Programme Grants for Applied Research (Feb 2025)
Enhanced recovery pathway for older people with hip fracture and cognitive impairment in acute hospitals: the PERFECTED research programme including an RCT
Abstract
Background Hip fracture has a substantial impact on the health, well-being and independence of patients and their families. In the 12 months after fracture, patients are at increased risk of cognitive and functional decline, admission to long-term care institutions and higher mortality. People with cognitive impairment are among the most vulnerable in acute hospital settings. They have lower short-term survival, with 24% mortality during admission. They are susceptible to suboptimal and inconsistent care standards that contribute to cognitive deterioration, increase risk of postoperative complications, prolong their length of stay and cause loss of independence. Objectives Establish best-practice from a systematic review of literature, observations of practice, perspectives of service users, carers, healthcare professionals, health service managers and experts in the field. Design the care pathway. Determine cultural/organisational changes necessary to implement and maximise adherence to the enhanced recovery pathway in hospital settings. Develop staff training and a training manual. Undertake a feasibility randomised controlled trial and collect outcomes to identify potential clinical and cost-effectiveness of the enhanced recovery pathway. Disseminate the findings and develop a definitive trial bid. Design A programme to develop an enhanced recovery pathway for people with hip fracture and cognitive impairment, tested for implementation and refined in the clinical environment. This refined enhanced recovery pathway was then tested in a feasibility study in 10 hospitals across the UK. Setting Acute care. Participants Hospital staff, people with cognitive impairment and hip fracture, carers and national and international experts in hip fracture or dementia. Interventions An enhanced recovery care pathway with checklist and an implementation process. Main outcome measures Mortality, patient and carer quality of life, cognition, activities of daily living. Data sources Clinical trial. Results A total of 284 participants were recruited, 132 to the PEFECT-ER intervention arm and 150 to the control arm, had good retention in the study and provided data for analysis. There was no evidence of any systematic between group difference at either the point of discharge from hospital or at 1-month follow-up. However, at 3 months, a relatively small effect of around one quarter of a standard deviation (0.071 units), was evidenced with respect to the health-related quality of life of the patient based on the EuroQol-5 Dimensions, five-level version by proxy in the intervention group (95% confidence interval 0.018 to 0.124; p = 0.009). A difference of 0.099 units in favour of the intervention group was also seen at the 6-month follow-up (95% confidence interval 0.001 to 0.198; p = 0.047). ‘Timed Up and Go’ and the Suitable Informant EuroQol-5 Dimensions, five-level version showed a no statistically significant difference except the model for length of stay. Those individuals in the intervention group had significantly longer lengths of stay, on average 1.22 times longer (95% confidence interval 1.02 to 1.45; p = 0.028). Mortality was similar in both groups, with a 6.1% mortality rate by 30 days post surgery. The process evaluation found that patients and carers were unable to comment on receiving the intervention. Limitations This was a feasibility study and was not designed as a definitive evaluation of the intervention. Lack of direct access to patient notes meant that researchers were unable to verify the Perioperative Enhanced Recovery hip FracturE Care of paTiEnts with Dementia-Enhanced Recovery check listing results. The relationship between changes in documentation of practices and changes in care practices is also unclear. Patient and suitable informants did not assist understandings of implementation, mechanisms of action or experiences of interacting with the intervention. Client Services Receipt Inventory data collection burden was an issue. Conclusions The Perioperative Enhanced Recovery hip FracturE Care of paTiEnts with Dementia-Enhanced Recovery feasibility trial demonstrated mean recruitment of 1.87 participant per centre per month. Retention at 1 month was over 80% and at 6 months approximately 50%. This information is useful for those wishing to design a definitive clinical trial. Although 30-day mortality was the same in both groups, the potential for reduction, by Perioperative Enhanced Recovery hip FracturE Care of paTiEnts with Dementia-Enhanced Recovery being implemented, exists from cumulatively increased good practices across a range of care domains. To compare longer-term survival of patients who received the intervention, we would recommend measuring 3-month (110-day) mortality in addition to 30-day mortality. These data are readily available from National Hip Fracture Database and are thus ideal for efficient trial design. Client Services Receipt Inventory