Physical Activity in Multiple Sclerosis: Real-World Data From Saskatchewan, Canada

Physical activity in people with multiple sclerosis (MS) has been associated with improved function,¹  participation,²  and quality of life,³  as well as an estimated 27% relative reduction in MS relapse rates,⁴  and neuroprotective effects.⁵  Comprehensive MS care recommendations include supporting physical activity, symptom management, rehabilitation, and disease-modifying drug treatments (DMTs).⁶  The first guidelines for physical activity in MS were published in 2013 for people with mild to moderate disability.⁷  In 2020, updated guidelines included recommendations for physical activity from earliest onset of MS through advanced MS.⁸  Exercise in combination with DMTs is an approach proposed to further improve health and disability outcomes in MS.⁹  DMT recommendations emphasize the importance of early treatment for preventing relapses, controlling the inflammatory aspects of MS, and delaying early disability progression.^10,11  However, high-level, quality evidence concerning the mechanistic effects of physical activity levels over the course of MS is needed.^12,13  Despite recent advocacy and research interest in physical activity,¹⁴  there is limited real-world data concerning physical activity levels in people with MS.

The primary objective of this study was to determine if physical activity levels have changed over time among people living with MS who were newly registered with the Saskatchewan MS Drugs Research Program¹⁵  between September 1, 2014, and February 22, 2022. The secondary objective was to explore associations and interactions between physical activity and DMT exposure, sex, age, age at MS onset, and MS disease duration.

Saskatchewan has a population of 1.2 million and a high prevalence of MS.¹⁶  In Canada, provincial/territorial governments are responsible for health services, including provincial drug plans.¹⁷  In 1997, the Saskatchewan MS Drugs Research Program began providing administrative and educational services for DMTs as well as research opportunities. The program implemented a voluntary prospective process to assess self-reported physical activity levels on September 1, 2014.

For this study, physical activity levels were evaluated by a one-time mailed questionnaire that was sent when a person first applied to the Saskatchewan MS Drugs Research Program. All individuals who initiated registration between September 1, 2014, and February 22, 2022, received a questionnaire. The study was approved by the University of Saskatchewan Biomedical Research Ethics Board (BEH 14-86).

During the study period, new program applicants were invited to participate if they had clinically definite MS and were ambulatory. Participants were either on a DMT funded through other sources (ie, private insurance or compassionate industry coverage) or they were DMT naive. Eligible participants were mailed a study invitation letter, a consent form, a Godin Leisure-Time Exercise Questionnaire (GLTEQ), and a self-addressed, postage-paid return envelope. They were encouraged to complete and return the GLTEQ within 1 week of receipt. Participation was voluntary and had no bearing on eligibility for DMT funding coverage through the program. All completed GLTEQs received prior to February 22, 2022, were included in our analysis. Participant demographics and DMT history were extracted from the program application, which includes information about MS history and DMT treatment history.

The GLTEQ is a self-reported, validated measure of physical activity in ambulatory people living with MS.^18,19  It provides an estimate of the metabolic output from physical activity in the proceeding week.²⁰  This study aimed to evaluate the change over time in physical activity levels of new registrants according to the total physical activity (TPA) and health contribution (HC) scores on the GLTEQs. The TPA and HC scores are both derived from the GLTEQ and calculated in arbitrary units according to established methods.^18,19  Higher TPA and HC scores indicate higher levels of physical activity. To complete the GLTEQ, participants fill in the number of self-reported 15-minute bouts of strenuous, moderate, and mild intensity physical activity in the preceding week. The HC score includes only the number of strenuous and moderate 15-minute bouts of physical activity multiplied by their respective metabolic equivalents. The TPA score includes the addition of the bouts of mild intensity physical activity. From the HC score, classification of an individual's activity level for substantial health benefits is defined categorically as either sufficiently active (score ≥ 24 units), moderately active (score of 14-23 units), or insufficiently active (score < 14 units).

Participant demographics included sex, age at MS onset, MS disease duration (defined as time from first MS symptom onset), and age and DMT treatment history at the time of questionnaire completion. Available data for DMT history included drug names with start and stop dates estimated to the nearest month. DMT exposure was divided into 2 broad categories: exposed if a person was currently on or was ever previously on a DMT, or naive if they were never previously exposed to a DMT at the time of questionnaire completion. Exposure time among those exposed to DMT was the total number of months on any DMT rounded up to the nearest month. An additional 12 months of exposure was added for alemtuzumab and 6 months for ocrelizumab from the date of last dose administration, as these are the respective drug dosing intervals. In addition, DMT exposure history among those exposed was further classified into 1 of the following 4 categories: stopped any prior DMT, currently on a first-line DMT for relapsing-remitting MS, currently on a second-line DMT for relapsing-remitting MS, or on a DMT for primary progressive MS. DMT classification was according to Saskatchewan MS Drugs Program criteria: first-line DMTs (interferon beta 1a, interferon beta 1b, glatiramer acetate, dimethyl fumarate, teriflunomide, or ocrelizumab); second-line DMTs (natalizumab, fingolimod, alemtuzumab, or cladribine), and ocrelizumab for primary progressive MS.¹⁵  Participants on second-line DMTs would have had prior exposure to a first-line DMT in order to meet funding eligibility for a second-line DMT through the program.

