Frailty evidence‐practice gaps in acute care hospitals

Frailty is common in hospitalised older people. Clinical practice guidelines for the management of frailty provide recommendations for identification and management; however, adoption into practice in hospitals is limited. This study identified and quantified the evidence‐practice gap between frailty guidelines and clinical practice in two hospitals using an audit tool.


Methods:
A cross-sectional audit of medical records of frail older patients admitted to two hospitals was conducted.Data were collected using an audit tool based on the Asia Pacific Clinical Practice Guidelines for frailty management.Data were analysed using descriptive statistics and inter-rater reliability of the tool was assessed.
Results: Auditing of n = 70 electronic medical records showed that assessment of frailty in the acute setting did not regularly occur (17%).Few participants received guideline-recommended interventions.Physiotherapy treatment was limited, with 23% of participants receiving progressive resistance strength training.
Gaps exist in provision of nutritional supplementation (26%) with limited recordings of weight during the admission for 10% of participants.Pharmacy review of medications was consistently documented on admission (84%) and discharge (93%).Vitamin D was prescribed for 57% of participants.Inter-rater reliability showed a high level of agreement using the audit tool.Conclusions: An audit tool was feasible to assess frailty evidence-practice gaps in the hospital setting.Further understanding of the contextual barriers is needed to inform implementation strategies (dedicated staffing, education and training and ongoing audit of practice cycles) for the uptake of frailty guidelines in hospital settings.

K E Y W O R D S
evidence-practice gaps, frail older adults, guidelines 1

| INTRODUCTION
Frailty is a clinical state with multiple causes, in which there is heightened vulnerability for increased dependency and mortality when exposed to stressors. 1 Frailty is common in hospital settings with an estimated prevalence in 27%-80% of older patients. 2,3][7] The Asia Pacific Clinical Practice Guidelines for the management of frailty provide evidence-based recommendations for frailty management, including comprehensive assessment using validated screening tools; physical activity incorporating resistance training; nutrition support; addressing polypharmacy; screening for reversible causes of fatigue; and Vitamin D supplementation. 8However, research suggests that frailty is not routinely assessed in hospital setting. 9,10At present, there are gaps in the adoption of guideline recommendations for the assessment and management of frailty in hospitalised older adults. 9he implementation of guidelines into clinical practice in hospital settings is well acknowledged with complex stakeholder and contextual factors influencing implementation. 11,12The aims of this study were to: 1. identify and quantify the evidence-practice gap between frailty guidelines and clinical practice in two acute care hospitals and 2. develop and pilot an audit tool for measuring evidencepractice gaps of frailty management in acute care hospitals.

| Study design
A retrospective audit of electronic medical records (EMR) was undertaken to identify current practice in acute care and quantify evidence-practice gaps relating to frailty guideline recommendations.Medical records were audited of frail older patients admitted to two acute care hospitals at Southern Adelaide Local Health Network in Adelaide and Northern Sydney Local Health District, Sydney, who were a subset of recruited control group participants in the Frailty in Older People, Rehabilitation, Treatment, Research Examining Separate Settings (FORTRESS) trial 13 from December 2020 to February 2022.Details of the FORTRESS study and methodology are published elsewhere. 13This study received ethics approval from the Northern Sydney Local Health District Human Research Ethics Committee, reference number: 2020/ETH01057.

| Participants and setting
The EMR of participants were retrospectively audited.The FORTRESS study used the FRAIL scale 14 to identify participants who were frail defined by a score of three and above indicating frailty. 14Inclusion criteria were as follows: hospitalised adults aged 75 years or older, who were living in the community and mobile prior to the acute admission, without significant cognitive impairment as measured by a score of 24 and above on the Mini Mental State Examination (MMSE). 15Participants were admitted to one of three acute general medicine hospital wards included in this study.One ward was located at Southern Adelaide Local Health Network in Adelaide, and two others were located at Northern Sydney Local Health District, Sydney.As patients in control group wards within the FORTRESS study, the participants included in this audit underwent frailty screening by the research team, separate to the treating team, without influencing usual care during the acute admission.

| Data collection
Consecutive patient electronic medical records were accessed to audit documented frailty assessments and

Policy Impact
There is a need for routine and sustained implementation of frailty guideline recommendations for improved quality care to hospitalised frail older people.Acute care and hospital policy should prioritise consistent and quality care to identify and manage frailty in hospital settings.

