First of all, one should notice that this report only describes the availability of the data. Besides timeliness, other aspects of the quality of data (e.g. accuracy, comparability, usefulness and relevance) need to be explored in a next step.
Regarding the availability of data for Belgium, there is a great difference between the WHO Health for All Database, the OECD Health Data, and Eurostat. While data for Belgium are only available for 27% of the indicators in the WHO Health for All database, this amounts to 51% in the Eurostat database, and to 72% in the OECD Health Data database (Table ) (even 78% after correction for the irrelevant and non compulsory indicators). An explanation for this difference, particularly between the WHO and OECD databases, is the number of indicators related to health status, and especially the proportion of mortality-based indicators in the whole dataset. For WHO, 42% of the database consists of mortality-based indicators, while only 10% of the indicators of OECD are related to health status (containing mortality and morbidity).
Overview of availability problems for Belgian data in OECD, WHO and Eurostat databases, by type of source
When looking at the sources, in all three datasets about 90% of the data comes from a national source.
One conclusion that can be made based upon this exploration of the availability in three international databases is the lack of recent Belgian mortality, or broader, health status data. All three datasets no recent data are available about mortality. This dimension accounts for a great proportion of the missing indicators, especially in the WHO database. Indicators concerning health status account for 69% (304/441) of the missing indicators in the WHO HFA-DB, and (after correction) for 46% (366/797) of the missing indicators in the OECD Health Data. The reform of the Belgian State has had consequences on the provision of health information and data. Over the last decades, Belgium has become a federal state. In the 1980s, collecting mortality data became a competency of the communities and since then the medical information on the forms of death certificates is encoded at community level [6
]. This results in a delay in provision of national mortality data but during the last years many efforts have been done to shorten the delay. The recent European regulation in this domain should enhance the capacity to have data on causes of mortality with a delay of less than 2 years [7
In the OECD Health Data database there is another dimension which scores badly, namely expenditure on health (21% of the missings). A comparison with WHO and Eurostat databases of expenditure on health is not very appropriate because of the low number of indicators about expenditure in these databases (OECD: 2141 (or 1853 after correction), WHO: 21, and Eurostat: 13).
Belgium should consequently urgently make an effort to catch up, especially on data about mortality. The availability of Belgian non-mortality data in the WHO dataset was quite good in comparison with neighbouring countries. If Belgium could solve this problem, a great deal of the lacking health data could be made available.
The OECD, EUROSTAT and WHO-Europe have been collecting data on monetary and non-monetary health care resources for many years and they agreed on a new joint data collection which has been launched in 2010. The most important goal is 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 will increase the use of international standards and definitions and improve the consistency of data reported by international organisations. This new joint data collection would initially cover key variables related to: 1) human resources in health; and 2) physical/technical resources. In order to achieve consistency in data submissions to the three international organisations, one Focal Point for the joint data collection on non-monetary healthcare statistics had to be designated in 2009 [8
]. In Belgium, this Focal Point will be assisted by a coordination group with representatives of the federal and federated entities [9
Initiatives have already been taken in the past to coordinate all activities concerning health statistics and to provide up-to-date information. To reflect reforms in the structures of the Belgian state, the composition of the Superior Council of Statistics was amended by the Royal Decree of March 31, 1998 (Belgian Official Gazette May 6, 1998) [10
]. This Council is a coordinating council, intra- and interfederal by its composition: representatives of federal, community and regional authorities, parastatal and subordinate institutions. Besides this, in March 2000, a protocol for data exchange between federal authorities and authorities responsible for health policy was signed in order to coordinate the collection of information [11
]. The crucial need for a national focal point with representatives from different institutions was also mentioned in a report published in 2008 in the framework of the implementation process of the ECHI indicators (European Community Health Indicators) [12
]. The aim of the ECHIM (European Community Health Indicators and Monitoring) Joint Action (2009-2011) is to advance health monitoring throughout Europe by developing relevant and comparable health indicators and by making them available in the EU and its Member States, as well as in other European countries.
Although lots of health data are available in Belgium, they are until now not integrated into a national health information system. Health data from different regions and communities have to be aggregated in order to get national data. An overview of the most important healthcare databases has been made by the KCE in 2006 [13
]. A problem is the comparability of data from different sources: different definitions, reference periods and calculations are used. This also makes it difficult to compare data over time. There is thus an urgent need for coordination between all concerned authorities to improve the situation. Special attention should be paid to the development of data-exchange systems, so that timely information can be provided.
The Belgian Focal Point for the data collection on non-monetary healthcare statistics should first make an inventory of possible data sources, with information on their timeliness, periodicity and representativeness. Inventoried sources could be clustered into five groups: disease-specific registers, annual statistical reports, administrative databases, surveys and studies. The definition, sources and methods (sampling, calculation,...) underlying the data should be documented. Problems/weaknesses related to data collection should be listed. Different problems could for instance arise when a source is used for another purpose than the original one (codes of diagnoses which are reimbursed at higher level could be "overused"). Another very important problem related to routine data collection systems is lack of quality assessment for some data sources.
The Focal Point should afterwards choose the most appropriate source and try to fill in the gaps. An information system should be developed to centralise the information. The establishment of a regularly maintained national database with aggregated data at national and regional levels, available online, would increase the availability, accessibility and use of health information by different kinds of users (policy makers, healthcare professionals, researchers, students,...). This database should also contain metadata to allow users to correctly interpret data, and include data presentation tools. Hopefully this valorisation of data will lead to an increase in their quality and also in the data exchange between all involved partners.