Redesigning Acute Care for Older People: The Start of Sheffield’s Journey Tom Downes MB BS, MRCP, MBA, MPH (Harvard) Clinical Lead for Quality Improvement Sheffield Teaching Hospitals The Health Foundation / IHI Quality Improvement Fellow 28th November 2012

UNSUSTAINABLE Healthcare inflation 4.3% per year over the last 30 years Driven by technology and expectation Only 0.4% attributable to ageing Need to deliver over 20% more care in 5 years’ time Need to deliver over 50% more care in 10 years’ time UNSUSTAINABLE Rises in healthcare spending: where will it end? Jon Appleby, BMJ 1st November 2012 Rises in healthcare spending: where will it end? Jon Appleby, BMJ 1st November 2012

‘We must redesign services. Decisions about service redesign must be clinically led and clinicians must be prepared to challenge the way services - including their own service – are organised.’ Hospitals on the Edge – The time for action Royal College of Physicians, 13th September 2012

Day 2127 as a consultant

A ‘system’ problem We have a problem. A combined health and social care of Sheffield challenge. This control chart shows the length of stay of consecutive patients discharged over 3 months last year. Over the whole year, the average length of stay in geriatric medicine was about 20 days. We have approximately 300 geriatric medicine beds with a length of stay roughly 50% longer than expected for our case-mix. Last year, our nurses measured that 48% of our inpatients are ‘waiting’ rather than receiving acute hospital care. This waiting and delay is harmful for older people: infection risk, immobilisation and pressure ulcers, infection risk, falls risk and institutionalisation. We have set out on a journey to redesign our system for our patients.

A complex system problem We mapped out our processes – they are very complex.

2003 Toyota Corolla

How do others design complex systems? Toyota Oobeya Room Our system is so complex we realised that we need a methodology to redesign a very complex service. Modern cars are very complex and involve many different processes and technology. We can see similarities to modern health care which also has many different processes and technology. We have decided to use a methodology from the car industry that we have translated to healthcare. We do not talk about tyres, engines and petrol. For us it’s about nurses, doctors, therapists and medication.

First find a room

The Room We meet regularly in a room set up to share our learning between all our health and social care interprofessional groups. It was a coffee room that we underused.

Board 1: The Business objectives: GSM weekly bed occupancy from April 07 with target lines Board Level Business objectives for GSM

Board 2: What do these objectives mean for our patients? A Future State diagram Of the GSM Process as it evolved Post-it note comments from stakeholders

Board 3: How are we doing against the GSM business and patients objectives?

Board 4: High Level GSM Process through the complex health and social care system & Board 5: Real time plan High Level: Current State Map of the GSM process Programme Plan time April 01 2012 Each row presents the tasks (yellow post-its)to be performed by each stakeholder group

Tests

Let me introduce ‘George’ PDSA tests of moving from ‘post take’ to ‘on take’ Let me introduce ‘George’ We have tested changing the way consultant geriatricians work at the front door. We demonstrated that we can reduce the average time from arrival at hospital to being seen by a consultant geriatrician from about 20 hours to near to 4 hours. George (not his real name, nor his real picture) was involved in one of our first tests of changing outpatient care. Usually it takes about 3 months for patients to navigate the process of getting from referral by their GP through a first visit, tests and a second visit to see the consultant for a definitive diagnosis. We tested whether we could deliver that 3 month process in less than 8 hours. George came to see us the next working day after we contacted him and his daughter. From arrival at hospital we did everything including; history, examination, test, physio assessment (he was fitted with a walking stick), consultant review. He left less than four hours after arriving with a letter in his hand documenting everything that had been done with all the results. He had lunch at home with his daughter. His daughter (who works) was delighted by the way the care was organised. 82 years old Lives independently and wants to continue doing so Widowed 5 years ago Has mild dementia Daughter lives locally Losing weight and finding walking more difficult

Challenge to UK geriatric medicine traditions: Split of inpatient / outpatient care Combined immediate delivery of specialist MDT care

This is a control chart of the demand for specialist geriatric care over 6 months. We have measured the demand of our service. We know that we have to design to be able to deliver a specialist multidisciplinary assessment including tests to 26 patients per day to be able to meet our demand 80% of the time. We know how important it is to not design for our average of 17 assessments per day as this would mean that half the time we wouldn’t be able to meet the demand and queues would form.

Batching patients for ‘Post-take ward round’ Real-time senior specialist review (7/7) Bedded medical assessment unit could be unnecessary for most geriatric medicine patients

Twice weekly senior clinician ward rounds Daily senior decision capability on every ward

MDT planning meetings Assess needs at home once acute hospital environment no longer adding value

Through the testing we have learned that the same nurses, doctors and therapists look after patients regardless of how they are referred to us. We have measured how fragmented the function of this team is. In our tests, this interprofessional team has to work in 5 different places. This fragmentation makes us less efficient.

Porter’s Value Based Design Integrated Practice Units Measure Outcome Bundled Pricing Integrate across Geography Expand Excellence Enabling IT platform VALUE What Is Value in Health Care? Michael E. Porter, Ph.D. N Engl J Med 2010; 363: 2477-2481December 23, 2010

Implementation headlines: April 2012 New discharge process from assessment units Consultant geriatricians ‘on take’ 7 days per week May 2012 Frailty Unit process initially virtually Frailty Unit opens mid-May July 2012 Ambulatory care area for work formerly considered to be outpatient

Outcome measure: 34% increase in discharge within 1 day

Outcome measure: Bed occupancy reduced by over 60 beds

Was reduction in bed usage due to reduced admissions? No

Would it have happened regardless? Balance measure: Would it have happened regardless?

Decrease in readmissions Balance measure: Decrease in readmissions

Balance measure: Decreased mortality

Value Value = Outcome / Cost Return on investment = Saving – Investment / Investment = (£3,000,000 - £750,000) – 140,000 / £140,000 = 2,110,000 / 140,000 = 15

Not hospital @ home Resources have started to move to the community Designing hospital@hospital and home@home

We feel like we are at the bottom of an enormous cliff. At the start it felt completely insurmountable. Through testing we have started to find some hand and foot holds to start climbing. Through the measurement and interprofessional discussions we are beginning to plot a route. We still have a long way to go.

‘Improvement in health care is 20% technical and 80% human’ Marjorie Godfrey The Dartmouth Institute

Conclusion Modern health care is complex Iterative testing and prototyping is required Cooperation between and health and social care is essential Our journey has only just started 35

Thank you Tom.Downes@sth.nhs.uk @sheffielddoc