SHA-Based Health Accounts in Twelve Asia-Pacific Economies : A Comparative analysis. Hyoung-sun Jeong and

Transcription

1 SHA Technical Paper(2010)1 SHA-Based Health Accounts in Twelve Asia-Pacific Economies : A Comparative analysis Hyoung-sun Jeong and Ravi P. Rannan-Eliya 10 OECD/Korea Policy Centre - Health and Social Policy Programme SHA TECHNICAL PAPERS

2 Unclassified OECD/Korea Policy Centre - Heatlh and Social Policy Programme SHA TECHNICAL PAPERS(2010)1 20-January-2010 English text only OECD/Korea Policy Centre Health and Social Policy Programme : TECHNICAL PAPERS NO. 10 SHA-BASED HEALTH ACCOUNTS IN TWELVE ASIA/PACIFIC ECONOMIES : A Comparative analysis Hyoung-sun Jeong and Ravi P. Rannan-Eliya JEL Classification : I10, H51 1

3 OECD/KOREA Policy Centre Health and Social Policy Programme SHA TECHNICAL PAPERS This series is designed to make available to a wider readership Health and Social Policy studies with a focus on the Asia/Pacific region. The papers are generally available only in English, and principal authors are named. The opinions expressed and arguments employed here are the responsibility of the author(s) and do not reflect those of the OECD, the Korean authorities, or the OECD/Korea Policy Centre Health and Social Policy Programme. Applications for permission to reproduce or translate all or part of this material should be made to: Director of the OECD/Korea Policy Centre - Health and Social Policy Programme 87 Hoegiro Dongdaemun Gu, Seoul, Korea Copyright OECD/Korea Policy Centre - Health and Social Policy Programme

4 ACKNOWLEDGEMENT The authors would like to express their great thanks to all experts from participating economies who carried out a project to present the initial results from the implementation of the System of Health Accounts. The results are presented in this comparative study and in a set of Joint OECD /Korea RCHSP SHA Technical Papers and OECD/Korea Policy Centre - Health and Social Policy Programme SHA-TECHNICAL PAPERS, presenting individual economy studies. The authors of the economy studies were: Ghulam Rabbani and Najmul Hossain (Bangladesh), Jui-fen Rachel Lu (Chinese Taipei), Keith YK Tin and Gabriel M Leung (Hong Kong SAR), Batbayar Chuluuzagd and Chimeddagva Dashzeveg (Mongolia), Hyoung-Sun Jeong (Korea), Kanjana Tisayaticom, Walaiporn Patcharanarumol, Viroj Tangcharoensathien, Artidtaya Tiampriwan and Hathaichanok Sumalee (Thailand), Tharanga Fernando, Ravi P. Rannan-Eliya and JMH Jayasundara (Sri Lanka), Yuxin Zhao (China) and Zailan Adnan (Malaysia). 3

5 SUMMARY 1. This comparative analysis covers twelve economies, providing a reasonably good picture of different health funding and spending patterns across the Asia-Pacific region. It includes four members of the OECD (Australia, Japan, Korea and New Zealand), two other high income economies (Hong Kong SAR and Chinese Taipei), as well as a number of relatively low income economies, such as Bangladesh, Mongolia, Sri Lanka and Thailand with GDP per capita less than USD 3,000. Health expenditure by financing agent 2. A wide variation in overall health spending can be observed ranging from Australia with a Total Health Expenditure (THE) per capita of 3,234 USD PPP to Bangladesh with 37 USD PPP. The level of an economy s wealth devoted to health varied from 3.2% in Bangladesh up to 9.9% in the case of New Zealand. The differences in health spending across economies are far greater than for economic performance (per capita GDP), but they show a logarithmic relationship indicating that health expenditures are income-elastic above a certain income level. The figures also seem to suggest that above a certain level of economic development, other factors (i.e., budgetary and health policies, as well as history and current structure of the health systems) exert important influences on the share of GDP spent on health care. 3. The public share of THE ranged from 28% in Bangladesh up to 81% in Japan. While in seven economies including New Zealand, Japan, Mongolia, Australia, Thailand, Korea and Chinese Taipei the majority share of health spending comes from public funds, in five economies including Bangladesh, China Tianjin, Malaysia, Hong Kong SAR and Sri Lanka from private funds. It is interesting to note that the difference between economies with the highest and the lowest per capita public health expenditure (in USD PPP) is greater than the range observed in the case of THE. In general, the public share of THE tends to increase with per capita GDP. 4. In all of the economies observed, private household out-of-pocket payments formed the largest part of private funding sources (except Malaysia). In the case of Bangladesh, 63% of THE comes from households out-of-pocket payments. Of the other economies, the share varies from lows of 14% and 15% in New Zealand and Japan up to 43% and 50% in Sri Lanka and China Tianjin respectively. Household out-of-pocket payments equated to an average of 302 USD PPP, ranging from 18 USD PPP in Mongolia to 635 USD PPP in Hong Kong SAR. With the exception of Korea, the more developed economies tend to fund a greater share of their private health expenditure from private, than the less developed economies. Health expenditure by function 5. Distribution of expenditure by functions in terms of the purpose of health care is as follows. Curative and rehabilitative care accounts for 60% of THE on average across the economies studied - ranging from 30% in Bangladesh to 78% in Thailand. Long-term 4

