JMIR MHEALTH AND UHEALTH Daly et al

Original Paper

Feasibility, Usability, and Enjoyment of a Home-Based Exercise

Program Delivered via an Exercise App for Musculoskeletal Health
in Community-Dwelling Older Adults: Short-term Prospective Pilot
Study

Robin M Daly, PhD; Jenny Gianoudis, PhD; Travis Hall, MClinExPhys; Niamh L Mundell, MClinExPhys; Ralph
Maddison, PhD
Institute for Physical Activity and Nutrition, School of Exercise and Nutrition Sciences, Deakin University, Melbourne, Australia

Corresponding Author:
Robin M Daly, PhD
Institute for Physical Activity and Nutrition, School of Exercise and Nutrition Sciences
Deakin University
221 Burwood Highway, Burwood
Melbourne, 3125
Australia
Phone: 61 392446040
Email: [email protected]

Abstract
Background: Many older adults choose and prefer to exercise at home, but to attain the greatest benefits, the correct type and
dose of exercise should be prescribed and adherence maintained. Advances in digital health technologies now provide the
opportunity for exercise professionals to deliver and monitor personalized, evidence-based exercise programs to anyone at any
time.
Objective: The aim of this study was to evaluate the feasibility, usability, and enjoyment of a web-based exercise prescription
app as a platform for exercise professionals to remotely deliver and monitor an individually tailored, home-based multicomponent
exercise program (delivered through tablet computers) to older adults living independently in the community.
Methods: This was an 8-week, prospective single-arm pilot study in 20 adults aged ≥65 years living independently in the
community: 10 owned a tablet computer (tablet owners) and 10 did not own tablets (tablet nonowners). All participants were
prescribed a home-based, muscle strengthening, weight-bearing impact and challenging balance/mobility program (3 days/week)
using a commercial exercise prescription app on a tablet computer. Study endpoints were feasibility (retention, adherence, adverse
events), usability (System Usability Scale), physical activity enjoyment (Physical Activity Enjoyment Scale), changes in lower
extremity function (Short Physical Performance Battery [SPPB]), and level of physical activity (questionnaire). Process measures
related to the participants' experiences and perceptions of the exercise program and web-based app were also included.
Results: A total of 19 participants (mean age, 70 years) completed the study (19/20, 95%), and mean adherence to the exercise
program was 84% (95% CI 70%-97%). There were 2 minor adverse events in 2 participants from 401 completed sessions. Mean
weekly walking time increased by 78 minutes (95% CI 0-156, P=.049) and moderate-to-vigorous physical activity time by 41
minutes (95% CI –8 to 90, P=.09). For SPPB scores, there was a 0.3 point (95% CI –0.1 to 0.7, P=.17) modest sized (effect size,
d=0.42) improvement after 8 weeks. Mean (SD) system usability was high (86 [10] with 100 best imaginable). There was no
change in the overall physical activity enjoyment scores after 8 weeks, but participants reported that they enjoyed using the
web-based exercise app and the exercise program (median score 4 on a 5-point Likert scale). For all measures, there were no
differences between previous tablet owners and nonowners.
Conclusions: This pilot feasibility study indicates that it is safe and feasible for community-dwelling older adults to participate
in a home-based, multicomponent exercise program targeting musculoskeletal health and function that was delivered and monitored
remotely by exercise professionals using a tablet-based exercise prescription app.

(JMIR Mhealth Uhealth 2021;9(1):e21094) doi: 10.2196/21094

http://mhealth.jmir.org/2021/1/e21094/ JMIR Mhealth Uhealth 2021 | vol. 9 | iss. 1 | e21094 | p. 1

(page number not for citation purposes)
XSL• FO
RenderX
 JMIR MHEALTH AND UHEALTH Daly et al

KEYWORDS
home exercise; multicomponent exercise; mobile health; musculoskeletal; adherence; usability; older adults; physical activity
enjoyment

have evaluated the feasibility, usability, and enjoyment of

Introduction exercise prescription apps targeting the key musculoskeletal
Every year, around 30% of people aged over 65 years living in health and function (eg, bone and fall-related) risk factors
the community fall at least once [1], and falls are a leading cause associated with fractures in older adults residing in the
of fragility fractures, injury-related hospitalization, mortality, community.
and health care costs for older adults [2,3]. By 2022, it is The aim of this feasibility study was to evaluate the feasibility
predicted that there will be 1 fragility fracture every 2.9 minutes (retention, adherence and adverse events), usability, and
(>500 per day) in Australia [4]. The current models of care for enjoyment of a commercial web-based exercise prescription
fracture risk reduction focus largely on pharmacological agents app (Physitrack) as a platform for exercise professionals to
targeting bone mineral density (BMD) [5], but there is a need remotely deliver and monitor an individually tailored,
for multifaceted approaches that can simultaneously target home-based multicomponent exercise program (via tablet
multiple fall and fracture risk factors. computers) for older adults living independently in the
Exercise is widely recognized as a safe and effective approach community. In addition, we explored participants’ perceptions
to improve nearly all modifiable fracture risk factors, including of the exercise program and Physitrack app and whether
BMD and falls risk [6,7]. Several meta-analyses of randomized outcomes differed between previous tablet computer owners
controlled trials provide compelling evidence to support the and nonowners.
benefits of exercise as a single intervention to prevent falls in
community-dwelling older people [7] and multicomponent Methods
resistance-based exercise programs for improving bone health
in postmenopausal women [8]. The findings from several of
Study Design
our previous randomized controlled trials conducted within The Seniors Made Active thRough Technology (SMART) study
community-based health and fitness centers have also shown was an 8-week community-based, prospective single-arm pilot
that multicomponent exercise programs incorporating study in which adults aged ≥65 years were prescribed a home
progressive resistance training combined with weight-bearing exercise program (accessed by the PhysiApp-patient portal) by
impact and challenging balance and mobility training are safe an accredited exercise physiologist (AEP) using the commercial
and effective for improving hip and lumbar spine BMD, muscle Physitrack (clinician portal) exercise prescription app. The trial
mass, strength, power, and function in healthy older adults and was managed through the Institute for Physical Activity and
those with low BMD or at increased falls risk [9-12]. Despite Nutrition at Deakin University, Australia, and was approved by
these positive findings, geographical location and access to the Deakin University Human Research Ethics Committee
affordable community-based exercise programs and qualified (HREC 2016-219).
exercise trainers and a general aversion to the gym environment
Participants and Recruitment
are key barriers to participation reported by many older people,
which has implications for intervention effectiveness [13-15]. Twenty relatively healthy men and women (convenience sample)
Thus, there is a need to consider alternative models of service aged 65 years and over living independently in the community
delivery to meet individuals’ exercise needs, preferences, and were recruited through our research trial database, and study
financial resources more broadly. flyers were sent to a number of community (Rotary) clubs in
the eastern suburbs of Melbourne, Victoria, Australia. In order
Advances in digital health apps have provided new opportunities to explore differences in the outcomes and the experiences in
for health care professionals to remotely deliver and monitor using mobile technology between those with tablets (tablet
evidence-based exercise programs tailored to the needs of older owners) and without tablets (tablet nonowners), we deliberately
adults and within their own home or community environment. recruited 10 participants who possessed a tablet computer (and
This is important, as a study of 240 community-dwelling adults had access to Wi-Fi at home) and 10 participants who did not
attending osteoporosis-related programs revealed that a lack of possess such a device. The tablet nonowners were provided
access to exercise programs that meet their needs and with an iPad and a SIM card (and adequate data capacity) for
preferences and limited resources, time, and trust in exercise the duration of the study and instructed on how to use it during
providers were some of the key barriers to participation [13]. the initial home visit.
For falls prevention, a study in 5440 older adults indicated that
a home-based strength and balance training program was Participants were initially screened over the telephone and
preferred over other prevention strategies [16]. However, to included if they were able to walk without the use of an aid,
ensure clinical effectiveness, it is important that any prescribed willing to use an iPad (their own or one provided to them) for
exercise programs adhere to current best practice guidelines the execution of the exercise program, and if they were able to
and incorporate behavioral strategies to promote long-term speak English. Participants were excluded based on the
adherence. Despite the rapid rise in the number of web-based following criteria (all self-reported): (1) aged <65 years, (2)
and mobile health apps available to health care professionals participation in resistance exercise >1 session per week for at
to deliver exercise programs to people at home, few studies least 20 minutes or moderate-to-vigorous intensity physical

