Das Gesundheitssystem als komplexe Intervention methodische Überlegungen zur Evaluation und ausgewählte Ergebnisse

Size: px

Start display at page:

Download "Das Gesundheitssystem als komplexe Intervention methodische Überlegungen zur Evaluation und ausgewählte Ergebnisse"

Kathleen Morris
6 years ago
Views:

Ergebnisse Reinhard Busse, Prof. Dr. med.

Universität Berlin (WHO Collaborating Centre for Health

1 Das Gesundheitssystem als komplexe Intervention methodische Überlegungen zur Evaluation und ausgewählte Ergebnisse Reinhard Busse, Prof. Dr. med. MPH FFPH FG Management im Gesundheitswesen, Technische Universität Berlin (WHO Collaborating Centre for Health Systems Research and Management) & European Observatory on Health Systems and Policies

3 Dummerweise kennt das Gesundheitssystem gar keine Interventions- und Kontrollgruppe meta-komplex

4 4,000,000 patient-physicians contacts in ambulatory care 2,300,000 packs of OTC drugs purchased 2,000,000 packs of prescription drugs dispensed 400,000 inpatients in hospital Wie metakomplex, zeigt diese Abbildung zu TÄGLICHEN Aktivitäten im deutschen Gesundheitssystem (erscheint demnächst im Lancet) 60,000 50,000 16,000 15,000 60,000 emergency department visits 50,000 hospital admissions 16,000 CT-scans in ambulatory care 15,000 CT-scans in inpatient care 700 stroke patients admitted 600 hip-implant operations

5 Erweiterung 1: breites Verständnis von Technologien und konsequente Aufarbeitung für deren Evidenz Gesundheitssystem-bezogene Interventionen Research Synthesis Appraisal Decision Utilization Evaluation Organisatorische Interventionen Research Synthesis Appraisal Decision Utilization Evaluation Bevölkerungsbezogene Interventionen Research Synthesis Appraisal Decision Utilization Evaluation Klinische Interventionen/ Verfahren Research Synthesis Appraisal Decision Utilization Evaluation Technologien i.e.s. (Arzneimittel, Medizinprodukte...) Research Synthesis Appraisal Systematische Reviews und HTA in der «Wissen-Kette» Decision Utilization Evaluation Vortrag Auf dem Weg zu einem evidenzbasierten Gesundheitssystem, IQWiG

6 unabhängig davon, wie wir jede Ebene nennen Gesundheitssystem EbGS Gesundheitssystem-bezogene Interventionen Gesundheitspolitik Research Synthesis Appraisal EbGP Decision Utilization Evaluation Organisatorische Interventionen Management Research Synthesis im Gesundheitswesen Appraisal Decision Utilization EbMiG Evaluation Bevölkerungsbezogene Interventionen Research Synthesis Appraisal Public health EbPH Decision Utilization Evaluation Klinische Interventionen/ Verfahren Technologien i.e.s. (Arzneimittel, Medizinprodukte...) Research Research Synthesis Synthesis Appraisal Klinische Medizin EbM Appraisal Systematische Reviews und HTA in der «Wissen-Kette» Decision Decision Utilization Utilization Evaluation Evaluation Vortrag Auf dem Weg zu einem evidenzbasierten Gesundheitssystem, IQWiG

7 und wer zuständig ist Gesundheitssystem EbGS Gesundheitssystem-bezogene Interventionen Research Synthesis Appraisal Decision Utilization Evaluation Organisatorische Interventionen Research Synthesis IQWiG? Appraisal Decision Utilization Evaluation Bevölkerungsbezogene Interventionen Research Synthesis Appraisal Decision Utilization Evaluation Klinische Interventionen/ Verfahren Technologien i.e.s. (Arzneimittel, Medizinprodukte...) Research Research Synthesis HTA IQWiG Synthesis Appraisal Appraisal G-BA Systematische Reviews und HTA in der «Wissen-Kette» Decision Decision Utilization Utilization Evaluation IQTiG Evaluation Vortrag Auf dem Weg zu einem evidenzbasierten Gesundheitssystem, IQWiG

Erweiterung 2: Erweiterung der akzeptablen Evidenz (und Berücksichtigung der Rückkopplung) Published literature + Grey literature + study data & responsiveness

8 Erweiterung 2: Erweiterung der akzeptablen Evidenz (und Berücksichtigung der Rückkopplung) Published literature + Grey literature + study data & responsiveness & financial protection & equity & efficiency Performance assessment as evidence Vortrag Auf dem Weg zu einem evidenzbasierten Gesundheitssystem, IQWiG

The starting point: 2000 World Health Report First attempt to rank performance of 191 national health systems Aimed at identifying and measuring performance of member

9 The starting point: 2000 World Health Report First attempt to rank performance of 191 national health systems Aimed at identifying and measuring performance of member states on key health system objectives Examined whether each health system is performing as well as it can, given existing resources Based on Murray & Frank framework (2000)

10 WHO Framework: strategy behind World Health Report 2000

11 Further development at WHO (2007): building blocks and intermediate goals/ outcomes SYSTEM BUILDING BLOCKS OVERALL GOALS / OUTCOMES SERVICE DELIVERY (DEL) HEALTH WORKFORCE (HW) ACCESS COVERAGE IMPROVED HEALTH (LEVEL AND EQUITY) INFORMATION RESPONSIVENESS MEDICAL PRODUCTS (MP), VACCINES & TECHNOLOGIES FINANCING (FIN) LEADERSHIP / GOVERNANCE (GOV) QUALITY SAFETY SOCIAL AND FINANCIAL RISK PROTECTION IMPROVED EFFICIENCY Source: World Health Organization (WHO) (2007) Everybody s business: Strengthening health systems to improve health outcomes. WHO s framework for action. Geneva: WHO Document Production Services.

