The aims of this thesis were to investigate the changes in muscle strength that occur in older adults during unplanned hospital admissions, and to investigate the feasibility and efficacy of in-hospital exercise interventions in improving muscle strength and functional outcomes.

To investigate changes in muscle strength during hospitalisation, I conducted a systematic review and meta-analysis, and a repeated measures cohort study. The review concluded that adult patients who undergo an unplanned hospital admission may, regardless of age, experience a reduction of around 10% in knee-extension strength. To corroborate these findings in an older general medical population, I conducted a cohort study, which included 70 participants aged ≥75 recruited within 24 hours of acute hospital admission. Knee-extension strength declined during hospitalisation by 11% (p <0.001) but did not change by follow-up four to six weeks after hospitalisation (p = 0.458). General functional ability as measured by the Barthel Index deteriorated by 9 points (10%) between two weeks before admission and follow-up (p < 0.001). Functional mobility as measured by the de Morton Mobility Index improved by 6.5 points (15%) during hospitalisation (p < 0.001) but had not changed at follow-up (p = 0.508). A repeated measures mixed model showed that greater loss in knee-extension strength during hospitalisation was associated with increased sedentary time, frailty, baseline strength and lower baseline inflammatory levels. To investigate the feasibility and efficacy of in-hospital exercise interventions, I conducted a feasibility study and a meta-analysis. The feasibility study found that it is possible to conduct a randomised control trial of a twice-daily exercise programme for older adults during an acute hospital admission in an NHS hospital. The exercise intervention was well received, with most participants describing health benefits in terms of both physical improvement and increased confidence. The meta-analysis concluded that there is low certainty of evidence that exercise interventions, and more specifically progressive resistance training, when compared to usual care or sham interventions, results in improved independence with activities of daily living and functional mobility at discharge from hospital, without increasing the risk of adverse events. The thesis has several limitations including: that the primary research studies were all single-centre studies using convenience samples thereby limiting generalisability; that causality between sedentary activity and loss of knee-extension strength could not be confirmed in an observational design; and that recruitment to the feasibility study achieved only half of the initial target. Nevertheless, the objectives of the thesis were met, and provide a solid foundation to continue the development of interventions to prevent hospital-associated deconditioning. In line with the MRC guidance on the development of complex interventions, next steps will include refinement of the intervention, including identifying or developing an appropriate behavioural change theory, before further feasibility work and evaluation of the intervention.