http://mhealth.jmir.org/2021/1/e21094/ JMIR Mhealth Uhealth 2021 | vol. 9 | iss. 1 | e21094 | p. 2

(page number not for citation purposes)
XSL• FO
RenderX
 JMIR MHEALTH AND UHEALTH Daly et al

activity for ≥150 min/week over the past 3 months, (3) recent balance/mobility exercises based on the principles of the
low trauma fracture (within the past 6 months), (4) inability to Osteo-cise: Strong Bones for Life program, which is an
stand unaided, (5) acute or terminal illness likely to compromise established and effective community-based osteoporosis
exercise participation, (6) unstable or ongoing prevention exercise program for older adults at increased risk
cardiovascular/respiratory disorder, (7) musculoskeletal or for falls and fracture [11,12]. Participants were prescribed 8-9
neurological disease or functional limitations disrupting targeted exercises (2-3 sets of 8 repetitions) at a moderate
voluntary movement or that might have limited training, or (8) intensity (3-6 on the 10-point modified RPE scale) to be
inability to commit to the study and its requirements. The completed on 3 nonconsecutive days per week. All participants
Exercise and Sports Science Australia adult pre-exercise were provided with an exercise equipment pack (box step,
screening tool was used to identify any individual who may be dumbbells, TheraBand resistance bands, stepping cones, foam
at increased risk for any adverse event(s) due to participation balance mat). The program was designed to be completed within
in our exercise program. Participants with signs or symptoms 30 minutes and consisted of 2 warm-up exercises (eg, marching,
of unstable or unmanaged disease were excluded from the study. side stepping, sit to stand), 2 challenging balance/functional
exercises (eg, alternating lateral steps, tandem walking, single
A total of 87 older adults expressed an interest and were
leg standing), 1 upper limb (eg, wall press up, triceps dips,
screened for the study, of which 20 were included. The reasons
overhead press), and 2 lower limb resistance exercises (eg, step
for exclusion were as follows: 18 due to the presence of a
ups, bodyweight squats, reverse lunge with weights), 2
cardiovascular, musculoskeletal, or neurological condition that
weight-bearing impact exercises (eg, vertical, lateral, and
could limit their ability to participate in the exercise program;
multidirectional jumping or hopping), and cool down
11 due to being too physically active; 10 due to expected travel
(stretching) activities.
during the study period; 3 due to age (<65 years); and 1 due to
terminal illness. The other 24 participants were not interested All participants received 3 home visits from the AEP during
or did not have the time to participate in the study upon the study. At the initial home visit, baseline assessments were
receiving further details about the requirements. completed and participants were educated on the PhysiApp and
safe exercise training and prescribed their initial exercise
Intervention program. Participants then received a weekly phone call for the
All participants were prescribed (by a single qualified AEP first 2 weeks of the study to monitor progress and address any
recruited to work on this study) two 4-week multicomponent questions. Thereafter, participants were encouraged to liaise
home exercise programs using the commercially available with the AEP directly via the PhysiApp, which was monitored
web-based Physitrack exercise programming app with the daily. A second home visit was conducted at week 4 to review
accompanying PhysiApp that was accessible via their progress, record any adverse events, and update the exercise
iPad/tablet. Physitrack is a cloud-based, digital platform that program, with the final home visit conducted at the end of the
allows health professionals to assign exercises and programs study (after week 8) to complete the follow-up assessments.
(with training dosage) to people remotely, track progress,
provide feedback in real time, and send reminders. Using this Feasibility: Retention, Adherence, and Adverse Events
program, the AEP formulated a personalized exercise program Retention was recorded as the number (proportion) of
for each participant by selecting from a battery of >3500 participants who completed the 8-week assessment. Adherence
exercises that includes narrated videos and descriptions about to the exercise program, including the number of sessions
how to perform each exercise. The Physitrack system allows completed, number of exercises, and sets and repetitions
the AEP and participants to set up automated reminders about completed (all expressed as a percentage) within each session
exercise times and record exercise completion, including sets, were recorded within the Physitrack system. We considered the
repetitions, and rate of perceived exertion (RPE) for each program to be feasible if at least 90% of the participants
exercise, as well as include feedback or messages that are sent completed the trial and if the adherence to the program was at
(in real time) to the AEP (or to participants from the AEP) for least 66% (equivalent to 2 out of 3 sessions per week).
monitoring and review. For each exercise, participants were Participants were asked to record any adverse events (including
prescribed a specific training dose (frequency, sets, and falls) directly into PhysiApp so that they could be reviewed by
repetitions) and asked to report on their RPE using the 10-point the AEP and research staff. Information on adverse events was
scale provided in the app. Each participant’s program was also collected at home visit 2 (week 4) and 3 (after week 8). An
reviewed and progressed weekly by the AEP if needed, by adverse event was defined as an intervention-related event
reviewing the self-reported RPE and sets/repetitions for every resulting in absence from or modification to the exercise
exercise completed via the web-based Physitrack platform. The intervention.
AEP also checked the Physitrack system daily for any urgent
alerts/messages from participants.
Anthropometry and Demographics
Height to the nearest 0.1 cm and body weight to the nearest 0.1
Exercise prescription was individualized based on each kg were measured using standard procedures. The following
participant’s initial functional capacity determined from the information was collected by the questionnaire (baseline only):
baseline assessment, medical and physical activity history, as date of birth, ethnic background, education, living arrangement,
well as the AEP’s clinical judgement. Each exercise program medical history, medication use, and history of falls. At the
included a combination of muscle strengthening (resistance) completion of the study, participants were also asked if they
exercises, weight-bearing impact activities, and challenging had experienced a fall(s) over the past 8 weeks.

