Efficiency in health care
Download
Report
Transcript Efficiency in health care
Efficiency in Health Care Cardiac Care in Sweden
Rosita Claesson Wigand
Project manager (MSc Economics, MPH)
National Board of Health and Welfare (NBHW)
Department of Statistics and Evaluation
Unit for Performance Assessment
Telephone : + 46 75 247 37 17
E-mail: [email protected]
Nordic Case Mix Conference , Helsinki, June 4, 2010
National Strategy for Good Care
Nationellt
Systematic reviews of
evidence-based knowledge
Monitoring and evaluation at
national level: transparent
and open comparisons
Monitoring and evaluation at
local and regional levels
Prerequisites
• Regulation on
systems for
quality and patient safety
• Patient Data Act
• IT-based documentation
• Health Data Registries
• National Information
Structure
• National Terminology
• National Indicators
Good Care
Regulations, National guidelines
and priorities
Steering and systematic
work of improvement at
local and regional levels
National Guidelines
• Evidence based reviews:
effectiveness and cost
effectiveness
• Recommendations
– priority 1 - 10
– ”not to do”
– R&D
• Indicators
• Expected consequences of
implementing the guideline
(economic and
organisational)
• Different versions
Performance Assessment Framework - Good Care
Effectiveness
Safety
Patientcenteredness
Timeliness
Equity
Costs of Care
Quality of Care
Efficiency
Efficiency
Efficiency
Costs
Processes
Productivity
Goal/
Desired Outcome
Outcome
Objective
to assess quality and
analyse efficiency of
cardiac care in Sweden
Methodology
45 process and outcome
indicators
Costs (estimations)
Data sources:
• six quality registers
• patient register
• prescribed drug register
• cause of death register
Data sources:
• DRGs
• cost per patient database
• prescribed drug register
• statistics, Swedish
Association of Local
Authorities and Regions
(SALAR)
Data from 2007 or earlier.
Assessment of Quality
45 process and outcome indicators:
• General indicators (2)
• Care and treatment at hospitals – coronary artery disease (7)
• Pharmaceutical treatment - coronary artery disease (12)
• Mortality and readmission - coronary artery disease (9)
• Other - coronary artery disease (2)
• Rhythm disorders (5)
• Heart failure (5)
• Valvular heart diseases and
congential heart diseases (3)
Results - Quality
Per indicator:
• background (choice of indicator,
importance)
• comments on results:
development over time,
geographical and gender
differences
• biases and interpretation issues
Figures depict data:
• over time
• country council/regional
comparisons
•hospital comparisons
Analysis of Efficiency
• How much cardiac care is consumed?
• What are the costs of cardiac care?
• Co-variation between costs and quality of
cardiac care?
• What are the costs of not choosing costeffective interventions/treatments? What are
the costs of non-compliance with the National
Guidelines? (Inefficient use of resources)
Consumption of cardiac care can
only partly be described
• In-patient care (38/50 DRGs in MCD 5 defined
as cardiac care: no of cases, total DRGweight, number of hospital days, ALOS)
• Out-patient visits in specialised hospital care
(physicians + surgery)
• Primary health care?
In-patient: Cases och total weight
Intervention-based categories:
No of cases
Total DRG weight
Medical
187 205
35 787
Invasive
24 944
41 817
Surgical
7 875
108 57
Others
5 413
3 800
Source: DRG/Patient Register
In-patient: cases och total weight
Diseases in National Guidelines:
cases, % total weight,%
Coronary artery
46,2
44,5
Rhythm disorders
26,6
19,1
Heart failure
21,4
21,4
Valvular & congenital 2,9
9,7
heart diseases
Other types
2,9
5,3
Total
100,0
100,0
Average Length of Stay
2007
3,7 days
2002
4,1 days
2002-2007,%
-9,8%
Note!
Measure hospital days per ward!
Says nothing about the ALOS per episode!
Organisational aspects have changed over time!
Costs of cardiac care varied, but
data on costs limited
County council agreed accounting - functional levels:
• primary health care
• specialised somatic care
• specialised psychiatric care
NOT possible to monitor costs of cardiac care or any
other type of care for specific group of patients
Cost Per Patient (KPP) for some hospitals!
Costs per case per hospital, KPP database
Costs per case, Insertion or replacement of pacemaker (DRG 115B), 2007
SEK/case
=> Estimated Costs per Case, Cardiac Care and
Acute Myocardial Infarction (AMI) Care
Costs per case, cardiac care and AMI care, 2007
cardiac care
AMI care
National
SEK/case
Co-variation between costs and process
quality? This is what we would want to do at hospital level if
data on costs.......
Costs per case (AMI care) and RIKS-HIA Quality Index (cardiac intensive care) ,
county councils with KPP, 2007
Score RIKS-HIA Quality Index, 2007
Regression line
SEK/case
Co-variation between costs and outcome
quality? What we would like to do if....
Costs per case, AMI care and Percentage of readmissions within 30 days,
county councils with KPP, 2007
Percent
Regression line
SEK/case
Additional costs due to non-compliance
with National Guidelines
Patients with AMI 1998-20007 treated with low cost statins or high cost statins
January- June 2008
Low cost statins
High cost statins
Possible prescription of low cost statins
Percent
Additional costs due to inefficient
use of resources
Costs of statins for patients with AMI 1998-2007
Low cost statins
High cost statins
Percent
Efficiency in cardiac care?
• differences in quality exists
• differences in costs exists
• no obvious co-variation between costs and
quality? But...........
• could use resources more efficiently for
instance by prescribing low-cost (generic)
drugs
Efficiency potential.............
.........but which hospital/county council is
the most efficient?
Based on existing data it is not possible
to determine which hospital/county
council has the most efficient cardiac
care!
Further developmental work
• Data on specialised out-patient care (not only
physicians) and primary health care
• Costs per patient/case (based on unique
patient identifier!)
• National health data registers and quality
registers: coverage, outcome indicators
(quality of life/health), case-mix
Thanks a million!
More information:
www.socialstyrelsen.se