 5.36)	+0.7%	1.000	No significant change

Notes: Results are based on Gaussian generalized linear mixed models with identity link, with time as a categorical fixed effect and participant as a random intercept. Baseline was used as the reference category. N refers to the number of participants with observed MSQoL-54 data at each follow-up timepoint. Higher MSQoL-54 scores indicate better health-related quality of life. P-values were adjusted for comparisons with baseline using Dunnett adjustment. Statistical significance was defined as p <0.001.

Adjusted sensitivity analyses yielded results consistent with the primary mixed-model analysis. After controlling for baseline age, sex, disease course, and disease duration, no statistically significant changes were observed in either the MSQoL-54 physical or mental health composite scores at 5, 8, or 10 years. Additional adjustment for baseline EDSS category did not alter the findings. However, the EDSS-adjusted model included fewer participants due to incomplete baseline EDSS and other model data, reducing the sample size from 237 to 174 participants.

Discussion

This is the first study to prospectively assess at population level long-term changes in disabilities and HRQoL in PwMS in the Netherlands. The study used a web-based population-based real-life open cohort design with PROMS as outcomes.¹⁸ We observed a worsened disability regarding limitations in activities of daily living (8 and 10 years), and a worsened disability (perception) due to limitations in basic movement activities (10 years), as well as an improved disability related to lack of support from environmental factors (5, 8, and 10 years). No consistent changes were found in the other disability domains or symptoms, nor in physical or mental MSQoL-54 scores or subgroups regarding sex, age category, disease course, disease duration and baseline EDSS category.

Using the MSIP, Wynia et al conducted a single-centre study in Groningen, Netherlands, over a 5-year period (baseline, 2004; follow-up, 2009) with 241 PwMS. Statistically significant and clinically relevant increases were observed in the seven domains and one symptom.¹⁵ Notably, in their study Lack of Support from Environmental Factors showed the largest increase (+2.3 [mean]) from baseline (3.9 [mean]),¹⁵ followed by Impairment in Speech Functions, Limitations in Basic Movement Activities and Limitations in Activities of Daily Living. In addition, in the WHO Quality of Life-abbreviated version, patients reported a small reduction in HRQoL concerning Physical Health and Autonomy.¹⁵

In contrast, in our study, the most conspicuous finding was a substantial decrease in the Lack of Support from Environmental Factors score, from 5.3 (mean) to 3.3 (mean). This decrease occurred in the first 24 months of the study, broadly in the period 2011–2014, and persisted throughout the 10-year follow-up. The extent of this positive change indicated that it was clinically relevant, meaning that the patients reported improved support from environmental factors. Additional analysis at the item level showed specifically an improved support from professional care and assistance, and from social security services.

From our data, we could not infer which specific factor(s) explained this substantial and lasting improvement. One potential factor concerning professional care and assistance might be the greater availability of MS nurses and multidisciplinary care,^26,27 and also indirectly a more widespread use of highly efficacious DMDs.²⁸ A recent nationwide observational cohort study of PwMS in the Swedish MS registry with disease onset between 2005 and 2015 found that in relapsing MS, clinic-level quality of care indicators, including visit density, mean time to start DMD therapy, and MRI density, benefitted physical symptoms (and EDSS), and that faster treatment and frequent visits benefitted psychological symptoms.²⁹

The introduction of the Social Support Act (Wet maatschappelijke ondersteuning) in 2007 may have played a role in the support from social security services. Under this act, municipalities are responsible for supporting the self-reliance and participation of people with disabilities. This support includes the coordination of services and the provision of a Personal Budget (persoonsgebonden budget), which is intended to enable more effective tailor-made care for patients. As the implementation of this far-reaching act was not without its difficulties, it may well be that the positive effects appeared only after a few years. These explanations are in line with the findings by Wynia et al in their regional Dutch MS cohort, as in the years 2004–2009 professional care, including DMD treatments, and assistance were markedly less available,²⁷ and the Social Support Act only started to be implemented. Moreover, a 10-year population-based study (2015) of 121 PwMS from the Stockholm County Council computerized register appears to confirm the supposed role of the Dutch Social Security Act and Personal Budget in our cohort. For, the Stockholm study concluded that the implementation of person-centred care could be a strategy to improve the efficacy and outcomes of challenging MS care.³⁰

