Perceptions of Exercise and Restless Legs Syndrome: Results from a nationwide survey

Abstract

Restless legs syndrome (RLS) is a prevalent, sensorimotor sleep disorder temporarily relieved by movement, with evidence of symptomatic improvement with regular exercise. The present study describes perceptions of the effects of exercise on symptoms of RLS. Participants (N=528) completed a mixed-methods (i.e., numerical and narrative), nationwide survey including items assessing personal experiences with exercise and RLS (both positive and negative) as well as RLS diagnosis, RLS severity, and demographic and clinical characteristics.

Responses varied widely on specific experiences with exercise, but a higher percentage of participants indicated positive experiences with exercise than those who reported negative experiences (72% to 40%, respectively) with exercise. Further, 54% of respondents reported that exercise only improves RLS while 24% reported exercise only worsens symptoms. Participants described that any abrupt change in exercise routine would almost always elicit RLS symptoms (e.g., hiking for a long time, stopping an exercise routine) and that a consistent pattern of exercise improved RLS symptoms with an overall beneficial effect on the frequency of symptomatic bouts. Participants further described time of day as impactful for their exercise experience, with >50% indicating morning exercise improves symptoms and evening exercise worsens symptoms. Participants described several questions that they wanted answered regarding the evidence for exercise in RLS and specific exercise prescription recommendations. The present study describes information crucial to the creation of stakeholder-informed health promotion programs for individuals with RLS to optimize personalized treatment plans that could prevent and manage symptoms.

Introduction

Restless legs syndrome (RLS) is a sensorimotor sleep disorder that affects roughly 10% of the adult population in North America and Europe (Ohayon,O’Hara,&Vitiello,2012). It significantly disrupts sleep (Budhiraja et al.,2012) and is associated with increased risk of cardiovascular disease, hypertension, stroke (Gottlieb,Somers,Punjabi,&Winkelman,2017; Janes et al.,2021). depression and anxiety (Becker&Sharon,2014), and causes overall poorer health (Allen,Bharmal,&Calloway,2011; Ohayon et al.,2012) and decreased quality of life (Allen et al.,2005; Dodel et al.,2010). A leading theory underlying the mechanisms of RLS is central iron deficiency and dopaminergic dysfunction (Didato et al.,2020), but despite the identification of genetic variants associated with RLS (Didriksen et al.,2020; Schormair et al.,2017; Tilch et al.,2020), the pathophysiology of RLS is not well understood, hindering the development of targeted disease modifying therapies. As two key diagnostic features of RLS include the worsening of symptoms at rest and relief by movement (Allen et al.,2014), there is increasing interest in utilizing physical activity and exercise to manage symptoms.

Previous research highlights exercise as one of the most promising non-pharmacological approaches for managing symptoms of RLS (Aukerman et al.,2006; K.L. Cederberg&Motl,2016; de Mello,Lauro,Silva,&Tufik,1996; Esteves,de Mello,Benedito-Silva,&Tufik,2011; Giannaki,Hadjigeorgiou, et al.,2013; Giannaki,Sakkas, et al.,2013; Giannaki et al.,2015; Mortazavi et al.,2013; Sakkas et al.,2008). Despite current evidence for benefits, there are anecdotal reports of certain parameters of exercise (e.g., high-intensity) exacerbating symptom of RLS. However, to date, such evidence has yet to be systematically studied. Understanding which aspects of exercise (e.g., intensity, timing) benefit and/or exacerbate symptoms may improve our approach in managing symptoms. Further, the development of exercise programs and prescription recommendations for managing RLS should consider the needs of people with RLS. Indeed, including the target population in the development of an exercise programs is an imperative step in clinical and translational research that increases success of subsequent implementation efforts (Goodman&Sanders Thompson,2017).

To that end, the present study utilized a mixed-methods, nationwide survey of adults with RLS to: (1) describe personal experiences regarding exercise parameters that are perceived to benefit and/or exacerbate RLS symptoms; (2) identify perceptions of the effect of time of day on symptoms of RLS; and (3) inquire about specific questions people with RLS have about exercise that can inform the development of future exercise-based research studies and programs. This exploratory study was intended to describe experiences and perceptions of potential interactions between exercise and RLS as well as to identify what people with RLS believe to be important considerations for future research. Such results will facilitate the design of future behavioral interventions examining the effect of exercise on RLS symptoms that can identify specific exercise prescription parameters for maximizing the benefit of exercise.

