EMPSCCaseStudyDelirium web“I had contact on a daily basis with pharmacy staff that prescribe patient medications on the day of admission. We had the opportunity to discuss possible amendments to the analgesia protocol, the possible need to initiate the alcohol withdrawal pathway or any other pharmacy issues relating to the patients which may not previously have been highlighted but are beneficial to the patient and ward staff receiving the patient following their surgery.

“In addition, my experience, knowledge and my ability to communicate with them has allowed me to refer to the ortho-geriatrician and or other services without challenge.”

Delirium Nurse Specialist

Background

The East Midlands Patient Safety Collaborative (EMPSC) supported a demonstrator project to prevent and manage Delirium in an acute surgical setting. This initiative has been successful in reducing hospital length of stay (LOS), nursing time and associated costs to the health and care system.

There is a clear model for calculating the return on investment of providing this service where it does not exist. In addition, if there are staff in place, this model also serves as a marker to be able to calculate saving from LOS reductions. The model further demonstrates the value of using this innovation to create a predictable service to prevent delirium and that will reliably reduce LOS to save costs. Results of £64,000 (minimum) of cost savings could be realised in the shortest period of time of 12 months, based on 400 patients.

Delirium is a common and serious complication of dementia with a poor prognosis. It is the most common surgical complication in the elderly (between 20-60%) and it increases the risk of them requiring long-term care.

Only one in eight admissions to hospital are correctly coded for delirium and people with delirium in hospital are six times more likely to be re-admitted within 30 days of discharge. Patients with delirium are also four times more likely to be transferred to another healthcare facility (not including care homes).

deliriumStakeholder analysis undertaken by the EMPSC indicated the need for better screening to identify those most at risk, support prevention where possible, and improve management of fully-manifested symptoms at the earliest stage.

Following an EMPSC call for proposals to stakeholders and the East Midlands health and care community, Chesterfield Royal Hospital were successful in receiving EMPSC funding and support for the implementation and evaluation of their Delirium demonstrator project.

Findings suggest that the intervention has impacted positively on the incidence of post-operative delirium in elective hip and knee replacement surgery by providing risk assessment and a tailored care management process delivered by a Delirium Nurse Specialist (DNS).

The median length of stay (LOS) was reduced from 5.6 to 4.2 and was made stable and predictable and associated costs of around £200,000 were saved through reductions in LOS and intensive nursing support.

Aims, objectives and scope

An ageing population, plus advances in anaesthetics and surgical procedures, has resulted in increasing numbers of people aged over 65 years undergoing surgery. Post-operative delirium is a serious complication in elderly patients with 10-50% of patients developing delirium following surgery. Incidence is estimated to be between 3.6-28% in elective orthopaedic patients and it is thought that 30% of the cases of delirium are preventable with appropriate management.

Delirium has serious adverse effects on mortality, functional outcomes, LOS, institutionalisation and is associated with increased nursing time and higher per day hospital costs. Pro-active multidisciplinary comprehensive geriatric assessment (CGA) has been shown to reduce post-operative delirium.  Delirium is a significant complication in this cohort of older patients, the majority of which have complex medical histories. The incidence of post-operative delirium in this group can be as high as 80% without appropriate identification and pre-operative planning. Due to a lack of awareness, delirium goes undetected and untreated in approximately 50% of cases.

Whilst treating the causes of delirium is vital, the evidence that specific programmes of delirium care can improve prognosis is limited. However, multi-component care planning aimed at delirium prevention has been shown to be effective.

Whilst the development and implementation of strategies for delirium prevention and diagnosis in high risk patients is shown to be effective, awareness and understanding of delirium prevention and the related positive impact on patient outcomes remains low amongst health professionals and carers.  Therefore a holistic education and training approach was chosen to increase the awareness of risk, symptoms and appropriate care planning amongst health and care professionals, patients, families and carers.

Aims and objectives of the Delirium Demonstrator Project included:

  • Increase prevention of delirium / reduce the number of patients developing delirium post-operatively.
  • Increase awareness and understanding of delirium, its symptoms and impact on patient prognosis and outcomes amongst health professionals, patients and their families / carers
  • Improve education and training to support awareness raising, risk assessment and prevention
  • Develop and implement strategies for delirium prevention and diagnosis in high risk patients to reduce rates of harmful impact on patients and facilitate speedy re-enablement
  • Change working practice and culture including engaging consultants in ownership of LOS rates and the potential patient safety impacts
  • Test a multidisciplinary / holistic approach to post op care in improving delirium prevention and management
  • Reduce length of stay and associated costs
  • Reduce nursing time required and associated cost
  • Prevent delirium-related re-admissions to hospital.