http://mhealth.jmir.org/2021/1/e21094/ JMIR Mhealth Uhealth 2021 | vol. 9 | iss. 1 | e21094 | p. 3

(page number not for citation purposes)
XSL• FO
RenderX
 JMIR MHEALTH AND UHEALTH Daly et al

Physical Activity level of importance about key elements that may have helped
Duration (minutes per week) of walking and participants maintain their motivation to continue with the
moderate-to-vigorous physical activity (MVPA) during the past program was assessed using a 5-point scale (1=extremely
week, which was truncated at 840 minutes, was assessed using important, 5=not all important). Finally, participants were asked
the validated Active Australia survey [17]. whether they would continue to use the Physitrack app to
exercise at home if it was made available. All open-ended
Physical Function questions were analyzed using a general inductive thematic
Physical function was assessed using the standardized Short approach [22].
Physical Performance Battery (SPPB), which is a composite Statistical Analysis
measure of 3 tasks: standing balance, habitual gait speed, and
repeated (5) chair rise [18]. A score of 0 to 4 was assigned to As this was a pilot feasibility study [23], a convenience sample
each test and added to yield a composite score ranging from 0 of 20 older adults was recruited with no formal sample size
to 12, with higher scores indicating better physical function. calculations [24]. However, the observed effect sizes (Cohen
The SPPB has been shown to be valid, reliable, and sensitive d) for the functional and physical activity measures were
to change with intraclass correlation coefficients of 0.88-0.92 calculated using the following formula: mean posttest minus
for tests performed 1 week apart [19]. mean baseline divided by baseline standard deviation. The
potential clinical meaningfulness of the results (in addition to
Physical Activity Enjoyment statistical significance) was based on the magnitude of the
At baseline and follow-up, participants completed the Physical effects: small (d=0.20), medium (d=0.50), and large (d=0.80)
Activity Enjoyment Scale (PACES) [20]. This 18-item (and P values) [25]. Nevertheless, the results and findings from
questionnaire asked participants to rate “how do you feel at the the hypothesis tests should be treated with caution, given our
moment about the physical activity you have been doing” using modest sample size.
a 7-point bipolar rating scale with scores ranging from 18 to All statistical analyses were conducted using SPSS Statistics
126 points. Eleven of the 18 items are reverse-scored with higher for Windows, version 26 (SPSS Inc). Baseline characteristics
scores representing higher levels of enjoyment. between the group (tablet owners and nonowners) were
System Usability compared using independent two-sided t tests for continuous
variables and chi-squared tests for categorical variables. Paired
Usability represents the participants’ experience with using the
sample t tests (two-sided) were used to assess within-group
app. At the final assessment, participants completed the System
changes for the continuous variables and the McNemar test was
Usability Scale (SUS) [21] to assess perceived usability of
used for categorical variables. Between-group differences for
Physitrack. The SUS is a standard 10-item questionnaire in
changes were assessed using analysis of variance or McNemar
which responses are measured on a 5-point Likert scale ranging
test for categorical variables. Between-group differences were
from 1 (strongly disagree) to 5 (strongly agree). Questions 1,
calculated by subtracting the within-group changes from the
3, 5, 7, and 9 are positive and questions 2, 4, 6, 8, and 10 are
baseline in each group. Within-group changes were presented
negative. A total SUS score is derived by summing the
as absolute changes from the baseline. All data were presented
individual scores and multiplying by 2.5, which yields a score
as mean (SD) or 95% CI (or median and interquartile range)
ranging between 0 (worst) and 100 (absolute best). A score >68
and the significance was set at P<.05.
is considered above average usability and >80 considered high
usability and a level at which participants are likely to
recommend the product to peers [21].
Results
Process Measures: Perceptions of the Program and Baseline Characteristics
System The characteristics of the cohort are shown in Table 1, with no
Upon trial completion, process measures were collected using marked differences between the tablet owners and nonowners.
an author-derived questionnaire completed by participants to The mean age of the 20 participants was 70 years (range 65-81
evaluate their experiences with and perceptions of the home years); 50% (10/20) of the participants were females, 40% (8/20)
exercise program and Physitrack system. Enjoyment about the were classified as overweight (BMI 25-29.9 kg/m2), and 45%
exercise program and using Physitrack was assessed on a 5-point (9/20) as obese (BMI ≥30 kg/m2), with 35% (7/20) reporting
scale (1=did not enjoy at all, 5=extremely enjoyable). In the presence of a chronic disease(s) and a median of 3
addition, participants were asked open-ended questions about medications.
what they liked and disliked most about using Physitrack. The

http://mhealth.jmir.org/2021/1/e21094/ JMIR Mhealth Uhealth 2021 | vol. 9 | iss. 1 | e21094 | p. 4

(page number not for citation purposes)
XSL• FO
RenderX
 JMIR MHEALTH AND UHEALTH Daly et al

Table 1. Baseline characteristics of the cohort.

