HEALTH CARE NON EXPENDITURE STATISTICS

EUROPEAN COMMISSION EUROSTAT Directorate F: Social statistics Unit F-5: Education, health and social protection DOC 2016-PH-08 HEALTH CARE NON EXPENDITURE STATISTICS 2016 AND 2017 DATA COLLECTIONS

In 2010, a new Joint OECD/Eurostat/WHO-Europe Questionnaire on Non-Monetary Health Care statistics (JQNMHC) was launched. The overall objective of this joint data collection on non-monetary health care statistics was to provide internationally comparable data on key aspects of health care systems, focussing initially on nonmonetary resources of health care systems. This questionnaire complements the existing joint OECD/EUROSTAT/WHO data collection on Health Accounts. This document summarises the steps made by the three organisations that lead to the Joint data collection on Non-Monetary Health Care statistics and presents the changes introduced in the 2016 data collection (seventh round of the Joint Questionnaire), as well as the timetable and changes for the upcoming 2017 data collection. Working Group members are asked to: Take note of the state of play of the seventh round of the Joint Questionnaire on Non-Monetary Health Care statistics and the additional Eurostat Module; Support the timetable of the eighth round of the Joint Questionnaire on Non- Monetary Health Care statistics and the additional Eurostat Module; Try to fill the data gaps on the breakdown on hospital beds by function of care; Try to fill the gaps on the data collection on health workforce migration; Take note that a new flag will be available on the data reporting ("P"). BACKROUND The OECD, EUROSTAT and WHO-Europe have been collecting data on non-monetary health care resources for many years. The most important goal of the new joint data collection was to reduce the burden of data collection for the national authorities responsible for the provision of statistical information to the international organisations. Moreover, a joint effort increases the use of international standards and definitions and improves the consistency of data reported by international organisations. Before launching this joint questionnaire, some preliminary work had to be done such as selecting the set of variables, harmonising the definitions, designing a focal point (and a backup person) for each country, freezing the three international organisations databases and harmonising the data submitted to the three international organisations. That work was done in close cooperation with Member States. Before 2009 Eurostat, OECD and WHO had already started to harmonise definitions for their non-monetary health care data collections, however those data were still collected separately, with different focal points for their Member States. In 2010, a new Joint OECD/Eurostat/WHO-Europe Questionnaire on Non-Monetary Health Care statistics (JQNMHC) was launched. This first JQNMHC was focussing on health care resources only, as that part of data was easier to harmonise than the one on health care activities. Therefore, Eurostat requested data for two separated data collections: - A joint data collection with OECD and WHO on health care resources (human and physical) using one focal point for all three organisations (JQNMHC); 2

- A separate data collection on health care activities (additional Eurostat questionnaire). In 2013, the health care activities were incorporated in the joint data collection. As a result, the Eurostat s data collection split into: - A joint data collection with OECD and WHO on health care resources (human and physical) and health care activities (with some additional Eurostat variables) using one focal point for all three organisations (JQNMHC); - A separate module on regional data and hospital technical resources (additional Eurostat module). In 2015, a new module on Health Workforce Migration was added to the JQNMHC. Joint Questionnaire on Non-Monetary Health Care Statistics Additional Eurostat Module Health employme nt and education Health workforce migration Health care activities Eurostat additional variables Physical and technical resources Health employment and hospital resources at regional level Physicians by categories and technical resources For the joint questionnaire Member States use the same focal point and the same data transmission via e-damis. Thereby, data is sent simultaneously to the three organisations. 2016 DATA COLLECTION The JQNMHC focuses on non-monetary resources and activities of health care systems. It is composed of four parts: - 'Health employment and education' which covers the number of physicians, dentists, pharmacists, nursing and caring professionals, physiotherapists, hospital employment and graduates; - 'Health Workforce Migration' which covers the migration movements of doctors and nurses; - 'Physical and technical resources' which covers the number of hospitals, hospital beds, residential care beds and medical technology; - 'Health care activities' which covers ambulatory care, hospital care and procedures. In this fourth part, Eurostat collects some additional variables: on hospital discharges of non-resident patients and on some extra surgical procedures. Eurostat also sends an additional module on variables related to health employment and education and physical and technical resources: 3

- The health employment and education variables cover the number of physicians (total number and in training) by detailed categories and the health personnel at regional level; - The physical and technical resources variables cover the number of hospitals, hospital and residential care beds at regional level and some hospital technical resources. All tables of the questionnaire are pre-filled with the data provided to the three organisations in previous years. National focal points are asked to check the data currently available and update them. If longer time series are not readily available, national correspondents are invited to report 5-10 years of the most recent data. If possible, in addition, data are requested to be supplied in five year intervals (1980, 1985, 1990, 1995, 2000). Time series in the 2016 JQNMHC were to cover the period up to 2014. The questionnaire also included the year 2015, for reporting of either final data or preliminary estimates. For this seventh round of the JQNMH, the questionnaires were sent on 18 of December 2015 with a deadline for return on 26 February 2016. The validation process lasted from end of March to June 2016. Data was published by Eurostat in July 2016. 4

