Improving Shared DECISion-making in oldEr Hip Fracture Patients Living With Frailty

WP1 - Functional outcomes and quality of life following hip fracture in older adults living with different degrees of frailty: A systematic review Aim To synthesise the literature on the relationship between frailty and functional and QoL outcomes in patients presenting with hip fracture.

Rationale Existing literature on the link between frailty and function after hip fracture is limited. This review will guide WP2's design, ensuring it addresses current gaps.

Methods and design I will work with my supervisory team and university librarians to develop a search strategy. Databases searched will include Embase, MEDLINE, CINAHL, and Web of Science Core Collection. Grey literature (e.g. conference abstracts) will be searched using Grey Matters. Bibliographies of included studies will be screened, and all databases searched from earliest record to present.

Inclusion and exclusion criteria Inclusion Criteria Exclusion Criteria Population Age ≥65 Frailty as measured by a validated screening tool Diagnosed with an acute hip fracture including proximal femoral shaft, head, and neck fractures whether this results in admission or not Age <65 Studies that do not report frailty status Distal femoral fractures, impacted femoral head fractures, periprosthetic fractures, pathological fractures Exposure Frailty defined by a validated scoring system within the study No measure of frailty Outcomes Mortality, Mobility, Functional status, Health-related QoL, Discharge destination Studies that do not report at least one of the relevant outcomes Study design Observational studies including prospective and retrospective cohort studies, Randomised controlled trials Case reports, review articles, animal and in vitro studies, editorials, book chapters or extracts Table 1. Inclusion and exclusion criteria for WP1 Data extraction Abstracts will be double screened by 2 researchers in Rayyan.ai, with duplicates removed and disagreements resolved by my supervisor. Full abstract texts will then be double screened against criteria. Extracted data will include title, journal, publication year, authors, study country, design, participant details, study aims, sample size, outcomes, measurement tools, results, and statistical analyses.

Assessment of bias The Joanna Briggs Institute (JBI) critical appraisal checklists for the respective study types included will be used to summarise the risk of bias for the included studies.

Data synthesis I will follow the PRISMA 2020 checklist. While meta-analysis is preferred, heterogeneity may preclude it. Cochran's Q test and the I² index will assess heterogeneity and synthesis without meta-analysis (SWiM) used if necessary.

WP2 - The predictive value of frailty on functional status following fragility fracture of the hip in older adults: A retrospective cohort study Aims To identify prognostic factors and assess the relationship between frailty and functional decline following hip fracture in older adults.

Rationale Most patients with hip fractures receive operative management regardless of frailty due to data showing high mortality with non-operative management. However, mortality is not the only concern to patients. My PPI work and existing studies highlight functional status as equally important. This study will provide functional outcome data across frailty levels to support informed decision-making.

Methods and design I will conduct a retrospective cohort study using routinely collected data from the Wessex Secure Data Environment (SDE) and the Clinical Practice Research Datalink (CPRD). I will identify older adults who had fragility hip fractures from August 2021-present to assess the relationship between frailty status and functional status. Prior to August 2021 hip fracture incidence was lower than expected due the Covid-19 pandemic(33), which may have influenced outcomes. The Barthel Index (BI) will be used assess functional status as it been validated in this setting(34).

Inclusion and exclusion criteria Eligible Ineligible Age ≥65 Fragility fracture of the proximal femur defined as a fall from standing height or less from August 2021 to present Age <65 Femoral shaft or distal femur fractures, periprosthetic fractures, or pathological fractures, stable impacted femoral head fractures, or high impact trauma Table 2. Inclusion and exclusion criteria for WP2 Data collection Data will be collected from the Wessex SDE and CPRD. Primary outcomes include functional status via the Barthel Index (BI) at discharge, 3, and 6 months. Secondary outcomes include mortality, days alive and at home, discharge destination and residential location at 3 and 6 months. I will identify other important secondary outcomes with PPI input.

I will collect data on variables that may influence the primary outcome, including age, sex, residential status, ethnicity, surgery type, time to surgery, admission day, fracture type, prosthesis type, surgeon and anaesthetist seniority, dementia diagnosis, baseline BI, and Charlson Comorbidity Index (CCI).

Data analysis I will perform linear regression to assess associations between frailty status and other risk factors, followed by linear regression to evaluate links between frailty and functional status at discharge, 3, and 6 months. A multivariate analysis will then explore relationships between all predictors and the primary outcome.