Baseline characteristics Tablet owners (n=10) Tablet nonowners (n=10) Total (N=20)
Age (years), mean (SD) 70.1 (3.1) 70.8 (5.3) 70.4 (4.2)
Sex (% male), n (%) 5 (50) 5 (50) 10 (50)
Height (cm), mean (SD) 165.1 (12.4) 165.7 (9.3) 165.4 (10.7)
Weight (kg), mean (SD) 83.5 (14.3) 81.5 (20.2) 82.5 (17.1)

BMI (kg/m2), mean (SD) 30.7 (5.0) 29.3 (5.1) 30.0 (5.0)

Ethnicity, n (%)
Caucasian 6 (60) 3 (30) 9 (45)
Other 4 (40) 7 (70) 11 (55)
Highest level of education, n (%)
Primary/High school 3 (30) 0 (0) 3 (15)
University or Tertiary level 7 (70) 5 (50) 12 (60)
Technical/Trade certificate 0 (0) 5 (50) 5 (25)
Living arrangement, n (%)
Alone 0 (0) 3 (30) 3 (15)
With adult without children 9 (90) 5 (50) 14 (70)
With adult with children 0 (0) 1 (10) 1 (5)
Retirement village/hostel 1 (10) 1 (10) 2 (10)
Marital status, n (%)
Married/De Facto 9 (90) 8 (80) 17 (85)
Separated/Divorced/Widowed 1 (10) 2 (20) 3 (15)
Number of medications, median (IQR) 3.5 (1.0-4.0) 2.5 (1.0-5.5) 3.0 (1.0-4.0)

Presence of chronic disease(s),a n (%) 5 (50) 2 (20) 7 (35)

Previous fall in past 12 months, n (%) 2 (10) 0 (0) 2 (10)

a
Presence of chronic disease include self-reported hypertension, cardiovascular disease, stroke, Alzheimer disease, Parkinson disease, chronic kidney
disease, liver disease, type 2 diabetes, or a neurological/brain disease.

messages (from 480 prescribed exercise sessions) were sent to

Feasibility: Retention, Adherence, and Adverse Events the AEP by 9 of the 20 participants (median 6 per person).
Study retention was 95% (19 of 20 participants completed the
study). Mean exercise adherence over the 8 weeks was 84% Physical Activity
(95% CI 70%-97%, median 94%) and was no different between In the total cohort, mean weekly time spent walking increased
tablet owners and nonowners (mean 95% vs 72%, P=.07). Mean on average by 78 minutes (95% CI 0-156, P=.049; d=0.66]
adherence to the prescribed number of exercises per session (Table 2). Thirteen participants (68%) reported an increase in
was 81% (95% CI 68%-95%), number of sets was 82% (95% the weekly walking time (range 10-360 minutes), 2 (11%)
CI 68%-96%), and the number of repetitions for each exercise reported no change, and 4 (21%) reported a decrease (range
was 81% (95% CI 68%-95%), with no differences between the 30-240 minutes). For MVPA, there was a mean change of 41
2 groups (P>.05). Over the 8-week program, 1 musculoskeletal minutes (95% CI –8 to 90, P=.09; d=0.35]. Nine participants
complaint (knee pain that was pre-existing) and 1 injury (47%) reported an increase in the weekly MVPA time (range
(strained calf muscle) was reported by 2 participants. The 20-240 minutes), 6 (32%) reported no change, and 4 (21%)
participant with knee pain continued to exercise with a modified reported a decrease (range 15-180 minutes). There were no
program, while the second participant sought treatment and significant between-group differences (tablet versus not tablet
subsequently withdrew from the study. No falls were reported owners) for the change in either physical activity variable.
by any participant over the 8-week study. A total of 72 in-app

http://mhealth.jmir.org/2021/1/e21094/ JMIR Mhealth Uhealth 2021 | vol. 9 | iss. 1 | e21094 | p. 5

(page number not for citation purposes)
XSL• FO
RenderX
 JMIR MHEALTH AND UHEALTH Daly et al

Table 2. Changes in physical activity and the Short Physical Performance Battery scores for the tablet owners, tablet nonowners, and all participants
combined.
Parameters Baseline (N=20), Week 8 (n=19), Mean change (95% CI) Effect size (Cohen d)
mean (SD) mean (SD)
Walking time (min/week)
Tablet owners 154 (84) 225 (155) 67 (–32 to 166) 1.20
Tablet nonowners 155 (154) 244 (251) 89 (–50 to 227) 0.58
All 155 (121) 235 (206) 78 (0 to 156)a 0.66

Moderate-to-vigorous physical activity (min/week)

Tablet owners 50 (88) 97 (164) 42 (–37 to 121) 0.54
Tablet nonowners 151 (152) 191 (217) 40 (–36 to 116) 0.26
All 100 (132) 147 (195) 41 (–8 to 90) 0.36
Short Physical Performance Battery score
Tablet owners 11.3 (0.7) 11.7 (0.7) 0.3 (–0.3 to 1.0) 0.59
Tablet nonowners 11.7 (0.7) 11.9 (0.3) 0.2 (–0.4 to 0.8) 0.30
All 11.5 (0.7) 11.8 (0.5) 0.3 (–0.1 to 0.7) 0.42

a
P=.049 within group change after 8 weeks.

SUS
Physical Function
Based on SUS, the Physitrack app was reported to be highly
After 8 weeks, there was a nonsignificant mean 0.3 point (95%
usable by all participants (mean score 86, SD 10), with no group
CI –0.1 to 0.7, P=.17) improvement in the composite SPPB
differences (Table 3). For the 10 individual questions,
score in all participants, which represented a moderate effect
participants most strongly agreed that they felt confident using
(d=0.42) (Table 2). Five participants (26%) had an improvement
it and that most people would learn to use the system very
of one or more points in SPPB performance, 12 (63%) had no
quickly. Participants also strongly disagreed that the system
change, and 2 (11%) experienced a reduction. There were no
was cumbersome and unnecessarily complex.
group differences for the change in the mean composite SPPB
scores.

Table 3. Means and standard deviation scores for the System Usability Scale.

