Physical Activity and Sedentary Behaviour In People With Myasthenia Gravis: A Cross-Sectional Study

Background Despite improvements in the medical management of myasthenia gravis (MG) in recent years, patients continue to report poor health and wellbeing outcomes such as high levels of fatigue, reduced quality of life, walking limitation and reduced balance condence. Physical activity has been shown to be associated with these outcomes in other populations, however there has been limited research in adults with MG. The primary aim of this study was to describe physical activity and sedentary behaviour in adults with MG and the secondary aim was to explore associations between these behaviours and fatigue, quality of life, balance condence and walking limitation. Methods A self-report online survey was used to assess physical activity, sedentary behaviour, fatigue, quality of life, balance condence and walking limitation in 85 community-dwelling adults with MG. Multiple linear regression was used to examine associations between these variables and descriptive statistics were used to analyse participant characteristics, physical activity and sedentary behaviour. Results Most participants (n=53, 62.4%) reported sucient physical activity to meet public health guidelines, however participants also reported an average of 9 h/day of sedentary behaviour (mean 8.9±3.5). Physical activity and fatigue (R 2 =0.196), quality of life (R 2 =0.330), walking limitation (R 2 =0.305) and balance condence (R 2 =0.304) were signicantly (p<0.05) and positively correlated, with no associations found for sedentary behaviour. When patterns of physical activity and sedentary behaviour were combined, greater fatigue (R 2 =0.213), lower quality of life (R 2 =0.364), increased walking limitation (R 2 =0.341) and lower balance condence (R 2 =0.279) was observed in patients who had greater physical activity levels and lower sedentary time. Conclusions Physical activity and sedentary behaviour is associated with favorable health and wellbeing outcomes in adults with MG. Findings highlight that physical activity and sedentary behaviour may be possible intervention targets, however longitudinal and intervention studies are needed to determine causality. assessed for normality using the Shapiro-Wilk test. Participant characteristics, physical activity (METmin.wk − 1 ) and sedentary behaviour (h/day) and clinical characteristics were analysed using descriptive statistics (mean and standard deviation for normally distributed variables, median and interquartile range for non-normally distributed variables and proportions for categorical variables).


Abstract
Background Despite improvements in the medical management of myasthenia gravis (MG) in recent years, patients continue to report poor health and wellbeing outcomes such as high levels of fatigue, reduced quality of life, walking limitation and reduced balance con dence. Physical activity has been shown to be associated with these outcomes in other populations, however there has been limited research in adults with MG. The primary aim of this study was to describe physical activity and sedentary behaviour in adults with MG and the secondary aim was to explore associations between these behaviours and fatigue, quality of life, balance con dence and walking limitation.
Methods A self-report online survey was used to assess physical activity, sedentary behaviour, fatigue, quality of life, balance con dence and walking limitation in 85 community-dwelling adults with MG. Multiple linear regression was used to examine associations between these variables and descriptive statistics were used to analyse participant characteristics, physical activity and sedentary behaviour.
Results Most participants (n=53, 62.4%) reported su cient physical activity to meet public health guidelines, however participants also reported an average of 9 h/day of sedentary behaviour (mean 8.9±3.5). Physical activity and fatigue (R 2 =0.196), quality of life (R 2 =0.330), walking limitation (R 2 =0.305) and balance con dence (R 2 =0.304) were signi cantly (p<0.05) and positively correlated, with no associations found for sedentary behaviour. When patterns of physical activity and sedentary behaviour were combined, greater fatigue (R 2 =0.213), lower quality of life (R 2 =0.364), increased walking limitation (R 2 =0.341) and lower balance con dence (R 2 =0.279) was observed in patients who had greater physical activity levels and lower sedentary time.
Conclusions Physical activity and sedentary behaviour is associated with favorable health and wellbeing outcomes in adults with MG. Findings highlight that physical activity and sedentary behaviour may be possible intervention targets, however longitudinal and intervention studies are needed to determine causality.

