Physical activity and frailty in older people with heart disease – learning from implanted cardiac device remote monitoring data

Student thesis: Phd


Remote monitoring of physical activity from accelerometers embedded in cardiac devices (pacemakers, defibrillators and cardiac resynchronisation therapy devices) is clinically feasible, yet evidence of utility is sparse. Prior to this thesis, research had demonstrated device measured activity predicted heart failure hospitalisations and death in certain device cohorts, with limited evidence of association with measures of physical performance and frailty. Investigation of activity trajectories over time had not been reported. This thesis aimed to examine the association between activity and frailty in an older cardiac device population and explore the existence of characteristic trajectory phenotypes. Data was selected from participants aged 60+ years from two observational studies: PATTErn and TriageHF Plus. Association with frailty was measured using multiple logistic regression, where OR = odds ratio. Latent class trajectory analysis was applied to investigate the presence of participant groups with characteristic activity trends, and relationship with hospitalisation and death measured by time to event analysis. Median activity in the 30-days prior to enrolment was 134.9±135.0 minutes/day for the PATTErn cohort (n = 140), and 137.7±118.9 minutes/day in TriageHF Plus (n = 247). Accounting for age, gender, body mass index, device type and recent hospitalisation, each additional hour of daily activity reduced the odds of frailty by 28% (OR 0.72, 95% CI: 0.57-0.90, p =0.004). Trajectory class characterisation for the majority was “steady� (79-85%), with additional small “peaked� and “downwards� trajectory groups. Clinical characterisation did not reveal distinctive profiles, and there were no significant associations with events in follow-up. Low activity (< 1 hour per day, 12.6% of participants) associated with increased risk of hospitalisation and death (adjusted hazards ratio 3.15, 95% CI 1.67 – 5.95, p
Date of Award1 Aug 2023
Original languageEnglish
Awarding Institution
  • The University of Manchester
SupervisorFozia Ahmed (Supervisor), Adam Greenstein (Supervisor) & Niels Peek (Supervisor)


  • Older people
  • Remote monitoring
  • Heart failure
  • Physical activity
  • Cardiac devices

Cite this