Questionsa Tablet owners Tablet nonowners All

I think I would like to use the app frequently 4.22 (0.67) 4.00 (1.05) 4.11 (0.88)
I found the system to be unnecessarily complex 1.33 (0.50) 1.40 (0.52) 1.37 (0.50)
I thought the system was easy to use 4.22 (1.30) 4.40 (0.97) 4.32 (1.11)
I think that I would need support of a technical person to be able to use the system 1.67 (0.87) 1.60 (0.84) 1.63 (0.83)
I found the various functions in the system were well integrated 4.22 (0.44) 4.10 (0.99) 4.16 (0.77)
I thought there was too much inconsistency in the system 2.00 (1.10) 1.50 (0.53) 1.74 (0.81)
I would imagine that most people would learn to use the system very quickly 4.56 (0.53) 4.30 (0.68) 4.42 (0.61)
I found the system very cumbersome to use 1.11 (0.33) 1.30 (0.48) 1.21 (0.42)
I felt very confident using the system 4.67 (0.50) 4.80 (0.42) 4.74 (0.45)
I needed to learn a lot of things before I could get going with the system 1.56 (0.73) 1.60 (1.27) 1.58 (1.02)
System Usability Scale total score (out of 100) 85.6 (7.6) 85.5 (11.8) 85.5 (9.8)

a
Responses were scored on a 5-point Likert scale: 1=strongly disagree, 5=strongly agree.

Process Measures: Perceptions of the Program and

PACES
System
In all participants, the mean (SD) physical activity enjoyment
(PACES) scores did not change over time (baseline 71.2 [8.6] Participants reported that they enjoyed using the Physitrack app
vs 8 weeks 69.2 [7.9], P=.45) nor differ between the groups and participating in the exercise program (median score 4 out
(P=.07). of 5), with ease of use and the narrated video demonstrations

http://mhealth.jmir.org/2021/1/e21094/ JMIR Mhealth Uhealth 2021 | vol. 9 | iss. 1 | e21094 | p. 6

(page number not for citation purposes)
XSL• FO
RenderX
 JMIR MHEALTH AND UHEALTH Daly et al

of the exercises within the app reported by participants as to improvements were rated as the most important factor for them
what they liked most. Overall, 71% (12/17) of the participants continuing with the program over the 8 weeks. Finally, 94%
reported no response in terms of what they disliked most about (16/17) of the participants reported that they would recommend
the app but a lack of clarity around the terminology (eg, the the home-based program to other older people and 88% (15/17)
terms reps and sets) was reported by 3 participants (17%). would continue to use the Physitrack app and exercise at home
Participants’ own motivation rather than the use of the app and if it were made available (Table 4).
interactions with the exercise trainer or potential health

Table 4. Participants’ perceptions of the exercise program and web-based Physitrack exercise programming system (n=17).
Questions Scores

How did you enjoy participating in the home-based exercise program?a, median=4.0, mean (SD) 3.50 (0.97)

How did you enjoy using the Physitrack online exercise system?a, median=4.0, mean (SD) 3.88 (0.86)

What did you like most about the Physitrack system (app)? n (%)
Ease of use 7 (41)
Narrated video demonstrations of exercises 7 (41)
Ability to track and modify program, if needed 3 (18)
What did you like least about the Physitrack system (app)? n (%)
Lack of clarity around terminology 3 (17)
Lack of flexibility if exercising on a different day 1 (6)
Just another piece of equipment (device) to use when exercising 1 (6)
No response 12 (71)

How important was each of the following to help you maintain your motivation to continue with the program?b,
median=2.0, mean (SD)
Own motivation 1.88 (0.93)
Physitrack system 2.18 (1.01)
Exercise trainer 2.24 (1.03)
Feedback from trainer and performance assessments 2.24 (1.03)
Improvements to health 2.47 (1.07)
Do you feel that taking part in the program improved your health and fitness? n (%) 15 (88)
Would you recommend the home-based program to other older people? n (%) 16 (94)
If the Physitrack system continued to be available, would you continue to exercise at home? n (%) 15 (88)

a
Responses scored on a 5-point scale: 1=did not enjoy at all, 5=extremely enjoyable.
b
Responses scored on a 5-point scale: 1=extremely important, 5=not important at all.

Most participants stated that they would recommend the

Discussion home-based exercise program to other older people and continue
Summary of the Main Findings to use the exercise programming app if it was still available,
which adds further support to the widespread usability and
Overall, the findings from our SMART prospective pilot acceptability of our intervention and the web-based exercise
feasibility study indicate that it was safe and feasible for exercise programming app.
professionals to prescribe and remotely monitor a thrice weekly
home-based, multicomponent exercise program targeting Comparison With Prior Work
musculoskeletal health and function delivered via the Physitrack Consistent with our findings, several previous interventions
app for older adults living independently in the community. conducted over 2-6 months have reported that the use of
This feasibility was evident by the low attrition, high adherence web/mobile exercise prescription apps represent a safe, effective,
to the exercise training, low number of adverse events, increased and feasible approach to deliver tablet-based, home muscle
weekly physical activity time, and the high reported usability. strength and balance training for older people [26-30]. However,
No differences were observed in the outcomes between there were marked differences in the attrition rates (8%-47%)
participants who were previous owners and nonowners of a in some of these studies, and adherence to the exercise training
tablet computer. Participants reported that they enjoyed using ranged from 61% to 73% [27-29]. This heterogeneity is likely
the Physitrack app and participating in the exercise program. related to factors such as whether additional behavioral change

http://mhealth.jmir.org/2021/1/e21094/ JMIR Mhealth Uhealth 2021 | vol. 9 | iss. 1 | e21094 | p. 7

