EUROPEAN COMMISSION EUROSTAT Directorate F: Social statistics Unit F-5: Education, health and social protection DOC 2013-PH-06 Annex 6D Towards a possible Out of Pocket (OOP) expenditure Indicator at macro-level based on the System of Health Accounts (SHA) for the Joint Assessment Framework (JAF) on Health Item 6.2.6 of the Agenda Joint Document Eurostat DG EMPL Meeting of the Public Health Statistics Working Group Luxembourg, 2-3 December 2013 Bech Building Room Quetelet
TOWARDS A POSSIBLE OUT OF POCKET (OOP) EXPENDITURE INDICATOR AT MACRO-LEVEL BASED ON THE SYSTEM OF HEALTH ACCOUNTS (SHA) FOR THE JOINT ASSESSMENT FRAMEWORK (JAF) ON HEALTH 1. INTRODUCTION The Joint Assessment Framework (JAF) was developed and agreed in 2010 by Member States and the Commission to provide a transparent framework for tracking progress in relation to Europe 2020 headline targets on employment and poverty and identify country-specific challenges in the areas of employment and social policy in the context of the European Semester. Therefore, developing a JAF for health would complement the assessment of MS' social policies and enable transparent identification of countries' challenges and progress in that area in the context of the European Semester process. Assessment of the country's performance in the area of health should be in line with the agreed principles of the social Open Method of Coordination (OMC) that highlight the importance of access, quality and sustainability of health care systems. The model proposed below thus takes account of the OMC principles. As it has been done in other areas of social policy, the Indicators Sub-Group (ISG) of the Social Protection Committee (SPC) proposed to develop an assessment framework based on the JAF methodology also in the area of health and use and complement the existing set of indicators on health underpinning the three goals of access, quality and sustainability agreed upon in the context of the Open Method of Coordination (OMC) on social protection and social inclusion 1. This evidence-based approach offers a possibility for analysis of a single Member State but also for a comparative analysis across Member States and with respect to the EU average. 2. ISG APPROACH TO MEASURING ACCESS TO HEALTH Access to good quality healthcare services is a prerequisite for social integration and inclusive growth. Achieving access to adequate health care and long-term care for all is embedded in the Open Method of Coordination for social protection and social inclusion, along with other health-related objectives, i.e. tackling health inequalities, with a further dedicated strategy in this field. 2 The right of access to preventive health care and the right to benefit from medical treatment are guaranteed in the EU s Charter of Fundamental Rights. It is also prominently listed as one of the four common values and principles of EU health systems, along with universality, solidarity and equity. Furthermore, ensuring people have easy access to primary care of good quality may generate savings and is likely to enhance efficiency by preventing ill health and avoiding use of more expensive services in secondary care. 3 1 http://ec.europa.eu/social/blobservlet?docid=3882&langid=en 2 http://ec.europa.eu/health/ph_determinants/socio_economics/documents/com2009_en.pdf 3 (WHO Health in crisis report) 2
The World Health Organization (WHO) defines accessibility as a measure of the proportion of the population that reaches appropriate health services. An equitable health service, therefore, is one where individuals access to and utilisation of the service depends on their health state alone. 4 When talking about equity of access in financial terms, one should consider, not only who is covered, but also, what services are covered and to what extent these are subject to out-of-pocket payments. This is illustrated in the WHO "cube" in the figure below. Dimensions of health coverage Source: WHO (2010) Organisational barriers, such as waiting lists or limited surgery opening hours also have a relatively greater impact on people with low incomes. If waiting lists are long, these people usually lack the means to turn to alternative providers in the private sector. People in blue collar jobs and/or working in shifts may have less flexibility to attend surgery hours. When they feel their job is at risk they may delay seeking care. Geographical barriers are especially relevant to older people and those with limited mobility. Such barriers may be exacerbated in rural areas, where poverty risk also tends to be higher. To capture the dimension of access, the ISG discussed the following set of indicators: - "self-reported unmet need for medical care" as well as "self-reported unmet need for dental care", descrbe barriers resulting from financial, geographical and waiting time issues. The gap between the top and bottom quintiles in the reported unmet needs for medical/dental care was considered as a good indicator of access to health despite the cultural/structural differences that may exist between. Member States. 4 Eurohealthnet (2012) Re-orienting Health Care Systems: Towards modern, responsive and sustainable health promoting systems, EuroHealthNet: Brussels 3
When analysing the information on reported unmet need for medical care, the indicator on care utilisation, defined as use of primary, secondary and out-patient care should be looked at as context information. Data for this indicator is available from EU-SILC offering good coverage across the EU with the information being updated on annual basis. - share of the population covered by health insurance Health baskets offered within the scope of public insurance programmes are fairly comprehensive, but people in vulnerable situations may still miss out on certain services. Dental care, physiotherapy and certain mental health services are often excluded from basic packages. However, taking out additional insurance cover may be a financial step too far for people on lower incomes, thereby severely restricting their access to such service. This indicator includes coverage by both public and private insurance. However, it comes with its shortcomings - range of services covered and the degree of cost sharing can vary across countries. Levels of health insurance coverage determine the extent to which people are protected from the financial consequences of ill health (financial protection) and have access to needed services. - an indicator on out-of-pocket payment has been proposed for complementing the access