integrating care by bundled payments
TRANSCRIPT
1 18 september 2012
Integrating care by bundled payments
Lessons from the Netherlands
JN Struijs
Implementing bundled payments in the Netherlands | 18 september 2012
2
Background
● Integrated care for diabetes is introduced in Netherlands since decades.
● The fragmentary funding hampered the establishment of long-term programs on a national level.
● In 2007 a bundled payment (BP) approach was introduced in the Netherlands to stimulate integrated care programs.
› 2007-2010: on a experimental basis
› 2010: structural implementation for dm type 2, COPD and Vascular Risk Management (VRM)
Implementing bundled payments in the Netherlands | 18 september 2012
3
Basic premises of the Bundled Payment (BP) model
• Comprehensive funding for one product
• Contents of the BP contract in conformity with Health Care
Standard (‘standard’/generic diabetes care)
• Health Care Standard describes activities (the ‘what’, not the
‘who’, ‘where’ and the ‘how’)
• Fees for BP contracts are freely negotiable
• Fees for subcontractors are freely negotiable
• Not simultaneously with a hospital-based ‘DBC
uncomplicated diabetes’
Implementing bundled payments in the Netherlands | 18 september 2012
4
‘Outline of BP model’
Insurance
companies
Care Group
GP SPEC LAB DIET
BP contract based on
Health Care Standard
PROVIDERi
Multidisciplinary protocol
contract
PN
contract contract contract employee
- Legal entity: Privat limited,
foundation, cooperative
- Ownership: GPs; some
case co-owners: GP lab,
hospital, home care
Implementing bundled payments in the Netherlands | 18 september 2012
5
Dutch Health care market
Insurer Patient/
consumer
Health care
providers
Health care
purchasing
market
Health care
delivery market
BP
Health care
insurance market
Implementing bundled payments in the Netherlands | 18 september 2012
6
Purchasing market superseded by two markets
Insurer Patient/
consumer
CG
Health care
purchasing
market 1
Health care
delivery market
GP LAB PROVn
Health care
insurance market
Health care
purchasing
market 2
Implementing bundled payments in the Netherlands | 18 september 2012
7
Effect BP on quality of care and health care costs
Implementing bundled payments in the Netherlands | 18 september 2012
8
Effect BP on quality of care
Health care delivery process
● Coordination more intensive and structured
● Additional training and education of care givers
● More attention for benchmarking
● IT hindering factor in most care groups, but BP is flying wheel for IT development
● Task delegation towards nurse practitioner (working in GP practice)
● Insulin dependent patient treated in Care Group instead of hospital
Process and outcome indicators
● Slight to modest improvements in most process and outcome indicators
Implementing bundled payments in the Netherlands | 18 september 2012
9
Year 3
BMI
92%
SBP
93%
Hba1c
91%
Creatinine
85%
LDL
84%
78%
BMI
92%
SBP
94%
Hba1c
91%
Creatinine
84%
LDL
83%
75%
Year 2
Composite process indicator: % patients who had all process measures checked in the last 12 months (N=23,088)
Implementing bundled payments in the Netherlands | 18 september 2012
10
T2 T1
SBP
<140 mmHg
46%
Hba1c <53 mmol/mol
66%
LDL cholesterol
<2.5 mmol/l
49%
24% 32%
25%
15%
SBP
<140 mmHg
48%
Hba1c <53 mmol/mol
66%
LDL cholesterol
<2.5 mmol/l
54%
27% 33%
27%
18%
Composite outcome indicator: % patients below target level on all indicators (N=5,623)
Implementing bundled payments in the Netherlands | 18 september 2012
11
T3
SBP
<140 mmHg
51%
Hba1c <53 mmol/mol
64%
LDL cholesterol
<2.5 mmol/l
59%
31% 34%
31%
21%
Composite outcome indicator: % patients below target level on all indicators (N=5,623)
Implementing bundled payments in the Netherlands | 18 september 2012
12
Effects BP on curative health care costs
● Objective: Insight in the effects of integrated care and BP on
curative health care costs
● One disease: DM
● Distinction between effects implementation of Integrated Care and implementation of BP
– Integrated care = efforts to enhance quality and continuity of care. Payment method stands apart
– BP= payment method for integrated care
Distinction possible: biggest insurer in the NL is not convinced of the BP approach and still contracts on the basis of a care management fee
Implementing bundled payments in the Netherlands | 18 september 2012
13
‘Nationwide implementation of care groups’
Method
• Data: Nationwide claim data from Vektis
• Curative health care costs: primary care, medicine and hospital-based specialist care
• Study population:
• DM2 Patients in care program (paid by BP or MF) or Care as Usual (CAU).