Background
Myasthenia gravis (MG) is a chronic autoimmune neuromuscular disease involving dysfunction at the neuromuscular junction resulting in fatigue and muscle weakness. It can present as either one of or a combination of ocular and/or generalised weakness (1). Despite increasing life expectancy and availability of treatments (2), in people with MG, self-reported quality of life remains low (3), fatigue remains high (4) and overall life expectancy is still lower than the general population, however few deaths are directly attributed to the disease itself (5). Due to the pathophysiology of the disease, there has been uncertainty about whether strenuous physical activities will exacerbate weakness and induce exertional fatigue and consequently advice regarding physical activity engagement has historically been conservative (6). More recent evidence has demonstrated that balance (7), strength and aerobic exercises (8,9) are safe and effective in people with MG and have been shown to positively affect disease processes (10). Given the established evidence for the positive effect of exercise and physical activity on health outcomes and all-cause mortality in healthy populations (11), physical activity in people with MG warrants further investigation. Physical activity and time spent sedentary (activities performed while awake in a sitting or lying position with an energy expenditure of < 1.5 METs; metabolic equivalents) (12), are important determinants of health. Evidence supports that higher levels of physical activity and lower time spent sitting is associated with reduced mortality, reduced occurrence of non-communicable diseases and improved quality of life (13). Public health guidelines recommend a minimum of 150 minutes per week of moderate-to-vigorous physical activity with minimising time in prolonged sitting in order to attain health bene ts (14). A recent meta-analysis of over 1 million participants has con rmed that the combination of these behaviours are important, demonstrating a dose-response relationship between physical activity and sedentary time, with increasing levels of physical activity mitigating the effects of sedentary behaviour on all-cause mortality (15).
People with diseases that compromise physiological capacity to move (such as in MG) require special consideration as to how best facilitate physical activity and minimise sedentary time. Impairments known to be present in MG such as fatigue, weakness (4,(16)(17)(18) and reduced exercise tolerance due to lowered respiratory capacity (19) can reduce movement capacity and subsequently lead to a cycle of deconditioning. As movement becomes more di cult, aversion to movement can increase and in turn physiological effects of inactivity compound, leading to further di culty with movement (20). This can subsequently increase the risk of non-communicable diseases and premature mortality associated with inactivity (21). This is an important consideration given the lower life expectancy of people with MG compared to the general population, with cause of death often due to factors shown to be modi able by physical activity (22) such as cancer and cardiovascular disease (23). Concurrently, the pharmaceutical management of individuals with MG places them at risk of further health complications that are also able to be mitigated with exercise, such as osteoporosis (24).
To date, there has been limited research investigating physical activity and sedentary behaviour in people with MG. One study (n = 27) has explored physical activity, sedentary behaviour and associations with disease severity in MG (25). Using accelerometry, this study found that most participants (n = 13, 78%) with MG met current physical activity recommendations, however they had signi cantly lower levels of physical activity than age-corresponding healthy individuals (based on World Health Organisation data) and that participants spent an average of 78% of the day sedentary. No associations were found between physical activity and disease severity. However, this study had a small sample size (n = 27) and possible recruitment bias reducing generalisability given the study led into a training intervention. The relationship between physical activity and sedentary behaviour and health and wellbeing outcomes that are commonly reported in individuals with MG, such as higher fatigue (4), reduced quality of life (3), walking limitation (26) and reduced balance con dence (27), are of interest given the demonstrated relationships in other neurological populations; research evidence in other progressive neurological groups have established positive associations between physical activity and fatigue (28,29), depression (28) and selfreported quality of life (28,29). Therefore, the aim of this cross-sectional study was to describe the patterns of self-reported physical activity and sedentary behaviour in adults with MG and to explore associations between these healthrelated behaviours and fatigue, quality of life, walking limitation and balance con dence.