(page number not for citation purposes)
XSL• FO
RenderX
 JMIR MHEALTH AND UHEALTH Daly et al

strategies or remote support were provided or not, differences recommending that such approaches replace traditional
in the stability of internet connections, and duration of the community-based exercise programs but provide an alternative
interventions. In our 8-week study, the low attrition and high option that might best meet some individuals’ needs,
exercise adherence are likely due to several components related preferences, and financial resources.
to both the intervention and web-based exercise programming
An interesting observation from our study was the increase in
system. This includes the prescription of individualized exercise
weekly walking time following the 8-week home-based, exercise
programs based on individual’s functional status and
program. It is important to note that participants were not
health/medical history, the initial telephone calls and home
specifically instructed to engage in any additional physical
visits by the AEP, the option for participants to communicate
activity outside of the intervention. Thus, the reason(s) for the
with the AEP via the app at any time to receive
mean 78 minutes per week increase in weekly walking time is
feedback/support, knowing that the AEP was remotely
difficult to explain. However, there is some evidence that
monitoring all exercise programs, and the shorter study duration.
participation in structured exercise programs is associated with
Further data to support the feasibility and usability of the
an increase in nonprescribed activity and energy expenditure
exercise programming system used in our study is highlighted
outside of the intervention [34]. It has been suggested that this
by the findings from a 3-week pragmatic randomized controlled
may be related to a number of factors, including exercise-related
trial in 305 adults being treated for a musculoskeletal
improvements in functional capacity, gains in muscle strength,
condition(s) [31]. In this study, it was found that the use of the
reduced levels of fatigue or feeling more energetic,
Physitrack system by physical therapists improved home
improvements in exercise self-efficacy, or mood [34]. However,
exercise adherence and confidence in the ability of patients to
others have observed that adoption of structured exercise leads
undertake exercise at home compared to usual care (eg, written
to no change or a decrease in habitual physical activity or energy
exercise instructions, printed exercise diagrams) [31].
expenditure, which has been attributed to some compensatory
Although our study was not designed nor powered to detect an behavioral adaptation [35]. Given the small sample size and
effect of the intervention on physical function, we did observe wide confidence interval for the change in mean weekly walking
a modest effect (d=0.42) on improving SPPB scores (mean time in our study, these findings must be interpreted with caution
change 0.3 points). Previous research has indicated that a change but warrant further follow-up to understand the reason(s) why
in SPPB of 0.3-0.8 points represents a minimally important home-based exercise training may improve habitual activity
(significant) change [32], with a change for 0.5 and 1.0 point levels.
classified as a small but meaningful change and substantial
change, respectively [33]. In our study, it is likely that the
Limitations
modest changes relate to the initial functional status of our This study has a number of limitations, including the small
participants, who had a mean SPPB score of 11.5 out of 12. sample size, convenience sample that limits generalizability,
Indeed, this may explain why only 26% (n=5) of the participants pretest-posttest study design, relatively short intervention
experienced an improvement of one or more points on the SPPB duration, lack of blinding of the assessor, lack of a nonexercise
test after 8 weeks. Nevertheless, our findings must be interpreted control group, the use of self-reported measures of physical
with caution given that there was no control (nonexercise) activity, and the inclusion of generally healthy and physically
comparison group. For comparison, a previous 12-week active older adults with normal functional capacity, which may
tablet-based, home strength and balance training program in 44 also affect generalizability. The 3 home-based visits by the AEP
independently living older adults found that the intervention to brief participants on the use of the app and exercise program
and control groups experienced similar significant improvements and to conduct the functional tests is a further limitation in terms
in SPPB scores [26]. of future widespread scalability. The need for the AEP to
regularly review and monitor the messages/alerts from
An important finding from our study was that participants were participants and their weekly progress using the Physitrack
enthusiastic about the web-based Physitrack system and its ease app/platform could be considered burdensome, but the daily
of use, independent of whether they had previously owned a time commitment was typically less than 5 minutes. Finally,
tablet computer and would be willing to continue to use such a the addition of semistructured interviews may provide further
system in the future. Despite these positive experiences, overall insights into the experiences and perspectives of the participants,
enjoyment in physical activity did not change over the 8 weeks, including potential differences between tablet computer owners
which may be related to the relatively short duration of the and nonowners.
exercise program or that the questionnaire we used to monitor
physical activity enjoyment may not measure the ideal constructs Conclusion
specific to the intervention. The lack of any marked changes in This pilot feasibility study indicates that it was safe and feasible
physical activity enjoyment (and SPPB performance) may also for older adults living independently in the community to
be due in part to the fact that participants recruited into the study participate in a tablet-computer–delivered, home-based,
were already habitually active (mean, 100 minutes of MVPA multicomponent exercise program targeting musculoskeletal
per week). While future studies are needed to evaluate the health and function that was developed and monitored remotely
long-term acceptability, adherence, and clinical effectiveness by exercise professionals using a web-based exercise
of remotely prescribed web-based exercise programs using apps prescription app.
for older adults, we wish to highlight that we are not

http://mhealth.jmir.org/2021/1/e21094/ JMIR Mhealth Uhealth 2021 | vol. 9 | iss. 1 | e21094 | p. 8

(page number not for citation purposes)
XSL• FO
RenderX
 JMIR MHEALTH AND UHEALTH Daly et al

Acknowledgments
This study was supported by an internal grant from the Institute for Physical Activity and Nutrition, Deakin University. The
sponsor was not involved in the design, conduct, collection, analysis, or interpretation of data; in writing of the manuscript; or
in the decision to publish the findings.

Authors' Contributions
RMD, JG, NM, and RM designed the study. JG and TH conducted this study. TH collected the data. RMD and JG performed
data analysis and interpretation. JG and RMD drafted the manuscript. RMD, JG, NM, and RM revised the manuscript content
and provided the final approval. RMD takes responsibility for the integrity of the data analysis.

Conflicts of Interest
None declared.

