The design of an exercise and nutrition intervention trial for middle-aged adults aimed at preventing muscle loss.

Background: Muscle loss is a characteristic of ageing. Mitigating this loss may have benefits for delaying or preventing the onset of the disease sarcopenia. Strength training and protein consumption have been suggested as intervention strategies for muscle loss prevention. The overarching objective of this thesis was to design a multi-component, remotely-delivered exercise and nutrition intervention trial protocol for middle-aged adults to prevent muscle loss. The research involved preceding steps that contributed to the protocol design.
Methods:
1) The first study involved a cross-sectional survey validating an amended protein screener tool. Participants responded to the survey online, with two separate cohorts completing this. The first cohort was asked to complete the protein screener questionnaire followed by a Food Frequency questionnaire (FFQ), to obtain readouts for protein intakes. The second cohort, used for further validation, completed the protein screener, followed by a 24-hour food and drink recall. This group also provided demographic information and completed an additional questionnaire to test their protein knowledge. The protein screener tool was validated using receiver operating curves and sensitivity and specificity calculations at three cut-off points, comparing predicted protein intake, based on screener prediction scores, with dietary reference measurements (FFQ and 24-hour recall.)
2) Associations between protein intakes, protein knowledge and demographic/lifestyle characteristics were explored using regression analysis based on the responses collected from cohort two.
3) The final study was a feasibility trial delivered to adults aged 40-65. This was conducted remotely, via video classes, and involved an exercise training component, predominantly using resistance bands. Depending on their randomisation arm, a subset of participants were asked to take one or two protein supplement(s). Quantitative and qualitative measurements of recruitment rates, adherence and acceptability were obtained, including semi-structured interview analysis.
Results: The protein screener tool reached an acceptable area under the curve (AUC) to be deemed appropriate for use in a UK adult population at a cut-off of 0.1 (AUC=0.731, sensitivity: 48.6 (CI:48.3-49.0)) when validated against an FFQ. However, further validation with 24-hour recalls failed to reach the 0.7 “fair” threshold (AUC=0.658.) Correlations between protein intakes, age, body mass index, physical activity levels and dietary habits were demonstrated. A significant relationship between protein knowledge and protein intake was only found in older adults (55 years+) (p=0.011.) Recruitment and adherence rates for both interventions were satisfactory. Acceptability of the exercise and protein intervention were high and low, respectively.
Conclusions: Based on pre-defined progression criteria, it was deemed feasible to progress to a full-scale randomised controlled trial. A protocol based on the insights and findings of this thesis has been developed, including participant inclusion criteria, and future work should test this.