6 care ranged from almost 0% in Malaysia, China Tianjin and Mongolia to Japan and New Zealand at the other end of the scale, with 16% and 14% of THE respectively. However, these statistics suffer from significant problems in comparability, with many economies attempting to measure these expenditures currently. Medical goods (pharmaceuticals and therapeutic appliances) amounted to 19% (14% and 2% respectively) of THE, although with a wide variation ranging from 4% in Thailand to 44% in Bangladesh. In per capita terms spending on medical goods ranged from 10 USD PPP in Thailand to 552 USD PPP in Australia. It is interesting to note that differences in per capita spending on pharmaceuticals/medical goods are far smaller than differences in THE. 6. Data on mode of production show that inpatient curative-rehabilitative care occupies a smaller share of health expenditure than is usually supposed ranging between 15% of current health expenditure (CHE) in Bangladesh up to 56% Mongolia. Out-patient care varies from 11% of CHE in China Tianjin up to a high of 51% in Chinese Taipei. Day care and home care are assumed to be developing in many of the less-developed economies, but health statistics are only now starting to register these services in many of these. Health expenditure by provider 7. In all economies, apart from Malaysia, hospitals account for the highest proportion of CHE among the health care providers. Almost half of CHE is directed towards hospitals - varying from 30% in Bangladesh up to a high of 72% in Thailand, while around 24% providers of ambulatory health care - ranging from 5% in China Tianjin up to 43% in Malaysia. The share of retail sale and other providers of medical goods obviously reflect the trend of medical goods. Health expenditure by function and financing agent 8. In financing inpatient care, public funds are the dominant source contributing, on average, 71% of the costs, leaving the private sector to fund the remaining 29%. The contribution varies from China Tianjin where 41% of inpatient services are financed through public funds to Mongolia and Japan at almost 90%. On average across the economies public finances continue to fund the greater share in out-patient care at around 55%, but in some economies including Bangladesh, Hong Kong SAR, Sri Lanka, Malaysia, China Tianjin and Korea, private financing tends to fulfil the greater role. Again Mongolia is an exception amongst the lower-income economies, owing to high level of overall public financing. The majority of the economies except Japan, New Zealand and Korea report private financing more than half of medical goods expenditure. At the same time, in all economies the role of private funding is more important in financing medical goods than in paying for inpatient and out-patient care. 9. Curative and rehabilitative inpatient care tends to account for higher share (45%) of current public expenditure than is the case for total expenditure. Out-patient care accounts for 28% of current public expenditure on health ranging from 13% in China Tianjin to 55% in Chinese Taipei. As to private funding, curative and rehabilitative inpatient care accounts for 22% of current private expenditure; while curative and rehabilitative out-patient care ranges from 10% in China Tianjin to 50% in Hong Kong SAR. A major difference between public and private health spending is with respect to medical goods: around 33% of current private expenditure on average directed to medical goods as opposed to 9% of public funds. 5

7 10. On average 38% of households spending on health pays for medical goods, 37% for out-patient care and 20% for inpatient care. plays a considerable role only in Malaysia and Hong Kong SAR where it finances inpatient care to a greater extent than pharmaceuticals. Comparing the functional structure of social security and private in economies with the social health scheme, an important difference is that the share of administration (HC. 7) is far higher in private spending. In Japan and Korea, private spent 16% and 14% on administration, respectively, while social security funds (HF.1.2) spent 3%. Health expenditure by function and provider 11. The cross-classification of health expenditure by function and provider is the feature of the SHA that is expected to reflect structural changes in the way health care services are delivered. However it requires longer time series to be able to exploit this opportunity. The tables in this study show a considerable difference in the share of different functions provided by hospitals. For example, inpatient care represents about 44% of hospital expenditure in China Tianjin, while it represents 85% in Mongolia. Out-patient care accounts for 13% of hospital expenditure in Mongolia, while almost 50% is devoted to out-patient care in Chinese Taipei. 12. The share by hospitals ranges from 19% of out-patient expenditures in Australia to Thailand, where 79% of out-patient expenditure is provided by hospitals, respectively. Health expenditure by provider and financing agent 13. Hospitals are the main recipients of general government health financing on average accounting for 65% of public health expenditure (from 42% and 43% in Korea and New Zealand respectively to 82% in Bangladesh). From the providers perspective: although general government plays the major financing role in hospitals expenditure in most economies, the variations are notable. In New Zealand, Japan, Mongolia and Australia between 80% and 90% of hospital spending is financed by public health funding, while at the lower end, the share of public funding is between 47 and 68% in China Tianjin, Malaysia, Korea, Thailand and Bangladesh. 14. On average 16% of public health funds are directed to ambulatory health care providers varying from 0% in Bangladesh, 4% in China Tianjin and 5% in Thailand up to 36% in Australia and 28% in Japan and Chinese Taipei, where doctors clinics play a considerable role in inpatient care as well as out-patient care. From the providers perspective, the private sector takes on a greater share of the funding of ambulatory health care providers in seven counties, while public funds provide the greater share of funding in five economies. 15. The share of public health finance directed to providers of medical goods is relatively small (6% on average). Korea has the highest public health finance directed towards providers of medical goods among 12 economies (22%). China Tianjin, Japan and Australia report 11-12%. In all other economies, public health finance is rarely allocated to retailers of medical goods. 16. As a matter of fact, the analysis in this paper has its limitations due to methodological and data problems as yet unresolved. Despite this, SHA-based health accounts provide a far more reliable and detailed insight into the way money is spent in the 6

