health problems, such as heart disease, diabetes, cognitive impairment, and dementia.

After a person suffers a stroke, they often experience difficulties in getting a good night's sleep. Approximately half of stroke survivors have insomnia, or have trouble falling and/or staying asleep. Poor sleep quality among stroke survivors increases the risk of recurrent stroke by 3-fold and the risk of early death by 76%. Hence, stroke survivors need strategies to promote better sleep.

Fortunately, evidence shows that sleep quality can be improved with exercise, even among those who struggle with insomnia. Whether exercise training can improve sleep quality in adults with chronic stroke (i.e., at least 12 months has passed since their stroke) and poor sleep quality is not known. In addition, it is unknown if improved sleep is associated with improved outcomes in those with chronic stroke.

This study will specifically evaluate the effect of twice-weekly targeted exercise training on sleep quality over a 6-month period in persons with chronic stroke and poor sleep quality. This study will also evaluate the effect of exercise on the following outcomes: 1) sleep structure; 2) fatigue; 3) daytime sleepiness; 4) mood; 5) physical function and capacity; 6) thinking abilities; 7) heart health; and 8) quality of life. Finally, this study will examine how changes in sleep quality may be related to changes in these outcomes. Our proposed research is timely as the importance of sleep to recovery, health, and wellbeing post-stroke is increasingly recognized.

Approximately half of stroke survivors have insomnia. Poor sleep quality among stroke survivors increases the risk of recurrent stroke by 3-fold and the risk of early death by 76%. Hence, stroke survivors are a target population in need of intervention strategies to promote sleep quality.

Current research and stroke rehabilitation do not address the sleep consequences of stroke - especially for those in the chronic phase (i.e., = or > 12 months since an index stroke). Adults with chronic stroke have lower sleep efficiency compared with sex- and age-matched controls. Poor sleep quality in chronic stroke is associated with depression, impaired physical function, and reduced cognitive function.

How to effectively treat insomnia in adults with chronic stroke is not well established. Intervention studies show exercise improves sleep quality and structure. Whether exercise can improve sleep quality in adults with chronic stroke and poor sleep quality is not established.

Primary Question: In adults with chronic stroke and poor sleep quality, can a 26-week multimodal exercise training program of moderate intensity (EX) improve sleep efficiency, as objectively measured by the actigraphy, compared with a 26-week cognitive and social activities program (CON; active control group)?

Secondary Question: What are additional benefits of EX vs. CON? Compared with CON, we will evaluate the effect of EX on: 1) actigraphy-measured sleep parameters of latency, duration, and wake after sleep onset; 2) sleep architecture by frontal electroencephalography; 3) subjective sleep quality; 4) fatigue; 5) daytime sleepiness; 6) mood; 7) functional capacity; 8) cognitive function; 9) cardiometabolic risk factors; and 10) quality of life.

Tertiary Question: Are changes in sleep quality associated with changes in: 1) fatigue; 2) daytime sleepiness; 3) mood; 4) cognitive function; 5) cardiometabolic risk factors; and 6) quality of life?

Pilot Data: Using actigraphy, the investigators acquired sleep data from 21 of 120 (i.e., subset) participants with chronic stroke enrolled in a RCT with cognitive function as the primary outcome. These 21 participants (mean age=69) were randomized to 26 weeks of: 1) multimodal exercise of moderate intensity (EX; n=12); or 2) cognitive and social activities (i.e., no exercise; CON; n=9). Of these 21 participants, 48% had a baseline actigraphy-measured sleep efficiency < 85%. Compared with CON, EX significantly improved actigraphy-measured sleep efficiency values (in %) at 13 weeks and 26 weeks.

Methods: A 26-week, assessor-blinded, single-site RCT of 62 community-dwelling adults with chronic stroke, WatchPAT measured sleep efficiency < 85%, Pittsburgh Sleep Quality Index global score > 5, and aged 55 years and older. Individuals will be randomized to one of two experimental groups: 1) twice-weekly EX (n=31); or 2) twice-weekly CON (n=31). Randomization will be stratified by sex and age. Measurement will occur at baseline, 13 weeks, and 26 weeks, unless otherwise stated.

Inclusion Criteria:

We will include community-dwelling adults who had an ischemic or hemorrhagic stroke at least one year prior to study enrolment and have poor sleep quality - defined as a sleep efficiency of < or = 85%, with allowance for minor upward deviations on individual nights consistent with known night-to-night variability and instrument accuracy limits, and a PSQI global score of > 5. Individuals must also meet these inclusion criteria: 1) aged 55 years and older; 2) a baseline MMSE score of = or > 22/30 and a MoCA score of = or > 19/30; 3) community-dwelling; 4) able to comply with scheduled assessments, classes, and other trial procedures; 5) read, write, and speak English with acceptable visual and auditory acuity; 6) not to start or stable (i.e., = or > 6 months) on a fixed dose of medication that impacts sleep (i.e., anti-arrhythmics, beta blockers, selective serotonin reuptake inhibitor, etc.) during the 26-week intervention period; 7) able to walk for a minimum of six metres with rest intervals with or without assistive devices; 8) based on interview, have an activity tolerance of 60 minutes with rest intervals; 9) not participating in any regular therapy or progressive exercise (e.g., treadmill or weight-lifting); and 10) able to safely engage in exercise as indicated by the PAR-Q+63 and written confirmation by family or study physicians.

Exclusion Criteria:

We will exclude individuals who: 1) are diagnosed with OSA and are not using CPAP or defined as potentially having undiagnosed OSA (AHI > 20) as a result of study screening; 2) have restless leg syndrome; 3) are diagnosed with dementia of any type; 4) are diagnosed with another type of neurodegenerative or neurological condition (e.g., Parkinson's disease); 5) are planning to participate, or already enrolled in, a clinical drug trial or exercise trial concurrent to this study; 6) are at high risk for cardiac complications during exercise and/or unable to self-regulate activity or to understand recommended activity level; 7) have clinically significant peripheral neuropathy or severe musculoskeletal or joint disease that impairs mobility, as determined by his/her physician; or 8) have aphasia as judged by an inability to communicate by phone.