can be reduced for a definitive trial, removing equipment questions and some community health use questions. Qualitative interviews with Perioperative Enhanced Recovery hip FracturE Care of paTiEnts with Dementia-Enhanced Recovery trial patient and carer should not take place. Future work Work to date shows that the intervention pathway for Perioperative Enhanced Recovery hip FracturE Care of paTiEnts with Dementia-Enhanced Recovery required considerable input from champions for delivery. We are exploring further funding options to facilitate work to understand these mechanisms and further test, pilot and produce the Perioperative Enhanced Recovery hip FracturE Care of paTiEnts with Dementia-Enhanced Recovery manual. Trial registration This trial is registered as Current Controlled Trials ISRCTN99336264. Funding This award was funded by the National Institute for Health and Care Research (NIHR) Programme Grants for Applied Research Programme (NIHR award ref: DTC-RP-PG-0311-12004) and is published in full in Programme Grants for Applied Research; Vol. 13, No. 1. See the NIHR Funding and Awards website for further award information. Plain language summary What was the problem? Hip fracture and thought/memory (‘cognitive’) impairments, such as dementia and delirium, are major challenges for older patients, their families/carers and the National Health Service and social care. The outcomes of medical treatment for elderly patients with hip fracture are often poor, and worse when patients have memory and thinking problems. There is little research on how best to look after this patient group in hospital. Our previous work shows that patients, families, carers and staff repeatedly seek more sensitive ways to look after this patient group. Staff have also highlighted the need for training to help them work more appropriately to meet these patients’ needs. What did we do? We have created, adapted and piloted a set of care actions called Perioperative Enhanced Recovery hip FracturE Care of paTiEnts with Dementia-Enhanced Recovery. Perioperative Enhanced Recovery hip FracturE Care of paTiEnts with Dementia-Enhanced Recovery comprises a best practice checklist, a staff training manual, staff time to put the checklist into practice and to train colleagues, and a process to improve care of this patient group continuously. We tested Perioperative Enhanced Recovery hip FracturE Care of paTiEnts with Dementia-Enhanced Recovery in five hospitals and asked whether staff found it acceptable. We wanted to find out if Perioperative Enhanced Recovery hip FracturE Care of paTiEnts with Dementia-Enhanced Recovery could be used in a bigger trial, testing whether Perioperative Enhanced Recovery hip FracturE Care of paTiEnts with Dementia-Enhanced Recovery improved outcomes for older people with hip fracture and memory problems. What did we find? We found that Perioperative Enhanced Recovery hip FracturE Care of paTiEnts with Dementia-Enhanced Recovery was practical to use, both in the trial and by ward staff. Some measures we planned to use to calculate its costs were not completed by enough people to analyse in the study, suggesting that a future trial should have different measures. We found that measuring the number of people dying and people’s quality of life should be considered for the trial. What does this mean? Our evidence suggests that Perioperative Enhanced Recovery hip FracturE Care of paTiEnts with Dementia-Enhanced Recovery can be used in hospitals and was acceptable to staff. There remains a need for a larger evaluation to investigate whether Perioperative Enhanced Recovery hip FracturE Care of paTiEnts with Dementia-Enhanced Recovery provides benefit for this patient group. Scientific summary Text in this section reproduces material from Cross JL, Hammond SP, Shepstone L, Poland F, Henderson C, Backhouse T, et al. PERFECTED enhanced recovery pathway (PERFECT-ER) versus standard acute hospital care for people after hip fracture surgery who have cognitive impairment: a feasibility cluster randomised controlled trial. BMJ Open 2022;12:e055267. https://doi.org.uea.idm.oclc.org/10.1136/bmjopen-2021-055267. This article is distributed under the terms of the Creative Commons Attribution 4.0 International Licence (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution and reproduction in any medium, provided that you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence and indicate whether changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0) applies to the data made available in this article unless otherwise stated. The text below includes minor additions and formatting changes to the original text. Background This programme of research aimed to develop a best practice care pathway for people with dementia and hip fracture in hospital using the enhanced recovery framework. As the programme progressed, the intervention was adjusted to include all people with cognitive impairment (CI) as we learnt about implementation in practice. Objectives What is the best care practice for people in hospital with dementia who fracture their hip? [work package (WP) 1 phases 1–4] Can an