Materials and Methods

Participants and Procedure

The procedure was approved by the University’s Institutional Review Board. The Restless Legs Syndrome Foundation (RLSF) distributed the Qualtrics (Provo,UT) survey to 3,644 registered members with RLS via an email notification (i.e., “eblast”) in October 2021. Persons interested in participating were instructed to proceed to the internet-based survey by clicking on the dedicated link provided in the eblast advertisement. An RLSF member who completed the survey also shared the survey link to the “RLS SUCKS! Restless Legs Syndrome – Willis-Ekbom Support Group” Facebook group, which included approximately 1,804 members at the time of distribution. Upon entry into the survey, all participants were provided with a detailed description of the survey and all participants provided electronic informed consent followed by anonymous completion of the questionnaire. All questionnaires were checked for completeness; however, the survey was conducted completely anonymously. Therefore, in the event of missing responses to quantitative items (<50% of the survey completed), a survey was considered incomplete and excluded from formal analyses. Of note, narrative questions were not required and therefore surveys were not excluded for missing narrative responses. The survey was closed to participants in June 2022, at which time 1,935(53%) of the 3,644 emails sent by the RLSF were opened with 727(38%) recorded unique clicks to proceed to the survey. We were unable to track the number of individuals in the Facebook group who completed the survey. In total, there were 587 survey responses with a total of 528 people (90%) who completed all relevant outcomes of the present study.

Survey

The survey was designed in collaboration with the Northern California Restless Legs Syndrome Support Group. The first author (KLJC) pre-tested the proposed survey items to 20 members of the group in June 2021, whereby stakeholders engaged in an active discussion and provided feedback and suggestions regarding specific survey items believed to be important to address for people with RLS. The final, distributed version of the survey included items assessing RLS diagnosis, RLS severity, personal experiences with exercise and RLS (both positive and negative), and demographic and clinical characteristics.

RLS Diagnosis.

The Cambridge-Hopkins Restless Legs Syndrome Short Form Diagnostic Questionnaire (RLS-SFDQ13)(Allen,Burchell,MacDonald,Hening,&Earley,2009) was used to assess a positive screen for RLS. A positive screen for RLS requires that participants present with the following criteria: (1) the desire to move the legs in association with uncomfortable sensations; (2) the need to move the legs in response to these sensations; (3) the worsening of sensations at rest; (4) the partial or complete relief of the urge with movement; and (5) the sensations occurring most frequently during the evening or early part of the night. All participants either had a positive screen for RLS or a previous diagnosis of RLS from a medical professional.

RLS Severity.

The Self-Administered version of the International Restless Legs Syndrome Study Group Scale (sIRLS) was used. It is a validated 10-question survey that provides a global score regarding the severity and frequency of symptoms over the previous week (Sharon et al.,2019). Overall symptom severity scores range from 0 to 40 and are determined by summing item scores, with higher scores indicating a greater severity of symptoms. Severity scores can further be categorized as no symptoms (0), and mild (scores 1–10), moderate (scores 10–20), severe (scores 20–30), and very severe (scores 30–40) symptoms.

Experiences and Perceptions of Exercise in RLS.

Seven items were included in the survey to assess participant’s personal experiences with exercise (Table 1). Questions related to positive and negative experiences with exercise were preceded by the following instructions: “The following questions ask about your own, personal experience with exercise and how exercise has affected your symptoms of RLS. Please consider your overall experience with exercise on average (not just the best or worst).” We further provided a narrative response item encouraging participants to report any questions they had specifically related to exercise and RLS.

Table 1:

Survey items examining the perception of exercise in adults with RLS.

