Towards more comparable data to assess the performance of health systems: Past, present and future work at OECD Gaetan LAFORTUNE, OECD Health Division EUPHA Conference, Helsinki, 12 October 2007 1 1
Overview of presentation Why OECD work on health system performance? What do we mean by performance of health systems? What do we know about the performance of health systems across countries? 1. Health expenditure data (System of Health Accounts) 2. Health care quality indicators How to link health expenditure with outcome measures? What are the data gaps for comparative analysis of health system performance at the international level? 2
Why is OECD doing work on health? OECD countries share common policy goals Health care that is accessible and of high quality Health systems that are affordable, responsive & good value for money And common challenges Health is large and growing share of OECD economies Ensuring sustainable financing is more and more challenging Increasing demand to improve systems performance (quality of care, responsiveness, reduce access disparities) Remarkable cross-country variation System design, inputs, outputs, outcomes Opportunities to learn from experience and share best practices OECD comparative advantages Focus on issues key to developed countries and comparisons between European and non-european countries Apply expertise in health economics 3
OECD Health Data OECD has built up, over 20 years, one of the leading international databases on health & health systems Database contains more than 1000 statistical series, with some series going back to 1960 Based on collaboration with national data correspondents in 30 OECD countries Cooperation also with WHO and Eurostat User-friendly database released every year on CD- ROM and online versions Descriptive analysis of key indicators released every two years in Health at a Glance 4
What do we mean by performance of health (care) systems? Performance needs to be assessed against (common) policy objectives: - Maintaining/improving access to care - Maintaining/improving quality of care (outcomes) - Maintaining/improving responsiveness to patients needs and expectations - Controlling costs, ensuring sustainable financing - Increasing value for money (improving efficiency) 5
Health System Performance Framework (WHO, 2000) Average level Distribution Health gains/outcomes (+) a a Responsiveness and access (+) a a Financial contribution/health expenditure (-) a Efficiency a Equity Source: Adapted from World Health Report 2000 6
Health system performance framework (OECD, HCQI Project, 2006) Health Non-health care determinants of health Healthcare System Performance Current focus of HCQI Project Health care needs Staying healthy Getting better Effectiveness Quality Safety Responsiveness/ Patient centered Access Accessibility Expenditure Equity Living with illness or disability Coping with end of life Efficiency Health system design, policy and context 7
1. Improving the comparability of health expenditure data 8
System of Health Accounts OECD System of Health Accounts (SHA) published in 2000 Common International Classification for Health Accounts includes breakdown of expenditure data: - by functions (e.g., prevention and care) - by providers of care (ambulatory, hospitals, nursing homes) - by funding sources (government, social insurance, private insurance and out-of-pocket payments) SHA implementation facilitated through joint data collection effort with Eurostat and WHO Nearly all EU member states and OECD countries have started at least pilot implementation of SHA 9
Impact of SHA implementation on national health expenditure estimates Austria: pre-sha implementation: 7.5% of GDP (2003) post-sha implementation: 10.2% (2003) France: pre-sha implementation: 10.1% of GDP (2003) post-sha implementation: 10.9% (2003) Portugal: pre-sha implementation: 8.2% of GDP (1999) post-sha implementation: 8.8% (2000) Due mainly to redefining boundaries of health care (e.g. longterm care), identifying other (private) sources of funding, etc. 10
Further Development of Health Accounts (supported by DG Sanco) 1) Refine definition and methodology for data collection on long-term care 2) Improve comparability and availability of private health expenditure under joint SHA data collection 3) Develop methodology for estimating reliable healthspecific Purchasing Power Parities (to separate out price and volume effects on expenditure) 4) Develop methodology for estimating expenditure by disease, age and gender under SHA framework 11
2. Development of health care quality indicators 12
OECD Health Care Quality Indicators Project Goal: Fill current gap on indicators of health care quality at the international level: - key to assessing value for money from health systems - developing a balanced set of indicators, representative of main diseases in developed countries Approach: Build on efforts to develop quality of care indicators in Nordic countries and Anglosaxon countries Initial attention on technical quality of health care (including both process and outcome measures) Work funded by DG Sanco, US and other countries 13
Progress to date on Health Care Quality Indicators Analysis of available measures and consensus on an initial set of indicators Application of scientific criteria for indicator selection: Scientific soundness Clinical and policy importance Feasibility of measurement Results of first data collection published in March 2006 (OECD Health Working Paper No. 22) Update from second round of data collection to be released shortly (in OECD Health Working Paper and new chapter in OECD Health at a Glance) 14
Examples of Initial Set of Health Care Quality Indicators Good Process Measures Good/Bad Health Outcomes Vaccination rates (measles pertussis for children, flu/influenza for elderly) Disease incidence (measles, pertussis) Screening rates for cancers Survival rates from cancers Diabetes care (rates of retinal exams) 30-day mortality following heart attack or stroke 15
Next Steps on Health Care Quality Indicators - Five priority areas 1) Patient safety (in coordination with WHO World Alliance for patient safety and WHO-Europe, and EU Working Party on patient safety) 2) Mental health 3) Primary care and prevention 4) Cardiac care 5) Diabetes care 16
Measuring value for money: How to link health spending with outcomes? 17
Three possible approaches to link health spending with outcomes 1) System-level analysis (macro-type) 2) Disease-level analysis 3) Sub-sector level analysis (hospital, primary care) [Reference: OECD Economics Working Paper No. 554, by Hakkinen and Joumard, May 2007] 18
Current availability and comparability of data to use these three approaches Health system level Disease level Sub-sector level Expenditures Total health expenditure data available and increasingly comparable through SHA implementation Cost of illness studies in some countries (limited comparability) Starting methodological work to develop more comparable cost of illness estimates in SHA framework Health expenditure data increasingly available by type of providers in SHA framework (hospitals, ambulatory providers) Outcomes Life expectancy, premature mortality, infant mortality, etc. But these outcome measures more influenced by non-health care determinants (need to control for other determinants) HCQI project developing comparable outcome measures for different diseases (cancer survival rates, mortality rates following AMI or stroke, diabetes care, etc.) Mainly related to hospital care (inhospital mortality rate, adverse events following treatment, nosocomial infections, etc.) Less developed for ambulatory care 19
Life expectancy and health spending across OECD countries, 2005 Life expectancy, years 84 82 80 78 KOR JPN ISL CHE SWE ESP AUS ITA FRA CAN NZL IRL AUT NOR FIN DEU NLD PRT GBR BEL GRC DNK LUX R 2 = 0.56 USA 76 MEX CZE POL 74 SVK HUN 72 TUR 70 0 2000 4000 6000 8000 Health spending per capita, USD PPP Source: OECD Health Data 2007 20
Conclusion Comparative analysis of health system performance requires comparable data on health spending and quality of care (notably in terms of outcomes), at different levels Through effective collaboration between OECD, European Union and WHO, the availability and comparability of data on health spending based on SHA has improved greatly in recent years Through the OECD Health Care Quality Indicators project and other activities, some progress has also been made to develop and collect comparable measures on quality of care/outcomes Further effort is needed to better align measures of expenditure and outcomes at system-wide level, disease-level and subsector level, for health system performance measurement 21
OECD health-related websites Health Division: www.oecd.org/els/health OECD Health Data: www.oecd.org/health/healthdata System of Health Accounts: www.oecd.org/health/sha Health Care Quality Indicators: www.oecd.org/health/hcqi 22