optimised care pathway Perioperative Enhanced Recovery hip FracturE Care of paTiEnts with Dementia-Enhanced Recovery (PERFECT-ER) be developed for people in hospital with dementia who fracture their hip? (WP1 phase 5) What are the staff training and cultural/organisational changes required to implement and maximise adherence to the optimised care pathway-enhanced recovery pathway (PERFECT-ER) in hospital settings? (WP2) What are the components of a training manual for the enhanced recovery pathway (PERFECT-ER) promoting cultural and organisational changes and maximising adherence? (WP2) What is the feasibility of undertaking a randomised clinical trial (RCT) to assess the clinical and cost-effectiveness of the enhanced recovery pathway (PERFECT-ER) to inform a definitive RCT? (WP3) Methods Work package 1: evidence for best practice This WP generated evidence for best practice from peer-reviewed and grey literature including access to national and international initiatives. We undertook a Cochrane review (CRD42012002047; WP1, phase 1a and b). We collected front-line national and international healthcare professional perspectives on best practice, current practice and explanations for implementation gaps between ‘best’ and ‘current’ practice (WP1, phase 2). We undertook observations of care of delivery (WP1, phase 3) to understand ‘usual care’ and collected the views and opinions of patients, carers and healthcare professionals of regarding current care (WP1, phase 4). From WP 1 phases 1–4 we identified and synthesised components to inform consensus development events attended by a number of experts (both by profession and experience) to develop the PERFECT-ER checklist (WP1, phase 5). Work package 2: optimising care for patients with dementia and hip fracture We used an action research approach, with a mixed-methods case study design to study implementation on an orthopaedic ward in each of three hospitals, using a series of plan, do, study, act (PDSA) cycles. Work package 3: feasibility trial We undertook a feasibility, multicentre, cluster RCT with integral economic evaluation. In line with Medical Research Council guidance for complex interventions, we also conducted an integrated multimethod multiperspective (from patients, suitable informants and NHS professionals) process evaluation. The trial ran from November 2016 to August 2018 in 11 hospitals in England and Scotland. Results (research findings) Work package 1 key findings Work package 1 phase 1a Our systematic review indicated that there was insufficient quality research on the rehabilitation of people living with CI following hip fracture surgery. Of the literature reviewed, the majority focused on people with mixed CI status and/or people with CI and hip fracture as a subgroup of larger studies. Studies lacked power to detect differences between intervention groups. The review suggested that models of rehabilitation could decrease the length of hospital stay and reduce admissions to care. No cost-effectiveness studies were located. We found that people living with CI are at greater risk of postoperative complications and higher mortality 12 months postoperatively. We also found uncertainty around rehabilitation provision, with no guidelines based on UK or international policy on how to deliver care to people living with CI following hip fracture. Components of enhanced care identified included screening for delirium and assessing pain for people with CI and hip fracture. Work package 1 phase 1b Our Freedom of Information Act survey highlighted the policy priorities that NHS trusts emphasised for people with dementia and hip fracture in March 2014. Numerous hospitals disclosed global dementia strategies and geriatric acute hip fracture pathways. However, no hospital disclosed an integrated dementia and hip fracture pathway or other documentation. Major themes affecting care elements were identified: antipsychotics, behaviours that challenge, cost consequences, communication (between staff and patients, between staff and carers/family, between staff), consenting, constipation, delirium, dementia assessment, deprivation of liberties, discharge processes, end-of-life care, falls, hydration, identification of patients with dementia, incontinence, involving carers, manual handling, minimising ward changes, non-pharmacological interventions, nutrition, pain relief, pressure ulcers, safeguarding vulnerable adults, training in dementia care and ward environment. These themes were then aligned with corresponding elements described in the enhanced recovery after surgery literature (admission, preoperative, intraoperative, postoperative, rehabilitation, discharge). Work package 1 phase 2 The telephone survey investigated domestic and international healthcare professionals’ perspectives on the care pathways and costs, length of stay in hospital and discharge destinations for people with dementia and hip fracture. The participants identified diverse needs for patients with dementia in acute settings but supplied documentation that was largely dementia or hip fracture focused but not integrated. Participants described that training