	Survey Item	Response Options
Positive Experiences	“Have you noticed that exercise has made your RLS symptoms better?”	Definitely No Sometimes Definitely Yes
	“Exercising at what time of day makes your symptoms better at night? (Select all that apply)?”	Early morning (4AM-8AM) Mid-morning (8AM-12PM) Mid-day (12PM-4PM) Afternoon (4PM-8PM) Evening/Night (8PM-11PM) Don’t Know
	“Based on your previous experiences, please explain and/or describe the type of exercises that make your RLS symptoms better (no word limit).?”	Open-ended (Narrative)
Negative Experiences	“Have you noticed that exercise has made your RLS symptoms worse?”	Definitely No Sometimes Definitely Yes
	“Exercising at what time of day makes your symptoms worse at night? (Select all that apply)?”	Early morning (4AM-8AM) Mid-morning (8AM-12PM) Mid-day (12PM-4PM) Afternoon (4PM-8PM) Evening/Night (8PM-11PM) Don’t Know
	“Based on your previous experiences, please explain what (if any) exercise or physical activity has made your symptoms worse (provide examples of the types of exercise that you have found to worsen your RLS symptoms) (no word limit)?”	Open-ended (Narrative)
Questions	“What questions do you have about exercise?”	Open-ended (Narrative)

Demographic and Clinical Characteristics.

Participants completed items for assessing age, sex, race, height, and weight. Several questions were included to examine RLS-specific clinical characteristics, including previous diagnosis of RLS, age at diagnosis, bilateral or unilateral symptomology, duration of RLS symptoms, and time of symptom onset on a typical day/night (e.g.,2AM). Female participants were also asked if symptoms began during pregnancy and stopped after pregnancy. Participants were further asked questions regarding current RLS-specific treatments, including the responsiveness of symptoms to the treatment as well as treatment type, dosage, timing, and regularity of administration. We further examined the presence of conditions highly associated with RLS including a single item assessing the presence of periodic limb movements(PLMs), described as repetitive involuntary leg movements that occur during or before sleep, and another item assessing the presence of secondary conditions considered to mimic RLS or be related to secondary forms of RLS (e.g., iron deficiency anemia, renal disease, diabetes, radiculopathy, peripheral edema, and peripheral neuropathy).

Statistical Analyses

Quantitative Analyses.

All quantitative analyses were conducted using Python in Jupyter Notebook using the statsmodels package (Seabold&Perktold,2010). Descriptive statistics are reported as mean and standard deviation(SD) for continuous variables and number and percentage for dichotomous variables. The perception of exercise improving and worsening symptoms as well as the perception of the optimal time of day for the benefit of exercise or the worsening of symptoms are presented as the percentage of respondents in each category.

Content analysis of open-ended responses.

Directed content analysis was used to analyze open-ended narrative responses, wherein responses are initially coded with a focus on validating or extending an existing theory (Hsieh&Shannon,2005). Open-ended questions aligned with anecdotal evidence suggesting exercise is associated with RLS experience, and each open-ended question was paired with related quantitative questions. This allowed for a targeted and rich description of the phenomena initially explored with quantitative metrics by those with lived experience of exercise and RLS. See Table 1 for specific questions asked and related to quantitative questions.

Content analyses on the experiential narrative data were conducted in 6 steps. In Step 1, the researcher (EMS) read the responses for each question in full, making notes on recurrent content for the purposes of creating initial codes. Each open-ended question was read and analyzed individually. At this phase, the Positive and Negative experience open-ended questions were merged into a single data set, and the Questions about Exercise data set was assessed individually. In Step 2, initial codes, or similar recurrent content categories, were outlined regarding exercise experiences (e.g., positive and negative experiences) within each open-ended question data set. In Step 3, responses were re-read, and each participant’s response for each question was organized into themes within the categories, or placed in a new, unlabeled, category. In Step 4, the unlabeled categories were organized and labeled to provide a clear section for responses. In Step 5, all responses within each open-ended question set were re-read to ensure participant responses were reflected in the themes. No new, unlabeled themes or categories emerged at this time from the data. Step 6 involved condensing responses into a cohesive story and reporting the data in this manuscript. All responses were coded and considered during data analyses. Multiple direct narrative quotes are provided to support each theme, and individual theme frequencies are not reported to support thorough, rich, and wide description of participant experiences, in line with suggested qualitative rigor techniques (Smith,Sparkes,&Caddick,2014). Of note, some participant responses included minor typographical errors that have been corrected for ease of reading.