The 10-year increase in Limitations in Basic Movement Activities and Limitations in Activities of Daily Living are most likely long-term effects of disease activity. However, age is also an important determinant of disability progression in MS. Trojano et al suggested that an age-adjusted progression index may be useful when comparing progression between MS groups.³¹ In the present study, the potential effect of aging was addressed within the mixed-model framework by including baseline age in adjusted sensitivity analyses and by testing time-by-age category interactions. For MSIP disability outcomes, no statistically significant age-by-time interactions were observed at the predefined threshold of p <0.001, suggesting no strong evidence that patient-reported disability trajectories differed by age group. As to the MSQoL-54 physical score, although trajectories differed by age category, disease duration, and baseline EDSS category, no individual subgroup showed a statistically significant change from baseline. Therefore, while we did not calculate a separate age-adjusted progression index, the role of age was addressed through age-adjusted sensitivity and interaction analyses.

The degree of the increase in Limitations in Basic Movement Activities and Limitations in Activities of Daily Living observed by us indicates that these changes were clinically relevant and therefore qualify as worsening. The patient-reported long-term decline in these domains is consistent with the changes in walking observed in an analysis (2018) of 10-year prospective data from 155 PwMS who attended the MS Center at Karolinska University Hospital, Stockholm,³² and another report (2013) from the 10-year Stockholm County Council cohort (n=166), mentioned above, stating that the proportion of patients with disability in walking and activities of daily living had increased.⁴ On the other hand, the apparently benign course of non-motor and non-activities of daily living-disabilities in our cohort contrasts not only with the earlier Dutch study by Wynia et al but also with other long-term patient-reported data from the pre-DMD era.^33,34

Neither the improved disability due to Lack of Support from Environmental Factors nor the worsened disability due to Limitations in Activities of Daily Living were matched by corresponding changes in the perception of these disabilities. Likewise, we observed no changes in physical or mental HRQoL over time. We believe that in these outcomes a response shift may have occurred because of adaptation to the disease.^35,36 Response shift in PROs is a well-known phenomenon that has also been reported in patients who actively experience symptoms on a consistent basis.¹² A response shift is thought to reflect, among others, an adjustment to the disease.³⁶ Factors that influence how one adjusts to a chronic health condition include contextual factors (demographics, dimensions of illness), personality variables (resilience, vulnerability), dyadic coping and social environmental factors; the latter include in particular social integration and social support.³⁷ In a meta-ethnographic synthesis of qualitative studies, Meek et al (2021) found that the success of adjustment to secondary progressive MS was influenced by a patient’s primary coping mechanism; coping mechanisms being determined by pre-existing individual differences, as well as social support networks and activities.³⁸ The authors state that services should ensure that PwMS are provided with informational support about their illness progression, and emotional support concerning coping strategies, social networks, and physical activity, as these are key determinants of successful adjustment.³⁸ Hence, we think that the improved support that our patients received from social security services and professional care / assistance, may have enhanced their adjustment to the disease, and thus contributed to a response shift.^37,38 Nevertheless, the increasing availability of high-efficacy DMDs throughout the study period may have also contributed to a stable HRQoL, as these drugs have a greater impact on HRQoL than low-efficacy DMDs.¹²

The strength of our study is that it provides long-term data on a broad range of disabilities and HRQoL in PwMS in the Netherlands. We believe that our findings can be generalized to the entire Dutch MS population for two reasons. First, an analysis showed that patients from all over the country were recruited and broadly distributed according to the population density. Second, additional analyses indicated that at the 5, 8 and 10-year follow-up, the demographic and disease characteristics of patients who were included in the respective analyses did not differ from those who had discontinued participation (p>0.05), regarding age, sex, disease duration, EDSS category, disease course and MSIP and MSQoL-54 baseline values, except for age in the 8-year group.