WP3 - Decision-making in the management of hip fractures in older adults living with frailty: a qualitative study Aim To explore decision-making processes in hip fracture management in older adults living with frailty, including exploring barriers to discussing non-operative care.

Rationale To support shared decision-making in hip fracture care, it is vital to understand key factors and barriers in discussing management options, including non-operative management. I will explore this through interviews with four groups: patients, their families/carers, NHS clinicians, and European clinicians.

Methods and design I will conduct a qualitative study using semi-structured interviews. Reflexive thematic analysis will enable me to critically examine interpretations of identified themes in the context of my perspective as a geriatrician(35).

Participant group Recruitment method Group 1 - Age ≥65 admitted to University Hospital Southampton (UHS) following a fragility hip fracture Screen new admissions to UHS orthopaedics using EPR system Group 2 - Families or non-paid carers in addition to group 1 participants Seek permission from group 1 participants to contact families and carers Group 3 - Registered healthcare professionals involved in acute hip fracture management decisions including physicians, orthopaedic surgeons, nurses and allied health professionals Email invitations to relevant professional groups using UHS departmental email address books and posters inviting potential participants displayed on orthopaedic wards Group 4 - Registered healthcare professionals (as per group 3) from other European countries Invitation emails to be disseminated via contacts from the British Geriatrics Society and the European Geriatric Medicine Society Table 3. Participants and recruitment for WP3 I will explore factors influencing hip fracture decision-making. Meetings with individual interviewees will last 45 minutes, using guides developed with PPI input and clinicians. Group 1 interviews will be at the bedside or by phone post-discharge; Groups 2 and 3 in a private office, by phone, or video call; Group 4 by phone or video call. Interviews will be recorded and transcribed.

Data analysis Interview transcripts will be thematically analysed using Braun and Clarke's steps(36). Themes will be compared within and across groups. NVivo will be used to manage data for thematic analysis. A PPI group will review coding and themes to ensure they reflect participants' views. I will be supported in qualitative data analysis by my supervisor Dr Natalie Cox, who has extensive experience in this field.

Sample size I will recruit until data saturation is reached. Previous reviews suggest this occurs at around 9 interviewees per group(37). However, exact numbers of interviewees will depend on data quality and richness. As groups 3 and 4 include varied professional roles, I will aim for at least 3 participants per profession, as suggested in a 2024 review, resulting in a target sample of 12 for these groups(38).

WP4 - Co-design and pilot testing of a patient decision aid to support shared decision-making for hip fracture management Aim Co-design a patient decision aid to support shared decision-making in hip fracture management with key stakeholders including patients, carers and clinicians, then pilot implementation locally to assess acceptability to these stakeholders.

Rationale To make informed decisions, patients need to understand all the possible treatment options, the likely outcomes with each option, and what the treatments involve. How information is presented to patients can vary between clinicians. Patient decision aids standardise information presented in accessible, comprehensible formats.

Patient decision aid development I will develop a paper-based patient decision aid following the development processes recommended by the International Patient Decision Aid Standards (IPDAS) framework(39). Decision support needs, patient priorities, and outcome data identified in preceding WPs will inform the design of a draft decision aid.

Patient, clinician, and carer participants from WP3 will be invited to form a focus group to review the draft decision aid. An iterative process of at least 3 cycles will be used to further develop the decision aid as per IPDAS(40). Nominal Group Technique (NGT), a structured consensus development technique, will be used to generate consensus on the final draft aid that will be pilot tested(41).

Figure 1. Model development process adapted from IPDAS Systematic Development of Patient Decision Aids flowchart(40).

Pilot testing A quasi-experimental pre- and post-test study will use the Decisional Conflict Scale (DCS), a validated tool to measure perceptions of uncertainty in health-related decision making(42), to assess the effectiveness of the patient decision aid. All older hip fracture patients over a 4-week period on orthopaedic wards at UHS will be invited to complete the DCS. Pilot implementation of the decision aid will then start, and subsequent patients, over 4 weeks, will be invited to complete the DCS. Difference in mean DCS scores before and after the pilot will be compared using a t-test.

Acceptability to clinicians of using the decision-aid will be assessed through a questionnaire co-designed with stakeholders, using the Theoretical Framework of Acceptability model(43) which has been used to inform development of acceptability measures in surgical settings(44).