Risk profiling, multiple long term conditions & complex patients, integrated care

Risk Profiling, LTC complex patients, Integrated Care teams

& EoLC

Dr Bruce PollingtonMedical Director

The Heart of Kent Hospice March 2012

LTC QIPP Workstream

• Risk profiling to identify the top 5% (or 1%), ie highest risk of admission to hospital in the next 12 months.

• Developing Integrated Care Teams, multi-professional, multi-agency providing holistic care to the neighbourhood.

• Self Care programmes, giving back control to people, imparting information and supporting knowledge, ‘expert on ones self’.

Risk Profiling

• Risk profiling to identify the top 5% (or 1%), ie highest risk of admission to hospital in the next 12 months, the top of the pyramid may be 20 times greater, then target selected populations for integrated intervention 1,2,3,8

• By combining hospital, opd, A&E, and GP data sets improves the PPV.

• Add in social care data (1) Stuck, Siu, Whieland et al. “Comprehensive geriatric assessment: a meta-analysis of controlled trials. Lancet 1992; 342: 1032-6 (2) Conn, Valentine & Cooper “Interventions to increase physical activity among aging adults: a meta-analysis”. Ann Behav Med 2002; 24(3): 190-200 S (3) Fagerberg, et al. “Effect of acute stroke unit care integrated with care continuum versus conventional treatment. Stroke 2000; 31(11): 2578-84 (8) COMBINED PREDICTIVE MODEL FINAL REPORT Kings Fund 2006

Integrated working

• Developing Integrated Care Teams, multi-professional, multi-agency providing holistic care to the neighbourhood. 6, 11

(6) Integrated care for patients and populations: Improving outcomes by working together Report to the Department of Health and NHS Future Forum from The King’s Fund and Nuffield Trust 2012 (11) Integrated team working: a literature review S. Maslin-Prothero International Journal of Integrated Care, 29 April 2010

SelfCare

• Self Care programmes, giving back control to people, imparting information and supporting knowledge, ‘expert on ones self’. 4

• Health and other outcomes5

–increase in life expectancy–better control over symptoms–reduction in pain, anxiety and depression levels• Implications for the care system5

–visits to GPs can reduce by 40 to 69% –hospital admissions can reduce by up to 50%

(4) RESEARCH EVIDENCE ON THE EFFECTIVENESS OF SELF CARE SUPPORT Department of Health 12 December 2007. www.dh.gov.uk/prod_consum_dh/groups/dh_digitalassets/@dh/@en/documents/digitalasset/dh_081251.pdf (5)Self Care Support summary of work in progress (2005-07) THE EVIDENCE PACK DH July 2007 www.dh.gov.uk/prod_consum_dh/groups/dh_digitalassets/documents/digitalasset/dh_076932.pdf

The EoLC National Strategy

• Definition of EoLC • By end of life care it is meant the services that

support those with advanced progressive incurable illness to live as well as possible until they die.

• These services enable the supportive and end of life care needs of both patients and their families to be identified and met throughout the last phase of life and into bereavement

Holistic Care

• It includes the management of pain and other symptoms and the provision of psychological support. It is not restricted to any specialist services and includes those services provided as an integral part of the practice of any health or social care professional in any setting.

The 6 Steps in the EoLC Pathway

• Starting the conversation• Assessment and care planning• Coordination of Care • Delivering high quality care• Last days of life• Care after death

Starting theconversation

Assessment andcare planning

Coordination of Care for the patient

Delivering high quality care

Last days of life

Care after death

•Open, honestcommunication• Identifying triggersfor discussion• Listening to cuesfrom patients

• Assessment and regular review ofpatients’ needs• Care planning• Assessing carers needs

• Coordination of individual patient care• Register-information shared across all sectors

• High quality careprovision in all settings

• Rapid response services

• Hospital, community, carehomes, hospices, communityhospitals, prison, secure hospitalsand hostels

• Ambulance Services•All OOH services

• Spiritual care

• Identification ofthe dying phase

• Review of needsand preferencesfor place of death

• Support for bothpatient and carer

• Recognition ofwishes regardingresuscitation andorgan donation

• Timely verificationand certification ofdeath• Viewing of thebody/mortuaryfacilities• Return of property• Care and supportof carer andFamily

The End Of Life Clinical Pathway

Coordination of Care for carers

End Of Life Care for all diseases

EOLC CANCER

DEMENTIA

LONG TERM CONDITIONS

COMPLEX FRAIL ELDERLY

NEUROLOGICAL CONDITIONS

Illustration of changing functional level over last year or so of life.

