Home rehabilitation after delirium reached older adults who could engage, but not at scale
Recovery after delirium remains one of acute medicine’s most stubborn gaps. The RecoverED feasibility study tested whether a theory-informed, person-centred rehabilitation intervention could be delivered after hospital discharge to adults over 65 years who had experienced delirium, and whether the pathway was workable enough to support a definitive randomised trial. The intervention was built around the practical consequences of delirium: impaired cognition, deconditioning, reduced confidence, disrupted routines, and the psychosocial strain that often follows an acute admission. Rather than treating discharge as the end of care, the programme extended recovery into the home, where many of the problems actually persist.
The study used a multicentre, single-arm feasibility design across six NHS acute hospitals in the UK. Nineteen patient and carer pairs were recruited, a small sample by design, but one that allowed the team to examine whether the intervention could be accepted and delivered in the setting where future evaluation would have to occur. The programme itself was intensive enough to matter clinically without becoming unrealistic for routine services: a comprehensive home assessment, collaborative goal setting, up to 10 personalised sessions over 12 weeks, and a recovery record used to structure progress, education and psychosocial support. Delivered by occupational therapists, physiotherapists and rehabilitation support workers, the package addressed cognitive, physical, physiological and psychosocial needs in parallel.
A single delirium episode can fracture recovery in ways that are easy to miss on ward rounds. Persistent disorientation, weakness and low confidence often outlast the acute event, and families are left to manage the consequences before the next clinic visit ever happens.
Recruitment was the main bottleneck, not bedside delivery
The feasibility results were strongest once participants entered the programme. Of 419 patients identified as having delirium, only 36 met full eligibility, which immediately narrowed the pool available for recruitment. Nineteen patient and carer pairs agreed to participate, giving a consent rate of 53% with a 95% CI of 35% to 70%. From there, 13 participants started the intervention, and 10 completed final follow-up. Those figures translated into a start rate of 68% and a completion rate of 53%, with corresponding confidence intervals that reflect the uncertainty expected in a feasibility study.
Once enrolled, participants generally engaged with the intervention in a way that supports further development. The mean number of sessions completed was eight, with a standard deviation of 2.9, suggesting that the planned dose was achievable for many who were able to begin treatment. Baseline assessments were conducted during hospitalisation or after discharge, with the initial assessment occurring a mean of 18 days after discharge, and 77% completed within 14 days. In a recovery pathway shaped by frailty, delirium and fluctuating function, that timing matters clinically because it places rehabilitation in the narrow post-discharge window when patients are still vulnerable yet potentially receptive to support.
The trial process itself also pointed to a workable measurement strategy. Nineteen participants completed the primary outcome at baseline, and 10 completed it at six months. For a feasibility study, those numbers are less about efficacy than about whether a future trial can collect enough data without overburdening participants or carers. The fact that follow-up remained possible at six months is important in older adults after delirium, where attrition can easily erode the value of a larger study if procedures are too complex or if the post-discharge burden is too heavy.
Costing suggests the intervention is measurable, while effectiveness still needs proof
The economic evaluation estimated a total cost of £1249.29 per participant, including assessments, intervention sessions and training costs. That figure does not establish value for money, but it does establish that the intervention can be costed in a structured way, which is essential before a full economic evaluation can be trusted. In rehabilitation research, especially for older adults with complex needs, a feasible cost model is often the difference between a future trial that can inform commissioning and one that cannot speak to service planning at all.
The clinical logic for this programme is straightforward. Delirium is not only an acute neuropsychiatric syndrome; it often marks a deeper decline in functional reserve. A home-based rehabilitation approach can, in principle, help reverse some of the secondary harms by restoring mobility, reinforcing daily routines, supporting cognition, and reducing the isolation that commonly follows discharge. The recovery record and personalised sessions turn that principle into a structured intervention rather than an informal promise of follow-up.
What remains unresolved is not whether the intervention can be delivered in a few selected patients, but whether recruitment can be made efficient enough for a definitive randomised controlled trial. The eligibility gap between the 419 identified patients and the 36 who met full criteria shows how quickly the recruitment funnel narrows once real-world delirium, comorbidity and discharge timing are applied. Any future trial will need stronger strategies to identify eligible patients earlier, secure consent at the right moment, and minimise drop-off before intervention start.
RecoverED therefore lands in an important but narrow clinical space. The intervention appears acceptable and operationally workable among those who entered the study, and the trial procedures produced usable feasibility, follow-up and cost data. The remaining barrier is scale: a definitive trial will have to solve recruitment before it can answer the more consequential question of whether post-delirium home rehabilitation changes recovery enough to justify wider adoption.