geraint_london_slides25jan 2011

Predictive case modelling in social care and health

www.nuffieldtrust.org.uk

Geraint Lewis FRCP FFPH

t: 020 7631 8450e: info@nuffieldtrust.org.uk

www.nuffieldtrust.org.uk

The Nuffield Trust

Case Finding• NHS predictive models• Models for social careEvaluationRemuneration

Why Predictive Modelling?

• BMJ in paper* in 2002 showed Kaiser Permanente in California seemed to provide higher quality healthcare than the NHS at a lower cost

*Getting more for their dollar: a comparison of the NHS with California's Kaiser Permanente BMJ 2002;324:135-143

• Kaiser identify high risk people in their population and manage them intensively to avoid admissions

• Inaccurate Approaches:o Clinician referrals o Threshold approach (e.g. all patients aged >65 with 2+

admissions)

Frequently-admitted patients

0

5

10

15

20

25

30

35

40

45

50

- 5 - 4 - 3 - 2 - 1 Intense year

+ 1 + 2 + 3 + 4

Average number of emergency bed days

Regression to the mean

0

5

10

15

20

25

30

35

40

45

50

Average number of emergency bed days

- 5 - 4 - 3 - 2 - 1Intense

year + 1 + 2 + 3 + 4

0

5

10

15

20

25

30

35

40

45

50

Average number of emergency bed days

- 5 - 4 - 3 - 2 - 1 Intense year

+ 1 + 2 + 3 + 4

Emerging Risk

Kaiser Pyramid

The Pyramid represents the distribution of risk across the population

Small numbers of people at very high

risk

Large numbers of

people at low risk

[Size of shape is proportional to number of patients]

