Increasing physical activity in stroke survivors using STARFISH,an interactive smartphone application: Protocol for a randomised controlled trial

Abstract

BACKGROUND:

Physical activity (PA) in stroke survivors is low however the ubiquitous nature of smartphone technology means smartphone-based interventions could be a feasible and cost-effective approach to increase PA. We developed STARFISH, a group behavioural change intervention delivered via a smartphone app to help increase PA.

OBJECTIVE:

Following the positive results of our pilot study the aim of this single-blind, Randomised Controlled Trial (RCT) is to examine the effect of the STARFISH intervention, on PA in stroke survivors, compared to usual care.

METHODS:

One hundred and twenty-eight community dwelling stroke survivors will be recruited, from four NHS boards in Scotland, and randomised to intervention or control groups. The interventional groups will use the STARFISH app. The control group will receive literature on PA post-stroke. Outcome measures will be taken at: baseline, four months (end of the intervention), two months post-intervention. The primary outcome will be objectively measured PA. Secondary outcome measures will be sedentary time, activity profiles, walking speed and endurance, fatigue, anxiety and depression, activities of daily living, quality of life and metabolic health-risk biomarkers.

CONCLUSION:

If improvements are found in the PA and health of stroke survivors then STARFISH could be deployed through app stores to allow implementation at scale.

Keywords

Physical activity stroke stroke survivor smartphone rehabilitation

1. Introduction and rationale

Stroke is one of the most common causes of disability, and the physical impairments caused by a stroke often encourage a physically inactive and sedentary lifestyle [1]. Physical inactivity following a stroke is also associated with lower muscle strength and cardiovascular fitness, reduced ability to perform activities of daily living and higher risk for recurrent stroke and cardiovascular disease [2]. Inactivity and low cardiovascular fitness are two modifiable risk factors associated with cardiovascular disease [3].

Previous research by our group has demonstrated that community dwelling stroke survivors are significantly less physically active ( $p<$ 0.001) and have higher sedentary time than healthy matched controls ( $p<$ 0.001) [4]. Accordingly, novel methods of supporting PA and exercise programmes following stroke should be developed [5]. Mobile devices, such as smartphones, are effective platforms for supporting changes in health behaviours, including Physical Activity (PA) as they can provide real-time feedback to the user, allow individualised content, and facilitate social support [6]. We developed STARFISH, a group behavioural change intervention delivered via a smartphone app, which uses the metaphor of a virtual fish tank, to help increase PA.

It is important that smartphone apps, which have the intention of improving health and well-being, are rigorously evaluated [7]. The step counting algorithm in the STARFISH app has been validated at typical walking speeds of stroke survivors [8], and, to date, we have carried out controlled pilot studies of effectiveness of the STARFISH app for increasing PA in older adults [9] and in stroke survivors [10]. The results of this latter study showed that stroke survivors in the STARFISH intervention group ( $n=$ 15) increased their daily step count by 39.3% whereas the control group ( $n=$ 8) decreased theirs by 20.2%. However, this was a small pilot study of six weeks and had no follow-up. A longer study was therefore needed to determine the adherence and efficacy of a longer-term intervention and with a follow up to investigate if changes in PA are maintained.

The primary aim of this Randomised Controlled Trial (RCT) is to investigate the effects of a four-month, group based, behavioural change intervention via STARFISH, on objectively-measured PA in community dwelling stroke survivors and a control group who receive usual care.

Secondary aims are to compare the effect of the intervention in terms of sedentary time, walking time and intensity, walking endurance, gait speed, metabolic risk factors and self-reported activities of daily living, quality of life, fatigue, anxiety and depression between the two groups. The study will also determine if any changes in outcomes are maintained two months post-intervention. Lastly, a purposive sample of 20 participants will be interviewed to explore their views of acceptability, usability and effectiveness of the intervention.

2. Methods

2.1 Design

This will be a single-blinded RCT. Whilst the assessor will be blinded to group allocation, due to the nature of the intervention, it will not be possible to blind participants.

