Shifting accountability An international comparison of payment reforms
Jeroen N. Struijs Financiering Eerstelijnszorg Jan van Es instituut Almere,10/09/2015
END OF PART 1
2
Verzekeraar
Zorginkoopmarkt Zorgverzekeringsmarkt
Zorgaanbieders Zorgverleningsmarkt
Patient/ consument
3
Ref: unknown
5
Juridische entiteit
Per verrichting (FFS)
Integrale Bekostiging
‘Shared Savings’
Bron: Averill et al., 2010, bewerkt voor Nederlandse bekostigingsmodellen
Uitkomst Bekostiging
Background • FFS: – flexible and easy – maximizing patient visits (volume?) – No incentive to deliver efficient care or prevent unnecessary care – No accountability across setting and multiple providers – Financial risk for payer
Background • Capitation: – flexible and easy – Minimizing patient visits (incentive to deliver efficient care or prevent unnecessary care – Stinting on care? – Financial partly for provider and provider (salary)
Background • Alternative models: – Pay for coordination – Pay for reporting – Pay for Performance – Bundled payment – Shared savings – Global payment – Combination of above
Centrale vragen 1. 2.
Hoe bekostigingshervormingen in verschillende landen vormgegeven? Hoe kunnen zorginkoopcontracten bijdragen aan kwaliteitsverbeteringen en kostenreducties in de zorg? (lessen van de AQC voor NL)
Casussen: - Zorggroepen / Integrale bekostiging (NL) - Alternative Quality Contract (AQC) (US) - MSSP ACOs (US) - Clinical Commissioning Groups (England) Methode: semi-gestructureerde interviews en literatuur
Dutch payment reform
Insurers Bundled Payment contracts (multiple single-disease care program contracts)
Care Group
capitation
capitation
capitation
GP
PROVIDERi
PROVIDERi
capitation PROVIDERi
FFS PROVIDERi
10
National monitor Pioneer sites • Shift from BP towards Population Management • 9 regions selected as pioneer sites of population management • Pioneer sites are enrolled in the National Monitor of Population Management • All aiming to improve the TA
11
Where are we?
Thema’s Thema’s & Thema’s & Themes updates & updates & updates updates
2013
2014
2015
2016
Follow-up report: Process and outcomes
2017
2018
Drewes et al. Samen werken aan duurzame zorg. http://www.rivm.nl/Documenten_en_publicaties/Wetenschappelijk/Rapporten/2015/juli/Samen_werken_aan_du 12 urzame_zorg_Landelijke_monitor_proeftuinen
12
Objectives National Monitor PM 4 overall research questions: 1. How is population management designed? 2. What are the barriers and facilitators in PM?
3. How is health, quality of care and costs developed over time? 4. What is the association between these outcome measures?
Mixed methods
13
Organisation: involved actors 1
Schools/ Sports Employer GGD Municipality Home care Youth care Mental care GPs Long term care Hospitals Others Patient representatives Health care insurers
2
3
4
5
6
7
8
9
A say Co-produce Advice Consult/ inform
16
Current Interventions 1
2
3
4
5
6
7
8
9
Prevention Lifestyle interventions Selfmanagement Awareness Selfmanagement capacity Integrated care Frailty Maternity care Mental care Diabetes COPD VRM Others Substitution Farmacy 1,5 care Concentration&specialisation Diagnostics Precondition Teamclimate Community involvement Data-infrastructure Transition towards population management | 26 maart 2015
17
Population 1
2
3
4
5
6
7
8
9
Rest NL
142990
491860
84470
222290
143560
321600
209455
400915
84640
10770140
Sexe (% male)
48,5
49,8
49,5
49,3
48,7
49
49,1
50,2
51,6
49,1
Age (% 65+)
23,6
23,2
27,6
25,2
18,4
21,4
23,1
22,8
23,2
20,8
Education (% low)
8
7,3
11
9,4
6,6
6,4
5,2
6,7
8,7
8
Income (% high )
20,9
16,7
24,2
17
30,8
23,5
26,4
25
17,4
24,9
Employed (%)
57,3
60,8
59,8
55,4
65,4
62,9
63
62,4
62,7
63,2
5,8
3,2
4,5
6,9
2,2
4,8
4,6
4
3,2
3,9
Population
Disablled to work totally disabled)
(%
Transition towards population management | 26 maart 2015
18
Population health Blauwe Friesland Goed PZF Mijn Zorg Pelgrim Zorg Voorop Leven Rijnland