dimension. Groups at risk of poverty and/or social exclusion are disproportionately affected by the financial burden of the impact of cost-sharing arrangements. This can impact negatively on the uptake of necessary services. In some countries, special arrangements exist to protect people on lower incomes for the relatively high costs incurred through exemptions. However, it should also be acknowledged that some of the out-of-pocket expenditure occurs because people choose to (rather than have to) pay for private services (perhaps wanting higher quality), therefore it may not always be associated with payments by those on lower incomes. Overall, this contributes to an improvement of health outcomes at population level. Therefore, two possible indicators on out-of-pocket payment are proposed to complete the access dimension. An indicator at the micro-level addressing the financial burden for health care at the household level which is not possible to define based on current data availability but is highlighted for future development. An indicator at the macro-level on aggregate out-of-pocket expenditure which can be built based on existing data from the System of Health Accounts and which can serve as context information. Consequently, Eurostat was asked to contribute towards a possible Out-of-Pocket (OOP) expenditure Indicator based on the System of Health Accounts (SHA). 3. EUROSTAT S PREPARATORY WORK 3.1. 10 April 2013: ISG meeting On 10 April 2013, Eurostat presented the joint Eurostat-OECD-WHO data collection on health accounts based on the System of Health accounts (SHA) methodology. More specifically Eurostat presented a short overview of the methodological framework and the main data collected for out-of-pocket (OOP) expenditure. The main methodological and practical difficulties in this area of statistics were rapidly evoked (which could lead to some comparability problems), as 4
well as the geographical coverage of available data (some missing countries). A first discussion on the possibility to develop an indicator took place. A preference for a micro-level indicator was mentioned, because this would allow the identification of risk-categories in the population. However, the possibility of developing an indicator on an aggregate level of OOP expenditure was discussed. Concerning the relevance of such an indicator, questions were raised about its usability to measure the (non-)accessibility to health care as opposed to other phenomena: health system financing indicator, background/context indicator, sustainability indicator, etc. Suggestions were given to explore the break down by functions or by providers. However the comparability of expenditure on pharmaceutical products was questioned, since the HC.5 Medical goods function in SHA2011 covers only the expenditures for medical goods where the function and mode of provision are not specified (e.g. medical goods that are part of inpatient, outpatient and day care consumption are not usually identified separately). Some possible denominators for the indicator were mentioned (e.g. total health expenditure (including capital expenditure), private expenditure, etc.). The ISG concluded inviting Eurostat to further explore the possibilities of using SHA to build an OOP expenditure macro indicator for the next discussion at the ISG which took place on the 26 th of June 2013. 3.2. 26 June 2013: ISG meeting The working document that was prepared by Eurostat for the 26 th June 2013 ISG meeting is provided in DOC 2013-PH-06 Annex 6E. During the meeting the floor was opened for any first reactions based on the draft paper circulated to delegates. While few delegates expressed no objection to retain a macro indicator on this issue as delivered by the WHO, a number of delegations raised concerns related to: whether moving from the micro to the macro level is a solution as comparison between countries would be very difficult and the necessary detail on the financing burden on households still won't be there; the difference between OOP and general private source of financing with the suggestion of the paper showing how the OOP are part of the general private source of financing; the fact that the SHA-based OOP indicator does not include reimbursements ex-post so this can lead to inaccurate figures; 5
the fact that in some countries OOP are mainly done by well-off households and are related to getting faster and quicker health care services, thus, making it difficult to establish the link between an aggregate OPP measure and the situation of the most vulnerable; After the meeting, Malta commented that, in their opinion, the indicator Unmet need for medical and/or dental examination or treatment due to financial barriers in SILC is a better indicator for measuring the financial barriers in accessing health than the indicator based on the SHA OOP expenditure. The Commission highlighted that health has become much more prominent in the European Semester and the country-specific recommendations and analysis shows that OOP represents a fairly big share in some lower income countries, suggesting that figures based on SHA could be useful in identifying possible areas for further investigation. 3.3. 24 th September 2013: Task Force on health care expenditure statistics The scope of the discussion that took place during the Task Force meeting was to explore the feasibility and contribute to the development of a possible OOP) expenditure indicator based on countries experience. The assessment of the feasibility of such an indicator should refer to three quality criteria: Accuracy: although clear steps have been moved forward with the introduction of SHA2011, OOP expenditure still remains one of the most difficult indicators to compile; Comparability between countries: difficulties in compiling an accurate OOP expenditure may result in reduced comparability over space; Comparability over time: countries have gradually followed more closely the SHA methodology; yet a number of breaks in the timeseries are still observed. First, Eurostat informed TF participants about the work that has been done so far in collaboration with the ISG and invited TF delegates to provide their feedback on data quality issues related to OOP measurement, as well as to provide any comments that would feed the discussion on an OOP indicator in the ISG meeting that was scheduled on the next day (25th September 2013). Discussion Task Force delegates expressed their satisfaction about the overall quality of the study submitted by Eurostat to the ISG. They noted that, indeed, OOP expenditure has attracted a lot of attention lately and that a lot of people working in this area try to address the challenges associated to its estimation by combining data at the macro level with data at the micro level. They 6