DATA TRANSMISSION This year 36 countries (28 Member States, 4 EFTA countries and 4 candidate countries) participated in Eurostat data collection. Most countries met the deadline for the data submission and send recent data (2013-2015) for the different parts of the questionnaire. The major problems concern the CSV files with Hospital Discharge Data (5 countries did not send data 1 ) and the Health workforce migration module (12 countries did not report recent data 2 ). Table 1 shows the availability of recent data (2013-2015) for each country and each part of the questionnaire. Some countries have an overall high rate of data reported (Romania: 97%, Lithuania: 94%, Liechtenstein: 88%), while other present very low figures (Sweden (39%), Slovakia (47%) and Hungary (49%)). In average, the availability of data considering all the countries is of 69%. In the health employment and education module, Slovakia (40%), Finland (47%) and the United Kingdom (48%) reported less than 50% of the data series. In the physical resources module, Norway (18%), United Kingdom (21%) and Hungary (21%) have very low figures for the availability of recent data, while in the health activities module, Greece is the most problematic Member State, not reporting any recent data. For the Eurostat module, Sweden (12%), Hungary (29%) and Slovakia (34%) have the lowest reporting rates. 1 NL, LI, MK, ME and TR. 2 ES, HR, CY, LV, LU, PL, SI, FI, IS, LI, ME and RS. From those, some have older data available: ES, SI and FI. 5

Table 1. Percentage of recent data (2013-2015) reported in each part of the JQ 2016 Empl. & Educ. Migration Phys. Res. Activities Eurostat module Total (58 series) (16 series) (39 series) (34 series) (89 series) 3 (236 series) Belgium 43 74% 10 63% 15 38% 20 59% 73 82% 161 68% Bulgaria 39 67% 2 13% 39 100% 20 59% 79 89% 179 76% Czech Republic 38 66% 6 38% 39 100% 31 91% 52 58% 166 70% Denmark 54 93% 11 69% 24 62% 26 76% 47 53% 162 69% Germany 46 79% 10 63% 24 62% 23 68% 46 52% 149 63% Estonia 48 83% 10 63% 37 95% 27 79% 59 94% 181 85% Ireland 47 81% 9 56% 18 46% 25 74% 77 87% 176 75% Greece 35 60% 8 50% 38 97% 0 0% 82 92% 163 69% Spain 45 78% 0 0% 39 100% 30 88% 86 97% 200 85% France 45 78% 14 88% 29 74% 29 85% 77 87% 194 82% Croatia 29 50% 0 0% 39 100% 31 91% 74 83% 173 73% Italy 43 74% 15 94% 36 92% 21 62% 87 98% 202 86% Cyprus 49 84% 0 0% 38 97% 28 82% 36 57% 151 71% Latvia 49 84% 0 0% 39 100% 15 44% 59 94% 162 76% Lithuania 56 97% 15 94% 39 100% 32 94% 57 90% 199 94% Luxembourg 45 78% 0 0% 36 92% 32 94% 26 41% 139 66% Hungary 39 67% 12 75% 8 21% 31 91% 26 29% 116 49% Malta 47 81% 7 44% 39 100% 31 91% 59 94% 183 86% Netherlands 44 76% 14 88% 22 56% 16 47% 56 63% 152 64% Austria 43 74% 8 50% 33 85% 24 71% 54 61% 162 69% Poland 43 74% 0 0% 33 85% 27 79% 85 96% 188 80% Portugal 34 59% 5 31% 21 54% 18 53% 47 53% 125 53% Romania 58 100% 14 88% 39 100% 31 91% 86 97% 228 97% Slovenia 44 76% 10 63% 39 100% 29 85% 78 88% 200 85% Slovakia 23 40% 0 0% 33 85% 25 74% 30 34% 111 47% Finland 27 47% 0 0% 37 95% 31 91% 47 53% 142 60% Sweden 37 64% 14 88% 14 36% 16 47% 11 12% 92 39% United Kingdom 28 48% 15 94% 8 21% 19 56% 60 67% 130 55% Iceland 50 86% 0 0% 39 100% 24 71% 25 40% 138 65% Liechtenstein 58 100% 0 0% 39 100% 26 76% 63 100% 186 88% Norway 58 100% 16 100% 7 18% 20 59% 47 53% 148 63% Switzerland 41 71% 4 25% 27 69% 16 47% 47 53% 135 57% Montenegro 33 57% 0 0% 9 23% 4 12% 38 60% 84 40% The former Yugoslav Republic of 24 41% 1 6% 23 59% 24 71% 53 84% 125 59% Macedonia Serbia 47 81% 0 0% 35 90% 31 91% 63 71% 176 75% Turkey 29 50% 12 75% 23 59% 16 47% 51 57% 131 56% Of course, the improvement some countries have been doing in the reporting of recent data is worth to be mentioned: Greece, Cyprus, Liechtenstein, and Bulgaria have made significant progresses (Table 2). 3 The number of series for which the countries have to report data in the Eurostat module depend on the existence or inexistence of NUTS regions (89 series and 63 series, accordingly) and the rates were calculated considering that difference. 6