SSiZ
SmZ
Vitaal Rest NL Vechtdal
Experienced health (% more or less-bad’)
26,2
19,2
26,7
32,2
19
22,7
22,4
24,5
21,9
23,6
Disabilities (% 1 or more)
14,8
12,4
17,5
18,9
11,2
13,8
11,9
15,5
14,9
14,9
Chronic conditions (% at least 1)
62,9
58
66,1
69,7
57
62,5
57,3
57,9
57,1
60,4
Anxiety and depression (% high risk)
6
3,9
5,2
7,7
4,7
6,4
4,3
5,8
4
5,7
BMI (% overgewicht)
46,5
47,4
53,5
54,4
42,1
48,1
45,1
48,4
51,3
48,3
Mortality (per 10.000)
104
91
110
110
78
89
82
82
87
84
*red = significant unhealthier; green= significant healthier compared to other regions **Not standardized results. Transition towards population management | 26 maart 2015
19
Population health – corrected for age and sexe (100=NL average)
Blauwe Friesland Goed PZF Mijn Zorg Pelgrim Zorg Voorop Leven Rijnland
SSiZ
SmZ
Vitaal Vechtdal
Experienced health (% more or less-bad’) 97
107
98
90
104
101
102
99
103
98
82
105
117
82
92
78
103
96
104
95
105
111
97
103
93
96
93
Anxiety and depression (% high risk)
101
89
90
109
96
100
96
106
89
BMI (% overgewicht)
96
97
106
109
90
99
92
99
104
Disabilities (% 1 or more) Chronic conditions (% at least 1)
*100= equal to expected health based on age and sexe red = significant unhealthier; green= significant healthier compared to other regions
20
Background US health care system
Affordable Care Act / ‘Obamacare’ • Why the ACA? • What is in it?
US Health Care System
age
65
US population
US Health Care System
age
65
US population
age
65 Veterans Health Administration, 1930
Employment based insurance 1940-1950 subsidized excluded from taxable income
US population
US Health Care System
age
65
Indian Health Services 1955
Veterans Health Administration, 1930
Employment based insurance 1940-1950 subsidized excluded from taxable income
US population
US Health Care System
age
65
Indian Health Services 1955
Veterans Health Administration, 1930
Employment based insurance 1940-1950 subsidized excluded from taxable income
Medicaid 1965
US population
US Health Care System Traditional Medicare Fee-for-service payment 1965
age
65
Indian Health Services 1955
Veterans Health Administration, 1930
Employment based insurance 1940-1950 subsidized excluded from taxable income
Medicaid 1965
US population
US Health Care System Traditional Medicare Fee-for-service payment 1965
age
65
Indian Health Services 1955
Veterans Health Administration, 1930
Employment based insurance 1940-1950 subsidized excluded from taxable income
Children Health Insurance Program 1997
Medicaid 1965
US population
US Health Care System Traditional Medicare Fee-for-service payment 1965
age
65
Indian Health Services 1955
Veterans Health Administration, 1930
Employment based insurance 1940-1950 subsidized excluded from taxable income
Children Health Insurance Program 1997
Medicaid 1965
US population
US Health Care System Traditional Medicare Fee-for-service payment 1965
age
65
Indian Health Services 1955
Veterans Health Administration, 1930
Employment based insurance 1940-1950 subsidized excluded from taxable income
Children Health Insurance Program 1997
Medicaid 1965 Uninsured (small firms, individuals, 2012)
US population
Uninsured Health Insurance Exchanges
In 2012, Nearly Half of Adults Were Uninsured During the Year or Were Underinsured
Insured all year, not underinsured^ 54% 100 million
Uninsured during the year* 30% 55 million Insured all year, underinsured^ 16% 30 million
184 million adults ages 19–64 Note: Numbers may not sum to indicated total because of rounding. * Combines “Uninsured now” and “Insured now, time uninsured in past year.” ^ Underinsured defined as insured all year but experienced one of the following: out-of-pocket expenses equaled 10% or more of income; out-of-pocket expenses equaled 5% or more of income if low income (<200% of poverty); or deductibles equaled 5% or more of income. Source: The Commonwealth Fund Biennial Health Insurance Survey (2012).
Why the ACA?