Study design
A cross-sectional design using a self-report electronic survey was used to evaluate physical activity and sedentary behavior in people with MG and the relationship with fatigue, quality of life, walking limitation and balance con dence. Ethical approval was provided by The University of Queensland Human Research Ethics Committee (reference number HREC/2019/000/774).

Participants
To be eligible for the study, participants needed to be aged between 18-80 years, able to provide informed consent, be independently ambulant with or without mobility aids and have stable self-reported MG symptoms for at least four weeks preceding participation. Community-dwelling adults who reported having MG were recruited via convenience sampling using yers and social media advertising through a disease speci c support group. Advertisements included a website link for the electronic online survey. The link rst included information about the study, eligibility criteria and contact details of the research team and potential participants were required to indicate that they met eligibility criteria and provide informed consent prior to commencing the survey.

Scales and questionnaires
Participants were asked to report demographic characteristics including age, gender, height, weight, employment, relationship status, disease duration, past medical history, treatments for MG and other medication and comorbidities.
Physical activity was measured using the Active Australia Survey (30), a self-report measure validated in adult populations (31) that asked participants to report the amount of time spent in various types and intensities of activity such as walking (including for recreation and/or transport), gardening and moderate and vigorous-intensity activity in the last week. From these values, an overall MET.min.wk − 1 score was determined using the following formula: 3.33 METs for brisk walking and moderate-intensity leisure activities and 6.66 METs for vigorous intensity leisure activities (32,33). Physical activity was then classi ed into two subcategories: insu ciently active (0 to 499 METmin.wk − 1 ) or su ciently active according to physical activity guidelines (≥ 500 METminutes.wk − 1 )(34).
Sedentary behaviour was measured using the Past-day Recall of Sedentary Time (PAST) (35), a selfreport measure of sedentary behaviour validated in adult populations (35,36). Participants were asked to report the number of minutes spent sitting or lying in various domains throughout the waking hours of the previous day (such as when working, for transport, while watching television and reading). Total sedentary time was calculated by summing the time reported in each of the domains of sedentary behaviour. The total time was then classi ed into two categories; <10 and ≥ 10 hours of sedentary time per day, given > 9.5 hours has been reported in the literature as associated with higher risk of early mortality (37) and adverse health outcomes (38).
Fatigue was measured using the Modi ed Fatigue Impact Scale (MFIS) (39), a 21-item self-report measure used to assess perceived impact of fatigue on quality of life in physical, social and psychosocial domains. There are no reliability or validity studies for this scale in MG speci cally, however the measure has been shown to be a reliable and valid measure of fatigue in other neurological populations (40). A higher score indicates a greater impact of fatigue on a persons' daily life.
Quality of life was measured using the Myasthenia Gravis Quality of Life 15-item (MG-QoL-15) (41) a valid survey assessing self-reported impact of disease-speci c signs and symptoms on quality of life (42), with a higher score indicating poorer perceived quality of life.
Walking limitation was assessed using the Walk-12G (43), a 12-item survey that assesses self-reported di culties with walking in everyday life. Validation has occurred in other neurological conditions (43), with a higher score indicating greater limitation of walking.
Balance con dence was assessed using the Activities-speci c Balance Con dence Scale (ABC Scale) (44), a 16 item self-reported measure of perceived con dence in balance when mobilising and performing a number of everyday tasks. This scale has been validated in neurological populations (45) with higher scores indicating greater balance con dence.

