Device-Based Measurements of Physical Activity in Malaysian and UK Children and Adolescents

Abstract

The World Health Organization (WHO) suggests an average of 60 minutes of
moderate to vigorous intensity physical activity (PA) daily for children and adolescents
to enjoy its health benefits. Given the numerous advantages of PA to children and
adolescents, it is essential to monitor and assess their PA accurately. This thesis
aimed to (i) explore the use of smartphone technology for PA surveillance in children
and adolescents and its accuracy in comparison with device-based measures; (ii)
explore the experiences of children and adolescents in the UK and Malaysia with
smartphone applications (apps) and wearable activity trackers (WAT) for PA
monitoring; (iii) examine the validity of smartphone apps and commercial WAT for
assessing PA in Malaysian adolescents.
A systematic literature review identified eight previous research studies utilising
smartphone technology for PA surveillance among children and adolescents globally.
Interestingly, all research was conducted in high-income countries, highlighting the
need for similar research in low- and middle-income countries (LMIC). It was revealed
that PA surveillance of children and adolescents using smartphone technology was
lacking in LMIC. Some of the apps used in the studies showed accuracy in PA
measurement; however, it was recommended that more validation studies be
conducted using smartphone apps.
The cross-sectional survey further revealed that smartphones were prevalent among
Malaysian adolescents (82.1%), suggesting their potential use for PA monitoring. For
Malaysian participants, smartphone ownership differed across socioeconomic groups,
with 85.0% for low family affluence scale (FAS), 41.9% for middle FAS, and 35.7% for
high FAS. For UK participants, it was 55.6% (low FAS), 25.7% middle FAS and 60.0%
high FAS. Commercial WAT was less commonly owned by UK and Malaysian
adolescents (15.7%). As smartphones and commercial WAT are increasingly
accessible to children and adolescents, further validation of these tools for PA
monitoring is required, especially in an LMIC such as Malaysia.
The Huawei Band 4 was selected for validation as an affordable commercial wearable
activity tracker (WAT). Fifty-four adolescents aged 14 to 18 wore both the Huawei
Band 4 and Axivity AX6 for five days in free-living conditions. The Huawei Band 4
showed acceptable agreement in estimating step count (MAPE = 14.7%, 𝑟 = 0.678)
and time spent in moderate-to-vigorous PA (MAPE = 35.2%, 𝑟 = 0.643) compared to
the Axivity AX6. Similarly, the smartphone apps Apple Health (MAPE = 48.5%, 𝑟 =
0.477) and Samsung Health (MAPE = 26.8%, 𝑟 = 0.530) also demonstrated
acceptable agreement for step counts. However, an equivalence test revealed that the
data from the Huawei Band 4 and smartphone apps were not equivalent to the Axivity
AX6, indicating caution in interpreting their accuracy as measures of PA.
This thesis concludes that smartphone apps and commercial WAT are promising tools
for monitoring PA among children and adolescents, especially in Malaysia. Future
work is needed to understand how such tools can support the national surveillance of
PA in children and adolescents in LMIC. Of key importance is how such tools can be
implemented to allow for the collection of cost-effective and valid PA data to inform
future national PA guidelines and policies.

Date of Award	10 Dec 2024
Original language	English
Awarding Institution	University of Bristol
Supervisor	Miranda E G Armstrong (Supervisor) & Joey Murphy (Supervisor)