Results

Participant Characteristics

The summary of demographic and clinical characteristics for the final sample of participants completing the survey (N=528) is presented in Table 2. The sample had a mean age of 68±10 years and were mostly white (97%) females (65%) with an average BMI of 26.5±6.0 kg/m². Regarding RLS characteristics, participants reported experiencing RLS symptoms for an average of 30±18 years with an average RLS severity of 20.0±8.2 (i.e., moderate RLS) and a median RLS peak onset time of 19:00 (i.e.,7:00PM). Most participants reported symptoms bilaterally (95%) and symptoms were typically unrelated to pregnancy (90%). Most participants (91%) were taking medications on a regular basis (95%) to manage their symptoms with 98% reporting that medications either partially or completely relieve symptoms. Only 27% of participants reported a concomitant, secondary condition (e.g.,anemia, peripheral neuropathy) and most participants (63%) reported experiencing PLMs in addition to having RLS, presence of PLMs being reported subjectively.

Table 2:

Summary of demographic and clinical characteristics in the final sample (N=528).

	Mean (SD)	Range
Age (years)	68.1 (10.0)	18–89
Sex (n (%) Female)	341 (65%)
Race (n (%) White)	512 (97%)
BMI	26.5 (6.0)	16.1–53.2
RLS Duration	30.0 (18.2)	1.0–81.0
RLS Severity	20.0 (8.2)	0.0–39.0
No Symptoms (IRLS 0)	13 (2%)
Mild (IRLS 1–10)	56 (11%)
Moderate (IRLS 11–20)	198 (38%)
Severe (IRLS 21–30)	213 (40%)
Very Severe (IRLS >30)	48 (9%)
RLS Onset Time (median [IQR])	19:00 [15:00–21:00]	1:00–24:00
Bilateral vs. Unilateral (n (%) bilateral)	504 (95%)
Pregnancy Related RLS (n (%) yes)	34 (10%)
RLS Medication Use (n (%) yes)	479 (91%)
Non-Responsive to RLS Medications (n (%))	8 (2%)
Medication Frequency (n (%) regularly)	454 (95%)
Presence of Secondary Conditions (n (%) yes)	141 (27%)
Presence of PLMs	332 (63%)

Notes: Data are presented as mean (standard deviation) unless otherwise specified. SD standard deviation; BMI body mass index; RLS restless legs syndrome; IRLS International Restless Legs Syndrome Study Group Scale; IQR interquartile range; PLMs periodic limb movements.

Perceptions of the overall effect of exercise on RLS symptoms.

Most respondents (72%) perceived that exercise sometimes (n=253; 48%) or definitely (n=126; 24%) improves their RLS symptoms. Only 28% of respondents (n=149) believed that exercise definitely did not make symptoms better. Regarding exercise worsening symptoms, almost half (n=248; 47%) of respondents reported that exercise definitely did not worsen RLS symptoms. Approximately 40% of respondents (n=212) reported that exercise sometimes makes their symptoms worse and only a small proportion of people (n=67; 13%) reported that exercise definitely makes symptoms worse. Of note, 54% (n=172) reported exercise sometimes or definitely makes symptoms better while also reporting that exercise definitely does not make symptoms worse (i.e., exercise only benefits symptoms) and 23% (n=72) reported exercise definitely does not make symptoms better while also reporting that exercise sometimes or definitely makes symptoms worse (i.e., exercise only worsens symptoms).

Perceptions of the timing of exercise effect on RLS symptoms.

The summary of perceptions for specific timing of exercise benefiting (gray) or exacerbating (black) RLS symptoms is presented in Figure 1. The largest proportion of responses indicated that participants were unsure of a time of day that made their symptoms better (45%) or worse (47%). Regarding the time of day for benefiting symptoms (gray lines Figure 1), a higher proportion of participants said that early morning (31%) or mid-morning (24%) exercise made their symptoms better with fewer indicating that afternoon (22%) and evening/night (11%) made symptoms better. The opposite was observed for time of day for exercise and worsening of symptoms (black lines Figure 1), whereby more respondents indicated that afternoon (27%) or evening/night (29%) exercise made symptoms worse and fewer indicated that early morning (15%) or mid-morning (14%) exercise made symptoms worse. Overall, of those who indicated a time of day affecting symptoms, most respondents (55%) perceived that exercise in the morning (4AM to noon) improved their symptoms and 56% perceived exercise in the afternoon/evening (4PM to 11 PM) made symptoms worse.

