Could technology revolutionise physical activity guidelines and interventions?

Symposium Purpose:
Technology is becoming ever more engrained in everyday life on a global level. Smart and wearable devices continue to evolve and offer new opportunities to monitor and change movement behaviours. Physical activity patterns can now be measured more precisely, in greater detail, and for longer periods of time, linking to other health indicators for better data-driven insights into health. Furthermore, effective behavioural change techniques can be more easily delivered in engaging and attractive ways and on a scale that was, until recently, unimaginable. But with these opportunities come important challenges, including data privacy and security, digital disparities, and sustaining physical activity behaviour change on a population level. Critical discussions are needed on how to move forward in this space, knowing that technology will only become more prevalent globally and that no one should be left behind. We must harness its potential to impact positive change for all.

Symposium Overview:
We will explore the opportunities and challenges afforded by technology in the physical activity and health space. We invite three presenters working in different domains of digital health to present a holistic overview of current innovations and future perspectives around technology for physical activity and health. These topics will include:

1. Guidelines: How technology can deliver data-driven insights into physical activity behaviours and health outcomes, and how these insights can inform future physical activity guidelines
2. Surveillance: New device-based methods to measure physical activity patterns and what that means for guideline surveillance
3. Interventions: How innovations in technology can be used to support and maintain population level physical activity behaviour change

A critical panel discussion with invited guests from the field and led by the World Health Organization will conclude the session by focusing on:

1. How technology will impact future guidelines and surveillance, and
2. What are the critical joint actions needed to realise positive change?

Chair: Dr Jacqueline Mair, Singapore-ETH Centre, Singapore

Presenter 1: Professor (Manos) Emmanuel Stamatakis, The University of Sydney, Australia

Presenter 2: Dr Elroy Aguiar, The University of Alabama, USA

Presenter 3: Dr Jacqueline Mair, Singapore-ETH Centre, Singapore

Discussant: Dr Fiona Bull, World Health Organization, Switzerland

Abstract 1: Guidelines; Emmanuel Stamatakis

Title: Wearable device-based evidence on physical activity and health: present and future

Background: Most evidence (>90%) used for physical activity and sedentary behaviour guideline development comes from questionnaire-based studies.

Purpose: To discuss recent epidemiological evidence on wearable device-based physical activity and mortality/non-communicable disease outcomes and provide directions for maximising the contribution of wearable-based evidence to forthcoming guidelines and clinical practice.

Methods: We will review recent (2015 -2024) landmark epidemiological accelerometery studies that estimated associations with health outcomes including mortality, CVD, and cancer. We will synthesise findings from individual studies, harmonised meta-analyses, and wearables data consortia, and discuss the implications of such evidence in future guideline formation. We will highlight implications for data collection and processing methods (including machine learning), and comment on the generalisability of existing findings.

Results: An increasingly large volume of device-based evidence is being published. Almost all evidence comes from western high-income countries. Existing studies have utilised accelerometric sensors placed on the waist, wrist, or thigh, while standard sensors found in consumer wearables such as heart rate monitors, gyroscopes, and altimeters are absent. Large-scale cohorts and consortia have provided novel and complementary insights into the health impact of device-measured physical activity. There is a clear trend towards niche aspects of movement behaviours (e.g. VILPA) and holistic examination of the 24-hour day, including sleep.

Conclusions: Wearable device-based research on the health effects of physical activity and sedentary behaviour has accelerated in recent years and will continue to do so in the foreseeable future. However, critical gaps remain, including research in LMICs where wearables evidence is virtually absent.

Practical implications: it is likely that wearable device-based evidence will shape major aspects of movement behaviour guidelines in the next decade. Tight collaborations between key organisations and such as the WHO, ISPAH, the ProPASS consortium, and industry/device manufacturers are needed to expand wearables evidence beyond high income countries.