expressed their belief that OOP expenditure is a very important indicator in spite of the shortcomings of the sources that are used for its computation. However, they drew attention to the possible misinterpretation that the share of this private expenditure may have. They highlighted the fact that SHA has brought a significant progress in the calculation of OOP expenditure, since, now, cost-sharing can be clearly separated from direct OOP expenditure. However, they expressed the belief that the macro dimension of OOP expenditure may not be enough to measure health accessibility and, moreover, they noted that private expenditure cannot be measured easily and accurately. On top of this, there could be problems of comparability between countries. Further work on meta-information on a country by country basis can be proven very beneficial if resources are available. France outlined that there are many limitations associated with the use of OOP as an indicator, even more than those already included in Eurostat s study. They noted that, in France, there has already been huge reflection on this issue. They informed TF participants about a number of studies that France has produced in this area. There was a general agreement that SHA data on OOP should not be used to measure heath care accessibility. However, they could be used as a background/context statistics for other more specific indicators. Eurostat commented that in addition to indicators on unmet needs for medical care based on SILC special attention should be paid to the ongoing efforts to collect additional information on health care accessibility issues in the framework of Modernisation of Social Surveys. 3.4. 25 th September 2013: ISG meeting Eurostat presented the issues for the development of an indicator for measuring health care accessibility based on Households OOP) expenditure. First, the main characteristics of OOP were demonstrated and the advancements brought by SHA 2011 were explained. Details were also provided for the boundaries of OOP expenditure, as well as for its relation to cost-sharing. Special focus was put on the difficulties in measuring private health expenditure, which are ascribed to the inherent limitations of surveys. Possible future improvements with regard to the geographical coverage and the quality and comparability of data were also presented, as a result of the forthcoming Commission Regulation that will be based on SHA 2011. Eurostat invited the ISG to provide more detailed information on what the indicator to be built on OOP expenditure is exactly meant to measure. Based on this information, a number of decisions have to be taken: (a) aggregation level of the phenomenon: total OOP expenditure, breakdown by function and/or by health care provider and (b) appropriate denominators for an OOP indicator (total health expenditure, current health expenditure, household 7
disposable income, final households consumption). Last, Eurostat informed the ISG about the Task Force (TF) meeting that took place on the 24 th of September 2013 and presented the opinion of TF delegates on the possible use of an indicator based on OOP to measure health care accessibility. France asked whether capital formation is also measured by SHA. Belgium noted that, despite the inherent limitations of SHA data related to OOP, the work of ISG should be the initiative to put more attention on OOP measurement. In the meantime, the SHA data could be used for over-time comparison (NAT indicator vs. EU indicator). Germany requested some further clarifications on the interpretation of an OOP indicator. Eurostat informed the ISG that the draft Commission Regulation includes only current health expenditure. However capital formation is included in the new SHA 2011-based Joint Questionnaire. Eurostat outlined the fact that, during the TF on the 24 th of September 2013, there was a general agreement that there are serious concerns with the interpretation of OOP and a possible JAF-indicator based on OOP should not be used on its own as such. A more appropriate use of such statistics would be as a background/context information. Eurostat agreed with Belgium that further emphasis should be put on the compilation of more accurate data for OOP; concerning the use of existing data for NAT-indicator, Eurostat proposed to consult national statisticians. ISG concluded that, overall, delegates were supportive of pursuing in-depth work on the potential of SHA for producing such an indicator, also in view of the development of an assessment framework on health. ISG noted that, in some countries, special arrangements exist to protect people on lower incomes for the relatively high costs incurred through exemptions. However, it should also be acknowledged that some of the outof-pocket expenditure occurs because people choose to (rather than have to) pay for private services (perhaps wanting higher quality), therefore it may not always be associated with payments by those on lower incomes. Overall, this contributes to an improvement of health outcomes at population level. Therefore, two possible indicators on out-of-pocket payment were proposed by ISG to complete the access dimension. A micro-level dimension addressing the financial burden for health care at the household level which is not possible to define based on current data availability but is highlighted for future development. A macro-level dimension on aggregate out-ofpocket expenditure which can be built based on existing data on health expenditure and which can serve as a context information to be defined. 8
The Working Group is asked to: take notice of policy needs for building an indicator measuring (non-) accessibility to health care take notice of the actions taken by Eurostat in cooperation with the Task Force on health care expenditure statistics towards the possible implementation of a possible indicator based on SHA OOP give their opinion on the use of SHA-based OOP expenditure as context/background statistics in the context of JAF-health give their opinion on the use of the currently available OOP statistics for time-series analysis at individual country level (NAT indicators) give their opinion on the possibility to dedicate some resources in 2014 to the improvement of the quality of SHA OOP statistics. 9