Data analysis
Statistical analyses were performed using SPSS v25 (IBM Corporation). Data and model residuals were assessed for normality using the Shapiro-Wilk test. Participant characteristics, physical activity (METmin.wk − 1 ) and sedentary behaviour (h/day) and clinical characteristics were analysed using descriptive statistics (mean and standard deviation for normally distributed variables, median and interquartile range for non-normally distributed variables and proportions for categorical variables). Associations between study variables (physical activity; su ciently active (≥ 500METmin.wk-1) and insu ciently active (< 500 MET.min.wk-1) and sedentary behaviour (≥ 10 hours of sitting per day; <10 hours of sitting per day) and fatigue, quality of life, walking limitation and balance con dence were assessed using multiple linear regression. To examine associations between patterns of physical activity and sedentary behaviour and fatigue, quality of life, walking limitation and balance con dence, multiple linear regression was performed using 'su cient physical activity and < 10 hours/day sitting' as the reference category. All models were adjusted for age, body mass index and disease duration and unstandardized regression coe cients are presented. Level of signi cance was set at p < 0.05. Apriori sample size calculations estimated that a minimum sample size of n = 36 was required based on previously reported correlations between fatigue and physical activity in multiple sclerosis (r = -0.45; n = 36) (29).

Results
Participant demographics and clinical characteristics are listed in Table 1. Eighty-ve participants with an average age of 48 ± 15.52 completed the survey. Seventy-four percent (n = 63) were female, with most participants reporting generalized (n = 82, 96.5%) versus ocular-only symptoms (n = 3, 3.5%). Over half of the participants (n = 48, 56.5%) reported full or part-time work or study, with 44 (51.8%) participants reporting they had changed their employment and/or study workload as a result of MG. 87% (n = 74) of participants were ambulant without a mobility aid.  A score out of 84, with higher scores indicating greater impact of fatigue on a person's activities † A score out of 60, with higher scores indicating poorer perceived quality of life ‡ A score out of 42, with higher score indicating greater limitation to walking § A percentage out of 100 of perceived balance con dence, with a higher score indicating greater con dence in balance Associations between physical activity and sedentary behaviour and fatigue, quality of life, walking limitation and balance con dence are presented in Table 3. Physical activity was signi cantly negatively associated with fatigue, quality of life and walking limitation, with su ciently active participants reporting reduced fatigue, higher quality of life and better walking compared to their less active counterparts. Being su ciently active was also signi cantly positively associated with balance con dence, with su ciently active participants reporting higher balance con dence. Sedentary behaviour was not signi cantly associated with any of the outcomes. R 2 values ranged from 0.196-0.330. Assessed using the Activities-Balance Con dence Scale, scored as a percentage out of 100 of perceived balance con dence, with a higher score indicating greater con dence in balance Associations between patterns of both physical activity and sedentary behaviour and fatigue, quality of life, walking limitation and balance con dence are reported in Table 4. Compared with the reference group (those who were su ciently active and spent < 10 h/day sitting), fatigue was signi cantly higher in the insu ciently active groups for both < 10 hours and ≥ 10 hours sitting. Quality of life was signi cantly lower across all groups compared to the reference group and in the su ciently active and ≥ 10 h/day sitting group, as was walking limitation. Balance con dence was reduced among those in the insu ciently active and ≥ 10 hours sitting group only when compared to the reference group. Generally, beta-coe cients were larger in those who were inactive and sat more compared to those that were inactive and sat less, indicating that sedentary behaviour alone may have an important effect on these outcomes irrespective of physical activity. All models were adjusted for age, disease duration and body mass index * Assessed using the Modi ed Fatigue Impact Scale, scored out of 84 with a higher score indicating greater impact of fatigue on a person's daily life † Assessed using the Myasthenia Gravis Quality of Life 15-Item, scored out of 60 with a higher score indicating poorer perceived quality of life ‡ Assessed using the Walk-12G, scored out of 42 with a higher score indicating greater limitation to walking § Assessed using the Activities-Balance Con dence Scale, scored as a percentage out of 100 of perceived balance con dence, with a higher score indicating greater con dence in balance

