Linking consumer health measures from smartphones and wearables with UK national data assets: Protocol for Physical Activity Patterns after Knee Arthroplasty (PAPrKA), a retrospective cohort study.
PAPrKA protocol
View resourceOur outputs
Evidence-based recommendations for future research using smartphones and wearables linked to national data assets. Helping you to navigate common challenges.
Main results: Linking consumer health measures from smartphones and wearables with UK national data assets: Physical Activity Patterns after Knee Arthroplasty (PAPrKA), a retrospective cohort study (COMING SOON)
Results from the PAPrKA study.
Resource currently being developed. Check back here once it is available
Opening the Black Box: Developing a Transparent Framework for Processing Consumer Smartphone/Wearable Data in Health Research (COMING SOON)
Smartphones and wearable devices count our steps every day, and researchers
increasingly use these records to study health. But before the data can be analyzed,
researchers have to make many hidden choices, such as how to handle repeated records
or what to do when a smartphone and a watch disagree. These choices are rarely
reported, yet they can change the results. We developed a framework that makes each
choice visible and easy to report. Using data from people who had knee replacement
surgery, we show that different reasonable choices can lead to different answers to
the same question.
Resource currently being developed. Check back here once it is available.
Using pre-collected consumer-grade accelerometry data on physical activity linked to the UK National Joint Registry knee replacement records for research: A feasibility study (COMING SOON)
A study examining recruitment process, wearable data completeness and
representativeness of samples enrolled into PAPrKA study.
Resource currently being developed. Check back here once it is available.
Impact of different levels and patterns of missing data on patterns of treatment response: a simulation study (COMING SOON)
People respond to treatment in different ways: after knee replacement, for example,
some recover their activity quickly, others slowly. Smartphones and wearable devices
can follow this recovery day by day, which helps reveal these different patterns of
response. But people do not always carry their phone or wear their device, so many
days of data go missing. This study asks how much missing data, and what kind of
missing data, can be tolerated before the patterns of treatment response we find
become misleading. We created simulator generated data for people recovering from
knee replacement, so the true patterns were known. We then removed data at different
levels, from a few days to almost all of them, and in different ways: at random,
mostly early or late in recovery, or more often on days when people were less
active. We then checked how well a standard statistical method could still find the
true patterns. The results will help researchers judge when their findings can be
trusted.
Resource currently being developed. Check back here once it is available.
Long COVID paper (COMING SOON)
A short description of this resource and what users will find when they open it.
Resource currently being developed. Check back here once it is available
How to think about ‘representativeness’ in population health studies using smartphones and wearables – and what to do about it (COMING SOON)
Mobile consumer devices such as smartphones and wearables offer significant opportunities for advancing population health research by providing real-time, objective data about disease, its causes and its consequences. However, unequal device ownership and usage due to digital literacy and connectivity - linked to factors like age and socioeconomic status - can limit participation.
This resource is an article consisting of a worksheet as well as empirical examples designed to help researchers consider representativeness when carrying out population health research using smartphones and wearables, reducing the potential for bias to affect final results.
Resource currently being developed. Check back here once it is available
Strategies to support the recruitment and retention of under-served population groups in smartphone and wearable health research: a scoping review (COMING SOON)
Paper investigating how best to recruit under-served populations.
Resource currently being developed. Check back here once it is available
Longitudinal Studies Using Patient-Generated Smartphones and Wearable Data for Population Health Research in Long-Term Conditions: A Scoping Review. (COMING SOON)
Resource currently being developed. Check back here once it is available.
View resource