Practice Impact
Using an audit tool based on the Asia Pacific Clinical Practice Guidelines for management of frailty, it was possible to identify and quantify evidence-practice gaps between guideline recommendations and acute care for frail older people.Findings indicate opportunities for improved multi-disciplinary care for the assessment and treatment frailty in line with guidelines in acute care settings.
interventions during the acute admission.An evidenceinformed audit tool was developed based on the Asia Pacific Clinical Practice Guideline recommendations for the management of frailty. 8The audit tool (Appendix S1) was developed and piloted by the study team prior to use and was used for consistent data collection for each patient record.Patient admission characteristics were collected, including hospital site, ward, admission and discharge date.Frailty audit criteria included documented use of a tool to identify frailty and documented frailty interventions, including physical exercise (delivered by a physiotherapist and/or an allied health assistant), recommendations for nutrition and recommendations for medication reviews.Allied health clinicians self-report and document their direct patient-attributed time on documented entries on the EMR.Therefore, we collected data on the documented time spent with treating allied health professionals.At the time of the study, hospital policies were not based on the Asia Pacific Clinical Practice guidelines for the management of frailty.Patient characteristics and audit criterion data were entered into a Microsoft Excel spreadsheet.

| Data analysis
Audit data were analysed using descriptive statistics in Microsoft Excel, and reported as means, medians and percentages.Inter-rater reliability was tested by two independent auditors who completed 10 audits on the same patient records.Level of agreement between the two auditors was determined by Cohen's kappa 16 in SPSS (p < .05,95% CI), with displayed percentage of agreement for each audit criterion.

| RESULTS
A total of 70 participant medical records were audited.Twenty-four records were from acute admissions at Southern Adelaide Local Health Network, and 46 from acute admissions at Northern Sydney Local Health District.Median length of stay of acute admissions was 6.5 days (range: 4-8.8 days).
Audit results are displayed in Table 1.Very few participants (17%) had documentation of frailty assessment in their electronic medical record (EMR).Where frailty was assessed, the FRAIL scale 14 was used in all cases (100%).Eleven participants (16%) did not receive any physiotherapy while on the acute ward.Daily physiotherapy was documented for 29% of acute participants, with most participants receiving physiotherapy input for assessment of mobility and transfers only.A quarter of participants (23%) had documentation of physical exercise programs incorporating strength or resistance training.Information T A B L E 1 Audit results of frailty guideline recommendations in acute care.

Audit criteria Identified audit criteria n, %
A measurement tool has been used to identify frailty 12, 17 The frail patient has been referred for an exercise program 17, 24 about physical exercise and/or referrals to services for exercise after discharge was documented for 49% of the participants.For each physiotherapy consultation, median time spent with acute participants was 25 min.Nutritional assessment was documented for 59% of participants, and use of a validated malnutritional screening tool was documented for 47% of participants.The Malnutrition Universal Screening Tool (MUST) 17 was the most frequently documented malnutritional screening tool, completed by nurses, allied health assistants or dietitians.A dietitian consultation was documented for 39% of participants, with 27% having documented weight loss (either prior or during admission), and nutritional supplementation and/or protein or caloric recommendations documented for 26% of participants.A total of 10% of participants had no documentation of weight during the acute admission.For each dietitian consultation, median time spent with acute participants was 45 min.
Pharmacists documented medication history on admission for 84% of participants, and reviewed medication changes at discharge for 93% of participants.Vitamin D was prescribed (as an existing or new medication) for 57% of participants.Documentation on education about medications was documented for 24% of participants, and very few (n = 1) were referred or recommended for a Home Medicines Review. 18For each pharmacist's consultation, median time spent with acute participants was 60 min.

| Inter-rater reliability
Inter-rater reliability of the audit tool was assessed by the percentage of agreement between two auditors who both completed audits for 10 of the same participant records.Cohen's kappa was run to determine the level of agreement between two independent auditors who completed audits for 10 of the same participant records.There was very good agreement between the two auditors, κ = .813(95% CI .73 to .90),p < .001.Table 2 displays the percentage level of agreement between the two auditors for audit criteria.The lowest level of agreement was for weight loss (60%), due to difficulty finding this detail in documented medical records.

| DISCUSSION
This study identified and quantified the evidencepractice gap between frailty guideline recommendations and clinical practice in two acute care hospitals T A B L E 2 Inter-rater reliability level of agreement between two auditors.