2.2 Patient population – inclusion and exclusion criteria

Participants will be included if they have experienced a cerebrovascular event at any point in the past, been discharged from active rehabilitation, are able to walk independently with or without aid or orthosis, and have the ability to comprehend instruction. Participants will be excluded if they have a history of serious cardiac disease, uncontrolled hypertension (blood pressure $\geqslant$ 190/100 mmHg), another neurological condition or a musculoskeletal condition that would be exacerbated by walking, or are currently participating in another stroke related trial.

2.3 Recruitment and randomization

Participants will be recruited, in groups of eight, from four NHS boards in Scotland: NHS Greater Glasgow & Clyde, NHS Ayrshire & Arran, NHS Lanarkshire, and NHS Forth Valley and from stroke support groups. After gaining written consent, baseline measurements will be taken, and the eight participants will be randomised equally to intervention and control groups using opaque envelopes. Overall, 16 blocks of 8 participants ( $N=$ 128) will be randomised.

Figure 1.

Flow diagram of the progress through the STARFISH RCT of two groups.

2.4 Description of the starfish intervention

Four participants will form a STARFISH group. Participants are represented by a fish in a virtual fish tank, which is displayed as wallpaper on the home screen. STARFISH utilises the in-built tri-axial accelerometer of the smartphone to record the participant’s step count and data is uploaded to the STARFISH server. This data is then relayed to the other members of the team in real-time so progress can be monitored. As the participant walks, their fish swims faster and blows bubbles which is seen in real-time by the other three participants. Each participant has a daily step count target tailored to their baseline PA. If all participants meet their daily step count target on five out of seven days/week, a reward of a new creature is added to the fish tank. The STARFISH intervention has been described in more detail elsewhere [9, 10]. As the intervention has been extended from six weeks to four months, to introduce variety, the wallpaper background of the fish tank for the first two months will represent Loch Ness and for the second two months will change to represent the Great Barrier Reef. All creatures gained as rewards reflect the environment and are reset at the end of the first two-month period to avoid overcrowding on the screen.

2.4.1 Intervention group

After baseline assessment and randomisation, each member of the intervention group will be given a Samsung Galaxy ${}^{\text{TM}}$ smartphone. For the first week STARFISH will record the step count of each participant to calculate the individual step count target for the following week.

At the end of the week the four members of intervention group will meet with the researcher. At this visit, individualised step count target for each participant will be determined by adding 10% to the mean number of steps per day recorded on the phone. Thereafter individual step targets will be reviewed from data on the STARFISH server and updated automatically as follow: if a participant reaches their target on five of seven days their step count target will be increased by 5% the following week, up to a maximum increase of 3000 steps above baseline [11]. Conversely, if a participant does not reach their target then the next week their target will remain unchanged. If all four members of the group reach their daily step count target on five out of seven days then a creature (for example, a diver or octopus) will be added to the group’s fish tank. The group will meet again with the researcher two months after baseline, to discuss progress and address any concerns (Fig. 1).

2.4.2 Control group

Control group participants will receive literature on post-stroke PA. At completion of the trial the control group participants will receive a summary of their outcome measures and a pedometer (Fig. 1).

Adverse events in both groups will be recorded and reported as appropriate.

2.5 Outcome measures

Outcome measures will be taken by the blinded assessor at baseline, four months (end of intervention), and 6 months (two-month post-intervention follow up).

2.5.1 Primary outcomes

The primary outcome will be mean daily step count objectively-measured by an activPAL ${}^{\text{TM}}$ (PAL Technologies, Glasgow, Scotland) which is valid and reliable in detecting step counts [12, 13]. The assessor will affix the activPAL ${}^{\text{TM}}$ to the participant’s mid anterior thigh using a tegaderm waterproof dressing. The activPAL ${}^{\text{TM}}$ will be worn for seven days, before being returned to the research team by post.