Table 2. Availability of recent data in JQ 2015 (2012-2014) and JQ 2016 (2013-2015) 2015 2016 (243 series) (236 series) Trend Belgium 65% 68% Bulgaria 59% 76% Czech Republic 69% 70% Denmark 66% 69% Germany 60% 63% Estonia 89% 85% Ireland 76% 75% Greece 33% 69% Spain 81% 85% France 77% 82% Croatia 77% 73% Italy 82% 86% Cyprus 41% 71% Latvia 78% 76% Lithuania 95% 94% Luxembourg 64% 66% Hungary 52% 49% Malta 85% 86% Netherlands 60% 64% Austria 67% 69% Poland 82% 80% Portugal 51% 53% Romania 96% 97% Slovenia 83% 85% Slovakia 47% 47% Finland 70% 60% Sweden 54% 39% United Kingdom 56% 55% Iceland 64% 65% Liechtenstein 64% 88% Norway 62% 63% Switzerland 56% 57% Montenegro - 40% The former Yugoslav Republic of Macedonia 62% 59% Albania 7% - Serbia 79% 75% Turkey 54% 56% 7

DATA COLLECTION ON HOSPITAL BEDS After two years of piloting the data collection with the new definitions on hospital beds by function of care, in this last round of the Joint Questionnaire the new classification was adopted as the reference one. The objective of this change is to make the data collection consistent with SHA 2011 regarding the collection of psychiatric care beds data. Member States were encouraged to revise their data series in order to adapt to the new definitions. There are two main changes compared to the previous data collection: - The psychiatric care beds are now included in the curative care, rehabilitative care and long-term care categories (but the previous category psychiatric care beds still exists, as an extra column); - There is a new variable: rehabilitative care beds, which allows the reporting of these beds (previously put in curative care beds, long-term care beds or other beds). The proposed guidelines for the potential problematic situations were the following: For the cases where there are difficulties to separate curative care beds from rehabilitative care beds: countries should report all their curative + rehabilitative beds in the curative care category; For the cases where countries are not able to allocate the psychiatric care beds between the curative, rehabilitative and long-term care beds categories: countries should allocate these psychiatric care beds in the curative category (if this allocation poses some problems, there will be the flexibility to report these beds in other beds category). Table 3 shows the results from the 2016 data collection (2014 data). All countries (except the Netherlands 4 ) are able to provide the total number of hospital beds and most of them can allocate those beds to the different categories. Some countries (Greece, Latvia, United Kingdom and Turkey) are still not able to allocate the psychiatric care beds in the different categories and put them in the "other hospital beds". Other countries put all those beds in the "curative care beds" category (Belgium, Denmark, Estonia, France, Ireland, Italy, Norway, Portugal, Slovenia and Sweden). The Former Yugoslav Republic of Macedonia put all those beds in the "rehabilitative care beds". But in general, the results of this data collection are positive. For the next round of the data collection, Member States are asked to try to fill the gaps on this new breakdown and to avoid allocating the hospital beds to the "other" category. 4 At the time of the data collection, the Netherlands were participating in the "Health Care Non Expenditure Statistics" project, which includes the data on hospital beds. In the framework of the project, they later send the request data. 8

Table 3. Hospital beds by function (2014) Total hospital beds Curative care beds Rehabilitative care beds Long-term care beds Other hospital beds Psychiatric care beds Belgium 69924 63683 1728 4513 19483 Bulgaria 51505 42851 6088 1619 947 4823 Czech Republic 67937 44810 4564 16228 2335 10172 Denmark 15174 14776 165 233 0 Germany 666337 500680 165657 102916 Estonia 6584 4824 339 1293 128 730 Ireland 11989 11241 161 587 0 1640 Greece 46160 38090 232 7838 7838 Spain 137938 110346 1740 25852 0 16825 France 410921 274462 104820 31639 57863 Croatia 25036 16974 4362 3700 0 4185 Italy ** 199474 165384 24506 9584 0 5734 Cyprus 2912 2912 0 0 0 186 Latvia 11279 6748 809 1209 2513 2513 Lithuania 21176 18504 1700 972 0 3164 Luxembourg 2746 2332 177 0 237 466 Hungary 68910 42413 9015 17482 0 8811 Malta 1994 1353 436 205 0 570 Netherlands * 55690 0 Austria 64815 49395 10645 4775 0 5263 Poland 251904 188342 63020 542 0 24770 Portugal 34522 33821 444 257 6522 Romania 133619 99547 11941 22131 0 16503 Slovenia 9356 8740 200 300 116 1365 Slovakia 31348 26642 800 3906 0 4431 Finland 24741 16853 273 7466 149 3316 Sweden 24603 22754 1758 91 4369 United Kingdom 176324 146543 29782 29782 Iceland 1041 875 59 107 0 145 Liechtenstein 61 61 0 0 0 0 Norway 19751 17633 2118 0 5968 Switzerland 37540 30799 6741 0 7431 Montenegro 2446 2446 307 The Former Yugoslav Republic of Macedonia 9088 6183 1162 915 828 1162 Serbia 39395 32539 6202 633 21 5384 Turkey 206836 200628 1949 0 4259 4259 *2012 **2013 9