U.S. Health in International Perspective: Shorter Lives, Poorer Health • Americans live shorter lives and are in poorer health at any age • Poor outcomes cannot be fully explained by poverty or lack of insurance • White, insured, college-educated, and upper income Americans are in poorer health than their counterparts in other countries
International Comparison of Spending on Health, 1980–2012 Average spending on health per capita ($US PPP) 10000 9000 8000 7000 6000 5000
US SWIZ NOR NETH GER CAN FRA SWE AUS UK NZ
20 $8,745
18 16 14 12
AUS NOR UK SWE NZ CAN SWIZ GER FRA NETH US
10
4000 3000
Total expenditures on health as percent of GDP 17.6%
8.9%
8 $3,182
6
2000
4
1000
2
0
0 80 82 84 86 88 90 92 94 96 98 00 02 04 06 08 10 12
80 82 84 86 88 90 92 94 96 98 00 02 04 06 08 10 12
Note: $US PPP = purchasing power parity. Source: Organization for Economic Cooperation and Development, OECD Health Data, 2013 (Paris: OECD, November 2013). US data from National Health Expenditure Accounts, adjusted to match OECD definitions.
Cumulative Increases in Health Insurance Premiums, Workers’ Contributions to Premiums, Inflation, and Workers’ Earnings, 1999-2013 250%
Health Insurance Premiums Workers' Contribution to Premiums 200%
196%
Workers' Earnings Overall Inflation
182%
150% 117% 119% 100%
56% 57%
50%
50% 34% 40%
14%
29%
11%
0% 1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
SOURCE: Kaiser/HRET Survey of Employer-Sponsored Health Benefits, 1999-2013. Bureau of Labor Statistics, Consumer Price Index, U.S. City Average of Annual Inflation (April to April), 1999-2013; Bureau of Labor Statistics, Seasonally Adjusted Data from the Current Employment Statistics Survey, 1999-2013 (April to April).
2010
2011
2012
2013
The ACA: What’s in it?
ACA Made Simple… Hundreds of provisions in two big buckets:
Coverage expansion
Delivery system reform
Coverage Expansion Cover the uninsured (26 million*): – Medicaid expansions (about half) – Subsidies to buy private insurance (about half)
Regulate private markets: – – – –
Insurance mandate Children to 26 No discrimination against sick Health insurance marketplaces
* Latest CBO estimate (April 2014).
Delivery System Reform Reduced Payments for Avoidable Complications Value Based Purchasing
Accountable Care Organizations
Medicare Advantage Plan Bonuses Bundled Payments
Hospital Inpatient Quality Reporting
Physician Quality Reporting System
Medical Homes
Meaningful Use
Delivery System Reform: Three Buckets Payment reforms: pay for performance • Hospital and physician quality • Medicare readmissions • Hospital acquired conditions
Organizational reforms • Accountable care organizations
• Patient centered medical homes • Increased training and payment for primary care
Information availability • Comparative effectiveness research ($500 million/year) • Health information technology
Cross-nation comparison of payment reforms
Dutch payment reform
Insurers Bundled Payment contracts (multiple single-disease care program contracts)
Care Group
capitation
capitation
capitation
GP
PROVIDERi
PROVIDERi
capitation PROVIDERi
FFS PROVIDERi
43
Medicare Shared Savings Program ACOs Medicare Shared savings contract
ACO
FFS PROVIDERi
FFS PROVIDERi
FFS PROVIDERi
Multiple PROVIDERi
Multiple PROVIDERi
FFS = Fee For Service 45
Alternative Quality Contract (AQC) BlueCross BlueShield AQC contract
AQC group
FFS PROVIDERi
FFS = Fee For Service
FFS PROVIDERi
FFS PROVIDERi
Multiple PRIMARY CARE PHYSICIAN
Mutliple
PROVIDERi
The AQC explained
Source: Blue Cross Blue Shields
England’s payment reform
NHS England Resource allocation formula + quality premium
CCG
contract
SPECIALIZED CARE
contract
contract
contract
contract
GP
PROVIDERi
PROVIDERi
PROVIDERi
CCG: Clinical Commissioning Group
contract
PROVIDERi
48
Results (I) • Primary care providers’ role is strengthened in all models: – ‘Rostering’ patients within primary care practice seems to be a key element (AQC, CG, CCG) – Up-scaling the organizational structures of primary care
• But applied to different markets: – ACO + AQC: price-sensitive referral system delivery market – England + The Netherlands: use of clinical knowledge purchasing market
• Under ACOs, AQCs and CCG no real ‘transformation’ of the way providers are paid, while CGs made some steps towards capitated fees • Quality improvement tied to payment incentive in most models (CCG, ACOs, AQC)
49
Results (II) • Different approach to shift providers’ financial risks across services: – The Netherlands: narrow services package but full financial risks – Other models: Broad services package but no / moderate financial risks
• Huge impact of contextual factors: – Data information for providers – Local market structure: diversity in AQC contracts – Voluntary (CGs, ACOs and AQC) vs. mandatory (England)
‘health care is local’ 50
Financial risk across services 100%
Financial risk
Up side
ACO (one sided) 0%
services AQC
Down side 100%
CG
51
Policy Implications General • ‘Joy of the workforce’ is neglected within payment reforms • How to evaluate payment reforms? United States • How ACOs incentivize their providers which still are paid on Fee For Service is unclear The Netherlands • How to tie quality improvement to payment model?