References
1. Bergen G, Stevens MR, Burns ER. Falls and Fall Injuries Among Adults Aged ≥65 Years - United States, 2014. MMWR
Morb Mortal Wkly Rep 2016 Sep 23;65(37):993-998 [FREE Full text] [doi: 10.15585/mmwr.mm6537a2] [Medline:
27656914]
2. AIHW. Hospitalisations due to falls by older people, Australia 2006-07. Injury research and statistics series no. 57. Cat.
No. INJCAT 133. Canberra, AIHW. 2012. URL: https://www.aihw.gov.au/getmedia/
260e721e-4d8b-4ec6-b3a9-dc9c82a723cb/12266.pdf.aspx?inline=true [accessed 2020-05-01]
3. AIHW. Trends in hospitalisations due to falls by older people, Australia 2002-03 to 2012-13. Injury research and statistics
series no. 106. Cat. no. INJCAT 182. Canberra, AIHW. 2017. URL: https://www.aihw.gov.au/getmedia/
5f84eadd-6f25-4429-82fc-5e9072278335/aihw-injcat-182.pdf.aspx?inline=true [accessed 2020-05-01]
4. Watts JJ, Abimanyi-Ochom J, Sanders K. Osteoporosis costing all Australians: A new burden of disease analysis-2012 to
2022, Glebe, NSW: Osteoporosis Australia. 2013. URL: https://www.osteoporosis.org.au/sites/default/files/files/
Burden%20of%20Disease%20Analysis%202012-2022.pdf [accessed 2020-06-01]
5. Eastell R, Rosen CJ, Black DM, Cheung AM, Murad MH, Shoback D. Pharmacological Management of Osteoporosis in
Postmenopausal Women: An Endocrine Society* Clinical Practice Guideline. J Clin Endocrinol Metab 2019 May
01;104(5):1595-1622. [doi: 10.1210/jc.2019-00221] [Medline: 30907953]
6. Daly RM, Dalla Via J, Duckham RL, Fraser SF, Helge EW. Exercise for the prevention of osteoporosis in postmenopausal
women: an evidence-based guide to the optimal prescription. Braz J Phys Ther 2019;23(2):170-180 [FREE Full text] [doi:
10.1016/j.bjpt.2018.11.011] [Medline: 30503353]
7. Sherrington C, Fairhall N, Wallbank G, Tiedemann A, Michaleff ZA, Howard K, et al. Exercise for preventing falls in
older people living in the community: an abridged Cochrane systematic review. Br J Sports Med 2020 Aug;54(15):885-891.
[doi: 10.1136/bjsports-2019-101512] [Medline: 31792067]
8. Zhao R, Zhao M, Xu Z. The effects of differing resistance training modes on the preservation of bone mineral density in
postmenopausal women: a meta-analysis. Osteoporos Int 2015 May;26(5):1605-1618. [doi: 10.1007/s00198-015-3034-0]
[Medline: 25603795]
9. Kukuljan S, Nowson CA, Sanders KM, Nicholson GC, Seibel MJ, Salmon J, et al. Independent and combined effects of
calcium-vitamin D3 and exercise on bone structure and strength in older men: an 18-month factorial design randomized
controlled trial. J Clin Endocrinol Metab 2011 Apr;96(4):955-963. [doi: 10.1210/jc.2010-2284] [Medline: 21209030]
10. Kukuljan S, Nowson CA, Sanders K, Daly RM. Effects of resistance exercise and fortified milk on skeletal muscle mass,
muscle size, and functional performance in middle-aged and older men: an 18-mo randomized controlled trial. J Appl
Physiol (1985) 2009 Dec;107(6):1864-1873 [FREE Full text] [doi: 10.1152/japplphysiol.00392.2009] [Medline: 19850735]
11. Gianoudis J, Bailey CA, Ebeling PR, Nowson CA, Sanders KM, Hill K, et al. Effects of a targeted multimodal exercise
program incorporating high-speed power training on falls and fracture risk factors in older adults: a community-based
randomized controlled trial. J Bone Miner Res 2014 Jan;29(1):182-191 [FREE Full text] [doi: 10.1002/jbmr.2014] [Medline:
23775701]
12. Daly RM, Gianoudis J, Kersh ME, Bailey CA, Ebeling PR, Krug R, et al. Effects of a 12-Month Supervised,
Community-Based, Multimodal Exercise Program Followed by a 6-Month Research-to-Practice Transition on Bone Mineral
Density, Trabecular Microarchitecture, and Physical Function in Older Adults: A Randomized Controlled Trial. J Bone
Miner Res 2020 Mar;35(3):419-429. [doi: 10.1002/jbmr.3865] [Medline: 31498937]
13. Ziebart C, McArthur C, Lee L, Papaioannou A, Laprade J, Cheung AM, et al. "Left to my own devices, I don't know":
using theory and patient-reported barriers to move from physical activity recommendations to practice. Osteoporos Int 2018
May;29(5):1081-1091. [doi: 10.1007/s00198-018-4390-3] [Medline: 29441402]
14. Cavill NA, Foster CE. Enablers and barriers to older people’s participation in strength and balance activities: A review of
reviews. JFSF 2018 Jun;03(02):105-113. [doi: 10.22540/jfsf-03-105]

http://mhealth.jmir.org/2021/1/e21094/ JMIR Mhealth Uhealth 2021 | vol. 9 | iss. 1 | e21094 | p. 9

(page number not for citation purposes)
XSL• FO
RenderX
 JMIR MHEALTH AND UHEALTH Daly et al