2.5.2 Secondary outcomes

Time spent in a sedentary position (sitting or lying), intensity of walking (see below) and number of sit to stand transitions will be obtained from the activPAL ${}^{\text{TM}}$ monitor. Incremental cadence bands will be used to determine intensity of walking; from 1–19 steps/min (incidental movement or standing) to $\geqslant$ 120 steps/min (faster locomotion) [14]. Light PA (LPA) will be defined as a walking cadence $<$ 100 steps/min and moderate to vigorous PA (MVPA) a walking cadence $\geqslant$ 100 steps/min [14].

The six-minute walk test [15] will be used to measure walking endurance. The 6MWT measures the maximum distance walked, in six minutes, on a flat smooth surface The 6MWT is both valid and reliable in stroke survivors [16, 17]. The test will be carried out in a corridor with a minimum 30 m in length [17]. Gait speed will be measured using the 10 m walk test [15]. Time (in seconds) taken for the participant to walk 10 m will be recorded, and the average of two trials will be recorded.

The Nottingham Extended Activities of Daily Living Scale (EADL) will be used to measure activities of daily life. The total is summed out of 22 with a higher score indicating greater independence in activities of daily life. The EADL is valid [18] and reliable [19] in stroke survivors.

The Fatigue Severity Scale (FSS) is a nine item self-report measure of fatigue where the answers are summed and divided by 9 to give a final score of between one and seven with a higher score indicating a greater level of fatigue. The FSS is valid and feasible to administer to stroke survivors [21].

Quality of life will be measured using the Stroke Specific Quality of Life Scale (SS-QOL) which is a 49 item self-report measure. The total score is between 49 and 245, with higher scores indicating better quality of life. The SS-QOL is valid and reliable in stroke survivors [22].

Anxiety and depression will be measured by the 14-item Hospital Anxiety and Depression Scale (HADS). Individual totals will be calculated for both anxiety (seven items) and depression (seven items). Scores of 0–7 indicate normal levels of anxiety or depression, 8–10 indicate borderline abnormal levels, and 11–21 suggest abnormal levels of anxiety or depression [23]. Participants who score 8 or more for either category, will be asked to contact their General Practitioner.

Blood pressure and resting heart rate will be measured in accordance with the British Hypertension Society guidelines and the mean of three readings on the unaffected side will be recorded [24].

Height and weight will be measured from which Body Mass Index (kg/m ${}^{2}$ ) will be calculated. Lipid profiles of total cholesterol, non-fasting triglyceride and HDL cholesterol will be obtained via blood samples. Liver function will be measured by ALT, AST and GGT. C-reactive protein and HBA1c will also be measured. Blood will be drawn into EDTA, Fluoride Oxalate and Serum separator vacutainer tubes. Samples will be kept at $<$ 5 ${}^{\circ}$ C until centrifugation within 12 hours of collection. Whole blood samples will be aliquoted from the EDTA vacutainer, before all vacutainers centrifuged at 3000 RPM for 20 minutes. Whole blood, fluoride oxalate, plasma and serum aliquots will be stored in labelled 0.5 ml cryogenic tubes at $-$ 80 ${}^{\circ}$ C.

A purposive sample of up to 20 participants will be selected to take part in a semi-structured telephone interview. Participants across the range of disability and from all four recruitment sites will be selected. A topic guide will be developed to explore participant views with regards the effectiveness of the intervention and ease of use of the STARFISH app.

2.6 Sample size estimates

Using our pilot study data, we estimated that the RCT will need 58 participants in each group to detect a 30% increase in step count between intervention and control groups, with 80% power and a 5% level of significance. Therefore to allow for 10% drop out, 128 people will be recruited to the study.

2.7 Statistical analyses

Descriptive statistics will be used to summarise all variables at each time point. Group differences and interaction effects for each variable will be assessed using a two-factor repeated measures ANOVA model with a Greenhouse-Geisser correction factor. Model assumptions will be checked and if required appropriate transformation will be used. Where a significant time effect is found, Tukey’s post hoc test will be used. Effect sizes will be reported as will estimates of observed power. All analysis will be performed on IBM SPSS v22 and the level of significance will be 5%.