to provide colleagues with the skills to care for this patient group was generic and ‘tick box’. They also found that outcome metrics such as length of hospital stay did not help in recognising recovery in this patient group. The documents received through the survey were coded using a developed coding matrix to identify potential intervention components. This provided initial insights into what participants perceived as current and best practice. Implementation gaps they identified helped generate the observation topic guides used in WP1 phase 3. Work package 1 phase 3 Key themes identified specific interruptions (‘disjunctures’) in routines or planned sequences in caregiving as: Disruptions – when usual or expected practices were interrupted impacting on the ease with which staff manage care delivery. Discontinuities – when divisions in culture, spaces and timing interrupt the smooth delivery of tasks. Dispersions – occasions when environment artefacts [object(s) and/or people] are displaced from designated space. Work package 1 phase 4 The staff focus groups and interviews reported emotive experiences, fatigue and constraints that staff reported experienced on a daily basis. Many participants identified failing to deliver appropriate care to patients with dementia in several ways, including: lack of staff and/or time required combined organisational barriers care spaces not fit for these patients care needs. Carer experiences supported these staff views. Some carers felt under pressure to assist staff, whom they perceived as needing help, to relieve shortages. Carers did not view this as empowering them to produce co-delivered care. The patients were mostly full of praise for staff efforts, in improvising to meet ongoing disjunctures in workflow by re-prioritising tasks. Work package 1 phase 5 The intervention PERFECT-ER was developed through stakeholder consensus events. Work package 2 key findings In WP2 we identified common barriers, facilitators, underlying mechanisms and work of service improvement leads (SILs) and PERFECTED process leads (PPLs) entailed in embedding PERFECT-ER across distinct hospital settings. While this proved challenging, staff found different ways to implement changes within their settings. It became clear that the practices surrounding such changes could easily break down. Components of PERFECT-ER that aligned to context-specific motivations, including the National Hip Fracture Database (NHFD) best practice tariff or trust policies, were easier to implement and longer lasting. WP2 enabled us to refine the PERFECT-ER intervention and develop its training manual, understanding the requirements of the SIL, PPL roles and using the checklist as part of the PDSA cycle. This informed implementation aspects of the trial (WP3) and the process evaluation. Work package 3 feasibility key findings We recruited 282 participants, 132 from intervention sites and 150 from control sites. The average recruitment rates did not differ between intervention and control sites, ranging between 1.2 and 2.7 participants per month. Average recruitment of 1.87 per month contrasted with the expected 4 per site per month anticipated. There was no evidence of any systematic between group difference at either the point of discharge from hospital or at 1-month follow-up. At 3 months, however, a potential beneficial effect of the intervention over control was evidenced for patient health-related quality of life (HRQoL) based upon the EuroQol-5 Dimensions, five-level version (EQ-5D-5L) by proxy: those in the intervention group had a mean EQ-5D utility score 0.071 units higher than control [95% confidence interval (95% CI) 0.018 to 0.124; p = 0.009], a relatively small effect of around one quarter of a standard deviation. A difference of 0.099 units, in favour of the intervention group, was also seen at the 6-month follow-up (95% CI 0.001 to 0.198; p = 0.047). Examination of the residuals from each of the models appeared to show violation of the normal distribution assumption for the length of stay, ‘Timed Up and Go’ and the Suitable Informant EQ-5D-5L models. Logarithmic transformation was applied to these data and the models refitted. None then showed any statistically significant difference except the model for length of stay. The individuals in the intervention group had significantly longer lengths of stay, on average 1.22 times longer (95% CI 1.02 to 1.45; p = 0.028). Over the trial’s duration, 57 participants (20.2%) died. A higher rate of all deaths was observed in the intervention group than in the control group (22.7% vs. 18.0%). Death in hospital was determined from the NHFD data and only available for participants in England, thus excluding 59 Scottish participants. Eleven participants (3.9%) died in hospital, with a higher rate in the control group (4.7% vs. 3.0%). Seventeen (8 intervention, 9 control) patients were known to have died within 30 days of surgery and 52 (28 intervention and 24 control) within 6 months. Process evaluation The process evaluation demonstrated that, under particular conditions, the PERFECT-ER intervention package can be implemented in