DATA COLLECTION ON HEALTH WORKFORCE MIGRATION The new module introduced in 2015 focus on the migration of doctors and nurses. Both for doctors and nurses, the following variables are considered: - number of domestically-trained professionals; - number of foreign-trained professionals; - number of native-born but foreign-trained professionals; - number of foreign-trained professionals by country of first qualification (stock and annual inflow). As it can be seen in tables 4 and 5, the majority of countries are able to send some data on doctors, nurses or both. The data reporting improved in this round of the data collection compared to the previous one, but some progress still have to be made. Croatia, Cyprus, Latvia, Luxembourg and Poland still didn't manage to send any data either on doctors or nurses. 10

Table 4. Availability of data on doctors on the Health Workforce Migration Module Domesticallytrained Foreigntrained Stock of Doctors Native-born but foreign-trained Unknown place of training Foreign trained by country breakdown Inflow of foreign-trained doctors Total inflow Inflow by country breakdown Belgium x x x x Bulgaria x Czech Republic x x x x x Denmark x x x x x Germany x x x x Estonia x x x x x Ireland x x x x x x Greece Spain y y y y y France x x x x x x Croatia Italy x x x x x x x Cyprus Latvia Lithuania x x x x x x x Luxembourg Hungary x x x x x Malta x x x x x x Netherlands x x x x x x Austria x x x x x x Poland Portugal Romania x x x x x x Slovenia x x x x x Slovakia y y Finland y y y y y Sweden x x x x x x United Kingdom x x x x x x x Iceland Liechtenstein Norway x x x x x x x Switzerland x x Montenegro Former Yugoslav Republic of Macedonia Serbia Turkey x x x x X: availability of recent data (2013-2015) Y: availability of older data 11

Table 5. Availability of data on nurses on the Health Workforce Migration Module Domesticallytrained Foreigntrained Stock of nurses Native-born but foreigntrained Unknown place of training Foreign trained by country breakdown Inflow of foreign-trained nurses Inflow by Total inflow country breakdown Belgium x x x x Bulgaria x Czech Republic Denmark x x x x Germany x x x x Estonia x x x Ireland x Greece x x x x x x Spain y y y y y France x x x x x x Croatia Italy x x x x x x Cyprus Latvia Lithuania x x x x x x Luxembourg Hungary x x x x x Malta Netherlands x x x x x x Austria x Poland Portugal x x x x Romania x x x x x x Slovenia x x x Slovakia Finland y y y y y Sweden x x x x x x United Kingdom x x x x x x Iceland Liechtenstein Norway x x x x x x x Switzerland x x Montenegro Former Yugoslav Republic of Macedonia Serbia Turkey x x x x x x X: availability of recent data (2013-2015) Y: availability of older data 12

DATA VALIDATION The Joint Questionnaire excel files include a set of automatic checks that are designed to improve data quality and minimise errors. The main purpose of the data checks is to verify the internal consistency in the data reported in the questionnaire. National correspondents are encouraged to use these data checking formulas and correct data inconsistencies before sending back the questionnaire, or provide explanation in the Sources and Methods when the data do not add up or are not consistent. During the data validation process, when needed, Eurostat contacted the countries for clarifications. 2017 DATA COLLECTION The five parts of the Joint Questionnaire will be kept without any major modification in the 2017 round. A new data flag will from now on be available in order to distinguish provisional data from estimates previously codded with the same letter. B D E P Break in series Difference in methodology Estimate Provisional New The "P" flag should be used only for provisional data that are expected to be revised and replaced by final data in the near future (e.g. in the next round of data collection). The timetable for the 2017 data collection is as follows: Timetable for JQNMHC 2017 round Launch of 2016 data request December 2016 Completion of the files by the countries February 2017 Data validation March May 2017 Data dissemination End of June / July 2017 13