Figure 1: Distribution of responses to the time of day for exercise on RLS symptoms.

Note: Gray represents the proportion of responses to “Exercising at what time of day makes your symptoms better at night? (Select all that apply)?” and black represents the proportion of responses to “Exercising at what time of day makes your symptoms worse at night? (Select all that apply)?”

Specific Exercise Parameters Perceived to Benefit and Exacerbate RLS Symptoms.

Experiences with exercise were organized into three primary categories: 1) specific exercise descriptions, 2) specific situational descriptions in which exercise either exacerbated or benefitted RLS, and 3) descriptions of no perceived impact of exercise on RLS. Codes and themes within these three broad categories that outline specific participant experience are described below.

Specific Exercise Descriptions.

Although responses varied widely, recurrent themes emerged during analysis regarding types of exercise that participants believed to benefit or exacerbate RLS symptoms. Importantly, some exercises were described by some participants as beneficial and by others as exacerbating.

Beneficial.

Types of beneficial exercise reported by respondents widely varied in type, duration, intensity, and frequency. Overall, participants described many individually perceived beneficial exercises included walking, strength training, aerobic exercise (e.g., running, vigorous walking, swimming, cycling, dancing, rowing, skiing, golf, organized sports (tennis, soccer, pickleball), home maintenance activities (gardening, housework or home repairs), yoga, other stretching activities.

“Hiking definitely helps me, when puffing and panting and getting oxygenated really helps…”

“If I’m not able to walk then I have problems. For me. Walking is my best exercise.”

“Stretching (like yoga), walking, weight lifting, cycling…”

“Balance exercises, strength training, walking, golf, elastic bands, weights, hiking, housework…”

“Anything aerobic for 25 to 40 minutes improves symptoms. Without exercise, I would have to double my medication. Fast walking, swimming, bike riding, and exercise classes.”

Many participants that described a beneficial experience also had reported specific prescriptions they follow for positive experiences.

“Cycling hard core 1 hour 500–800 calories seems key.”

“Treadmill 7–8% incline, 3.7–3.8 mph every night before bed. Doubles tennis once a week for 2 hours.”“Knee bends holding 30 secs; exercycle 15 min, 5 days/week; quad and calf stretches; walking”

“… It [exercise regimen] includes 12 exercises on difference machines, 3 sets, 40 reps total. These exercises take me about 75 min to complete.”

Exacerbate.

Many respondents indicated any type of exercise with heavy lower extremity involvement, including high duration or intensity of walking/jogging/hiking/cycling, would elicit a perceived increase in RLS symptoms. Single leg exercises, or instances in which all your weight was on a single leg for any amount of time, were also described by many as specifically exacerbating such as single leg stances in yoga, single leg kneeling, or balance exercises.

“Any high exertion of my legs will create pain in my joints. My RLS symptoms can be triggered by this pain.”

“Heavy use of leg muscles makes my symptoms worse.”

“Anything pertaining to legs (e.g, long walks or hiking).”

As in the beneficial experience category, participants described many individually-perceived exacerbating exercises including heavy weight lifting or strength training (including CrossFit-type exercises), cardio machine (stair climber, elliptical, exercycle, treadmill, Pilates reformer), water aerobics and swimming), sports (golf, tennis, ice hockey, downhill skiing, pickleball, softball), home maintenance activities (mopping tile floors, walking to mow grass, lawn work and gardening, moving furniture), yoga (specifically Hatha), and dance (including Zumba and Jazzercise).

“When I do extremely tiring exercise lasting hours in the yard doing heavy labor, my RLS is much worse in the evening.”

“Long hikes, long bike rides, running, generally exercise relying on legs.”