Percentage of agreement
A measurement tool has been used to identify frailty.If yes, name the tool 100 The frail patient has been referred for an exercise program 100 • Progressive physical activity program with resistance training component was prescribed 80 • Daily physiotherapy and/or exercise physiology (Mon-Fri) 100 • Information and/or referral for ongoing exercise was provided on discharge 90 Average time per physiotherapy session (min) 100 A dietitian consultation has occurred during admission 90 • Malnutrition screening has been completed and documented 100 • A tool was used for malnutrition screening 100 • Weight loss in the past 12 months was identified 90 • Nutritional supplementation/protein was recommended for those with weight loss 90 • Weight was recorded at least 2x week during admission 60 • Information and/or referral on nutrition services was provided on discharge 90 Average time per Dietetics session (min) 90

Pharmacy reviewed medication history on admission 90
Pharmacy reviewed medication changes prior to discharge 90 • Vitamin D was prescribed (Cholecalciferol) 90 • Education was provided to the patient about their medications 90 • A home medicines review was recommended 90 Average time per pharmacy session (min) 90 in Australia.Assessment of frailty in the acute setting did not regularly occur and few participants received guideline-recommended interventions for frailty management.Findings indicate more needs to be done to assess and treat frailty in line with evidence-based guidelines.
The number of frail older people requiring acute care is expected to increase in line with population ageing in Australia. 19While there is evidence and guidelines for the treatment of frailty, there is less understanding of the implementation of frailty management guidelines into clinical practice. 20The implementation of guidelines is often challenging due to the complexity of contextual factors within health systems. 11Previous studies have identified contextual barriers to guideline implementation including time and resource limitations 21 and a lack of education and training about frailty. 22,23Within primary care settings, barriers to implementing integrated care for frailty relate to the complexity of care needs of frail older people, difficulties with system navigation and access, and limited consumer involvement in care decisions. 24Frailty is commonly perceived by health professionals as a cycle of decline and as such there may be scepticism about whether or not the condition can be reversed. 25Many health professionals believe frailty can be identified based on appearance, thus impacting the uptake of validated screening tools. 10,25Previous studies conducted within surgical wards in acute hospital settings have found key health professionals' limited knowledge and understanding of frailty influences adoption of frailty identification, management and communication in hospital settings. 10here are opportunities to identify and manage frailty within the acute hospital setting; however, implementation is constrained by perceptions of health professionals, organisational boundaries within the acute setting, the need to focus on patient-orientated outcomes, and the need for ongoing education and role clarity. 26Our findings provide pilot data quantifying evidence-practice gaps for the management of frailty in two acute hospital settings.The evidence-practice gaps are unsurprising given the Asia Pacific Clinical Practice Guidelines for the management of frailty were not embedded in the hospital policies at the time of the audit.Further understanding and contextualising of implementation of frailty guidelines in acute care is needed. 20Trialling and evaluating tailored implementation strategies, for example clinical champions who can influence decision-makers, dedicated staffing, education and training, and ongoing audit of practice cycles are imperative for the successful uptake of guideline recommendations into acute clinical practice. 26,27his will better inform comprehensive policies and funding to address frailty reduction in acute settings. 28e identified evidence-practice gaps in frailty identification and management within acute general medicine wards at two hospital settings in Australia.The included sample for this audit was a subset of FORTRESS control group participants; therefore, results are not reflective of all control participants within the trial.Furthermore, reflecting on the inclusion criteria, the audit was limited to mobile, cognitively intact, hospitalised older people, therefore is not representative of frail older people with mobility or cognitive impairments and those from non-English speaking backgrounds.A further limitation of this study is that the audit relied on staff documenting care in the EMR; it is possible that additional care was provided but not documented in the notes and therefore not captured.This audit was undertaken at two hospital settings and analysed descriptively, thereby limiting generalisability more broadly.

| CONCLUSIONS
There is a substantial evidence-practice gap for the adoption of frailty guideline recommendations for frail older people admitted to two acute care hospitals in Australia.Acute care and hospital policy should prioritise consistent and quality care to address frailty in hospital systems.
The audit tool tested in this study was shown to be feasible, with a high level of inter-rater reliability.The audit tool (Appendix S1) can be used for other acute care hospitals to evaluate their provision of frailty services, evidence-practice gaps and opportunities for quality improvement.