Discussion
The aims of this study were to describe the patterns of self-reported physical activity and sedentary behaviour in people with MG and to explore associations between these health-related behaviours and fatigue, quality of life, walking limitation and balance con dence. The ndings of this study indicate that despite over two-thirds of the participants meeting physical activity guidelines, over a quarter also spend over 10 hours sitting per day. Further, participants largely met guidelines through accruing moderateintensity physical activity, with less than half of all participants engaging in any vigorous activity at all. There were signi cant associations found between meeting physical activity guidelines and less fatigue, greater quality of life, less walking limitation and greater balance con dence, while conversely no associations were found between these variables and sedentary behaviour alone. However there were also signi cant dose-response relationships between patterns of physical activity and sedentary behaviour, with those that were inactive but accruing < 10 h/day of sedentary time signi cantly associated with less fatigue, greater quality of life, less walking limitation and greater balance con dence compared to those that are inactive but reported sitting ≥ 10 h/day.
The Australian Physical Activity Guidelines recommend 150 minutes per week of moderate-vigorous physical activity in order to attain signi cant health bene ts (46), which equates to 500 METminutes.wk − 1 . Two-thirds of the participants in our study were classi ed as su ciently active, however less than half of this su ciently active group engaged in any vigorous physical activity. This is slightly lower than a previous study investigating physical activity and sedentary behaviour in 27 adults with MG (mean age 62 ± 16 years), which reported that 78% of participants reached physical activity guidelines, however only 11% engaged in ≥ 20 min/day of vigorous activity for a minimum of three days per week (25). It is important to note that this previous study used device-based measures of physical activity, whereas the current student utilised self-report which typically results in higher estimates of physical activity (47).
There is a possible recruitment bias as the cross-sectional study was a part of a larger study that included a training opportunity for participants. Subsequently, those recruited may have had an interest in physical activity that in uenced the cross-sectional data. Given the additional health bene ts of vigorous physical activity (48), further investigation into the barriers to vigorous activity in this population is warranted.
Regarding sedentary behaviour, similar ndings were reported in both our own and the aforementioned study, with high levels of sedentary behaviour in both (25). Our ndings indicate the median daily sedentary time among participants was approximately 9 h/day (8.98 ± 3.53), 2.5 hours more than found in the general adult population in one study (n = 3,699), where average self-report sedentary behaviour was 6.5 h/day (95% CI 6.2-6.7)(49). This is also higher than ndings in other neurological populations; one study in adults with multiple sclerosis reported average self-report sedentary time of 7.5 ± 3.7 h/day (n = 1081) (50). The medical advice and bene ts of exercises in multiple sclerosis has evolved greatly over the past two decades and the overall lower sedentary time for this group may be re ective of this. Given the evidence for higher sedentary time and poor health outcomes including increased mortality (51), further interventional research to reduce sedentary time in people with MG is needed to reduce these risks.
In the present study, signi cant associations were found between physical activity and fatigue, quality of life, walking limitation and balance con dence. This is the rst study, to our knowledge, that has explored this in people with MG. In other populations physical activity interventions have shown positive effects on fatigue (52)(53)(54), quality of life (55), walking ability (55,56) and balance (56). Pathologies associated with impaired movement such as in MG may result in reduced capacity for physical activity and more sedentary behaviour, thereby risking physiological effects from disuse. The positive bene ts in the above studies may be a direct consequence of breaking this disuse cycle, restoring the physiological de cits from disuse and/or having an impact on the primary symptoms of the pathologies (20). At this stage for people with MG it is unclear whether the participants who engage in more physical activity, do because they are less fatigued, have better quality of life, less walking limitation and/or greater balance con dence and thus able to break this disuse cycle, or the other way around. Longitudinal research is warranted to investigate these relationships further and determine direction of associations and treatment impacts. Nonetheless, these preliminary ndings are of important clinical consideration given the historical concerns of increased activity leading to provocation of symptoms in this population. Overall, our ndings support the growing body or literature that higher levels of exercise and physical activity do not appear to be associated with negative outcomes in people with MG and may in fact have similar positive outcomes as seen in healthy populations. However, despite the statistical signi cance, R 2 values in our study are low and so our ndings should be interpreted with caution.