A few respondents reported that any exercise with higher intensity than basic stretching would increase their RLS symptoms.

“Stretching only. All else makes it worse.”

“All exercise makes my RLS worse.”

Interestingly, some participants stated activities in which they stood very still (e.g., standing desk) or basic prolonged standing while performing another task (e.g., cooking, washing dishes, working in a woodshop, standing on a ladder) would exacerbate their RLS.

“…standing or being on my feet for several hours a day.”

“Working in my woodshop when project requires me to stand for long periods without sitting, or I am so involved that I don’t take breaks to sit.”

Specific situations.

Change.

Participants described an abrupt change in activity type, frequency, duration, or intensity would almost always elicit RLS symptoms. This perception was extended by some to describe specific situations: e.g., hiking for a long time, higher effort than normal in a gym setting, stopping an exercise routine, and simple changes in exercise patterns.

“I have definitely found that an *unaccustomed* amount of exercise makes RLS worse: this has happened when I have taken an unusually long hike, even though it was slow…”

“If I exercise too close to bedtime or do significantly more or less than my typical exercise routine, my restless legs seem worse.”

“My pain worsens when I do too much e.g. when I swim for a mile and later take a several mile walk.”

Indeed, the alternative situation was thoroughly described by respondents as well. Consistent patterns, varying widely in type, intensity, and frequency, seemed to be associated with improvements in RLS symptoms and overall beneficial effects on frequency of symptomatic bouts.

“The type of exercise is not significant. The significant variable is the consistency of the amount and intensity of effort. That is, if there is EITHER an increase or decrease of exercise, RLS will worsen.”

“… for me, consistent and moderation seems to be the key.”

No Perceived Impact of Exercise On RLS.

Uncertainty.

A noticeable portion of respondents described uncertainty regarding the effect of exercise on their RLS experience, such that exercise may reduce symptoms when completed during an RLS symptom attack, but the influence of the exercise is not long-lasting. Some reported having contradictory experiences depending on the time of day (e.g., exercise at night exacerbating symptoms but exercise in morning benefitting).

“The effect (of exercise) is unpredictable. Sometimes that evening, I’ll have no symptoms and sometimes more severe symptoms.”

“Exercise outside of walking while symptoms occur does not affect whether I have RLS when resting.”

“If I get an episode and I am able to walk about 2 miles I have a success rate of about 50% of completely ending the episode.”

“If I exercise too close to bedtime or do significantly more or less than my typical exercise routine, my restless legs seem worse.”

No perceived impact.

Some described little to no effect of exercise on RLS symptomology.

“Exercise does not really seem to make a difference with me. I can have attacks on days with or without exercise.”

“… I used to be an athlete. I did hard exercise every day and it had no effect on RLS.”

Participant Questions About Exercise.

Participant-reported exercise questions were organized into two primary categories: ‘Evidence’ and ‘Specific Exercise Prescriptions’. Specific themes within the categories are described below.

Evidence.

A major recurrent question from participants focused on research supporting or contradicting the role of exercise for persons with RLS. These questions described an interest in evidence related to general mechanisms, specific influences, and abrupt changes in activity level.

General mechanisms.

Participants described an interest in understanding the physiological mechanisms underpinning RLS and movement, in general.

“Why does exercise affect RLS (what is the mechanism?)”

“Is there any actual evidence that exercise has an effect on RLS?”

“Exercise seems to be tied to my RLS symptoms, but often in contradictory ways.”

“I would be interested in hearing about the effects of exercise on RLS at the neurological level.”

“Why does weight training on legs help but aerobic exercises worsen RLS?”

Specific influences.

Additionally, specific questions were raised regarding mechanisms by which exercise might influence RLS including the role of electrolytes, muscle tightness, blood flow, circulation, water immersion, pharmacological mediators, iron levels, spinal involvement, proprioceptive input through the lower body, nerve inflammation, exercise-neurotransmitter relationship, psychological state, and past trauma.

“Does one’s iron levels make a difference as to the impact of exercise on RLS symptoms?”

“…I wonder if the pounding of my legs doing something is a problem.”

“Do other exercises cause inflammation of the nerves in the legs?”