52
Lessen van de AQC AQC
-
-
Shared savings contract (twee zijdig) FFS omgeving Shared saving afhankelijk van resultaat indicatoren
‘Key drivers’
‘Transferable Lessons’
‘Succesfactoren’
Randvoorwaarden (context afhankelijk en context onafhankelijk)
Bron: Ruwaard, et al. Transferring key drivers in provider-payer contracts: Lessons from the AQC (in preparation)
‘Succesfactoren’ Het AQC geeft de aanbieder een prikkel en de mogelijkheid om patiënten te sturen naar hoog-kwaliteit en betaalbare aanbieders • Prikkel om te sturen: – (1) Vormgeving bekostigingsmodel
• Mogelijkheid om te sturen: – (2) Verplichte verwijzing van vaste huisarts – (3) Data
• Inzet van verzekeraar Maar, - Het vervullen van deze 3 criteria in een andere setting hoeft niet automatisch tot soortgelijke succesen te leiden - Daarnaast zijn deze 3 criteria niet per se de enige manier om aanbieders een prikkel en de mogelijkheid te geven om te sturen
Lessen voor het ‘buitenland’: context-onafhankelijk 1. Het bekostigingsmodel werkt het beste op grote schaal: het succes van de hervorming neemt toe naarmate aanbieders soortgelijke prikkels ervaren 2. Haal eerst de ‘grote’ spelers binnen: deze aanbieders (1) kunnen het risico het makkelijkst dragen, (2) spelen een belangrijke rol in het reduceren van kosten, (3) en kunnen het moeilijkst zijn om te overtuigen 3. Start met genereuze en lange-termijn contracten: aanbieders moeten de middelen en tijd hebben om de hervorming door te voeren 4. Vooraf afgebakende populatie: Het ex ante definiëren van een populatie geeft een verhoogde prikkel om juist deze populatie proactief te managen
Lessen voor het buitenland: context-afhankelijk 5. Stel verwijzing verplicht: Als het niet mogelijk is om patiënten aan een vaste huisarts te koppelen, dan kan het verplicht stellen van een verwijzing van een huisarts een alternatieve oplossing bieden
6. De marktstructuur moet sturen toelaten: De structuur van de aanbiedersmarkt moet zo georganiseerd zijn dat er voldoende concurrerende aanbieders zijn die eenzelfde geografische markt delen
7. Beleid, wet- en regelgeving: regelgeving mag het introduceren van het bekostigingsmodel niet belemmeren
Discussie • Een soortgelijk bekostigingsmodel kan in andere settingen geïntroduceerd worden zolang aanbieders een ‘prikkel en de mogelijkheid hebben om te sturen’ • Contextuele lessen: Voldoet Nederland aan de randvoorwaarden? • Hoe sturen juridische entiteiten hun ‘downstream providers’? • Hoe duurzaam zijn de AQC resultaten? • Arbeidssatisfactie van zorgaanbieders onderbelicht
Conclusie • Juridische entiteiten die financiele risico’s op zich gaan nemen staan nog steeds in de kinderschoenen maar sterk in ontwikkeling • In alle modellen een beperkt risico voor zorgaanbieders maar bevatten wel prikkels om zinnige en gepaste zorg te bieden • ‘Uitkomstbekostiging’ nog ver weg
58