8 health care sector in the Asia-Pacific region than has been previously available. Moreover, the comparative tables allow researchers and experts to conduct a more detailed or specific analysis. Current data will provide a basis for a comparison over time, when longer time series become available for more economies. 7

9 ABBREVIATIONS CHE GDP ICHA ICHA-HC ICHA-HF ICHA-HP NHA NPISH OECD ROW SHA THE WHO Current Health Expenditure Gross Domestic Product International Classification for Health Accounts ICHA classification of health care functions ICHA classification of sources of funding ICHA classification of health care providers National Health Accounts Non-Profit Institutions Serving Households Organization for Economic Co-operation and Development Rest of the World System of Health Accounts Total Health Expenditure World Health Organization 8

10 TABLE OF CONTENTS ACKNOWLEDGEMENTS... 3 SUMMARY... 4 ABBREVIATIONS... 8 CHAPER Ⅰ. INTRODUCTION CHAPTER Ⅱ. EXPERIENCE AND CHALLENGE IN IMPLEMENTING SHA-BASED HEALTH ACCOUNTS IN THE ASIA-PACIFIC REGION APPLICATION OF SHA-BASED BOUNDARIES TO NATIONAL HEALTH EXPENDITURES APPLICATION OF THE SHA CLASSIFICATIONS CHAPTER Ⅲ. A COMPARATIVE ANALYSIS OF HEALTH EXPENDITURES ACROSS 12 ECONOMIES HEALTH EXPENDITURE BY FINANCING AGENT HEALTH EXPENDITURE BY FUNCTION HEALTH EXPENDITURE BY MODE OF PRODUCTION HEALTH EXPENDITURE BY PROVIDER HEALTH EXPENDITURE BY FUNCTION AND FINANCING AGENT HEALTH EXPENDITURE BY FUNCTION AND PROVIDER HEALTH EXPENDITURE BY PROVIDER AND FINANCING AGENT CHAPTER Ⅳ. CONCLUSIONS BIBLIOGRAPHY ANNEX ANNEX COMPARATIVE TABLES

11 Tables Table 1. Status of SHA-based health accounts in the economies covered by this publication Table 2: Status of SHA-based health accounts in selected other Asia-Pacific economies Table 3: Sub-categories of total health expenditure Table 4: Main differences between the estimates of total expenditure as presented in SHA-based health accounts (THE) and as reported in official statistics (NHE) Table 5. Total health expenditure per capita, by financing agent (US$ PPP per capita) Table 6. Total health expenditure by financing agent (% of GDP) Table 7. Total health expenditure by financing agent share of THE Table 8. Total health expenditure by function share of THE Table 9. Total health expenditure per capita by function (US$ PPP per capita) Table 10. Current health expenditure per capita by mode of production (US$ PPP per capita) Table 11. Current health expenditure by mode of production share of CHE Table 12. Current health expenditure per capita by provider (US$ PPP per capita) Table 13. Current health expenditure by provider share of CHE Table 14. Expenditure on personal health services by financing agent (Current expenditure on personal health services = 100%) Table 15. Expenditure on curative-rehabilitative care by financing agent (Current expenditure on C&R care = 100%) Table 16. Expenditure on long-term care by financing agent (Current expenditure on LTC = 100%) Table 17. Expenditure on inpatient care by financing agent (Current expenditure on inpatient care = 100%) Table 18. Expenditure on out-patient care by financing agent (Current expenditure on out-patient care = 100%) Table 19. Expenditure on medical goods by financing agent (Current expenditure on medical goods = 100%) Table 20. Expenditure on pharmaceuticals by financing agent (Current expenditure on pharmaceuticals = 100%) Table 21. Expenditure on prevention and public health by financing agent (Current expenditure on prevention and public health = 100%) Table 22. Total public and private health expenditure by function (Total public/private health expenditure = 100%) Table 23. Current public and private health expenditure by mode of production (Current public/private health expenditure = 100%) Table 24. Current government/social security health expenditure by mode of production (Current government/social security health expenditure = 100%) Table 25. personal health services expenditure by mode of production ( personal health services expenditure = 100%) Table 26. Current private health expenditure by mode of production (Current private health expenditure = 100%)