“I personally believe that repressed trauma and guilt and shame are a large part of the problem. Not nearly enough effort has been put into the psychological underpinnings of RLS.”

Abrupt changes.

Some participants reported a specific desire for research to be conducted on how the change in activity levels influence RLS symptomology.

“The exercise-neurotransmitter relationship would be interesting to study and I’m guessing it may be that extremes, i.e. no exercise or excessive levels, may affect the relationship.”

“Well, in an ideal world, someone would test the theory that it is more an abrupt change in intensity that worsens symptoms, not the level of intensity.”

Specific Exercise Prescriptions.

Respondents described a desire for specific exercise prescriptions related to RLS, including time of day, type of exercise, duration, intensity, and frequency of exercise. Time of day, in particular, was mentioned by several participants as a desirable area for suggestions.

“Do you have any suggestions as to what I can do in the evening?”

“I would like to know if there is some type of exercise that would relieve my restless leg syndrome symptoms.”

“Best time of day, optimum no. of minutes, mild vs moderate amount.”

“How much exercise and what kind of exercise should I be doing?”

“It might be helpful to have some general guidelines as to type of exercises that aren’t too extensive.”

“What are the best exercises for preventing RLS.”

“Please send suggestions as to what exercise to begin and the best time of day to do them.”

Discussion

This is the first study to describe personal perceptions of exercise effects on RLS symptoms and assess the needs of adults with RLS regarding exercise prescription. This information is useful to better inform exercise-based research in RLS. Several important themes emerged: (1) most respondents perceived that exercise improves RLS symptoms with only a small proportion perceiving exercise makes symptoms worse and provided specific descriptions in which exercise either benefitted or exacerbated RLS; (2) most perceived exercise in the morning (4AM to noon) improved symptoms while exercise in the afternoon/evening (4PM to 11PM) made symptoms worse; and (3) people with RLS had questions about the evidence for exercise in RLS and specific exercise prescription recommendations.

The overarching perception for the positive effect of exercise on RLS is consistent with previous research that highlights exercise as a promising non-pharmacological approach for managing symptoms of RLS (K.L. Cederberg&Motl,2016; de Mello et al.,1996; Giannaki,Sakkas, et al.,2013; Giannaki et al.,2015; Mortazavi et al.,2013; Sakkas et al.,2008). However, responses varied widely regarding specific exercises that were believed to benefit or exacerbate symptoms. Some participants described specific exercises to be beneficial while others described the exact same exercises as being detrimental. Of note, participants described exercises with heavy lower extremity involvement would elicit a perceived increase in RLS symptoms. Importantly, participants described any abrupt change in exercise would almost always elicit RLS symptoms and that a consistent pattern of exercise seemed to be associated with improvements in RLS symptoms and overall beneficial effects on frequency of symptomatic bouts. An interesting theme that emerged was related to the interaction between exercise and pharmacological treatment, whereby it was thought that medication dosage would need to be increased without exercise. Although it is widely thought that exercise could be used in complement to pharmacological treatment, additional systematic research is necessary to understand this interaction. Collectively, these perceptions suggest that the type of exercise may not be as important, but rather having a consistent routine that is individualized to the person may be most beneficial to the management of symptoms.

An additional novel finding of the present study was the perceived effect time of day exercise was performed had on symptoms. Specifically, most people (55%) perceived exercise in the morning improved symptoms with fewer perceiving benefits as the day progressed. Similarly, few (29%) indicated that morning exercise made symptoms worse; most (56%) perceived exercise in the afternoon/evening made symptoms worse. However, due to the cross-sectional nature of the study, we cannot rule out that these trends were related to the circadian or diurnal appearance of RLS symptoms. Importantly, there was uncertainty and/or confusion expressed by participants regarding the perception that moving while symptomatic helps RLS, but exercising later in the day, when symptoms are typically present, is perceived to worsen symptoms. Only two studies have systematically examined the effect of time of day on RLS symptoms. One study in adults with RLS demonstrated that taking more steps in the morning (6AM-noon) was associated with fewer periodic limb movements at night (Reimers,Heidenreich,Bittermann,Knapp,&Reimers,2021); that study did not include an outcome of RLS severity. Another study in adults with RLS and multiple sclerosis demonstrated that more step counts and higher light physical activity in the evening (6PM-midnight) was associated with lower RLS severity (K.L.J. Cederberg et al.,2022). This is contradictory to the perceptions of most people with RLS in the present study and highlights the need for additional studies. Future research should also include systematically examining the effect of exercise at different times of day on symptom variation to understand the role of exercise timing on symptoms of RLS.