Scope

The project focused on a cohort of patients aged over 65 undergoing elective orthopaedic surgery at Chesterfield Royal Hospital. A total of 400 patients attending orthopaedic clinic with a view to being listed for elective knee or hip replacement surgery were reviewed by an experienced orthopaedic matron in clinic between January and December 2017.

Method and approach

Background and launch

EMPSC undertook a three-month stakeholder engagement and scoping exercise across the East Midlands region to gain an understanding of existing patient safety initiatives and partner priorities in relation to patient safety. One of the priorities highlighted was delirium, with a particular emphasis on prevention, diagnosis and management.

In response to the insight collated, the EMPSC held an awareness raising and call-to-action event in January 2016, ‘Delirium – Dementia’s Deadly Twin’, to discuss and broaden the knowledge across the East Midlands health and care system related to delirium.  Following the event a call for proposals went out to the health and care community and Chesterfield Royal Hospital were successful in being chosen to run a 15 month demonstrator project supported by the EMPSC.  The project included the funding of a Delirium Nurse Specialist (DNS) post and ran for 15 months, ending in January 2018.

Cost and recruitment

At the centre of focus for the care pathway innovation was patient-centred care.  This was designed to be led by a Delirium Nurse Specialist (DNS) with the training and / or experience to engage with a wide range of medical colleagues, the patient and the relatives within the pathway.  This key element of the innovation could be provided by someone already in post with the provision of the necessary coaching and support to develop their role into the DNS.

The role was scoped as being provided by a senior nurse – from Band 6 through to Matron at Band 8a – so the staff cost profile could be within the range of:

Band Direct Employment Costs
6 £34,502 – £45,072
7 £40,863 – £52,940
8a £52,169 – £61,462

Additional costs included those associated with increased referrals to the ortho-geriatrician service, assuming this provision already was in place.

Assessment tools and scoring against criteria

This project used a range of screening tools and resulted in a ‘best fit’ application of the Montreal Cognitive Assessment tool (MOCA).  This new screening process constituted a bolt-on to the front end of the existing pathway and was supported by the creation of the Delirium Nurse Specialist (DNS) post.

Screening commenced in January 2017 and an individualised care plan based on the screening tool assessment outcomes was produced for each patient in the intervention group prior to surgery.  The patient’s tailored care plan was used for the duration of their admission.

The screening process used the Edmonton Frail Scale (EFS), 4AT, (MOCA) as well as taking a detailed past medical and social history.  A Delirium Elderly At-Risk (DEAR) score was then calculated for each patient with the information obtained. Patients were counselled on delirium and written information provided on this. Patients were also screened daily for signs of delirium using the Delirium Observation Scale score (DOS) and managed according to NICE guidelines (Delirium: Prevention, Diagnosis and Management. NICE CG103, July 2010).

Patients with any of the following; DEAR>2, EFS of >7, unstable medical co-morbidity, polypharmacy or any patient whom the matron had concerns about was referred for orthogeriatric assessment. In addition, patients, were referred to the therapy led pre-operative hip and knee classes at the Trust (if this had not already been arranged) and advice was given on arranging respite or additional help post-discharge where needed.

Methods of raising awareness and education

An education package for ward staff was developed and delivered to 32 staff from across the multi-disciplinary team.  The Delirium Nurse Specialist (DNS) gave education sessions to the nursing staff on recognition of delirium and use of the Delirium Observation Screening paperwork.  Education sessions were also delivered to ward staff to improve their knowledge of delirium, ways in which to assist in its prevention, how to identify delirium, its possible causes and ways in which to care for and treat patients with a diagnosis of delirium.

Staff received information on how to use the Inpatient Delirium Management tool, which helps both medical and nursing staff differentiate the process concerned with assessment and how to complete a Delirium Assessment Tool and Care Bundle – including how to appropriately action findings.

Patients and their families and carers were given an information leaflet on the risks and symptoms of post-operative delirium.

A key element to the successful enrolment and engagement of the patients’ families in their care was the amount of time taken to explain the assessment process and outcomes and then to listen to the families’ questions – and therefore helping them form a decision based on the information known about risk.

Patient handover and escalation

Weekly meetings were held between the DNS and ward matron to identify potential intervention patients coming onto the ward and to follow up the previous week’s admissions.

This process was key to identifying high risk patients and identifying any concerns or management requirements. This also enabled arrangements to be in place for relatives to remain with patients throughout their stay and participate in care delivery if they should wish to do so.