Although there were no associations found between sedentary behaviour alone and the outcomes assessed in this study, there were signi cant associations with these variables when patterns of sedentary behaviour were analysed in combination with su cient versus insu cient physical activity.
Our ndings indicate that even in those who are inactive, lower levels of sedentary time (< 10 h/day) are associated with less fatigue, higher quality of life, less walking limitation and greater balance con dence compared to those with higher sedentary time (≥ 10 h/day). This is a promising nding, potentially suggesting that sedentary behaviour may be an appropriate intervention target even when physical activity levels are low. The potential protective effect of physical activity against the detriments of sedentary behaviour is another consideration in this population. A recent meta-analysis found that the increased mortality risk associated with high sedentary time appears to be mitigated with increasing physical activity levels and even reversed in those who are in the highest quartile for physical activity (> 2130 METminutes.wk − 1 ) (15). These ndings indicate that changing sedentary time has a positive effect on mortality irrespective of physical activity levels, and so warrants further investigation as to a potential target intervention for people with MG, given that the ndings of our study indicate high levels of sedentary behaviour in this population. As these bene ts are most impactful when combined with higher levels of physical activity, interventions targeting whole of day activity and thus incorporating both physical activity and sedentary behaviour may be of bene t in improving health and well-being outcomes in adults with MG.
The ndings of this study should be considered in light of a number of limitations. The cross-sectional design gives insight into these behaviours at only one point in time and therefore limits the ability to determine causality. Further longitudinal and/or interventional research is needed to determine direction of associations and e cacy, respectively. Most of our participants reported generalised (n = 92, 96.5%) compared with ocular-only symptoms (n = 3, 3.5%); an underrepresentation of people with symptoms limited to the ocular muscles. Ocular-only MG is estimated to occur in approximately 17% of MG cases and around half will then progress to generalised symptoms (57). Despite symptoms only affecting the ocular muscles, associated symptoms such as diplopia are common in this type of MG (3) and can impact on function in daily tasks (58), and as such we opted to include both ocular and generalised in the analysis.
Self-report measures were utilised for both physical activity and sedentary behaviour, which can lead to over-reporting by participants (47) and be inaccurate when compared with objective device-based monitoring (59). Additionally, these measures were not speci c to MG as no such MG speci c measures of physical activity or sedentary behaviour are currently available. Despite limitations in the use of selfreport measures, a strength of this type of measure is the ease of distribution which facilitated a larger sample size for our study (n = 85) than what has been used in prior related research (n = 27) (25). Given the low incidence of MG and subsequent recruitment di culties, using an electronic self-report survey enabled recruitment across a larger geographical area. This method, however, limited our capacity to classify disease severity amongst participants given the lack of face to face clinical assessment and absence of a single universally acceptable classi cation of severity in this population (60). Prior research, although limited, suggests there is no relationship between disease severity and physical activity in this population (25), however further high-quality evidence is required to investigate this further. Finally, the R 2 values of our models was generally low, indicating that there may be a number of other variables contributing to these relationships.

Conclusions
Although physical activity guidelines were met by most participants with MG in our study, a large proportion of participants also reported high levels of sedentary behaviour. Our ndings provide preliminary evidence that physical activity and sedentary behaviour may be potential intervention targets given our ndings of associations between higher physical activity levels and lower sedentary time and less fatigue, higher quality of life, lower walking limitation and higher balance con dence. There is a need for high quality, large sample size studies in individuals with MG that can further investigate and con rm the direction of the relationship between these variables.

Consent for publication
Not applicable Availability of data and materials The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Competing interests
The authors declare that they have no competing interests.

Funding
There was no funding associated with this study.

Authors contributions
All authors contributed to the study conception and design. Material preparation and data collection was performed by Tahlia Alsop, and analysis were performed by all authors. The rst draft of the manuscript was written by Tahlia Alsop and all authors commented on previous versions of the manuscript. All authors read and approved the nal manuscript.