12 Table 27. Households' health expenditure by mode of production (Current households' health expenditure = 100%) Table 28. Current hospitals' expenditure by mode of production (Current expenditure on hospitals' service = 100%) Table 29. Expenditure on inpatient care by provider (Current inpatient expenditure = 100%) Table 30. Expenditure on out-patient care by provider (Current out-patient expenditure = 100%) Table 31. Expenditure on ancillary services by provider (Current expenditure on ancillary services = 100%) Table 32. Current expenditure on hospitals' services by financing agent (Current expenditure on hospitals=100%) Table 33. Current expenditure on nursing and residential facilities by financing agent (Current expenditure on ambulatory health care providers = 100%) Table 34. Current expenditure on ambulatory health care providers by financing agent (Current expenditure on ambulatory health care providers = 100%) Table 35. Current public and private health expenditure by provider (Current public/private health expenditure = 100%) Table 36. Current government and social security health expenditure by provider (Current government/social security health expenditure = 100%) Table 37. Current households' health expenditure by provider (Current households' health expenditure = 100%) Figures Figure 1. Total health expenditure, as per capita US$ PPP and % of GDP Figure 2. Relationship between the log of per capita expenditure on health (US$ PPP) and the log of per capita GDP (US$ PPP) Figure 3. Share of total health expenditure by financing agent Figure 4. Share of total health expenditure by function of care Figure 5. Share of expenditure on personal health care services by mode of production Figure 6. Share of current health expenditure by provider Figure 7. Share of expenditure on in-patient care by financing agent Figure 8. Share of expenditure on out-patient care by financing agent Figure 9. Share of expenditure on pharmaceuticals by financing agent Figure 10. Share of expenditure on prevention and public health by financing agent Figure 11. Share of current public health expenditure by mode of production Figure 12. Share of current private health expenditure by mode of production Figure 13. Share of households out-of-pocket expenditure by mode of production Figure 14. Share of current hospitals' expenditure by mode of production Figure 15. Share of expenditure on inpatient care by provider Figure 16. Share of expenditure on inpatient care by provider

13 Figure 17. Share of current expenditure on hospitals' services by financing agent Figure 18. Share of current expenditure on ambulatory health care providers by financing agent Figure 19. Share of current public health expenditure by provider Figure 20. Share of current households' health expenditure by provider

14 CHAPTERⅠ. INTRODUCTION 17. Economic and social development and changes in health systems in the Asia-Pacific region have led to increasing challenges for health policy and increasing questions about the performance of health systems. How much are economies spending on health? What are health expenditures being spent on? How are expenditures and the financing mix changing? What is the impact of reforms? These are common questions, to which have been added new ones about how spending compares across economies. How does the level of expenditures compare between economies? How much does the spending pattern vary across economies? These other questions require that data on health expenditures be not only reliable, but also comparable. 18. In 2000, the OECD published the System of Health Accounts (SHA) to provide a foundation for producing health expenditure statistics that would be comparable across economies and across time. This first version of the SHA version 1.0 was developed principally for use in OECD economies, but comments and feedback from health accounts experts in the Asia-Pacific region were also sought and incorporated during its development. Subsequently, WHO endorsed the use of the SHA by all economies as the appropriate standard for international reporting of health accounts statistics, in its Guide to Producing National Health Accounts with special applications for low-income and middle income economies (World Health Organization, 2003). Box 1. What is the System of Health Accounts? The System of Health Accounts (SHA) proposed an integrated system of comprehensive and internationally comparable accounts and provides a uniform framework of basic accounting rules and a set of standard tables for reporting health expenditure data. Boundaries of the health system are defined in a wider sense by the SHA, including long-term nursing care. It provides an International Classification for Health Accounts (ICHA) that in its 1.0 version - covers three dimensions: - health care functions (ICHA-HC); - health care service provider industries (ICHA-HP); - sources of funding health care (ICHA-HF). Standard SHA tables cross-classify expenditures under the three basic classifications providing new and deeper analytic possibilities of how services are financed and provided. The SHA allows for the incorporation of further dimensions of health expenditure into national health accounts: for example, regions, age and gender groups, and disease categories, in order to more adequately answer the question of Who gets what, where, and how? Source: Orosz and Morgan (2004) 13

15 19. Since the publication of the SHA, OECD and Eurostat member economies have widely adopted the SHA as the basis for regular reporting of health accounts statistics. This has led to expansion in the availability of comparable health expenditure data for these economies, as first documented by (Orosz and Morgan, 2004), with significant benefits in terms of the comparative health systems analysis that is now facilitated. The use of the SHA has also enhanced the comparability of health expenditure statistics that are reported annually in OECD Health Data. 20. In the late 1990s, health accounts experts in both developed and developing Asia- Pacific economies had also identified the need to collaborate in improving comparability of health expenditure statistics in the region. During , WHO funded the Asia-Pacific NHA Network (APNHAN), which brought together these experts, to look at options for developing a regional NHA standard that would help in improving comparability of health accounts statistics in the region. This effort eventually concluded that adoption of the new SHA as a basis for regional comparison would be a better strategy than attempting to develop a separate regional standard. 21. This consensus to recommend the SHA standard as a basis for comparison even by non-oecd economies in the region was partly in recognition of the rapid take-up of the SHA by Asia-Pacific economies. In fact, the first two economies to publish health accounts using the SHA standard were Korea, an OECD economy, and Sri Lanka, a developing non-oecd economy. Since then, there has been substantial take-up of SHA-based health accounts in Asia-Pacific region, by both developed and developing economies. In several economies, systems based on the SHA have been established for routine reporting of national health accounts (NHA) estimates, whilst in others pilot studies have been conducted to test and demonstrate the potential for SHA-based NHAs. In others that had existing NHA systems, adaptations have been made to these in order to facilitate reporting using the SHA framework. 22. In 2005, the OECD Korea Centre and health accounts experts from the Asia-Pacific region, meeting in Seoul, agreed to collaborate in the compilation of regional health accounts statistics, and in documentation of SHA implementations in the region (Joint OECD/Korea Regional Centre on Health and Social Policy, 2006). Following this, two parallel activities were launched. The first involved documenting the SHA implementations in different economies using a standard descriptive template, which is based on the one that was used for documenting the pilot SHA implementations in OECD economies. The OECD Korea Centre has published these as SHA Technical Papers for Bangladesh, Chinese Taipei, Hong Kong SAR, Mongolia, Korea, Thailand, Sri Lanka, China and Malaysia. These complement the similar country papers that have been published by OECD for Australia, Japan and Korea in its Health Technical Papers Series. The second activity has involved a collaborative effort to compile health accounts estimates for regional economies, with the eventual objective of establishing a regional health accounts database. This activity has involved collaborations between the OECD Korea Centre, the Asia-Pacific NHA Network (APNHAN) and WHO, and has used a SHA-based questionnaire to compile health accounts statistics. Two economies have participated in the latter exercise, which have not yet been covered by the OECD Korea Centre SHA Technical Paper Series; these are Tonga and Viet Nam. 23. As a consequence of these developments, comparable health expenditure estimates are now available in the SHA format for several Asia-Pacific economies. This paper provides an assessment of the first examples of these, and brings together for comparison the health expenditure data reported for these economies. The paper chiefly makes use of the SHA Technical Papers published by the OECD Korea Centre, as well as data collected for the 14