12 Inspired by OECD, the European Commission s Joint Assessment Framework A Q R

13 High performing? Difficulties in deciding what to measure and how to operationalize it Another problem The health of nations MODERN medicine may be good at gauging the health of patients, but it has proved less successful at taking its own pulse. Assessing the performance of a country s health-care system is no easy task, because deciding what to include from doctors to drugs to diet is difficult, and because some chosen criteria, from infant mortality to patient satisfaction, are themselves hard to define. Making comparisons between countries is even trickier, because health-care systems differ radically in their financing and organisation, and in the social goals they set out to achieve. June 22, 2000

14 My combined performance framework (incl. costs/ efficiency and relationship to WHO dimensions) Access(ability) incl. Financial protection Quality x (for those who = receive services) Population health outcomes (system-wide effectiveness, level & distribution) Responsiveness (level & distribution) Inputs (money and/or resources) Health system performance (Allocative) Efficiency (value for money, i.e. population health and/ or responsiveness per input unit)

15 My combined performance framework (incl. costs/ efficiency and relationship to WHO dimensions) Access(ability) incl. Financial protection Quality x (for those who = receive services) Population health outcomes (system-wide effectiveness, level & distribution) Responsiveness (level & distribution) Inputs (money and/or resources) Health system performance (Allocative) Efficiency (value for money, i.e. population health and/ or responsiveness per input unit)

16 The framework (without costs/ efficiency) Population-/ systemwide performance dimensions Access(ability) incl. Financial protection Quality x = (for those who receive services) Population health outcomes (system-wide effectiveness, level & distribution) Responsiveness (level & distribution) Both population health outcomes and responsiveness are the multiplicative effect of accessability and quality: high accessability but low quality as well as low accessability but high quality lead, on the population level, to inferior performance (but pointing to the problem is important for deciding on reform need)

) coverage (financial issues) availability of care waiting,

17 The access(ability) component Need (by socio-economic status, ethnicity/ migration status etc.) coverage (financial issues) availability of care waiting, acceptability etc. Unmet need Realised access Unmet need x Quality= Outcomes (populationhealth& responsiveness)

18 The first Coverage Cube was born 10 years ago

19 picked up by WHO only a year later

20 and again in 2010

21 1 st dimension/ population coverage: the importance is known usually by U.S. data; here: access problems in 2012 for U.S. adults x Experienced costrelated access problem Serious problems/ unable to pay medical bills Uninsured Insured all year Spent $1,000 or more out-of-pocket % Source: 2013 Commonwealth Fund International Health Policy Survey in Eleven Countries.

22 2 nd dimension/ covered benefits also matter: e.g. gaps in dental care x % Did not visit dentist/hygenist/dental clinic in past two years Skipped dental care because of cost in past year GER SWE NOR NETH SWIZ CAN UK FR US AUS NZ Covered in basic package Complementary coverage high Not covered Own elaboration based on data from 2013 Commonwealth Fund International Health Policy Survey in Eleven Countries.

23 3 rd dimension/ cost-sharing: size and protection mechanisms are important x % 60 Experienced cost-related access problem* Cap for cost-sharing Cost-sharing uncapped Spent US$1,000 or more out-ofpocket * Did not fill/skipped prescription, did not visit doctor with medical problem, and/or did not get recommended care. Source: modified from 2013 Commonwealth Fund International Health Policy Survey in Eleven Countries.

24 Usually a combination of 2 nd & 3 rd dimension: public expenditure for pharmaceuticals (in comparison to health services) x x

25 availability of care Urban-rural discrepancies are vary drastically between countries with definite scope to learn from another x3.2 x1.3 x1.3 x2.5 x1.3 x1.0

26 waiting, acceptability etc. 70 Waiting (here: >4 weeks for a specialist appointment) is a general problem, but some countries see improvements and others not Own elaboration, data: Commonwealth Fund International Health Policy surveys,

27 16,0 14,0 12, ,9 11,3 12,5 10, ,0 7,0 7,8 6,0 4,0 2,0 0,0 0,2 0,5 0,6 1,1 1,5 1,6 2,1 2,1 2,8 3,5 3,6 Unmet need Unmetneedin EU-27 (for costs, distance, waiting), Own elaboration, data: EU-SILC, 2015 Unmet need