Survey response rate was the number of completed, returned questionnaires divided by the number of invitations sent. Descriptive analyses characterized the study population. The change in TPA and HC scores over time was explored using scatter plots with fitted linear regression. Since there were fewer than 8 weeks of data collection in 2022, those GLTEQ scores were plotted together with the 2021 calendar year data. Change in the percentage of participants annually who were sufficiently, moderately, or insufficiently active for substantial health benefits were examined using the χ² test. Univariate, followed by multivariable, linear regression analyses examined the associations between independent variables (demographics and DMT history; TABLE 1) and TPA and HC scores. Variables significantly associated with TPA and HC scores in the univariate analysis (P < .05) were included in the multivariable regression model. Variables with P values less than .05 after a stepwise backward elimination were retained in the final multivariable regression model. Collinearity, confounding, and interaction effects among independent variables in the final multivariable model were also explored. Statistical tests were significant at a level of P values less than .05 (2-sided α = .05). SAS v9.4 software was used for data analysis.

A total of 993 surveys were sent with a response rate of 50.8% (n = 505). Median MS disease duration was 6.0 years (IQR, 2.0-14.0; Table 1); 37.8% (n = 191) had an MS disease duration of less than 3 years. Men were older in general (44.4 years vs 42.1 years; P = .04) and older at MS onset (35.3 years vs 32.7 years, P = .01) than women. There were no differences in MS disease duration between the sexes (women: median duration = 7.0 years, IQR, 2.0-14.0 years; men: median duration = 5.0, IQR, 2.0-14.0 years; P = .54).

Over approximately 7 years, physical activity levels according the GLTEQ TPA or HC scores improved by a small, nonsignificant amount (FIGURE 1A, 1B). The TPA scores increased by 0.71 units per year (P = .26) and the HC scores increased by 0.61 units per year (P =. 21). There was no significant change over 7 years in the percentage of participants considered sufficiently, moderately, or insufficiently active for substantial health benefits over time (P = .07), despite the observed fluctuations (FIGURE 2).

Longer time on DMT (if exposed to DMTs), female sex, older age at time of GLTEQ completion, older age at MS onset, and longer MS disease duration were significantly associated with lower (less active) TPA and HC scores on the univariate analysis (TABLE 2). There were no significant associations between those naive to DMT versus exposed to DMT with the TPA or HC scores (all P values > .05). The small number of participants on DMT for primary progressive MS (n = 5) and those who had discontinued MS drug treatment (n = 48) (Table 1) had lower TPA and HC scores compared with those who were treatment naive in the univariate analysis. However, in the multivariable analysis, none of the DMT independent variables remained significantly associated with physical activity levels (Table 2).

In the multivariable regression analysis, male sex was independently associated with higher physical activity levels; men scored 12 units higher on the TPA score (95% CI, 7.09-16.80; P < .001) and 6.7 units higher on the HC score (95% CI, 2.91-10.40; P < .001) compared with women (Table 2). There were also significant interaction effects for age and MS disease duration on physical activity scores. The interaction indicated that increased disease duration was associated with lower (less active) HC scores for participants below the age of 54 years and lower TPA scores for those below the age of 59 years. Increased disease duration above these ages was associated with higher activity scores.

Given the association with sex, and the interactions between age and MS disease duration and activity scores, we performed additional analyses (χ² tests and scatter plots with fitted linear regression; data not shown) to determine if participant demographics changed over the 7 plus years when questionnaires were completed. The results indicated that there was no significant change in sex, age, age of MS onset, or MS disease duration over the study period (P range, .17-.97).

Physical activity levels in people with MS newly applying to a provincial DMT program did not significantly increase over a 7-year period. Prior exposure to DMT vs being naive to DMT was not associated with physical activity levels in this sample. Men reported higher physically active levels than women, and there was an interaction between age and MS disease duration. Increased disease duration at younger ages was associated with lower physical activity levels. However, increased disease duration in older individuals was associated with higher activity levels. Overall, the majority of participants were not sufficiently active for substantial health benefits.