16 regional health accounts database by OECD Korea Centre and APNHAN. Table 2 presents the current status of SHA in the economies covered by this publication, and the relationship of the SHA estimates reviewed to existing national health accounts statistics for the same economies. 24. The SHA estimates reviewed in this report are for the most recent years reported by the economies. These comprise a range of years between 2002 and 2007, and so they are not always for the same year. Although the ideal situation would be to make comparisons using data for the same year, this is not currently feasible or realistic given that SHA adoption is still in the developmental stage in most economies in the region. Nevertheless, the small difference in timing in the estimates being used do not detract from the value of making this initial set of comparisons, because in practice the changes in both the levels and composition of health expenditures over 3 to 5 years in most economies tends to be minimal. 25. SHA-based health accounts have also been developed in several other Asia-Pacific economies. These are not covered in this publication owing to the unavailability of systematic documentations or sufficient data. Table 2 provides details of SHA implementation in some of these other economies not covered by this publication. Table 1. Status of SHA-based health accounts in the economies covered by this publication Australia* Bangladesh China Chinese Taipei Hong Kong SAR Japan* Korea* Malaysia Mongolia Status of institutionalization of SHA-based health accounts Institutionalized as basis for annual reporting Institutionalized as basis for national system Institutionalized as basis for annual reporting Pilot SHA study completed, with efforts ongoing to establish routine system Institutionalized as basis for annual reporting Institutionalized as basis for annual reporting Institutionalized as basis for annual reporting Institutionalized as basis for annual reporting Adopted as basis for national system as part of pilot project. SHA tables available as of July Full SHA tables only available for China Tianjin for New Zealand* Pilot SHA study completed Sri Lanka Institutionalized as basis for annual reporting Thailand Institutionalized as basis for national system * OECD member countries: OECD Health Data 2009 (June version) 15

17 Table 2. Status of SHA-based health accounts in selected other Asia-Pacific economies Federated States of Micronesia Fiji India Philippines Samoa Tonga* Vanuatu Viet Nam* Status of institutionalization of SHA-based health accounts First NHA estimates have used SHA as basis, but institutionalization ongoing First NHA estimates have used SHA as basis, but institutionalization ongoing NHA system institutionalized, but full compliance with SHA being implemented NHA based on local statistical framework, but pilot project has mapped NHA estimates to SHA tables Institutionalized, with regular updating Institutionalized, with regular updating First NHA estimates have used SHA as basis, but institutionalization ongoing Institutionalized using SHA framework as basis SHA tables available as of July Full SHA tables not available, but incomplete ones available Pilot estimates not yet published * Tonga and Viet Nam have participated in the reporting of SHA statistics to the regional health accounts collections, and so are referred to occasionally in this paper. 26. Following the approach of the earlier review of the first SHA implementations in the OECD region (Orosz and Morgan, 2004), this paper is organized in two parts: - Part I presents an assessment of the overall experience and comparability of implementation of the SHA framework in the economies reviewed, including a summary of the challenges faced. - Part II presents a comparative overview of the health expenditures reported, and a cross-economy comparison of the expenditure patterns revealed. 27. The related OECD Korea Centre SHA Technical Papers and the OECD Health Technical Papers (for Australia, Japan and Korea) contain the 12 individual studies that were used for preparation of this paper, and further details of the data sources and methodologies utilized. Additional information was also gathered from the SHA questionnaires of economies that participated in the various rounds of the regional SHA collections. 16