Measurement plan

400 patients were reviewed by the orthopaedic matron between January and December 2017. Data was collected on patients who received hip or knee replacement surgery capturing information on age, sex, length of stay and whether delirium was diagnosed during their stay as a comparison.

Results and evaluation

The project has been sustained at Chesterfield Royal Hospital via an internal business case following a demonstrated return on investment (see attached ROI model) related to reductions in LOS and enhanced nursing costs.

The Trust is now providing funding for an Arthroplasty Nurse Specialist role – the name of the DNS role has been changed to reflect the focus on orthopaedics – after the successful initial support and funding provided by the EMPSC.

Length of stay

Comparison of the length of stay (LOS) data between the two hip and knee groups (intervention and non-intervention) indicates that the intervention has contributed to a reduction in the median length of stay of the knee replacement patients by 1.205 days (Graph 1) and in the hip patients by 0.655 days (Graph 2).  This combined LOS represents a median reduction from 5.6 to 4.2 days a reduction of 1.4 days for all patients in these groups.

delirium graph 1

Graph 1. LOS for elective knee replacement patients Apr 15 – Dec 17

 

Graph 2. LOS for elective hip replacement patients Apr 15 - Dec 17

Graph 2. LOS for elective hip replacement patients Apr 15 – Dec 17

The data monitoring enabled the identification of periods when the process was sub-optimal and has been the key to further process improvement.

In this case study, the LOS at Chesterfield started from a relatively stable baseline which was then distorted when high risk patients were not screened, or the screening process did not identify the patient. These outliers from the norm do impact on the cohort of patients being observed because they distort the average LOS quite significantly.

The identification and focus on these outliers provided an opportunity to learn about process variation and methods to improve the reliability. These factors are associated with; a) staff compliance with the delirium assessment tool and care bundle (checklist in Appendix C) and b) when the screening process is suspended because the DNS was on leave. As a consequence some patients were not screened and the risks were not identified leading to longer LOS associated with a delirium event.

The demonstrator project has shown that the improved management by the DNS is directly related to the reliability of the process and the maintenance of care bundle standards and translates to a reduced LOS and better patient experience.

 In summary, the project has demonstrated the following outcomes:

  1. LOS has fallen from 5.6 to 4.2 days a median reduction of 1.4 days for the service.
  2. This is a saving of 1,015 bed days
  3. This saved the trust between £89,300 and £179,200 (optimism bias factor rating)
  4. Set against the direct nursing costs of a Band 7 DNP at £40,863 – £52,940, this project on this basis is cost effective in the first 12 months.
  5. Reduced enhanced care (HCA cover 24 hrs) days from a baseline of 228 (2015/16) down to 73 days in 2017. This co-relates to a,
  6. Reduced nursing costs of £56,000 (HCA Costs)
  7. No increase in falls, when they might have been expected.
  8. 63% of patients returned a questionnaire with 78% finding the information about delirium useful.

 

 Additional impacts include:

  • Improved medicines management – prior to this project there was only one pharmacy protocol for patients in this age range. With the exchange of knowledge and harm impacts to patients from this one protocol, it has now been reviewed and there are now three standard protocols which aim to reduce patient harm and lower medication costs.
  • Improved efficiency and capacity – the improved delirium prevention and management process has enable LOS to be better controlled which has allowed for more work and income to the service.
  • Positive culture change – the process has improved communication and trust within the multi-disciplinary team. Consultants are more engaged, see the value of the DNS role and are more aware of the impact their interactions can have on managing LOS and patient safety.
  • Co-production – The stories of patients and staff have been filmed to highlight some of the insights from patients’ and families being involved in the decisions about their care. This co-production has been instrumental in positively changing the culture of consultants and surgeons at the Trust.
  • Shared decisionmaking – None of the patients were denied surgery based on their screening results however five patients decided against surgery following screening, counselling and orthogeriatric input.

Financial impact

Costs

  •  Agenda for Change band 6 Nurse Specialist (1.0 WTE at £46,000)

 

Benefits

The follow table presented as scenario 1 demonstrates the potential cost savings set against Chesterfields’ data. Our choice of 0.4 OB makes the assumption that there is evidence but it is low level. That is no reflection on our confidence in our study, it is position to take until further evidence of other projects allows for a benchmark to be established. As a consequence we are presenting figures that are both pessimistic (minimum savings) and optimistic (maximum savings) for you to consider.