People living with MS are known to be less active than controls without MS²¹  and they may experience barriers to physical activity.²²  A 2023 meta-analysis of baseline accelerometer or self-reported physical activity data of people with MS participating in physical activity research suggests activity levels may be improving.^23,24  However, physical activity data from research studies may not accurately reflect real-world physical activity behavior. Of concern is a possible growing division between people already physically active for health benefits and those who are inactive at the general population level.²⁵  Participants included in the present Saskatchewan study were not recruited for exercise research at the time they were invited to complete the GLTEQ. There was no significant sustained change in the proportion of people physically active enough for substantial health benefits from 2014 to early 2022. However, the percentage of people sufficiently active for substantial health benefits ranged from 18% in 2017 to 47% for the last 4 months of 2014. Of interest, the high point in 2014 was 1 year after the publication of the first physical activity guidelines for MS.⁷  In comparison, a pooled secondary analysis of earlier physical activity accelerometer data from 2005 to 2013 from research studies conducted in the Illinois area reported that, at their baseline 20% of the participants were sufficiently active for substantial health benefits.²⁶ 

Similar to the general population, we found men to be more physically active than women.²⁷  In contrast, a Swedish MS registry study found that men were less physically active than women.²⁸  The people in the registry had a mean age of 52 years (SD, ± 14.5) and men had higher disability levels and greater mobility restrictions than women. These differences likely explain the lower activity levels in men from the Swedish registry.²⁸  In the present study from Saskatchewan, there was less discrepancy in mobility between the sexes since participants were ambulatory.

We also found interaction effects between age and disease duration on physical activity GLTEQ scores: Increased disease duration at younger ages was associated with lower activity levels. However, increased disease duration at older ages (ie, age > 54 for the HC score) was associated with higher activity levels. It is possible that older adults have more time to participate in physical activity compared with younger adults or that those with longer-lived experience with MS may prioritize physical activity. A Canadawide survey found older adults with MS to be more physically active than older adults without MS.²⁹  Another explanation for the increased activity observed in older adults may be the limitations of the GLTEQ as a self-report measure of perceived physical exertion. The GLTEQ may be capturing change in perceived or actual physical exertion levels for the same activity into older age. The age and disease duration of participants in this Saskatchewan study remained stable over the enrollment period; therefore, changes in these demographics over time are unlikely to have influenced the observed interaction effects or the physical activity GLTEQ scores.

Riemenschneider et al suggest that an uninvestigated “window of opportunity” in the earlier stages of MS exists for physical activity research.³⁰  Early intervention with DMT is critical to delay early disability progression in real-world settings.³¹  A similar window of opportunity may exist to maximize the effects of physical activity on disability outcomes by initiating exercise prior to the occurrence of irreversible degeneration.³⁰  The 2020 physical activity guidelines for MS across the disease course recommend “early evaluation by a physical or occupational therapist or exercise or sport scientist, experienced in MS, to establish an individualized exercise and/or lifestyle physical activity plan.”⁸  The guidelines provide a recommendation of 150 minutes of exercise or lifestyle physical activity per week. Progress toward these targets “should be gradual, based on the person's abilities, preferences, and safety.”⁸  Additional advanced aerobic exercise targets are described and supported by increasing evidence for safety, feasibility, and improved functional outcomes for ambulatory people with MS.⁸  A large proportion of people in our sample, irrespective of their DMT exposure history, were not meeting the minimum recommendations for health benefits. Ploughman et al propose exercise as a primer for central nervous system repair, potentially critical to the success of future remyelinating drug therapies.³²  Current DMT decision-making and monitoring involves contact with health care systems. People living with MS want information on exercise from health care providers.³³  An unrealized opportunity for earlier support for exercise may be feasible in combination with DMT services.

This study has several limitations. Only 1 physical activity questionnaire from each participant was completed at the point in time when they first applied to the provincial MS drugs program. Future research could include repeated measures to understand how physical activity may be changing at the individual level. While data collection continued without interruption through the COVID-19 pandemic, the pandemic itself may have affected the lack of change in activity levels. The physical activity uptake when the new physical activity guidelines were published in 2020 may also have been influenced by the pandemic.⁸  Further follow-up of current, real-world physical activity levels is warranted. Only a small number of participants with primary progressive MS were included in this study, as DMTs for primary progressive MS were not available for the majority of the study period. First Nations people and members of the Canadian Armed forces do not use the Saskatchewan MS Drugs Program, as their health care funding falls under federal programs. These individuals, as well as people with more advanced levels of disability, were not included in the present study. Social demographic data and other factors associated with physical activity behavior are also not captured in this study. It is possible that a change in the social or racial demographics occurred over the study period among people contacting the MS drugs program. Future research analyzing physical activity levels in real-world settings should aim to include data on race, ethnicity, and social economic status.

At first contact with a provincial MS drugs program, a majority of people with MS remain insufficiently active for substantial health benefits. However, shortly after the publication of the first MS physical activity guidelines, we observed the highest percentage of people meeting criteria for activity levels sufficient for substantial health benefits. Future research could include exploring how health care systems can help support increased physical activity levels for optimal health outcomes.

The authors thank Walter Hader Sr, MD, FRCPC, PE, for his leadership with the Saskatchewan MS Drugs Research Program from 1997 to 2013. The authors also thank Shirley MacGowan, nurse educator, and Sam David Kodamanchali and Monique Reboe Benjamin for research support.