Protocol for a Low Threshold Exercise Intervention for Women Experiencing Homelessness and Addiction.

Previous linked research of exercise interventions in people experiencing homelessness, demonstrated high retention with excellent engagement and feedback from its female participants, yet caution was observed around engaging in a mixed exercised programme with men. Women have traditionally been underserved in research and policy, yet multiple sources cite women as the fastest growing homeless demographic. No study has examined how targeted women-only exercise programmes with nutritional supplementation work in practice in this population.

The overall aim of this study is to explore physical functioning and frailty outcomes of an exercise intervention with protein supplementation in a group of (non-geriatric) women experiencing homelessness and addiction challenges.

The objectives are:

1. To evaluate recruitment, retention and adherence rates, and any adverse effects of the intervention.
2. To evaluate pre- and post-intervention levels of physical, nutritional and frailty status, pain and general health status.
3. To evaluate programme feedback and gain deeper insights into participants perspectives and experiences of the exercise intervention.

This mixed-methods study will involve a single-arm intervention which will take place in Dublin city centre, in a day-service centre which provides services for women who are homeless and/or in addiction. Following the intervention qualitative research in the form of individual exit interviews will be conducted with 20 participants or up to the point of data saturation.

The 10-week intervention will involve a low threshold, three-times weekly exercise programme (two exercise classes with protein supplementation and an outdoor 'Park Walk') delivered over 12 weeks to offer flexibility to participants. It will be delivered by two research physiotherapists who will adhere to all safety procedures. Further flexibility will be facilitated with the provision of a four-week window for pre-intervention evaluations prior to commencement of the programme, and group or one-to one sessions will be delivered based on participant preference. The exercise classes will be multi-modal, with a primary focus on strength and based on core set of resistance exercises. Aerobic, balance and flexibility work will be integrated into the class and the exercises will be individualised based on initial assessment results and presentation of participants. Using a gender-based perspective, the following considerations and adaptions will be built into the class; (i) core-stability exercises to target pelvic floor and/or abdominal muscle weakness, (ii) bone building exercises to target peri-and postmenopausal bone loss and (iii) age-associated muscle mass loss. Music, dance and fun orientated physical activity games, an important feature of prior linked studies, will be incorporated to optimize enjoyment and self-esteem. Borg's Rate of Perceived Exertion Scale will be used to monitor effort and scale the intensity of the workout. To promote post-exercise muscle protein synthesis, a nutritional supplement (200ml pre-prepared 'protein shake', Fresubin) which consists of 20g of protein will be offered immediately post exercise. The 'Park Walk' will focus on the aerobic component of the intervention, using green open space to maximise physical and mental health outcomes. It will be a flexibly arranged 20-30 minute self-paced group or one-to-one walk. To build sustainability beyond the ten week programme brief health promoting and physical activity educational interventions will be included in the exercise class setting to empower people to engage in unsupervised exercise following the study. Following the intervention one-to-one in-person exit interviews will be conducted to enable participants to provide feedback and express personal views, in order to enhance the understanding of behaviours, adherence and retention challenges and to provide insights for future programmes in this area. Interviews will continue until 20 participants are seen or until data saturation is reached. The interviews will be audio-recorded and transcribed verbatim and member checking will be utilised.

Analytic plan: For quantitative data, nominal or ordinal variables will be reported as frequencies and percentages. Continuous variables will be summarised as mean and standard deviation if normally distributed and median and inter-quartile range if non-normally distributed. Data will be tested for normality using the Kolmogorov-Smirnov/Shapiro Wilk test and will be compared across timepoints using the general linear model procedure (normally distributed data) and the Friedman's test (non-normally distributed data). Chi-squared t-tests will be used where appropriate and some data may be categorised to investigate relationships between variables. Exploratory regression models will be developed to explore correlates and predictors of frailty and poor physical functioning. A p-value of <0.05 will be considered significant. For the qualitative data, Braun and Clarke thematic analysis methodology will be employed to provide an in-depth analysis of the exit interview data.