Ageing versus frailty: older adults' physical activity habits

  • Dominique Radcliffe

Student thesis: Phd


Introduction: Frailty can be defined as a reduction in person’s ability to maintain efficient homeostasis. Pre-frailty is state where person is neither frail nor robust and could be the stage where it is most effective to intervene to prevent decline. This thesis explores pre-frailty and the development of an intervention to prevent decline into frailty, following the MRC Framework for complex interventions. Methods: following the MRC Framework for complex interventions, a systematic style review was first undertaken aiming to establish whether interventions aimed at pre-frail community-dwelling adults were effective in reducing risk factors associated with frailty and preventing the transition in to frailty. Following the systematic style review, two studies were undertaken using statistical techniques in secondary analysis of existing datasets to further understand the population of interest. The first, used Latent Class Growth Modelling to identify trajectories of functional decline (FD) in a representative population of English 60 to 70 year olds using the English Longitudinal Study of Ageing (ELSA). The second also used ELSA and Latent Class Analysis with distal outcomes to identify whether baseline functional decline score was associated with frailty status. Results: The systematic style review identified seven eligible studies, with 918 total participants (513 receiving interventions). Six studies had a physical activity component and one study had physical activity as a possible treatment. All multiple component interventions were combined with a nutritional component. A high risk of bias was identified for most studies and it was difficult to draw conclusions due to small numbers of participants and lack of between group analyses. Nonetheless, there was some evidence to suggest physical activity could be beneficial in preventing transition into frailty in a pre-frail population. Study one identified trajectories of functional decline; a four-trajectory class was found to be the model of best fit. Trajectory classes were: (a) Continuously healthy (40.29% participants), (b) Little decline (26.06% participants), (c) Intermediate decline (23.27% participants), (d) High decline (10.38% participants). Only class (c) clearly deteriorated over time. The functional decline trajectory classes were associated with frailty status, lower trajectory classes were less likely to be frail and higher trajectory classes more likely to be frail. Study two identified that age, gender and smoking status were positively associated with Frailty Index score and marital status and drinking regularly were negatively associated. Discussion: Results from these studies have identified that further research and interventions targeting pre-frail populations are required. The statistical studies identified that there was a trajectory class of intermediate decline which deteriorated over time and was likely to benefit from interventions to prevent decline. These trajectory classes were associated with frailty and therefore an intervention aiming to prevent functional decline may also work to prevent frailty. Next Steps: As part of the next steps in intervention development I designed an intervention and developed two protocols for studies to be undertaken in the future. The intervention was a signposting/behaviour change intervention linking participants to specific types of exercise groups/classes based on frailty phenotype criteria identified. (i) The first protocol outlined a qualitative study, which aimed to assess the perceived acceptability and ease of implementation of a behaviour change and signposting intervention through semi-structured interviews with professionals and focus groups with older adults. (ii) The second protocol was for a pilot feasibility RCT of a signposting and behaviour change intervention in pre-frail older adults. This would identify participants through the use of the electronic Frailty Index in primary care and explor
Date of Award1 Aug 2023
Original languageEnglish
Awarding Institution
  • The University of Manchester
SupervisorChristopher Todd (Supervisor) & Helen Hawley-Hague (Supervisor)

Cite this