• All patients were during study period (2008-2009) in the same payment system (N=64,139)
• Dependent variable: difference in costs 2008–2009
• Adjustments: 2008 baseline costs, age, gender, comorbidity
• Analyses: multivariate regression analyses (diff-in-diff analyses)
• multilevel modelling (2-level): GP-patients, insurer-patients
Implementing bundled payments in the Netherlands | 18 september 2012
14
Difference in curative health care costs 2008-2009 (€) adjusted for 2008 baseline costs, age, gender, comorbidity
Implementing bundled payments in the Netherlands | 18 september 2012
15
Model 1 SE
Intercept -882 *** 46
2008 baseline costs -0,7 *** 0
Payment system BP 288 *** 55
MF -74 63
CAU ref.
Age (centered)# 17 *** 2
Gender (ref= female) 296 *** 48
Comorbidity 1,813 *** 36
#= age is 67.8
Implementing bundled payments in the Netherlands | 18 september 2012
16 16
Hospital use of diabetes patients enrolled in a care program (paid via BP or MF) (OR; 95%CI) (ref.= routine care patients) #
0,831
0,583 0,588
0,724
0
0,1
0,2
0,3
0,4
0,5
0,6
0,7
0,8
0,9
1
Totaal Diabetes dbc Diabetes dbc
poliklinisch
Diabetes dbc klinische
opname
# adjusted for age (centered), gender, comorbidity and 2008 hospital utilization
0,831
0,583 0,588
0,724
0
0,1
0,2
0,3
0,4
0,5
0,6
0,7
0,8
0,9
1
Total hospital use Diabetes related
hospital use
Diabetes related
outpatient visits
diabetes related
inpatient utilization
Implementing bundled payments in the Netherlands | 18 september 2012
17
0,748
0,875
0
0,1
0,2
0,3
0,4
0,5
0,6
0,7
0,8
0,9
1
IB KT
Difference in 2009 hospital use of diabetes patients enrolled in a care
program (paid via BP or MF) (OR; 95%CI) (ref.= routine care patients) #
# adjusted for hospital utilization 2008, age, gender, comorbidity
Implementing bundled payments in the Netherlands | 18 september 2012
18 Integrale bekostiging | 22 augustus 2012 18
Difference in costs of hospital-based specialist care 2008-2009 for patients in BP and MF-group
-36
142
-35
-128 -150
-100
-50
0
50
100
150
200
Diabetes related Total
BP
MF
Implementing bundled payments in the Netherlands | 18 september 2012
19
Conclusions
● Nationwide implementation of care groups
● The organization and process of care improved
● Slight to modest increase in process and outcomes indicators; however effects difficult to interpret due to IT and transparency problems
● Less patients enrolled in a care program on the basis of BP used hospital care
● BP resulted in an increase of curative health care costs which is mostly to an increase of hospital care costs and the initial investment costs of the BP model
Implementing bundled payments in the Netherlands | 18 september 2012
20
Discussion
● Conflict of interest of GPs (general contractor + subcontractor)
● Risk for market power of Care Groups (no competition)
● Comorbidity vs. single-disease care programs
● Neglect of developments on purchasing market 2, while behaviour of care providers are influenced by the way they are remunerated on purchasing market 2
● Questionable whether modifications to purchasing market are possible without interfering the health insurance and health care delivery market
● Experiments with global payments and shared savings are underway (2013-2017)
Implementing bundled payments in the Netherlands | 18 september 2012
21
Thank you for your attention
M: +31 6 46312583
Implementing bundled payments in the Netherlands | 18 september 2012
22
Further reading ● Dinny H de Bakker, Jeroen N Struijs, Caroline A Baan, Joop Raams, Hubertus J.M. Vrijhoef, Frederik T.
Schut. Early Results From Adoption Of Bundled Payment For Diabetes Care In The Netherlands Show Improvement In Care Coordination. Health Affairs, 31, no.2 (2012):426-433.
• Jeroen N. Struijs. and Caroline A. Baan. Integrating Care through Bundled Payments — Lessons from the Netherlands. NEJM (2011)364;11
● JN Struijs, SM Mohnen, CCM Molema, JT de Jong-van Til, CA Baan. Effects of bundled payment on curative health care costs in the Netherlands. An analysis for diabetes care and vascular risk management based on nationwide claim data, 2007-2010. RIVM Report 260013002. Bilthoven 2012.
available: October 2012
● JN Struijs, JT de Jong–van Til, LC Lemmens, HW Drewes, SR de Bruin, CA Baan. Three years of bundled payment for diabetes care in the Netherlands. Impact on health care delivery process and the quality of care. RIVM Report 260013001. Bilthoven 2012.
available: October 2012
● Struijs JN, van Til J, Baan CA. Experimenting with a bundled payment system for diabetes care in the Netherlands. The first tangible effects. RIVM Report 260224002. Bilthoven 2010. http://www.rivm.nl/dsresource?objectid=rivmp:11840&type=org&disposition=inline
Integrale bekostiging | 22 augustus 2012 23
‘Care management fee’
Insurancies
companies
Care group
GP internist Lab Dietician
Care management fee
(Coordination, IT
Education and further
training, etc.)
Care provideri
contract
PN
contract contract contract contract