Physical Activity Surveillance Through Smartphone Apps and Wearable Trackers: Examining the UK Potential for Nationally Representative Sampling.
United Kingdom
adult
exercise
fitness trackers
health surveys
mobile phone
smartphone
surveys and questionnaires
Journal
JMIR mHealth and uHealth
ISSN: 2291-5222
Titre abrégé: JMIR Mhealth Uhealth
Pays: Canada
ID NLM: 101624439
Informations de publication
Date de publication:
29 01 2019
29 01 2019
Historique:
received:
10
08
2018
accepted:
22
09
2018
revised:
19
09
2018
entrez:
30
1
2019
pubmed:
30
1
2019
medline:
30
1
2019
Statut:
epublish
Résumé
Smartphones and wearable activity trackers present opportunities for large-scale physical activity (PA) surveillance that overcome some limitations of questionnaires or researcher-administered devices. However, it remains unknown whether current users of such technologies are representative of the UK population. The objective of this study was to investigate potential sociodemographic biases in individuals using, or with the potential to use, smartphone apps or wearable activity trackers for PA surveillance in the United Kingdom. We used data of adults (aged ≥16 years) from two nationally representative surveys. Using the UK-wide 2018 Ofcom Technology Tracker (unweighted N=3688), we derived mutually adjusted odds ratios (ORs; 95% CI) of personal use or household ownership of a smartwatch or fitness tracker and personal use of a smartphone by age, sex, social grade, activity- or work-limiting disability, urban or rural, and home nation. Using the 2016 Health Survey for England (unweighted N=4539), we derived mutually adjusted ORs of the use of wearable trackers or websites or smartphone apps for weight management. The explanatory variables were age, sex, PA, deprivation, and body mass index (BMI). Furthermore, we stratified these analyses by BMI, as these questions were asked in the context of weight management. Smartphone use was the most prevalent of all technology outcomes, with 79.01% (weighted 2085/2639) of the Technology Tracker sample responding affirmatively. All other outcomes were <30% prevalent. Age ≥65 years was the strongest inverse correlate of all outcomes (eg, OR 0.03, 95% CI 0.02-0.05 for smartphone use compared with those aged 16-44 years). In addition, lower social grade and activity- or work-limiting disability were inversely associated with all Technology Tracker outcomes. Physical inactivity and male sex were inversely associated with both outcomes assessed in the Health Survey for England; higher levels of deprivation were only inversely associated with websites or phone apps used for weight management. The conclusions did not differ meaningfully in the BMI-stratified analyses, except for deprivation that showed stronger inverse associations with website or phone app use in the obese. The sole use of PA data from wearable trackers or smartphone apps for UK national surveillance is premature, as those using these technologies are more active, younger, and more affluent than those who do not.
Sections du résumé
BACKGROUND
Smartphones and wearable activity trackers present opportunities for large-scale physical activity (PA) surveillance that overcome some limitations of questionnaires or researcher-administered devices. However, it remains unknown whether current users of such technologies are representative of the UK population.
OBJECTIVE
The objective of this study was to investigate potential sociodemographic biases in individuals using, or with the potential to use, smartphone apps or wearable activity trackers for PA surveillance in the United Kingdom.
METHODS
We used data of adults (aged ≥16 years) from two nationally representative surveys. Using the UK-wide 2018 Ofcom Technology Tracker (unweighted N=3688), we derived mutually adjusted odds ratios (ORs; 95% CI) of personal use or household ownership of a smartwatch or fitness tracker and personal use of a smartphone by age, sex, social grade, activity- or work-limiting disability, urban or rural, and home nation. Using the 2016 Health Survey for England (unweighted N=4539), we derived mutually adjusted ORs of the use of wearable trackers or websites or smartphone apps for weight management. The explanatory variables were age, sex, PA, deprivation, and body mass index (BMI). Furthermore, we stratified these analyses by BMI, as these questions were asked in the context of weight management.
RESULTS
Smartphone use was the most prevalent of all technology outcomes, with 79.01% (weighted 2085/2639) of the Technology Tracker sample responding affirmatively. All other outcomes were <30% prevalent. Age ≥65 years was the strongest inverse correlate of all outcomes (eg, OR 0.03, 95% CI 0.02-0.05 for smartphone use compared with those aged 16-44 years). In addition, lower social grade and activity- or work-limiting disability were inversely associated with all Technology Tracker outcomes. Physical inactivity and male sex were inversely associated with both outcomes assessed in the Health Survey for England; higher levels of deprivation were only inversely associated with websites or phone apps used for weight management. The conclusions did not differ meaningfully in the BMI-stratified analyses, except for deprivation that showed stronger inverse associations with website or phone app use in the obese.
CONCLUSIONS
The sole use of PA data from wearable trackers or smartphone apps for UK national surveillance is premature, as those using these technologies are more active, younger, and more affluent than those who do not.
Identifiants
pubmed: 30694198
pii: v7i1e11898
doi: 10.2196/11898
pmc: PMC6371078
doi:
Types de publication
Journal Article
Research Support, Non-U.S. Gov't
Langues
eng
Sous-ensembles de citation
IM
Pagination
e11898Subventions
Organisme : Medical Research Council
ID : MC_UU_12015/3
Pays : United Kingdom
Informations de copyright
©Tessa Strain, Katrien Wijndaele, Søren Brage. Originally published in JMIR Mhealth and Uhealth (http://mhealth.jmir.org), 29.01.2019.
Références
Med Hypotheses. 2018 Oct;119:32-36
pubmed: 30122488
Prev Med Rep. 2015 Dec 30;3:90-7
pubmed: 26844194
Nature. 2017 Jul 20;547(7663):336-339
pubmed: 28693034
Prev Med Rep. 2016 Oct 28;5:124-126
pubmed: 28101443
Digit Health. 2017 Nov 09;3:2055207617740088
pubmed: 29942617
J Neurosci Methods. 2014 Jul 15;231:22-30
pubmed: 24091138
Lancet. 2012 Jul 21;380(9838):247-57
pubmed: 22818937
Int J Behav Nutr Phys Act. 2015 Dec 18;12:159
pubmed: 26684758
Am J Epidemiol. 2017 Sep 15;186(6):648-658
pubmed: 28486584
Vital Health Stat 2. 2014 Apr;(165):1-53
pubmed: 24775908
Int J Gen Med. 2017 Sep 12;10:293-303
pubmed: 28979157
BMC Public Health. 2017 Nov 15;17(1):880
pubmed: 29141607
J Med Internet Res. 2018 May 02;20(5):e177
pubmed: 29720359
J Med Internet Res. 2017 Apr 05;19(4):e101
pubmed: 28381394