28 30,0 25,8 25,0 20,0 Highest income quintile Fünftes Quintil des Äquivalenzeinkommens Average Gesamt Lowest income quintile Erstes Quintil des Äquivalenzeinkommens 17,3 15,0 14,4 11,0 12,4 10,0 6,6 7,2 6,4 5,0 0,0 0,4 1,6 1,7 2,5 2,5 3,0 2,0 3,8 Unmet need Unmet need in EU-27 by income quintiles (for costs, distance, waiting), 2014 Own elaboration, data: EU-SILC, 2015 Unmet need

29 x1.2 x3.7 x2.7 x2.5 x1.5 x2.3 x3.3 x1.8 x1.5 Unmet need Unmetneedin selectedcountries (for cost reasons), by income level, 2013 Unmet need

with poor seeing GPs and rich seeing specialists

30 Inequity of physician visits by income (and equal need) in many countries and a real problem in certain ones with poor seeing GPs and rich seeing specialists Realised access

31 The condensed OECD report card for Access: does it really tell us much?

32 Access(ability) incl. Financial protection Health-service only performance dimensions Quality (for those who receive services): x = Q1. Effectiveness Q2. Safety Q3. Patient experience Population health outcomes (system-wide effectiveness, level & distribution) Responsiveness (level & distribution)

33 Responsiveness / patientexperience Time to relookat Responsiveness an expanded version to the original WHO concept (our Responsiveness in ambulatory care project) Respect for persons Dignity Autonomy/ Participation Confidentiality Clear communication added: Trust Client/ patient orientation Choice of provider Prompt attention Quality of basic amenities Access to social support added: Coordination

34 Responsiveness / patientexperience Autonomy/ participation Clear communication

35 Ambulatory caresensitive conditions/ avoidable hospitalisations Inpatient mortality

36 Inpatient AMI mortality? +0.4% +3.8% +0.9% +3.8% and taking 30 days follow-up into account

37 Extending the time horizon to 5 years for cancer patients

38 The condensed OECD report card for Quality: does it really tell us much? 38

39 Access(ability) incl. Financial protection Quality x (for those who = receive services) The area with the least agreement but highest political relevance Population health outcomes (system-wide effectiveness, level & distribution) Responsiveness (level & distribution) Inputs (money and/or resources) (Allocative) Efficiency (value for money, i.e. population health and/ or responsiveness per input unit)

40 How can we calculate the health system contribution to health? Environment Lifestyle Mortality/ (healthy) life expecancy Socio-economic status/ education etc. Health care = Avoidable mortality (amenable to health care) Medical errors

41 The concept of avoidable mortality (AVM; also amenable to health care ) Deaths from certain causes that should not occur in the presence of timely and effective health care Introduced by David Rutstein in the 1970s (originally for quality assurance purposes) Walter Holland published European Community Atlas of Avoidable Deaths in 1988; intends to provide warning signals of potential shortcomings in health care delivery Mackenbach et al. argue that associations between AVM and health care services are rather weak and inconsistent. Most health care measures only reflect quantity and not quality. Many studies use insufficient set of covariates. Nolte and McKee (2002) reviewed list of amenable causes of death

42 Amenable mortality, all persons, 0-74 Age-standardized rates per 100,000 Now to the cost-effectiveness of the complex intervention Decrease in avoidable mortality per persons aged 0-74, /14: Austria -54 (-42%) Denmark -55 (-40%) France -30 (-33%) Germany -49 (-37%) Netherlands -50 (-41%) United Kingdom -60 (-41%) Canada -32 (-29%) Total health expenditure, US$ PPP, per capita Calculations by Observatory and author, unpublished Austria Denmark France Germany Netherlands United Kingdom Canada

43 Incremental cost-effectiveness (death rate decrease per $1000 spent more): Amenable mortality, all persons, 0-74 Age-standardized rates per 100, Austria 25 Denmark 25 France 17 Germany 19 Netherlands 16 United Kingdom 30 Canada Total health expenditure, US$ PPP, per capita Calculations by Observatory and author, unpublished Austria Denmark France Germany Netherlands United Kingdom Canada

44 and including the U.S. 170 Amenable mortality, all persons, 0-74 Age-standardized rates per 100, / persons (-25%) -8/ $1000 spent more Total health expenditure, US$ PPP, per capita Austria Denmark France Germany Netherlands United Kingdom Canada United States Calculations by Observatory and author, unpublished

45 Zusammenfassend, das Gesundheitssystem als ganzes zu evaluieren erfordert (1) eine Erweiterung der Ergebnisdimensionen und (2) Nutzung auch von anderen Studientypen(keine Kontrollgruppe); es gibt eine gute Grundlage und auch Datenquellen; vermutlich wird sich das IQWiG aber auch zukünftig nicht mit Performance assessment des Gesamtsystems befassen. Warum trotzdem dieser Vortrag beim IQWiG? Verständnis für andere Bestandteile der Evidencebased -Familie fördern & Lücken/ Überschneidungen/ Widersprüche abbauen

Multinational Comparisons of Health Systems Data, 2016

Attachment F Multinational Comparisons of Health Systems Data, 216 Dana Sarnak The Commonwealth Fund Health Care Spending per Capita, 198 214 Adjusted for Differences in Cost of Living Dollars ($US) 9,