There is good evidence of the effectiveness of regular physical activity in the primary and secondary prevention of stroke. STARFISH is a smartphone-based application designed as a behavioural change intervention to encourage the user to become more physically active. The study will investigate if using STARFISH for 6 weeks can increase physical activity in people after stroke. Subjects will be randomised to the STARFISH intervention or a control group. Control subjects will not have access to their daily activity count.

Introduction Stroke is one of the leading causes of long-term disability in western countries (Donan et al. 2008). Stroke survivors are often deconditioned and predisposed to a sedentary lifestyle that limits performance of activities of daily living, increases the risk for falls, and may contribute to a heightened risk for recurrent stroke and cardiovascular disease (Ivey et al. 2005). There is good evidence of the effectiveness of regular physical activity (PA) in the primary and secondary prevention of several chronic diseases e.g. stroke, cardiovascular disease, diabetes, colon cancer, breast cancer, hypertension, obesity and osteoporosis (Warburton et al. 2006). Walking may be one of the most accessible forms of exercise in the stroke population. STARFISH is a mobile phone application designed to encourage the user to become more physically active. Phase 1 of the project will involve the development of the current STARFISH to ensure it is fit for purpose for people following stroke. This will be done in collaboration with stroke survivors and their family if necessary. The study will investigate if using STARFISH for 6 weeks can increase physical activity and if the programme has the potential to improve health and well being in people following stroke, including those with residual, physical, sensory and visual impairment.

Description of STARFISH The current version of STARFISH is a smartphone-based programme designed as a behavioural change intervention to encourage the user to become more physically active. The sensors within a standard smartphone record the number of steps taken per day by the individual. Each person is represented by a fish within a fish tank which the participant can see on their mobile phone display. When the participant is active their fish blows bubbles and swims faster and, as the participant reaches their target number of steps per day, their fish's fins and tail grow.

Aims and Objectives The overall aims are to adapt STARFISH for use in stroke survivors and to undertake a feasibility study to evaluate the acceptability and potential effectiveness of STARFISH in this group.

The objectives are:

1. To use a participatory co-design approach to develop the STARFISH software and interfaces to ensure that STARFISH is fit for purpose for use by people with stroke.
2. To explore the likely effectiveness of STARFISH in increasing levels of PA in people post stroke.
3. To explore the likely acceptability of STARFISH by people post stroke.
4. To provide sufficient data on effect sizes of STARFISH on outcomes related to amount of physical activity undertaken, quality of life, general well-being and fatigue, walking speed, and metabolic health biomarkers (BMI, blood pressure, lipid profile, resting heart rate) to undertake power calculations to establish the sample size required for a fully powered, randomised controlled study.

Design & Methods Design Following MRC guidance, the proposed mobile phone application development and pilot evaluation comprises a 2-phase (sequential) mixed-methods programme of research.

Phase 1 will involve the development of the current STARFISH mobile phone application. This will involve participatory co-design of the application to ensure suitability for stroke survivors.

Phase 2 will involve piloting the revised STARFISH application with participants to assess the feasibility to recruit the target populations, potential to support participants to become more physically active, improve health behaviours and deliver physical and mental health benefits . The Phase 2 findings will inform a decision on whether to proceed to larger scale evaluation.

Methods Up to 6 stroke survivors attending Western Infirmary physiotherapy outpatient clinic will be recruited to the phase 1 of the study. Another 24 patients with mild to moderate stroke-related walking dysfunction and who have had a single stroke will be recruited through Chest Heart &Stroke affiliated groups. Potential patients will be given verbal and written information about the study by the research team member. Patients who are willing to take part in the study will be advised to telephone the researcher via the contact details on the Participant Information Sheet. Thereafter, if the participant is willing to participate in the study a suitable date/time for the assessment will be arranged by telephone, and then confirmed by letter.

Phase 1. An initial co-design session of up to six people with stroke will be undertaken to elicit their views on the current STARFISH application and how this should be adapted to meet their needs. For example, the fish may require to be larger to allow stroke survivors to see them or for those with physical impairments of their upper limb voice activated commands may be beneficial. The software will require more modelling work to sense the different gait patterns of people with stroke and thus ensure accurate counting of steps. Thereafter, the STARFISH application will be refined and prototype user trial (for up to 3 weeks) will be undertaken. Another co-design session will be then organised to evaluate the changes and final revision of the application.