There is a clear call from individuals living with RLS to explore mechanisms of exercise impact on RLS, and other evidence providing conclusions or explanations for their lived experience. Respondents indicated frustration with lack of knowledge, lack of evidence, and general lack of understanding on the part of medical professionals in relation to non-pharmacological management of RLS and exercise specific supports. Indeed, individuals reported a specific desire for an exercise prescription, or at least general guidelines, for those living with RLS in relation to time, intensity, duration, and frequency of exercise that may benefit, or at the very least not specifically exacerbate, RLS symptomology. Although current literature demonstrates evidence for the general benefit of exercise for RLS (K.L. Cederberg&Motl,2016; de Mello et al.,1996; Giannaki,Sakkas, et al.,2013; Giannaki et al.,2015; Mortazavi et al.,2013; Sakkas et al.,2008), there remains a critical gap in research regarding specific exercise prescription parameters that can be used to benefit symptoms while avoiding exacerbations.

The present study has important limitations that should be considered when interpreting our results. The cross-sectional design of this study precludes any inferences on causality in the effect of exercise on symptoms of RLS. We did not evaluate the level of physical activity or exercise in relation to the positive or negative perception of exercise. As we sought to describe personal experiences with exercise, all outcomes are subjective in nature, thus increasing risk of recall bias. Selection and confirmation bias should also be considered as most people understand that exercise is good for overall health, and responses may have been tailored to the perceived expectation of researchers. We did not assess the amount or type of exercise routinely performed by participants. Although we recruited from an RLS-specific organization and we utilized the validated RLS-SFDQ13 (Allen et al.,2009) in assessing the presence of RLS, we did not include a physician confirmation of RLS diagnosis. As a consequence, we did not formally exclude people with conditions that can mimic RLS. Almost half of participants reported severe RLS notwithstanding the regular use of medications and the mean age of the study population was older; these factors could impact the characteristics of physical activity participation and the perceived interaction between exercise and RLS. The use of open-ended questions for participants to share their experiences rather than a formal qualitative approach precludes the ability to prompt respondents to elaborate on specific responses. Therefore, this method may have led to more contradictory responses than an interview method (e.g., some participants reported that exercise worsens RLS, but only described exercise as improving symptoms in narrative). The phrasing of “sometimes” and “definitely” worsens/improves symptoms may be confusing as “sometimes” refers to an increment of frequency while “definitely” refers to a certainty. There is no validated or standardized method of assessing the self-reported presence of PLMs and people may not be aware of PLMs. The survey was conducted through Qualtrics using the “Anonymize responses” mode, whereby we did not record respondents’ IP address, location data, or contact information; therefore, we were unable to contact participants for missing data or unclear responses. Lastly, our cohort was primarily White, limiting the generalizability to people of different racial and ethnic backgrounds who have different prevalence estimates and may experience RLS differently (Alkhazna,Saeed,Rashidzada,&Romaker,2014; Manconi et al.,2021).

This is the first study, to our knowledge, to describe personal perceptions of the effect of exercise on symptoms of RLS. Our results suggest that there is individual variation in the response to exercise, whereby some people experience improvements in symptoms with exercise, while others experience worsening of symptoms with exercise. Such individual differences and specific attributes should be considered in exercise-based management to further optimize personalized treatment plans to prevent and manage symptoms in people with RLS. Importantly, our sample had a mean BMI in the overweight range and, apart from the known link between obesity and RLS, this is a group in which exercise would be recommended for general health and wellness, irrespective of any effect on RLS. However, there is a critical need for additional research to examine specific exercise parameters for maximizing the benefit of exercise in people living with RLS. Objective measures of exercises (e.g., actigraphy) and correlation with symptomatic reports could also be helpful.