18 CHAPTER Ⅱ. EXPERIENCE AND CHALLENGE IN IMPLEMENTING SHA-BASED HEALTH ACCOUNTS IN THE ASIA- PACIFIC REGION APPLICATION OF SHA-BASED BOUNDARIES TO NATIONAL HEALTH EXPENDITURES 28. The major goal of developing the SHA is to ensure the comparability of health expenditure estimates reported by different economies. This requires that economies use a common boundary in defining the health systems and that there is consensus on what the elements of a health system are. The SHA achieves this by providing a consistent functional approach to define the boundaries of the health system. This approach is functional in that it refers to the goals and purposes of health care such as disease prevention, health promotion, treatment, rehabilitation and long-term care. The SHA requires the measurement and inclusion of expenditures spent on these functions regardless whether their providers are considered as health care organisations or institutions outside the health sector in national statistics. Table 3 presents the main functional components and the sub-totals of health expenditure as defined in the SHA. Table 3. Sub-categories of total health expenditure ICHA Code HC.1; HC.2 HC.3 HC.4 HC.5 HC.1-HC.5 HC.6 HC.7 HC.1-HC.7 HC.R.1 HC.1-HC.7 Description Services of curative and rehabilitative care (inpatient care, day-cases, out-patient and home care) Services of long-term nursing care (inpatient, day-cases and home care) Ancillary services to health care Medical goods dispensed to out-patients Total expenditure on personal health services and goods Services of prevention and public health Health administration and health Total current expenditure Investment (gross capital formation) in health TOTAL HEALTH EXPENDITURE 29. In order to implement the boundaries of health care and develop comprehensive and internationally comparable data on total expenditure, the SHA needs the following requirements to be fulfilled: (i) (ii) (iii) (iv) The functional classification of health care (ICHA-HC) is applied in an internationally harmonised way; Expenditure by all the financing agents defined by the SHA is accounted for; All primary and secondary providers of health care are included regardless of whether they are classified as health care institutions in national industry statistics or not. Furthermore, providers health, health-related and non-health expenditure are distinguished (and the latter two are excluded); Foreign trade of health services is estimated; 17

19 (v) Common methods for valuation of health services are applied following the SHA framework. 30. The review of the first SHA implementations in OECD economies assessed the extent to which these were able to adhere to these requirements (Orosz and Morgan, 2004). It found that whilst the use of the SHA did significantly improve comparability of estimates of Total Health Expenditures (hereinafter, THE) in relevant OECD economies, adherence to the full SHA requirements for comparability was not achieved in most economies. Improved comparability of health expenditure statistics was also achieved in the first SHA implementations in the Asia-Pacific region, but adherence also has not been complete. The question is then to what extent have the first implementations of SHA in the Asia-Pacific region adhered to these requirements? and to what extent does the pattern of adherence differ from that observed in the OECD region? The following issues are addressed: (i) the differences between THE and estimates of total health expenditure presented in national statistics; and (ii) application of the SHA classifications. Differences between existing and SHA estimates of total health spending 31. Adoption of the SHA standard has substantially improved the comparability of overall estimates of national health expenditure (NHE) that are reported by economies (Table 4). Several economies, such as Sri Lanka and Mongolia, have developed their first health accounts systems explicitly using the SHA framework, and in these the SHA-defined THE is used as the aggregate to report total health spending. In some which had older systems, such as Bangladesh, Hong Kong SAR and Thailand, official definitions have been changed to adopt the THE concept when reporting overall health spending. In Korea, NHE estimates produced in the existing NHE reporting system was replaced by the THE aggregate mapped to SHA format. In a few others, a different NHE aggregate continues to be reported locally, whilst the SHA THE aggregate is reported for international comparison, e.g., Australia and Japan. Table 4. Main differences between the estimates of total expenditure as presented in SHA-based health accounts (THE) and as reported in official statistics (NHE) Australia Bangladesh China Chinese Taipei Hong Kong SAR Differences NHE includes all the health and health-related functional classifications, except HC.R.2 Education and training of health personnel. NHE previously included education and research expenditures in the national total. China first initiated work developing its own NHA system and methodology in the early 1980, and the NHA system has since evolved to dual OECD SHA compatible classifications from Currently, China has completed nationwide spending estimates classified by financing source and by provider for 1990 through 2006 while health accounts with all three dimensional classifications are only available for Tianjin municipality for the year NHE excludes payments for private health, and includes cash benefits paid to patients by government agencies. Total domestic health expenditure (TDHE) previously included expenditures on education and training, research, environmental 18

20 Japan Korea Malaysia Mongolia New Zealand Sri Lanka Thailand health, and administration of cash benefits, but its definition has been revised to be compatible with the SHA definition of THE. NHE (based on so called National Medical Expenditure) excludes services not covered by public health and services financed by long-term care. 1 THE in the SHA estimates eliminated double counting under the item of private health in the case of previous versions of NHE, which are not used any longer. NHE includes Education and training (HC.R.2) No difference, as health accounts designed to be SHA-compliant. No differences reported between SHA and NHE estimates. Gross fixed capital formation (HC.R.1) is not measured in both estimates. No difference, as health accounts designed to be SHA-compliant. NHE previously included some water and sanitation expenditures. Compliance of SHA-based total health expenditure (THE) figures with the SHA definitions 32. Although all the economies covered are using THE as the overall indicator of aggregate health spending, problems in comparability still remain, owing to differences in the completeness of coverage of the estimates. As in the OECD, regional economies have faced challenges in identifying new data sources and methods to measure all the expenditure components that must be included when reporting THE in the SHA framework. The three major areas where gaps in coverage are observed are in the reporting of expenditures on longterm nursing care (LTC), and the inclusion of expenditures by non-profit institutions and by companies. This pattern is quite similar to that reported for the OECD economies as a whole, indicating that the data and methodological challenges in developed and developing economies are not too dissimilar. 33. The low and middle-income economies in the Asia-Pacific tend to not report any expenditures on LTC (HC.3), or to report only negligible amounts. The ones reporting no such expenditures include Bangladesh, Sri Lanka, Thailand and Tonga. In most cases, this is because data sources and methods have not been developed, and because these expenditures are thought to be currently so small, that efforts to address the gap are not a priority. For the developing economies in the region, LTC expenditures may in fact well be small given that they still have relatively young populations, and so this lack of inclusion may only be resulting in a small under-estimation of THE, of the order of less than 1%. 34. Several economies (Australia, China, Japan and Mongolia) do not report expenditures by non-profit institutions, and this reflects lack of data sources and methods. Two economies also fail to include expenditures by corporations (Chinese Taipei, Tonga), and this is due to lack of suitable routine data sources. It is uncertain how large an impact these gaps have on overall estimates of THE, since it is likely that the expenditure contributions of non-profit institutions and corporations are highly variable across economies. Nevertheless, they probably only account for less than 1% of THE, since both types of financing account for less than 1% of THE in all the economies in the region that do report such expenditures. 1 Refer to Jeong and Hurst (2001) for the difference between NME and THE in Japan. 19

