Why the investigators are doing this study? The best approach to rehabilitation after a broken hip is not known. A new approach could improve outcomes by tailoring rehabilitation to patient needs. This approach identifies subgroups of patients within a population who have different risks of poor outcomes. These subgroups are then matched to treatments better tailored to their needs. Survivors of a broken hip describe a tailored approach as key to recovery. Further, the NHS recommends this approach as central to healthcare progress.

This study wants to see if it is possible for the NHS to deliver this new approach to rehabilitation for older adults who break their hip.

What will be done? The investigators worked with patients to plan this study. Patients will keep helping the investigators during the study. Sixty older people who had surgery to fix a broken hip will be invited to take part. Participants will be given a level of risk (low, medium, or high) based on an online calculator (www.stratifyhip.co.uk).

All 60 participants will get usual care provided locally. Half, selected by chance, will get extra rehabilitation during their hospital stay including a self-managed exercise programme for the low-risk subgroup, education, a goal-orientated mobility programme and enhanced discharge planning for the medium-risk subgroup, and education, a goal-orientated activity of daily living programme, orientation, and enhanced assessment for the high-risk subgroup.

The investigators will collect information from the 60 people taking part, at the beginning, middle, and end of the study and again 12- weeks later.

What will the next step be? If this small study shows this extra rehabilitation can be provided in the NHS, and it may help patients, then the investigators plan to do a larger study. The larger study will see if this extra rehabilitation works to help older people get back home and feel happier.

Each year, United Kingdom (UK) hospitals admit 70,000 men and women over the age of 60 years with hip fracture. Even with surgery, 30% of patients die within a year. Among survivors, 25% never walk again, and 22% transition from independent living to nursing homes. This led 81 global societies to endorse a call to action to improve acute multidisciplinary care after hip fracture.

A recent Cochrane systematic review supports rehabilitation in hospital as an effective approach to reduce mortality after hip fracture. However, the nature of the rehabilitation interventions varied considerably limiting conclusions on the optimal components. This uncertainty has translated to NICE guidance being limited to daily mobilisation and regular physiotherapy review.

The generalisability of the evidence is also limited as many rehabilitation trials attempted to account for differences in the hip fracture population by targeting homogenous subgroups such as patients with cognitive impairment, women, or from nursing homes. It is therefore uncertain whether interventions deemed 'effective' are so for all patients, or for the targeted subgroup. This in turn poses challenges as to how these interventions may be implemented when subgroups compete for finite National Health Service (NHS) resources.

An intervention based on stratified rehabilitation may provide answers to these uncertainties. Stratified rehabilitation considers an entire population competing for resources to identify subgroups of patients with different risk of poor outcomes. Subgroups are then matched to rehabilitation tailored to their needs to optimise outcomes across the entire population. Hip fracture survivors recently described this tailored approach as key to successful recovery. Further, a stratified approach is regarded central to the progress of healthcare according to the NHS and House of Lords Science and Technology Committee. Although an approach of matching rehabilitation to patient subgroups with different risks of poor outcomes is intuitive and effective for other conditions, it has not been tested for rehabilitation after hip fracture.

A stratified approach to rehabilitation after hip fracture was subsequently developed. The study has been designed in collaboration with patients and carers, healthcare professionals working in the hospital setting, health services researchers and statisticians. The protocol has been scientifically appraised through the funding application process (with external peer review) and by experts in the field of orthogeriatric rehabilitation research.

First, the investigators developed and validated a subgrouping tool 'the stratify hip algorithm' (based on three multivariable prediction models) to identify patients at low-, intermediate- and high-risk of death and/or change in residence (to a higher level of care) using records for over 170,000 patients admitted to one of 173 hospitals in England and Wales. The approach requires website entry of 5 pieces of information (age, sex, prefracture mobility, prefracture residence, and dementia diagnosis) to generate a risk assignment.

Next matched interventions were designed for each subgroup identified by the algorithm. These interventions were informed by qualitative interviews with patients, physiotherapists, and the multidisciplinary team, an umbrella review of systematic reviews of older adults who underwent rehabilitation interventions in the acute setting, recent systematic (including Cochrane) reviews, international guidelines, consultation with our public and patient involvement group 'TROOP' (further detail of TROOP available at www.ppitroop.co.uk), and an intervention development workshop and survey (following a nominal group technique) to prioritize components with key stakeholders. The interventions comprise self-managed exercise (low risk subgroup); education, a goal orientated mobility programme, and enhanced discharge planning (medium risk subgroup); and education, a goal orientated activities of daily living programme, and enhanced assessment (high risk subgroup). Carers of participants in the medium and high risk subgroups will be invited to take part to receive training in supporting participants with their programmes.

The overarching purpose of a future main study is to determine the clinical- and cost-effectiveness of adding an intervention to usual care designed to improve outcomes of acute care for older adults after hip fracture. Several uncertainties will first be addressed through this randomised feasibility trial.

The primary objective of this feasibility and pilot randomised trial is to determine the treatment fidelity of the proposed intervention.

Secondary objectives seek to determine:

1. The acceptability of the intervention to participants, carers and therapists.
2. Barriers and enablers to intervention delivery.
3. Count of screened, eligible, approached, recruited and retained participants (and carers).
4. Acceptability, completeness, and descriptive comparison of outcome data collection.
5. Count of inadvertent unblinding of outcome assessors.
6. Count of adverse events (AE) and serious adverse events (SAE).
7. Indicative sample size for a definitive trial.

Sixty participants will be recruited (30 per treatment arm) with assessments at baseline, intervention end, and 12-week follow up. The intervention will start within 72 hours of surgery and end on discharge from the acute hospital. Feasibility criteria have been specified a priori to determine whether the trial will progress to a definitive trial.

Inclusion Criteria: We will include adults

• aged 60 years or more.
• admitted to hospital for surgical repair of a hip fracture.
• who are willing and able to provide consent or assent depending on the level of cognitive impairment.

Exclusion Criteria: We will exclude adults

• less than 60 years, to align with the National Hip Fracture Databases definition of the target population.
• not surgically treated, as this treatment approach is reserved for around 2% of patients in the UK who are often at the end of life.
• who broke their hip in hospital following admission for a different illness/injury as their anticipated care pathway and outcomes will vary from those who are admitted for hip fracture.
• participating in other treatment trials and without agreement of both trial teams.

An anonymised dataset will be preserved on the King's Open Research Data System. This dataset will include anonymised data collected which does not breach any agreement for outcome measure use.

Following publication of the primary paper, an anonymised dataset will be preserved indefinitely on the King's Open Research Data System.

Proof of ethical approval as a condition of access.