Phase 2. Participants will have an initial visit to the Clinical Research Facility (CRF) where the study will be explained and they will be given a mobile phone and asked to carry the phone for seven days to establish their normal level of PA. They will return to the CRF at the end of the seven day period for baseline measures (see below) and for agreed activity targets to be set. They will then be randomised to either the intervention (n=12) or control group (n=12). The intervention group will follow the STARFISH programme for six weeks. After three weeks the group will be invited to return to the CRF to informally discuss progress and for PA targets to be reviewed and amended as necessary. During the intervention period participants will be given contact numbers of the research assistant to call if they experience any problems. Control participants will carry the phone for six weeks, to allow recording of their levels of physical activity. Control participants will not have access to their daily step nor will they have access to the STARFISH app. All participants will return to the CRF after six weeks for final measures to be taken. All participants will then be given their daily activity results in a form of short report. A summative focus group will be undertaken to establish the views of the participants on areas of technical feasibility including accessibility, ease of use and effectiveness of the system.

Participants:

Phase 1. Up to six people with stroke, who will not be participating in the phase 2 of the study, will be involved in the development/co-design stage.

Phase 2. Twenty four adults with mild to moderate stroke-related walking dysfunction and who have had a single stroke will be randomised to the intervention (n=12) or control group (n=12). The randomisation process will be carried out using sealed opaque envelopes containing participant's unique numbers. All participants will be provided with a standard smartphone. Control participants will carry the phone for six weeks but without STARFISFH.

STARFISH is undertaken in groups of four and each member of the group is represented by a fish within the tank, each fish is distinguishable by colour as belonging to a specific individual thus each participant gets feedback on the activity level of each member of the group, in real time, via the visual representation of the fish in the fish tank. Different goals can be set for different individuals in the group, along with goals for the whole group. The study has been designed around a framework based on Control Theory, a technique known to be successful for behavioural change, which includes: goal setting, action planning, self-monitoring, reviews of goals and providing feedback on performance; social support is likely to act by influencing all of these.

Data collection

Data will be collected as follow:

Demographic:

Age, sex, height, weight, BMI

Disease specific:

Resting heart rate, blood pressure, Blood analysis; blood lipids, LFTs, CRP, HBA1c The Ten-Meter Walking Test (10 MWT) assesses walking speed (m/sec) over a 10m distance, measured by research assistant.

Fatigue Severity Scale (FSS) is a self reported questionnaire which evaluates the impact of fatigue. It contains 9 statements scored on a 1 to 7 scale. A total score of less than 36 suggests that patient may not be suffering from fatigue. A total score of 36 or more suggests that patient may need further evaluation by a physician.

Instrumental Activities of Daily Living Scale contains 8 items to assess more complex activities (termed "instrumental activities of daily living") necessary for functioning in community settings. The higher the score, the greater the person's abilities. The research assistant will examine the scale.

Stroke Specific Quality of Life Scale (SS-QOL) is a self-report questionnaire consisting of 49 items in the 12 domains of energy, family roles, language, mobility, mood, personality, self-care, social roles, thinking, upper extremity (UE) function, vision, and work/productivity. The domains are scored separately, and a total score is also provided. Higher scores indicate better functioning.

Psychological General Well-Being (PGWB) Index. A measure of self-representations of intrapersonal affective or emotional states reflecting a sense of subjective well-being or distress consists of 22 items that measure components of psychological well-being such as anxiety, positive well-being, self-control, depression, general health and vitality. Scores range from 0 to 110 and higher scores suggest better well-being. The PGWB index is self-administered.

Descriptive statistics will be used to summarise all variables. Where appropriate, comparisons between patients and controls will use Students t-test or UMann-Whitney analyses for parametric and non-parametric distributions respectively.

Inclusion Criteria:

• Post stroke survivors of single unilateral stroke
• discharged from rehabilitation
• the ability to walk independently, with or without using an aid or orthosis
• sufficient cognitive ability,
• the ability to comprehend instruction

Exclusion Criteria:

• history of serious cardiac disease (e.g. myocardial infarction, unstable angina)
• uncontrolled blood pressure
• significant neurological or musculoskeletal conditions in addition to stroke