Scenario 1: AHSN ROI Methodology Application to Chesterfields’ current LOS 4.2 Days. Down from 5.6 to 4.2 a shift of 1.4

Range of benefit linked to quality of evidence: Option Population Treated Reduction in LOS Cost per day in patient stay Optimism bias Value of reduced LOS
A 400 1.4 400 0.4% 89,600
B 400 1.4 400 0.8% 179,200

There is a clear model for calculating the return on investment of providing this service where it does not exist. In addition, if there are staff in place, this model also serves as a marker to be able to calculate saving from LOS reductions. The model further demonstrates the value of using this innovation to create a predictable service to prevent delirium and that will reliably reduce LOS to save costs. Results of £64,000 (minimum) of cost savings could be realised in the shortest period of time of 12 months, based on 400 patients.

Potential / unquantified savings

  • Savings based on a potential further reduction of LOS by an additional 1.2 days is £125,160 (Chesterfields’ own figures and based on 840 bed days saved)
  • There will be savings from reduction in cost of unnecessary antibiotic treatment in the community, however this is difficult to quantify
  • There is also likely to be reduced readmission to hospital, but again this is difficult to quantify.

There is potential for more cost savings associated with this project in terms of pharmacy spend at Chesterfield, but also importantly the potential for reduced costs post-discharge either with the provision of other services or at care home and GP costs. It was outside the scope of this project to model these potential cost savings.

Learning points

 Key learning points from the demonstrator and implementation included:

  • The education sessions given by the DNS to the nursing staff was valuable in providing the knowledge across the MDT to enhance care and reduce the need for longer periods of care associated with delirium.
  • Engagement with the Lead Pharmacist as part of the demonstrator led to the development of two more pharmacy protocols more sensitive to individual patient – prior to this project there was only one pharmacy protocol for patients in this age range.
  • Further research maybe needed to identify the most accurate screening tool to assess patient’s risk of post-operative delirium and the impact of delirium prevention interventions in matched case control groups. However, in this study MOCA seemed the most accurate predictor.
  • Further studies also might examine the differences in the length of stay between the hip and knee patients and the reasons for this.

Re-admission data was also reviewed and showed no difference in the re-admission rates between the two groups (intervention and control).

Plans for spreading learning and encouraging adoption

If further refined and embedded, it is hoped that the benefits of the initial demonstrator project can potentially further reduce LOS to three days at the Chesterfield Royal Hospital.  This could be achieved through the Arthroplasty Nurse Specialist (previously Delirium Nurse Specialist) delivering post-operative classes allowing discharged patients to attend the nurse-led class as outpatients.

The Trust is planning to spread the model of risk assessment, prevention and management of delirium to other specialities within the hospital including hip fracture and trauma wards.

Wider spread across the region commenced in June 2018 with a view to enrolling two new sites.

The EMPSC has engaged with acute trusts in Nottingham and Derby and provided learning and spread events in Chesterfield and at Sheffield Teaching Hospitals.

Which national clinical or policy priorities does this example address?

The demonstrator and its subsequent sustainability addresses the national frailty, deterioration and patient centred care agendas.

Find out more

Dr Cheryl Crocker, Regional Lead: Patient Safety Programme, East Midlands Academic Health Science Network
T: 07772353064
E: cheryl.crocker@nottingham.ac.uk

Supporting materials

A USB toolkit resource, filmed interviews, leaflets, case studies and additional Delirium resources can be found at http://emahsn.org.uk/psc-priority-areas/delirium.

References

  • Delirium: Prevention, Diagnosis and Management. NICE CG103, July 2010
  • Bruce AJ et al. The Incidence of Delirium Associated with Orthopaedic Surgery: A Meta-analytic Review, Int Psychogeriatr, 2007,Vol19 (pg 197-214)
  • Siddiqi N et al. Occurrence and Outcome of Delirium in Medical Inpatients: A Systematic Literature Review. Age Ageing 2006 Jul; 35 (4): 350-64
  • Harari D et al. Pro-active Care of Older People Undergoing Surgery (‘POPS’): Designing, embedding and Funding a Comprehensive Geriatric Assessment Service For Older Elective Surgical Patients. Age ageing 2007 Mar;36(“): 190-6
  • Freter S et al. Predicting Post-operative Delirium in Elective Orthopaedic Patients: The Delirium Elderly At-Risk (DEAR) Instrument. Age Ageing 2005; 34: 169-184
  • Schuurmans MJ. Delirium Observation Screening Scale: a screening instrument for delirium. Res Theory Nursing Prac, 2003 Spring;17 (1): 31-50