E2CM: WHAT’S NEXT?

World Bank Group

October 2017

e2cm = Estonian Enhanced Care Management

1

Areas of further work

Risk-stratification

PHC provider competencies and skills

Involvement of other stakeholders

E-Health

Contracting and monitoring

2

Payment

systems

Impact

evaluation

Adjust to group practices

e2cm

Areas of further work

Risk-stratification

PHC provider competencies and skills

Involvement of other stakeholders

E-Health

Contracting and monitoring

3

Payment

systems

Impact

evaluation

Adjust to group practices

e2cm

Risk-Stratification Model for EECM

1 2 3 4 5

Vet registry

Clinical data B/S data

1 4 5

A

B

Metabolic Triad

(at least 1)

Include Exclude

CVD (1 or 2)

Respiratory disease(1 or 2)

Mental Health /

Disability (1)

Patients with no or

limited potential to benefit

from CM at PHC level

Favorable / unfavorable

Tailor response

Intuition

Exclusion: Not

amenable to change

through CM

Inclusion: “Missed” patients

Metabolic Triad

(at least 1)

Non-utilizers

Exclusion: as above

Known

Unknown

E2CM

Design principles of the risk-stratification approach

▪ Health maximization / potential to benefit from EECM

Archetypes

▪ Dynamic

▪ KISS (Keep It Simple, Stupid) at the beginning

Modular structure

▪ Mixed approach – (Data and intuition-based)

– Data: Clinical, behavioral, social

Design principles of the risk-stratification approach

▪ Health maximization / potential to benefit from EECM

Archetypes

▪ Dynamic

▪ KISS (Keep It Simple, Stupid) at the beginning

Modular structure

▪ Mixed approach – (Data and intuition-based)

– Data: Clinical, behavioral, social

Risk-Stratification - areas for further development

▪ Two tracks of patients

▪ Disease severity and amenability to EECM

▪ Sources of behavioral & social data

▪ National variations in the disease burden

▪ Dynamic risk-stratification

Two tracks of patients

1 2 3 4 5

Vet registry

Clinical data B/S data

1 4 5

A

B

Metabolic Triad

(at least 1)

Include Exclude

CVD (1 or 2)

Respiratory disease(1 or 2)

Mental Health /

Disability (1)

Patients with no or

limited potential to benefit

from CM at PHC level

Favorable / unfavorable

Tailor response

Intuition

Exclusion: Not

amenable to change

through CM

Inclusion: “Missed” patients

Metabolic Triad

(at least 1)

Non-utilizers

Exclusion: as above

Known

Unknown

E2CM

Two tracks of patients

A

B

Known

Unknown

E2CM

Different risk profiles for the same objective▪ Example 1: Elderly patients, multiple a/o unstable chronic conditions▪ Example 2: Relatively young patients, some chronic condition

Disease severity and amenability to EECM

1 2 3 4 5

Vet registry

Clinical data B/S data

1 4 5

A

B

Metabolic Triad

(at least 1)

Include Exclude

CVD (1 or 2)

Respiratory disease(1 or 2)

Mental Health /

Disability (1)

Patients with no or

limited potential to benefit

from CM at PHC level

Favorable / unfavorable

Tailor response

Intuition

Exclusion: Not

amenable to change

through CM

Inclusion: “Missed” patients

Metabolic Triad

(at least 1)

Non-utilizers

Exclusion: as above

Known

Unknown

E2CM

Disease severity and amenability to EECM

1 2 3 4 5

Clinical data

1 4 5

A

B

Metabolic Triad

(at least 1)

Include Exclude

CVD (1 or 2)

Respiratory disease(1 or 2)

Mental Health /

Disability (1)

Patients with no or

limited potential to benefit

from CM at PHC level

Favorable / unfavorable

Exclusion: Not

amenable to change

through CM

Inclusion: “Missed” patients

Metabolic Triad

(at least 1)

Non-utilizers

Exclusion: as above

Known

Unknown

E2CM

Severity of disease ->

Example:< 4 admissions / 12

months

Am

enab

ility

->

Importance of Behavioral and Social Data

Who are the high-need patients that are amenable to EECM?