15. Timmons JF, Griffin C, Cogan KE, Matthews J, Egan B. Exercise Maintenance in Older Adults 1 Year After Completion
of a Supervised Training Intervention. J Am Geriatr Soc 2020 Jan;68(1):163-169. [doi: 10.1111/jgs.16209] [Medline:
31647585]
16. Yardley L, Kirby S, Ben-Shlomo Y, Gilbert R, Whitehead S, Todd C. How likely are older people to take up different falls
prevention activities? Prev Med 2008 Nov;47(5):554-558. [doi: 10.1016/j.ypmed.2008.09.001] [Medline: 18817810]
17. Australian Institute of Health and Welfare. The Active Australia Survey: A guide and manual for implementation, analysis
and reporting. 2003. URL: https://www.aihw.gov.au/getmedia/ff25c134-5df2-45ba-b4e1-6c214ed157e6/aas.pdf.
aspx?inline=true [accessed 2020-06-01]
18. Guralnik JM, Ferrucci L, Pieper CF, Leveille SG, Markides KS, Ostir GV, et al. Lower extremity function and subsequent
disability: consistency across studies, predictive models, and value of gait speed alone compared with the short physical
performance battery. J Gerontol A Biol Sci Med Sci 2000 Apr;55(4):M221-M231. [doi: 10.1093/gerona/55.4.m221]
[Medline: 10811152]
19. Ostir GV, Volpato S, Fried LP, Chaves P, Guralnik JM. Reliability and sensitivity to change assessed for a summary measure
of lower body function. Journal of Clinical Epidemiology 2002 Sep;55(9):916-921. [doi: 10.1016/s0895-4356(02)00436-5]
20. Kendzierski D, DeCarlo K. Physical activity enjoyment scale: two validation studies. Int J Sport Exerc Psychol 1991;13:50-64.
[doi: 10.1123/jsep.13.1.50]
21. Bangor A, Kortum PT, Miller JT. An Empirical Evaluation of the System Usability Scale. International Journal of
Human-Computer Interaction 2008 Jul 30;24(6):574-594. [doi: 10.1080/10447310802205776]
22. Braun V, Clarke V. Using thematic analysis in psychology. Qualitative Research in Psychology 2006 Jan;3(2):77-101.
[doi: 10.1191/1478088706qp063oa]
23. Abbade LPF, Abbade JF, Thabane L. Introducing the CONSORT extension to pilot trials: enhancing the design, conduct
and reporting of pilot or feasibility trials. J Venom Anim Toxins Incl Trop Dis 2018;24:4 [FREE Full text] [doi:
10.1186/s40409-018-0142-2] [Medline: 29434634]
24. Whitehead AL, Julious SA, Cooper CL, Campbell MJ. Estimating the sample size for a pilot randomised trial to minimise
the overall trial sample size for the external pilot and main trial for a continuous outcome variable. Stat Methods Med Res
2016 Jun;25(3):1057-1073 [FREE Full text] [doi: 10.1177/0962280215588241] [Medline: 26092476]
25. Cohen J. Statistical power analysis for the behavioral sciences. New York: Lawrence Erlbaum Associates; 1988.
26. van Het Reve E, Silveira P, Daniel F, Casati F, de Bruin ED. Tablet-based strength-balance training to motivate and improve
adherence to exercise in independently living older people: part 2 of a phase II preclinical exploratory trial. J Med Internet
Res 2014 Jun 25;16(6):e159 [FREE Full text] [doi: 10.2196/jmir.3055] [Medline: 24966165]
27. Silveira P, van de Langenberg R, van Het Reve E, Daniel F, Casati F, de Bruin ED. Tablet-based strength-balance training
to motivate and improve adherence to exercise in independently living older people: a phase II preclinical exploratory trial.
J Med Internet Res 2013 Aug 12;15(8):e159 [FREE Full text] [doi: 10.2196/jmir.2579] [Medline: 23939401]
28. Geraedts HAE, Zijlstra W, Zhang W, Spoorenberg SLW, Báez M, Far IK, et al. A Home-Based Exercise Program Driven
by Tablet Application and Mobility Monitoring for Frail Older Adults: Feasibility and Practical Implications. Prev Chronic
Dis 2017 Feb 02;14:E12 [FREE Full text] [doi: 10.5888/pcd14.160227] [Medline: 28152361]
29. Nikitina S, Didino D, Baez M, Casati F. Feasibility of Virtual Tablet-Based Group Exercise Among Older Adults in Siberia:
Findings From Two Pilot Trials. JMIR Mhealth Uhealth 2018 Feb 27;6(2):e40 [FREE Full text] [doi: 10.2196/mhealth.7531]
[Medline: 29487045]
30. Hong J, Kong H, Yoon H. Web-Based Telepresence Exercise Program for Community-Dwelling Elderly Women With a
High Risk of Falling: Randomized Controlled Trial. JMIR Mhealth Uhealth 2018 May 28;6(5):e132 [FREE Full text] [doi:
10.2196/mhealth.9563] [Medline: 29807877]
31. Bennell KL, Marshall CJ, Dobson F, Kasza J, Lonsdale C, Hinman RS. Does a Web-Based Exercise Programming System
Improve Home Exercise Adherence for People With Musculoskeletal Conditions?: A Randomized Controlled Trial. Am J
Phys Med Rehabil 2019 Oct;98(10):850-858. [doi: 10.1097/PHM.0000000000001204] [Medline: 31021823]
32. Kwon S, Perera S, Pahor M, Katula JA, King AC, Groessl EJ, et al. What is a meaningful change in physical performance?
Findings from a clinical trial in older adults (the LIFE-P study). J Nutr Health Aging 2009 Jun;13(6):538-544. [doi:
10.1007/s12603-009-0104-z] [Medline: 19536422]
33. Perera S, Mody S, Woodman R, Studenski S. Meaningful change and responsiveness in common physical performance
measures in older adults. J Am Geriatr Soc 2006;54:743-749. [doi: 10.1111/j.1532-5415.2006.00701.x]
34. Drenowatz C, Grieve GL, DeMello MM. Change in energy expenditure and physical activity in response to aerobic and
resistance exercise programs. Springerplus 2015;4:798 [FREE Full text] [doi: 10.1186/s40064-015-1594-2] [Medline:
26702387]
35. Melanson EL. The effect of exercise on non-exercise physical activity and sedentary behavior in adults. Obes Rev 2017
Feb;18 Suppl 1:40-49 [FREE Full text] [doi: 10.1111/obr.12507] [Medline: 28164451]

Abbreviations
AEP: accredited exercise physiologist

http://mhealth.jmir.org/2021/1/e21094/ JMIR Mhealth Uhealth 2021 | vol. 9 | iss. 1 | e21094 | p. 10

(page number not for citation purposes)
XSL• FO
RenderX
 JMIR MHEALTH AND UHEALTH Daly et al

BMD: bone mineral density

MVPA: moderate-to-vigorous physical activity
PACES: physical activity enjoyment scale
RPE: rate of perceived exertion
SMART: Seniors Made Active thRough Technology
SPPB: short physical performance battery
SUS: system usability scale

Edited by L Buis; submitted 05.06.20; peer-reviewed by N Gell, H Hawley-Hague; comments to author 04.08.20; revised version
received 25.08.20; accepted 29.10.20; published 13.01.21
Please cite as:
Daly RM, Gianoudis J, Hall T, Mundell NL, Maddison R
Feasibility, Usability, and Enjoyment of a Home-Based Exercise Program Delivered via an Exercise App for Musculoskeletal Health
in Community-Dwelling Older Adults: Short-term Prospective Pilot Study
JMIR Mhealth Uhealth 2021;9(1):e21094
URL: http://mhealth.jmir.org/2021/1/e21094/
doi: 10.2196/21094
PMID: 33439147

©Robin M Daly, Jenny Gianoudis, Travis Hall, Niamh L Mundell, Ralph Maddison. Originally published in JMIR mHealth and
uHealth (http://mhealth.jmir.org), 13.01.2021. This is an open-access article distributed under the terms of the Creative Commons
Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction
in any medium, provided the original work, first published in JMIR mHealth and uHealth, is properly cited. The complete
bibliographic information, a link to the original publication on http://mhealth.jmir.org/, as well as this copyright and license
information must be included.

http://mhealth.jmir.org/2021/1/e21094/ JMIR Mhealth Uhealth 2021 | vol. 9 | iss. 1 | e21094 | p. 11

(page number not for citation purposes)
XSL• FO
RenderX