Enabling health and maintaining independence for older people at home: the HomeHealth RCT

Background
Frailty services commonly target moderate-severely frail older people, despite evidence suggesting frailty can be reduced when addressed at earlier stages. HomeHealth is a new home-based, voluntary sector service supporting older people with mild frailty to maintain independence through behaviour change. Support workers discuss the older person’s priorities and supports them to set and achieve goals around mobility, nutrition, socialising and/or psychological wellbeing.
Aims
We tested the clinical and cost-effectiveness of HomeHealth for maintaining independence in older people with mild frailty in a randomised controlled trial (RCT).
Methods
Design: Single-blind parallel RCT open between 18/01/2021 and 04/07/2023, with mixed methods process evaluation.
Setting: Community-dwelling older people aged 65+ years with mild frailty from 27 general practices, community groups and sheltered housing in London, Yorkshire and Hertfordshire.
Randomisation: Participants were randomised 1:1 to receive HomeHealth or treatment as usual (TAU).
Outcomes: Our primary outcome was independence in activities of daily living (modified Barthel Index, BI), analysed using linear mixed models. Secondary outcomes included frailty phenotype score, extended activities of daily living, wellbeing, psychological distress, loneliness, cognition, falls and mortality. Health economic outcomes included quality of life, capability and service use including hospital admissions. Cost-effectiveness acceptability curves and cost-effectiveness planes were used to represent the probability of cost-effectiveness compared to TAU.
Process evaluation: We conducted semi-structured interviews with participants receiving the intervention, HomeHealth workers and other stakeholders supporting service delivery. Interviews were thematically analysed. Fidelity of audio-recorded appointments was assessed by two independent raters. We evaluated potential mechanisms of impact using data from appointments attended, types of goals set and progress towards goals.
Findings
We recruited 388 participants, mean age 81.4 years (SD6.5), 64% female and 94% White British/European. HomeHealth did not improve BI scores at 12 months (0.250, 95%CI -0.932 to 1.432). At 6 months, we found small significant reductions in psychological distress (-1.237; 95%CI -2.127, -0.348), and frailty phenotype score (-0.252; 95%CI -0.487, -0.017). At 12 months, we found significant improvements in wellbeing (1.449; 95%CI 0.124, 2.775), reduced unplanned admissions (IRR: 0.65; 95%CI 0.54, 0.92) with lower associated costs (-£586/participant; 95% CI -351, -821). There were no differences in other outcomes. HomeHealth dominates TAU with a negative point estimate for incremental costs (-796; 95% CI -2016, 424), positive point estimate for incremental quality adjusted life years (0.009; -0.021, 0.039) and high probability of cost-effectiveness.
For our process evaluation, 64 semi-structured interviews were completed, including 49 participants, 7 HomeHealth workers and 8 stakeholders. The service was acceptable and safe, with overall good fidelity of delivery. Participants made progress on personalised goals, with most working on enhancing mobility. They found the service empowering, and received emotional and practical support. Engagement was more challenging when participants identified no need for change, had significant memory impairment or where there was new/declining illness. Flexibility around varying symptoms and incorporating behaviour change into existing routines promoted engagement.
Conclusion
HomeHealth did not improve independent functioning for older people with mild frailty. There were however small significant improvements in frailty status, psychological distress and wellbeing and a 35% reduction in unplanned admissions, with high probability of cost-effectiveness.
Limitations
We used a pragmatic design where the intervention was delivered by voluntary-sector partners in real-world settings during and after the Covid-19 pandemic, potentially leading to more variability in delivery and more challenging to demonstrate effectiveness. Our findings might not apply to other geographical settings with different health-care systems.