Importance of Behavioral and Social Data

Who are the high-need patients that are amenable to EECM?

▪Social and behavioral factors are as important as actual medical needs

▪Behavioral/mental illnesses impact how well patients can handle their other chronic conditions

▪Social needs (loneliness, financial worries, etc.) critically define a patient’s health condition

Behavioral Social

x/✓ x/✓

Behavioral & social data - Tapping the great data existing data sources

Amenability to EECM

Behavioral▪ Info on missed visits (to become available)▪ Number of medications/number of medications picked up▪ Emergency visits▪ Avoidable specialist visits

Social▪ Living alone▪ Mother tongue & knowledge of Estonian▪ Socio-economic status (household income)▪ Education level▪ Employment status▪ Place of residence

Medical complexityPatient

behaviorSocial

determinants

Clinical data Behavioral & social data

National variations in the disease burden

0%

10%

20%

30%

40%

50%

60%

70%

80%

1

14

27

40

53

66

79

92

10

5

11

8

13

1

14

4

15

7

17

0

18

3

19

6

20

9

22

2

23

5

24

8

26

1

27

4

28

7

30

0

31

3

32

6

33

9

35

2

36

5

37

8

39

1

40

4

41

7

43

0

44

3

45

6

46

9

48

2

49

5

50

8

52

1

53

4

54

7

56

0

57

3

58

6

59

9

61

2

62

5

63

8

65

1

66

4

67

7

69

0

70

3

71

6

72

9

74

2

75

5

76

8

78

1

79

4

For about 500 GP lists, the % is 20 <-> 40. For the rest the % is lower or higher…

% of patients with >= 1 condition(s) from metabolic triad (DM/HTN/Hyperlipidemia)

National variations in the disease burden

0%

10%

20%

30%

40%

50%

60%

70%

80%

1

14

27

40

53

66

79

92

10

5

11

8

13

1

14

4

15

7

17

0

18

3

19

6

20

9

22

2

23

5

24

8

26

1

27

4

28

7

30

0

31

3

32

6

33

9

35

2

36

5

37

8

39

1

40

4

41

7

43

0

44

3

45

6

46

9

48

2

49

5

50

8

52

1

53

4

54

7

56

0

57

3

58

6

59

9

61

2

62

5

63

8

65

1

66

4

67

7

69

0

70

3

71

6

72

9

74

2

75

5

76

8

78

1

79

4

For about 500 GP lists, the % is 20 <-> 40. For the rest the % is lower or higher…

% of patients with >= 1 condition(s) from metabolic triad (DM/HTN/Hyperlipidemia)

▪ Homogeneity or heterogeneity of patients across FP practices▪ Balancing the size/risk-profile of patient lists across FP practices

▪ Dynamics of risk profiles– Changes in risk profiles over time (short-term acute vs. longer-term risk), need for graduation of

people from the registry according to rules

▪ Graduation/discharge rules:– How to balance the influx vs. outflow of patients through the right graduation/discharge criteria?

– How to determine whether someone stays in the CM program?

▪ Updates of the risk-stratification algorithms– Gradual but constant improvements to target the right patients

The goal - dynamic risk-stratification

Areas of further work

Risk-stratification

PHC provider competencies and skills

Involvement of other stakeholders

E-Health

Contracting and monitoring

18

Payment

systems

Impact

evaluation

Adjust to group practices

e2cm

Introduction to payment systems

19

Capitation FFS QBS Allowances

Payment methods

Classifying Counting Costing Pricing Monitoring

Payment system functions

Challenges of paying for E2CM:

• Activities are not considered under traditional payment methods (e.g. capitation, FFS,

allowances, pay for performance)

• Since they are historically not expected from primary health care providers, e.g.:

• Comprehensive assessment of care needs, considering patient’s social and contextual environment

• Development of comprehensive care plans (dual-facing, anticipatory)

• Coordination of care transitions (beyond referrals, including follow-up care)

• Coordination of social care services

20

Key principles of payment method design

• Payments can be made before services are delivered (ex-ante) or after (ex-post)

• Ex-ante payments tend to encourage efficiency, but may result in “skimping” or “cream-skimming”

• Ex-post payments encourage the delivery of priority services, but may result in unnecessary care

• Payment recipients include either:

• Structures/facilities (e.g. primary care practice) or

• Individual health care professionals (e.g. primary care physicians, nurses, primary and/or secondary care

specialists, etc.)