Slide courtesy of Whole Systems Partnership

Some issues to consider

• It has been estimated that 42% may follow a frail elderly functional decline trajectory, 9 Often this group has no diagnostic pigeon hole and their deterioration goes ‘largely’ unrecognised

• 73% of patients on GP Palliative Care QOF have a diagnosis of cancer, but cancer represents at most only 30% predictable death, 10

(9) Whole Systems Partnership, National EoLC Programme Et al Nov 2010 (10) NEoLCIN

Some issues to consider

• At any one time about 25% of inpatients in acute hospitals are in their last year of life

• On average people are admitted to hospital three times and spend nearly a month of the last year of their life in hospital

Complex Patients

• Risk Profiling really is identifying the complex patients. Or to put it another way, the people that would benefit most from a holistic and integrated approach to their care.

• A High proportion of this group will also be the patients making the transition into End of Life

Which Percentage Groups should we be looking at?

• Due to considerations called regression to the mean, LTC case management may need to look just below the highest risk to have time to make a difference

• But who is in the top 1% and is there still time to make a difference?

• Many in the top 1% actually need something different.

Risk profiling will help find the 1%

• 1% being the percentage of the population who die each year. And a National Campaign

• As it stands, timely identification of the non-cancer end of life care patients is presenting a challenge.

• Risk profiling can help us leap frog over this hurdle.

The top 1% need a change in focus

• For the top 1% integrated teams need to seriously consider if their patient is entering EoL or they will continue to have an average of 3 admissions in that last year.

The transition from LTC to EoLC

• Need to move away from the handing over the baton model.

• This relay race model of service provision leads to late referral and perpetuates the silo culture.

• We need to integrate disease management with symptom management

People Call Ambulances because of a deterioration in their symptoms.

We give disease management advice but as a collective we tend to

‘withhold’ symptom management until we diagnosis a person as at end of life.

Traditionally Palliative Care services would decline to see non-EoLC

patients as not meeting the criteria.

The COPD suffer is a good example or indeed anyone suffering with

SOB, there is good EBM for the use of morphine yet we withhold its use, you have to be dying before

your symptom of SOB is treated.

So what should the model of care look like.

• Integrated Care where the treatment and support provided is based on assessed needs with absolutely no arbitrary boundaries

• Also we need to front load assessment and care so people can gain the maximum benefit, hence retaining the maximum independence.

• An earlier OT home assessment being a good example. It is rare to need repeating.

So what should the model of care look like.

Physical

Psychological

Social

Spiritual

Disease management Symptom ManagementIntegrated Care Team for complex high need people

Front load assessment with

earlier intervention

LVF

COPD

CKD

OA

Dementia

How can technology help us? • Risk profiling to find the complex patients. Additional modelling may be developed to

identify the frailty group, who are at high risk of dying but perhaps not at high risk of admission, this group may otherwise deteriorate relatively un-noticed.

• Integrated Care Teams (NHS, social services, voluntary sector) to provide holistic assessment, with both disease and symptom management aiming to support independence. Use video-conferencing to link patients and families with the team and experts as needed bringing them closer to the decision making

• Self Care and escalation planning or personalised care plan. Deploy telehealth devices to the high risk group, these can have first tier self-care escalation advice tailored to the individual.

• Link teleheath device to urgent care dash board, efficient monitoring to alert the integrated team of changes to health status directing early intervention.

• “The key is to integrate these technologies into the care and services that are delivered. Going forward this evidence [WSD] gives us confidence that we can transform the way services are delivered”7

(7) Whole System Demonstrator Programme Headline Findings – December 2011 DH

We can use Risk Profiling to predict the future risk.

• How can you get ahead of the curve?• We want to identify the patients with

increasing risk before they are the highest risk.

How to get ahead of the curve?

• For any individual the risk of admission is dynamic.• The rate of change in risk will vary from one

individual to another.

Top 1%

Top 5%

Top 30%

• The white arrows represent what is happening to the admission risk for individuals over a given time.

• While for the majority the risk alters little over any given time period, for some the risks are escalating rapidly

Top 1%.

Top 5%

Top 40%

The risk is static for this individual

Risk

of h

ospi

tal a

dmis

sion

Time

On the way up there is plenty of scope for a rapid increase while at the top of the curve there is little head room left for further increase in risk, hence the rate of increase tails off

How to get ahead of the curve

At this point the individual enters the top 1%

5000th

1st

500th

Risk

of h

ospi

tal a

dmis

sion

Time

By regularly (or preferably constantly) running the risk tool we can plot individuals change in risk over time, we get new graph, the rate of change of risk.

Rate

of c

hang

e in

risk

of

hosp

ital a

dmis

sion

Risk

of h

ospi

tal a

dmis

sion

Time

As you can see the peak in rate falls ahead of the absolute maximum, roughly speaking a quarter cycle ahead.

Rate

of c

hang

e in

risk

of

hosp

ital a

dmis

sion

It buys this much time to prevent the admission

Risk

of h

ospi

tal a

dmis

sion

Time

On a population basis we should profile for those with the highest rate of increase in risk of hospital admission, they are the most unstable patients.

Rate

of c

hang

e in

risk

of

hosp

ital a

dmis

sion

It buys this much time to prevent the admission

Top 1%

Dr Bruce [email protected]

01622 792200