Our study has some limitations. First, there was an attrition of study participation over time, resulting in 8-year data from 35.7% of the participants who had reached this time point, and data of 32.3% of the participants who had reached the 10-year time point. This degree of attrition is partly related to the study design, in which study participation was not actively integrated into patient care, and two (very) long questionnaires were used. It is also conceivable that there was a relatively higher dropout of patients who experienced a worsening, notably of cognitive, visual or hand function, or fatigue, as these persons may have had more difficulties in filling in the questionnaires; and this may have resulted in an overestimation of the study outcomes. Second, the MS diagnosis and clinical disease course were patient-reported and were not verified. By giving the option “I don’t know” we tried to minimize false classifications of the disease course. Importantly, other observational studies with voluntary patient-driven participation, such as the North American Research Committee on Multiple Sclerosis registry, have obtained valuable information such as characterizing disease progression over time and establishing the comparative effectiveness of different disease-modifying therapies.³⁹

Conclusion

This 10-year prospective observational study in PwMS in the Netherlands found worsened patient-reported disabilities related to activities of daily living and basic movement activities, as well as an improved disability related to support from professional care, professional assistance, and social security services (environmental factors). Other disabilities and symptoms, and mental and physical HRQoL had remained unchanged.

We assume that in the Netherlands in the study period the increase in support from professional care and assistance, and from social security services may have led to more effective (high-efficacy DMDs) treatment and more personalized and coordinated care. The improvement of support from environmental factors occurred between 2011 and 2014 and may be related to the nation-wide implementation of the Social Support Act with Personal Budgets. In addition to these circumstances adjustment to the disease may also have played a role in the stabilization of most patient-reported disabilities over time, their perceptions, and mental and physical HRQoL. In all, this study provides important information on population level about the long-term changes in disabilities and HRQoL in PwMS in the Netherlands in the era of (high) efficacy DMSs. Its findings may be useful to policymakers, healthcare providers, and patient organizations, for and foremost to ensure that any decrease in patient support is prevented.

Funding Statement

This study was funded by the DNMSF, Rotterdam, Curavista bv, Raamsdonksveer, and MS4 Research Institute, Maastricht, the Netherlands. The DNMSF contributed to the recruitment of PwMS by providing information about the study on its website and sending an information brochure to its donors as an attachment to its journal and mailings.

Abbreviations

CI, confidence interval; CIS, clinically isolated syndrome; DMD, disease-modifying drug; DNMSF, Dutch National Multiple Sclerosis Foundation; e-CRF, electronic case report form; EDSS, expanded disability status scale; EU, European union; HRQoL, health-related quality of life; ICD, international classification of diseases; ICF, international classification of functioning, disability and health; M, month; min., minimum; max., maximum; MS, multiple sclerosis; MSIP, multiple sclerosis impact profile; MSVN, Multiple Sclerose Vereniging Nederland; PPMS, primary progressive multiple sclerosis; PROM, patient-reported outcome measure; PwMS, person with MS; RRMS, relapsing remitting multiple sclerosis; SD, standard deviation; SPMS, secondary progressive multiple sclerosis; VPN, virtual private network; WHO-FIC, World Health Organization international family of classifications; XML, extensible markup language.

Data Sharing Statement

Data supporting the results reported in the manuscript are available upon request.

Disclosure

A previous version of the abstract by Kremer et al was part of the ECTRIMS 2022 conference proceedings (P458) and was published in the Multiple Sclerosis Journal 2022; 28: (3S) 451-452. Ayshe Yaylali, Ingrid Kremer, Ghislaine van Mastrigt, Jitse van Dijk, and Lloyd Brandts have no competing interests. Esther van Noort is the co-owner and director of Curavista bv. Peter Joseph Jongen is the founding-director of the MS4 Research Institute.