21 35. Another almost universal departure from the SHA framework in the region is that the export and import of health services are not explicitly taken into account. The typical reason for this is the lack of suitable data sources and methods. Only Hong Kong SAR and Korea account for some of these expenditures, as they partially include health expenditures by residents abroad, and exclude some expenditure on health by persons within their territories, who are not considered to be part of the resident population. However, this gap in coverage is similar to that in OECD economies, most of which also currently fail to explicitly account for such expenditures. APPLICATION OF THE SHA CLASSIFICATIONS 36. Implementation of the SHA requires consistent application of the three ICHA classifications for sources of financing (HF), functions (HC) and providers (HP). All regional economies using the SHA framework do apply all three classifications, and in general they do not report major difficulties in doing so in using such a tri-axial approach. However, a number of deviations from the classifications are found, although the pattern is not that dissimilar to that seen in the OECD economies as a whole. Applying the health care financing classification (ICHA-HF) 37. SHA implementations in the Asia-Pacific region appear able to fully apply the ICHA-HF classification for public financing sources, and no deviations or problems are apparent. This is similar to the experience in the OECD economies. 38. As already noted, several economies are not able to report expenditures by nonprofit institutions. There is no apparent tendency for this problem to be worse in the developed or the developing economies, indicating that this is largely a problem related to coverage of non-profit institutions by routine statistical information systems. 39. With the exception of Bangladesh, none of the non-oecd economies in the Asia- Pacific region report figures for the different cost-sharing sub-categories (HF HF.2.3.9) of the private households out-of-pocket expenditure category (HF.2.3). This is in marked contrast to the OECD economies in the region, which all report these expenditures separately. This difference is the only systematic difference between the OECD and non-oecd economies in their compliance with the SHA classifications as a whole. The common reason appears to be the lack of suitable data to disaggregate household out-of-pocket expenditures in this way, largely because the main data sources used, such as household surveys, rarely have the necessary disaggregation. Applying the health care providers classification (ICHA-HP) 40. The developing economies in the region do not report or report only minimal expenditures by nursing and residential care facilities (HP.2). This partly reflects a definite scarcity of such institutions in the developing economies, with their younger populations and lack of organized systems for providing LTC or institutionalized care of the elderly. However, it also appears due to the lack of adequate data sources, and difficulties in separating such institutions from other private inpatient institutions. The lack of data sources in turn is probably a reflection of the absence of public financing schemes for LTC in these economies, since in most OECD economies these schemes generate much of the required data. However, this difference in reporting may eventually disappear as the elderly population in the developing economies increases, and as public policies evolve. 20

22 41. Either zero or very small levels of expenditures at other providers of ambulatory care (HP.3.9) are reported by the developing economies in the region, and for the subcategories for ambulance services (HP.3.9.1) and blood and organ banks (HP.3.9.2), the developing economies tend not to report any expenditures. This may reflect the absence of free-standing ambulance organizations and independent blood and organ banks in these economies, but is also due to the lack of data sources to estimate these expenditures. 42. Most of the economies in the region (Australia, China, Chinese Taipei, Hong Kong SAR, Japan, Thailand, Tonga) do not report any expenditures by providers providing occupational health care (HP.7.1), or expenditures by private households as providers of healthcare services (HP.7.2) at home (all but Sri Lanka). This reflects lack of data sources and appropriate methods. The gap for households as providers of healthcare services at home is likely to require significant development of new methodological approaches. 43. Some economies special features might lead to ambiguity in how to map certain services to ICHA-HC. For example, in Korea, oriental medical services and herbal medicine play a far greater role than in other economies. The Korean health accounts followed the manual, and thus, oriental medicine doctor s clinics are classified as HP. 3.3 Offices of other health practitioners. But they could be classified as HP.3.1 Offices of physician (with threedigit items such as HP and HP formed to distinguish the two). Applying the health care functions classification (ICHA-HC) 44. For international comparability of data, the functional classification is the most crucial element of the SHA. It is also of vital importance for national health policy-making. It is the breakdown of expenditure by function describing what is purchased. The SHA applies two approaches in functional classification: (i) functions in terms of the purpose of health care (curative care, rehabilitation and long-term care, etc.); and (ii) the mode of production that reflects characteristics of technical and managerial organisation of health care. 45. As noted before, most SHA implementations in the region are not able to report expenditures on long-term nursing care (HC.3). Most economies also do not report separately the subcategories of expenditures related to rehabilitative care and home care. In general coverage of these items by the developing economies tends to be worse than the OECD economies. Lack of data sources and appropriate methods appear to be the main reasons for these deviations. 46. The other important issues relate to differentiation of the modes of production in relation to separating day-care and long-term care and in or out-patient care provided in hospitals. Estimating LTC affects both the magnitude of total expenditure and its structure, while the other problems affect only the comparability of structure of health spending across economies. 47. Overall, the experience of non-oecd, Asia-Pacific economies in implementing the SHA approach generally mirrors that of OECD economies. All are able to implement the basic tri-axial classifications, but problems are commonly encountered with respect to measurement of expenditures on long-term nursing care, rehabilitative and home care, and in explicitly accounting for export and import of healthcare services. However, in contrast to the OECD economies, the developing economies in the region do not report detailed disaggregation of cost-sharing expenditures by households, and are more likely to be unable to separate out expenditures on nursing and residential care facilities. Some of these 21