diverse NHS contexts. General barriers to implementation including: staffing issues, low staff morale, staff sickness, staff movement between wards and lack of senior staff, impacted negatively on implementation. However, the protected SIL resource, PDSA processes, networking, and using key staff members and trust processes, along with the resourcefulness, determination, commitment and ingenuity of SILs, PPLs and others with whom they worked, were facilitators. We demonstrated that implementing an intervention encouraging more standardised practice and its documentation to improve patient, carer and staff outcomes via consistently amalgamating marginal gains was welcomed, necessary and seen as valuable. Health economics Data completeness was comparable between suitable informant-reported (SIR) and hospital records but slightly higher in the SIR data. Comparing agreement in the data, we found sources agreed on ‘non-use’, but suitable informant over- and under-reporting of ‘use’ compared with hospital records did not follow a consistent pattern. Comparing SIR ‘hospital use’ over the 3-month periods pre-baseline and pre-6-month follow-up were identical at both assessments. Data suggest that sources yielded inconsistent estimates for inpatient days and for emergency department visits but more consistent estimates for outpatient attendances. Total hospital costs yielded inconsistent estimates. Individual items of resource use were relatively well completed, with missing rates below 12%. However, the cumulative impact of missing data decreased the availability of complete costs at all-time points. A combination of missing resource use and unpaid care data from participants/suitable informants completing the trial and high attrition rates led to small or very small samples available for calculating 6-month costs. Low completion of self-reported HRQoL instruments (EQ-5D-5L and DEMQOL) and attrition led to small samples available for calculating 6-month participant-reported quality-adjusted life-year (QALY). There were 64 intervention participants and 79 controls at 6 months but on combinations of societal costs and self-completed HRQoL measures, only 25 cases were available for analysis. In line with the original proposal, cost-effectiveness estimates were produced. The evidence of these analyses points to substantial uncertainty as to the size of the incremental cost-effectiveness ratios produced. No assumptions were made as to the costs and QALY of participants that had died during the trial; instead, complete cases were analysed. The mechanisms underlying incomplete data were not necessarily observed or recorded and so no imputation strategy was possible. Conclusions The PERFECT-ER feasibility trial demonstrated mean recruitment of 1.87 participant per centre per month. Retention at 1 month was over 80% and at 6 months approximately 50%. In patients with CI and hip fracture, we estimated that 10–20% died within 30 days of sustaining a hip fracture. Our data suggest that short-term mortality could be reduced with implementation of PERFECT-ER from a culmination of increased good practice across a range of care domains. Thirty-day mortality is more commonly used as an outcome measure in the evaluation of enhanced recovery pathways; discussion with the patient and public involvement group indicated that this was an appropriate primary outcome in future trials. The process evaluation found patients and carers unable to comment on receiving the intervention, so did not assist in answering questions about PERFECT-ER implementation. Equipment costs contributed little to overall costs and these questions should be removed. Hospital records-extracted data were used to estimate costs for use in the cost-effectiveness analyses presented here. These are the ‘gold-standard’ source compared with SIR data. However, records did not include information on hospital stays outside the trusts providing the records, so use of other hospitals might have been omitted. The feasibility PERFECT-ER trial provided valuable information and evidence to future work. Thirty-day post-surgery mortality is appropriate primary outcome for future trials, but to compare longer-term survival, we recommend also measuring 3-month mortality. In a definitive PERFECT-ER trial, patient and carer interviews should not be undertaken. Community health use questions should be reduced, removing health care that might be routine and little affected by the intervention, such as dentistry. We consider that hospital records extraction proformas and trial database design in a definitive trial could address potential shortcomings. We would not recommend comparing SIR and hospital records sources in a larger study. Trial registration This trial is registered as Current Controlled Trials ISRCTN 99336264. Funding This award was funded by the National Institute for Health and Care Research (NIHR) Programme Grants for Applied Research Programme (NIHR award ref: DTC-RP-PG-0311-12004) and is published in full in Programme Grants for Applied Research; Vol. 13, No. 1. See the NIHR Funding and Awards website for further award information.
Keywords
