Physical activity pre- and post-total knee replacement, using NHS and consumer health and fitness tracker data

This research project aims to:

• Conduct a systematic literature review to summarise what is known about physical activity pre-and-post-TKR
• Conduct a qualitative study in pre-operative and post-operative TKR patients to explore physical activity behaviours and determine what digital health information should be used to summarise patterns of activity pre-and-post-TKR
• In partnership with Fitbit and Janeiro Digital, to identify patients who have undergone TKR, collect consent, and link their Fitbit data with their NHS data
• Analyse the linked data to:
  • Understand the representativeness of Fitbit-owning TKR patients;
  • Generate descriptive statistics of activity levels pre-and-post-TKR;
  • Examine patterns of activity in those with good and bad outcomes.

The details

The uptake of consumer technology has been significant in the last 10 years. Mobile devices collect digital traces from our everyday lives, for example tracking patterns of activity. Such data may be successfully reused for health research. One disease area where mobility is notably impacted by symptoms is arthritis, making this an interesting use case to understand how we might make use of retrospectively collected consumer data for health insights. Total knee replacements (TKRs) are a common surgical procedure for knee osteoarthritis.

Prior to surgery, clinicians and patients need to make shared informed decisions about the likely benefits and potential harms of surgery. Benefits commonly cited include improvements in pain and discomfort and improvement in mobility. Improvements in pain are supported by strong evidence, however, changes in mobility and activity levels are less clear. Understanding the patterns of physical activity pre-and post-op have historically been limited to physical activity questionnaires. Some studies are starting to use more objective measures including accelerometry – although this typically requires the provision of devices and prospective data collection. To date, the numbers of such studies are limited, outcome measures vary, and assessment is at fixed intervals such as 6- or 12-months post-op.

In the population of 80k UK total knee replacement patients per year, a subset will have been tracking their physical activity data via consumer devices (Fitbit, Google Fit, Strava etc). This provides a major opportunity to learn about changes in activity from real-world data. It is important to understand, though, whether this subset is representative of the whole population of interest, and how the findings can (or cannot) be extrapolated. We also need to know what aspects of activity are of interest to stakeholders, especially patients, so that the analysis can be usefully translated into meaningful information to guide shared informed decision making in the future.

The graduate researcher on this project is: Ayobami Esther Olanrewaju

Supervision team

• The University of Manchester: Professor William Dixon
• The University of Melbourne: Doctor Stephanie Filbay, Professor Kim Bennell and Doctor Mark Merolli