All interviews will be audio recorded, transcribed and analysed using thematic analysis.

2.8 Study organization and funding

This trial was funded by Chest Heart Stroke Scotland, has been approved by the West of Scotland Research Ethics Committee (REF: 15/WS/0173) and registered with ClinicalTrials.gov: NCT02494245.

3. Discussion

STARFISH, a behavioural change intervention, aimsto increase PA of community dwelling stroke survivors and, this will be the first RCT in community dwelling stroke survivors to include objectively-measured sedentary time as an outcome measure.

There have been calls to utilise smartphone sensors in the rehabilitation of stroke survivors [25]. Delivering a group PA behavioural change intervention via smartphone technology to stroke patients has a number of advantages which include the real-time group interaction of participants, and the ability to monitor progress remotely. During pilot studies of STARFISH, we mapped the behaviour change techniques (BCTs) in the intervention to the behaviour change technique taxonomy of Michie et al. [26]. The STARFISH intervention had the BCTs of: feedback and monitoring, goal-setting and planning, social support, comparison of behaviour, repetition and substitution of graded tasks, and rewards [9].

This RCT is unique as a stroke PA intervention as it includes a full lipid profile and other biomarkers as outcome measures. While the effects of aerobic training [27] and resistance training [28] on lipid profiles have been investigated in stroke survivors no study has investigated the effects of a walking intervention on these cardiovascular risk factors.

3.1 Summary

In summary, the STARFISH intervention has the potential to increase PA, reduce sedentary time, and decrease cardiovascular risk factors. If improvements are found in the PA and health of stroke survivors then STARFISH could be deployed through app stores to allow implementation at scale.

Footnotes

Conflict of interest

None to report.

References

Michael

Allen

Macko

. Reduced ambulatory activity after stroke: the role of balance, gait, and cardiovascular fitness. Archives of Physical Medicine and Rehabilitation. 2005; 86(8): 1552-6.

Ivey

Macko

Ryan

Hafer-Macko

. Cardiovascular health and fitness after stroke. Top Stroke Rehabil. 2005; 12(1): 1-16.

Celis-Morales

Lyall

Anderson

Iliodromiti

Fan

Ntuk

, et al. The association between physical activity and risk of mortality is modulated by grip strength and cardiorespiratory fitness: evidence from 498 135 UK-Biobank participants. Eur Heart J. 2017; 38(2): 116-22.

Paul

Brewster

Wyke

Gill

Alexander

Dybus

, et al. Physical activity profiles and sedentary behaviour in people following stroke: a cross-sectional study. Disabil Rehabil. 2016; 38(4): 362-7.

Moore

Hallsworth

Plotz

Ford

Rochester

Trenell

. Physical activity, sedentary behaviour and metabolic control following stroke: a cross-sectional and longitudinal study. PLoS One. 2013; 8(1): e55263.

Fanning

Mullen

McAuley

. Increasing physical activity with mobile devices: a meta-analysis. J Med Internet Res. 2012; 14(6): e161.

Burke

Azar

Bennett

Peterson

Zheng

, et al. Current Science on Consumer Use of Mobile Health for Cardiovascular Disease Prevention: A Scientific Statement From the American Heart Association. Circulation. 2015; 132(12): 1157-213.

Dybus

Paul

Wyke

Brewster

Gill

JMR

Ramsay

, et al. Validation of smartphone step count algorithm used in STARFISH smartphone application. Technol Health Care. 2017.

Paul

Brewster

Wyke

McFadyen

Sattar

Gill

, et al. Increasing physical activity in older adults using STARFISH, an interactive smartphone application (app); a pilot study. Journal of Rehabilitation and Assistive Technologies Engineering. 2017; 4(1): 1-10.

10.

Paul

Wyke

Brewster

Sattar

Gill

Alexander

, et al. Increasing physical activity in stroke survivors using STARFISH, an interactive mobile phone application: a pilot study. Topics in Stroke Rehabilitation. 2016; 23(3): 170-7.

11.

