Sedentary time and physical activity at key timepoints after stroke

Background: Stroke is a leading cause of disability, and risk of a second stroke is high. In Australia there are currently over 420,000 people living with the effects of a stroke, with numbers projected to increase due to population growth. Although physical activity is important in both recovery and in reducing cardiovascular risk, high levels of sedentary time and low levels of physical activity are common. Substituting sedentary time for physical activity of any intensity can attenuate cardiovascular risk and may be a realistic and achievable target for stroke survivors. Recovery occurs on a continuum after stroke, during which there are key timepoints which may influence physical activity behaviours. One such timepoint is the transition from hospital rehabilitation to home. A greater understanding about the impact of this transition may assist clinicians to target sedentary time and promote physical activity early, before contact with health services reduces. Additionally, there is limited understanding about physical activity participation, or the factors that might influence it, over the first 10 years after stroke. Aims: The broad aim of this thesis is to explore sedentary and physical activity time at crucial timepoints after stroke. The specific aims addressed in each chapter are to: 1) examine the long-term prevalence of exercise and factors associated with exercise participation up to 10-years after stroke; 2) determine the impact of change in a) the environment from hospital discharge to home and b) the first 3 months at home on activity behaviours; and to explore whether physical and psychological factors are associated with the activity behaviours examined in a) and b); 3) explore factors that influence activity behaviours during the transition from hospital to home from the perspective of the stroke survivor and 4) determine whether the use of novel technology is feasible to promote physical activity using a functional exercise in the home. Methods: To address these aims four studies were conducted. Participants in Study 1 were from the North East Melbourne Stroke Incidence Study. Participants in the further three studies were recruited from health services and the community in southern Tasmania. Study 1. Data (n=520 at 5 years, n=326 at 10 years) were from an observational study of first ever cases of stroke with 10-year follow up. Exercise was measured using a standardised self-reported questionnaire that asked about current exercise, exercise before stroke and whether participants had discussed the benefits of exercise with their doctor. Baseline measures included age, sex, residence, stroke severity and occupation. Covariates measured at follow up were functional ability, advice to exercise, depression, anxiety, and quality of life. Descriptive statistics were used to determine 5 and 10-year exercise prevalence. For 5 and 10-years cross-sectional data a log binomial regression model was used to determine factors associated with self-reported exercise. To examine change in exercise between 5 and 10-years four categories of exercise were created: no exercise, ceased exercising, commenced exercising and continued exercising. A log multinomial model was used to explore the relative risk of the previously listed covariates and change in exercise participation over time. No exercise at either timepoint was used as the reference category for the analysis. Study 2. I conducted a longitudinal observational study (n=34) with 3 timepoints: the final week of inpatient rehabilitation (time 1), the first week at home (time 2) and 3 months after hospital discharge (time 3). Sedentary time and walking activity (time and steps per day) were obtained with an ActiPAL3 accelerometer at each timepoint. Covariates (gait speed, walking endurance, lower limb strength, pain, functional independence measure, depression, anxiety and cognition) were collected at time 1 and 3. Linear mixed models were used to examine whether sedentary behaviour and walking activity changed in relation to a) the environment (time 1 to time 2) and b) over the first 3-months at home (time 2 to time 3). Interaction terms between the timepoint variable and each covariate were added to the models to determine if they modified any change in activity. Study 3. I conducted a qualitative study. Participants were stroke survivors who had recently been discharged from hospital rehabilitation (n=15) and their informal carers (n=7). Semi-structured interviews were used to explore the experience of physical activity during the individual's inpatient rehabilitation admission and soon after discharge home. Interviews were transcribed and inductively coded prior to completion of a thematic analysis. Study 4. I conducted a pilot clinical trial (n=10) after discharge home that involved 1) the development of a novel technology system (app, sensor, tablet) and 2) testing the feasibility of using the technology system to prescribe a functional exercise (sit-to-stand), and allow a therapist to remotely monitor, provide feedback and update the exercise program. Feasibility measures included testing of the study design, recruitment/withdrawals, adherence, safety, participant satisfaction and estimates of effect on function. Descriptive statistics were used to summarise participant characteristics, exercise adherence to the prescribed program and to provide an estimate of the effect of the intervention on measures of physical function (short performance physical battery and timed 2-minute sit-to-stand test). Results: Study 1. The prevalence of exercise was low at 5 years (18.5%, n=96) and at 10 years (24%, n=78) after stroke. In those with data at both timepoints (n=276), 15% (n=42) continued exercise between 5 and 10-years, 10% (n=27) commenced exercise, 14% (n=38) ceased exercise and 61% (n=169) reported no exercise at either timepoint. Younger age, better functional ability, independence in walking, better quality of life, exercising before stroke and recall of advice to exercise were associated with continued exercise between 5 and 10-years. Study 2. During the first week at home, participants spent less time sedentary (45 minutes/day) and more time walking (12 minutes/day, 724 additional steps/day) compared to the final week in hospital. Depression at discharge was associated with greater sedentary time in the first week at home. During the first 3-months at home, sedentary time reduced further (39 minutes/day) and walking increased (21 minutes/day, 1112 additional steps/day). No factors were found that predicted change in activity over the three-month period. Study 3. Five themes were identified as influencing early physical activity in hospital rehabilitation and soon after discharge home: Understanding ‚Äö- Participants had variable understanding of the role of physical activity in recovery. People with greater understanding reported being more physically active in hospital and at home; Safety and risk ‚Äö- Communication by health professionals about safety and risk in hospital was reported to be confusing for participants. People prioritised safety over physical activity participation if communication appeared conflicted; Sense of purpose ‚Äö- a desire to resume valued life roles was reported to support greater physical activity participation; Social influences ‚Äö- Professional, family and peer support were important to support physical activity in hospital and after discharge. Aspects identified as supporting opportunities for physical activity included structured and unstructured therapy time in hospital, physical function, and the environment. Study 4. It was feasible to deliver and monitor exercise remotely using an app, sensor and tablet device. Adherence to the prescribed program (sessions and repetitions) was high. Participants rated the system usability, enjoyment and benefit as high. No adverse events were reported. The estimated effect of the intervention showed positive improvements in measures of physical function between baseline and four-weeks later. Conclusion: Few stroke survivors reported long-term participation in exercise that could counteract the health risks of prolonged sedentary time or reduce the risk of secondary cardiovascular events. Greater physical function and recall of advice regarding exercise were associated with continued exercise in the long term. Health professional should use opportunities in the years after stroke to ask about exercise behaviour and provide exercise advice. Targeting factors associated with higher levels of activity early during recovery may help to support substituting sedentary behaviour for physical activity. The change in environment from hospital to home was associated with reduced sedentary time, but depression modified the benefit of the environmental change, suggesting it may be an important factor to target. Furthermore, stroke survivors and their carers reported that a better understanding of the role of physical activity, balancing safety and risk without overly restricting physical activity, and improving self-efficacy may be important enablers that clinicians can target. Finally, it was feasible to use technology to promote physical activity using a functional exercise program at home. Key features that supported adherence to the program included connecting stroke survivors and therapists, remote monitoring, feedback and exercise progression.