• Payment rates may vary based on patient severity level

21

Principle options for enhanced care management payment

22

Option 1:

Fixed-rate

payment

Option 2:

Fees for

services

Option 3:

Fees for

service

bundles

Option 4:

Performance

based

payments

Option 1: Fixed-rate payment

• Key features:

• Single payment covering costs of all program services for an average patient during a

business cycle

• Flat or risk-adjusted payment

• Ex-ante

• Example: “Experimentation of New Modes of Remuneration (ENMR)” in France

➢ Payer: National Health Insurance Fund - scheme for salaried workers

➢ Design:

➢ Separate flat, fixed-rate payments for three domains

• Care coordination

• Patient education

• Multidisciplinary approach

➢ Payments to the facilities not professionals

• 3 types of multidisciplinary primary care facilities with different governance structures and contracting arrangements

➢ Combined payments constituted approx. 5% of facility revenue

23

Option 2: Fees for services

• Key features:

• Fee-for-service

• Ex-post

• Example: “Chronic Care Management (CCM) Services” in the United States

➢ Payer: Medicare

➢ Design:

➢ Fees for two types of services:

i. initial assessment and care planning, and

ii. care coordination activities per month, including ongoing, non-face to face oversight, direction and management

➢ Fees vary depending on complexity of patient

➢ Payments to primary care physicians, certified nurse midwives, clinical nurse specialists, nurse

practitioners or physicians assistants

24

Option 3: Fees for service bundles

• Key features:

• Fees for bundles of services

• Ex-post

• Example: “Cardio-Integral” program in Germany

➢ Payer: AOK Plus - statutory sickness fund of Saxony and Thuringia

➢ Design:

➢ Three service bundles:

• Enrollment of patients, compliance monitoring (visits, treatment), adherence to clinical pathways

• Coordination of invasive diagnostic and therapeutic procedures

• Compliance with treatment guidelines, including drug lists

➢ Tariffs and billing frequencies dependent on care needs of patients with different cardiovascular

conditions

➢ Payments to structures – GP practices and cardiac department in University Hospital Dresden

25

Option 4: Performance-based payment

• Key features:

• Single payment, with amount contingent on provider performance

• Process or outcome indicators

• Ex-post or mixed

• Example: Second phase of “ENMR” program in France

➢ Payer: National Health Insurance Fund - scheme for salaried workers

➢ Design:

➢ Performance-based payment for care coordination domain only

➢ Distinguishes sub-domains of mandatory and optional services

➢ For each of these sub-domains, fixed and variable payments

• Fixed payments based on number of patients

• Variable component reflects improvement in care coordination processes (24 key indicators, three dimensions: (i)

quality of care, (ii) coordination and continuity and care, (iii) efficiency)

➢ Mixed: providers receive 60% of expected payment in advance with the remaining share paid at the end of

the year.

26

Some lessons

• Payment methods for enhanced care management can create strong incentives for providers to

enhance care management

• Focus is on improvements in processes rather than outcomes

• Typically constitute small share of provider revenue, however independent of the payment

method combined with close provider monitoring

• Administrative burden depends on overall payment system design

27

Potential integration with current payment methods

Presentation Title28

Capitation FFS QBS

Option 1:

Fixed-rate

payment

Option 2:

Fees for

services

Option 3:

Fees for

service

bundles

Option 4:

Performance

based

payments

Likely best design fit for Estonia

• Fixed-rate payment covering costs of all program services for an average patient during a

business cycle (Option 1)

• Reflects the comprehensive approach to enhanced care management

• Supports focus on coaching and supervision rather than billing and reporting in early stage of program

• Keeps administrative burden low (prior to a comprehensive e-health solution)

• Payment to structures (both solo and group practices)

• Facilitates shift from solo to group practices

• Eventually risk-adjusted dependent on improvements to risk-stratification

• Combined with development of contracting and provider monitoring model

29

30