Fitzsimons

Baker

Wright

Nimmo

Ward Thompson

Lowry

, et al. The ‘Walking for Wellbeing in the West’ randomised controlled trial of a pedometer-based walking programme in combination with physical activity consultation with 12 month follow-up: rationale and study design. BMC Public Health. 2008; 8: 259.

12.

Dahlgren

Carlsson

Moorhead

Hager-Ross

McDonough

. Test-retest reliability of step counts with the ActivPAL (TM) device in common daily activities. Gait Posture. 2010; 32(3): 386-90.

13.

Ryan

Grant

Tigbe

Granat

. The validity and reliability of a novel activity monitor as a measure of walking. Brit J Sport Med. 2006; 40(9): 779-84.

14.

Tudor-Locke

Camhi

Leonardi

Johnson

Katzmarzyk

Earnest

, et al. Patterns of adult stepping cadence in the 2005–2006 NHANES. Prev Med. 2011; 53(3): 178-81.

15.

Wade

Wood

Heller

Maggs

Langton Hewer

. Walking after stroke. Measurement and recovery over the first 3 months. Scand J Rehabil Med. 1987; 19(1): 25-30.

16.

Flansbjer

Holmback

Downham

Patten

Lexell

. Reliability of gait performance tests in men and women with hemiparesis after stroke. J Rehabil Med. 2005; 37(2): 75-82.

17.

Dunn

Marsden

Nugent

Van Vliet

Spratt

Attia

, et al. Protocol Variations and Six-Minute Walk Test Performance in Stroke Survivors: A Systematic Review with Meta-Analysis. Stroke Research and Treatment. 2015; 2015: 28.

18.

Sarker

Rudd

Douiri

Wolfe

. Comparison of 2 extended activities of daily living scales with the Barthel Index and predictors of their outcomes: cohort study within the South London Stroke Register (SLSR). Stroke. 2012; 43(5): 1362-9.

19.

Nouri

Lincoln

. An extended activities of daily living scale for stroke patients. Clinical Rehabilitation. 1987; 1(4): 301-5.

20.

Krupp

LaRocca

Muir-Nash

Steinberg

. The fatigue severity scale. Application to patients with multiple sclerosis and systemic lupus erythematosus. Arch Neurol. 1989; 46(10): 1121-3.

21.

Mead

Lynch

Greig

Young

Lewis

Sharpe

. Evaluation of fatigue scales in stroke patients. Stroke; a Journal of Cerebral Circulation. 2007; 38(7): 2090-5.

22.

Williams

Weinberger

Harris

Clark

Biller

. Development of a stroke-specific quality of life scale. Stroke; a Journal of Cerebral Circulation. 1999; 30(7): 1362-9.

23.

Snaith

. The Hospital Anxiety And Depression Scale. Health and Quality of Life Outcomes. 2003; 1(1): 29.

24.

O’Brien

Asmar

Beilin

Imai

Mallion

Mancia

, et al. European Society of Hypertension recommendations for conventional, ambulatory and home blood pressure measurement. Journal of Hypertension. 2003; 21(5): 821-48.

25.

Zhang

Chew

Yeo

. The untapped potential of smartphone sensors for stroke rehabilitation and after-care. Technol Health Care. 2016; 24(1): 139-43.

26.

Michie

Abraham

Whittington

McAteer

Gupta

. Effective techniques in healthy eating and physical activity interventions: a meta-regression. Health Psychol. 2009; 28(6): 690-701.

27.

Moore

Hallsworth

Jakovljevic

Blamire

Ford

, et al. Effects of Community Exercise Therapy on Metabolic, Brain, Physical, and Cognitive Function Following Stroke: A Randomized Controlled Pilot Trial. Neurorehabilitation and Neural Repair. 2015; 29(7): 623-35.

28.

Zou

Wang

. Resistance training improves hyperglycemia and dyslipidemia, highly prevalent among nonelderly, nondiabetic, chronically disabled stroke patients. Archives of Physical Medicine and Rehabilitation. 2015; 96(7): 1291-6.