Inpatient data

A&E data GP Practice

data

Outpatient data PARR

Patterns in routine data

Combined Model

Census data

Scotland• SPARRA• SPARRA-MD

Wales• PRISM model• Welsh Predictive Risk

Service

J7KA42

J7KA42

J7KA42

J7KA42

J7KA42

J7KA42 76.4

131178 76.4

Encrypted, linked data

Decrypted data with risk score

attached

131178

131178

131178

131178

Inpatient Outpatient A&E GP

Name, Address, DOB

Name, Address, DOB

Name, Address, DOB

Name, Address, DOB

10 Million Patient-Years of Data

5 Million Patient-Years of Data

5 Million Patient-Years of Data

Development

Validation

J7KA42

YH8TPP

G8HE9F

3LWZ67

2NX632

LG5DSD

3V9D54R

J7KA42

YH8TPP

G8HE9F

3LWZ67

2NX632

LG5DSD

3V9D54R

J7KA42

YH8TPP

G8HE9F

3LWZ67

2NX632

LG5DSD

3V9D54R

Year 1 Year 2 Year 3

Development Sample

Inpatient Outpatient A&E GP

J7KA42

YH8TPP

G8HE9F

3LWZ67

2NX632

LG5DSD

3V9D54R

J7KA42

YH8TPP

G8HE9F

3LWZ67

2NX632

LG5DSD

3V9D54R

J7KA42

YH8TPP

G8HE9F

3LWZ67

2NX632

LG5DSD

3V9D54R

Development Sample

Year 1 Year 2 Year 3

Inpatient Outpatient A&E GP

J7KA42

YH8TPP

G8HE9F

3LWZ67

2NX632

LG5DSD

3V9D54R

J7KA42

YH8TPP

G8HE9F

3LWZ67

2NX632

LG5DSD

3V9D54R

J7KA42

YH8TPP

G8HE9F

3LWZ67

2NX632

LG5DSD

3V9D54R

Development Sample

Year 1 Year 2 Year 3

Inpatient Outpatient A&E GP

A89KP5

833TY6

I9QA44

85H3D

6445JX

233UMB

RF02UH

A89KP5

833TY6

I9QA44

85H3D

6445JX

233UMB

RF02UH

A89KP5

833TY6

I9QA44

85H3D

6445JX

233UMB

RF02UH

Validation Sample True

Positive

False Positive

False Negative

True Negative

Year 1 Year 2 Year 3

Inpatient Outpatient A&E GP

A89KP5

833TY6

I9QA44

85H3D

6445JX

233UMB

RF02UH

A89KP5

833TY6

I9QA44

85H3D

6445JX

233UMB

RF02UH

Using the Model

Last Year This Year Next Year

Inpatient Outpatient A&E GP

Distribution of Future Utilisation

£0

£500

£1,000

£1,500

£2,000

£2,500

£3,000

£3,500

£4,000

£4,500

0 10 20 30 40 50 60 70 80 90

Predicted

Risk (centile rank)

Actual

Average cost per patient

NHS Combined Model

Clinical Profiles

Tackling the Inverse Care Law

Developing Business Cases

How the output of predictive models are used

• Case Management

• Intensive Disease Management

• Less Intensive Disease Management

• Wellness Programmes

Potential Misuses

Dumping

Cream-skimming

Skimping

Evaluation of Preventive Care

5

Start of intervention

Overcoming regression to the mean using a control group (1)

Overcoming regression to the mean using a control group (2)

Start of intervention

Overcoming regression to the mean using a control group (3)

Start of intervention

Start of intervention

Overcoming regression to the mean using a control group (4)

Person-Based Resource Allocation

• Historically, GP practice budgets set on area-based variables

• New approach is person-based• Exclude certain variables to avoid

perverse incentiveso Procedureso Disease severity

geraint.lewis@nuffieldtrust.org.uk

t: 020 7631 8450e: info@nuffieldtrust.org.uk

Model predicts:

Details

Examples

Trend

Model predicts: Cost

Details Model predicts which patients will become high-cost over next 6 or 12 months

Examples Low-cost patient this year will become high-cost next year

Trend

Model predicts: Cost Event

Details Model predicts which patients will become high-cost over next 6 or 12 months

Model predicts which patients will have an event that can be avoided

Examples Low-cost patient this year will become high-cost next year

Patient will be hospitalized

Patient will have diabetic ketoacidosis

Trend

Model predicts: Cost Event Actionability

Details Model predicts which patients will become high-cost over next 6 or 12 months

Model predicts which patients will have an event that can be avoided

Model predicts which patients have features that can readily be changed

Examples Low-cost patient this year will become high-cost next year

Patient will be hospitalized

Patient will have diabetic ketoacidosis

Patient has angina but is not taking aspirinPatient does not have pancreatic cancer (Ambulatory Care Sensitive)

Trend

Model predicts: Cost Event Actionability Readiness to

engage

Details Model predicts which patients will become high-cost over next 6 or 12 months

Model predicts which patients will have an event that can be avoided

Model predicts which patients have features that can readily be changed

Model predicts which patients are most likely to engage in upstream care

Examples Low-cost patient this year will become high-cost next year

Patient will be hospitalized

Patient will have diabetic ketoacidosis

Patient has angina but is not taking aspirinPatient does not have pancreatic cancer (Ambulatory Care Sensitive)

Patient does not abuse alcohol

Patient has no mental illness

Patient previously compliant

Trend

Model predicts: Cost Event Actionability Readiness to

engageReceptivity

Details Model predicts which patients will become high-cost over next 6 or 12 months

Model predicts which patients will have an event that can be avoided

Model predicts which patients have features that can readily be changed

Model predicts which patients are most likely to engage in upstream care

Model predicts what mode and form of intervention will be most successful for each patient

Examples Low-cost patient this year will become high-cost next year

Patient will be hospitalized

Patient will have diabetic ketoacidosis

Patient has angina but is not taking aspirinPatient does not have pancreatic cancer (Ambulatory Care Sensitive)

Patient does not abuse alcohol

Patient has no mental illness

Patient previously compliant

Patient prefers email rather than telephone

Patient prefers male voice rather than female

Readiness to change

Trend