23 deviations and gaps are likely to be resolved in future with improvement of data sources and methods, especially for such items as occupational healthcare expenditures, but the pattern of gaps also indicates that some ICHA subcategories, especially for cost-sharing in household out-of-pocket expenditures, may be too detailed given data limitations and current policy priorities. 22

24 CHAPTER Ⅲ. A COMPARATIVE ANALYSIS OF HEALTH EXPENDITURES ACROSS 12 ECONOMIES HEALTH EXPENDITURE BY FINANCING AGENT 48. Table 5 and 6 display the total health expenditure (THE) by financing agent for the latest years which data are available for, expressed in USD PPP and as a percentage of gross domestic product. Figure 1 highlights the fact that differences in the share of economies national income devoted to health are considerably smaller than in absolute spending levels. A wide variation in overall health spending levels can be observed, ranging from Australia with a total health spending per capita of 3,234 USD PPP, which is eighty eight times that of the lowest spending economy, Bangladesh (37 USD PPP). Some way behind Australia is Japan and New Zealand which spent 2,477 and 2,608 USD PPP respectively. Hong Kong SAR, Korea and Chinese Taipei follow, spending 1,840, 1,710 and 1,553 USD PPP respectively. A final group of economies each spent approximately 500 USD PPP per capita or less: Malaysia (516 USD PPP), China (452 USD PPP), Thailand (239 USD PPP), Sri Lanka (158 USD PPP), Mongolia (119 USD PPP), and finally Bangladesh. Figure 1. Total health expenditure, as per capita US$ PPP and % of GDP Per capita PPP 3,500 GPD(%) 12.0% 3, % 9.9% 10.0% 2, % 7.4% 8.0% 2,000 1, % 6.3% 4.2% 4.4% 3.5% 4.2% 5.8% 6.0% 1, % 4.0% % - AUS NZL JPN TPE HKG KOR MYS CHN THA LKA MNG BGD Public Share of GDP 0.0% 23

25 Box 2. What is the Purchasing Power Parities (PPPs)? Purchasing Power Parities (PPPs) are currency conversion rates that both convert to a common currency and equalise the purchasing power of different currencies. In other words, they eliminate the differences in price levels between countries in the process of conversion. Per capita volume indices based on PPP converted data reflect only differences in the volume of goods and services produced. Comparative price levels are defined as the ratios of PPPs to exchange rates. They provide measures of the differences in price levels between countries. The PPPs are usually given in national currency units per US dollar. The price levels and volume indices derived using these PPPs have been rebased on the OECD average. ( Essentially PPPs derive from price relatives that show the ratio of the prices in national currencies of the same good or service in different countries. PPPs are calculated in three stages. The first is at the product level, where price relatives are calculated for individual goods and services. The second is at the product group level (i.e. basic heading), where the price relatives calculated for the products in the group are averaged to obtain un-weighted PPPs for the group. And the third is at the aggregation level, where the PPPs for the product groups covered by the aggregation level are weighted and averaged to obtain weighted PPPs for the aggregation level. The weights used to aggregate the PPPs in the third stage are the expenditures on the product groups. Purchasing power parity (PPP) is a theory which states that exchange rates between currencies are in equilibrium when their purchasing power is the same in each of the two countries. This means that the exchange rate between two countries should equal the ratio of the two countries' price level of a fixed basket of goods and services. When a country's domestic price level is increasing (i.e., a country experiences inflation), that country's exchange rate must depreciated in order to return to PPP. The PPP theory uses the long-term equilibrium exchange rate of two currencies to equalize their purchasing power. ( 49. In considering health expenditure as a percentage of gross domestic product (Figure 1 and Table 6), the level of a economy s wealth devoted to health varied from 3.2% in Bangladesh up to 9.4% in the case of Australia. In general, health expenditures as a share of GDP increase with per capita GDP. Figure 2 shows the relationship between the log of per capita health expenditures and the log of per capita GDP, and this demonstrates a close relationship between income and health spending, which parallels that seen amongst OECD economies as a whole. These findings also confirm that health expenditures are incomeelastic in the Asia-Pacific region, generally increasing more than GDP per capita. 24