Funding: None

Abstract 2: Surveillance; Elroy J Aguiar, Cristal J. Benitez, Katherine Sullivan, Michael V. Fedewa

Title: Beyond steps per day: Characterization of physical activity using a comprehensive system of step-based metrics

Background: Steps per day is a widely reported metric that captures ambulatory physical activity (PA). However, it is often criticized for only reflecting daily volume of PA, while ignoring intensity, a central tenet of PA guidelines. Step-based metrics, encompassing both steps per day and effort/intensity-based expressions of stepping (peak-cadence and cadence bands), are becoming increasingly popular as a comprehensive method of evaluating ambulatory PA.

Purpose: To characterize PA levels of a sample of young women from the United States using step-based metrics.

Methods: Young women (N=50, age=22.0±4.1 years, BMI 24.4±4.4 kg/m2) wore an accelerometer on their waist for 7 days. Accelerometer data were downloaded in 1-min epochs and processed into step-based metrics (steps/day, peak-cadence indices, and cadence bands [steps/min]) using a custom R package. Summary statistics (median [IQR]) were calculated, and daily volume of PA was classified using a Graduated Step Index: basal activity <2500; limited activity 2500–4999; low active 5000–7499; somewhat active 7500–9999; active 10,000–12499, and highly active ≥12500 steps/day.

Results: Summary statistics were as follows: steps per day 8709 [5934-11275], peak 1-min cadence 121.0 [114.6-130.5] steps/min, and peak 30-min cadence 103.6 [80.6-115.2] steps/min. Participants spent 298.4 [254.5-346.3] min in incidental stepping (1–19 steps/min), and 31.1 [15.1–51.1] min at a cadence corresponding to moderate-to-vigorous PA (MVPA; ≥100 steps/min). Participants were classified (count, %) as: basal activity (n=0, 0%), limited activity (n=8, 16%), low active (n=15, 30%), somewhat active (n=8, 16%), active (n=13, 26%), highly active (n=6, 12%).

Conclusions: Approximately half (54%) of young women were classified as physically active (7500 steps/day), and half performed between 15 and 51 min/day of MVPA.

Practical implications: Step-based metrics can be used to comprehensively describe daily volume, intensity, and pattern of step accumulation. Further studies should explore associations between step-based metrics and health outcomes among young women.

Funding: None

Abstract 3: Interventions; Jacqueline Mair, Sarah Edney, Esther Na, Thitikorn Topothai, Xin Hui Chua, Müller-Riemenschneider

Title: Population-wide mobile health interventions promoting healthy movement behaviours: a systematic review of real-world evidence

Background: Despite the development of numerous mHealth interventions promoting healthy movement behaviours, few have achieved real-world implementation at the necessary scale for population health impact.

Purpose: To summarise findings from a systematic review of mHealth interventions promoting healthy movement behaviours in real-world settings.

Methods: Systematic searches of five databases (Medline, Embase, Scopus, Web of Science, PsycINFO) and Google Scholar were conducted. Included studies were those implemented at scale, using mHealth, and targeting physical activity, sedentary behaviour, or sleep. Data were extracted on study and intervention design, movement behaviour outcomes, and implementation indicators. Risk-of-bias was assessed using the Effective Public Health Practice Project tool.

Results: After screening 7134 records, 21 interventions reported in 39 studies were included. Most interventions (92%) were conducted in high-income countries, and in terms of scale, 5% were implemented citywide, 48% across part of a country, 29% across an entire country, and 19% globally. Smartphone apps and wearables were the most used (71% each), followed by text messages (14%) and tablets (10%). The median number of participants reached was n=7,944. Where data were available, participation rates ranged from 1.2% to 28.6% and adherence rates (actual use versus intended use) ranged from 20%-83%. Fifteen interventions (71%) reported positive effects on movement behaviours and 1 (4.8%) reported no change. Almost half (48%) of the interventions reviewed are currently still active.

Conclusions: mHealth interventions promoting healthy movement behaviours are being implemented at scale, effectively reaching large populations. Short-term effectiveness in changing movement behaviours is evident but sustaining long-term change requires attention. However, current evidence is limited to high-income countries and heterogeneous study designs and methodologies pose challenges for direct intervention comparisons.

Practical Implications: While mHealth interventions can effectively influence movement behaviours at scale in the short term, addressing their sustainability is crucial to achieve population health impact.

Funding: None