Access to data from European registries for epidemiological research: results from a survey by the International Epidemiological Association European Federation

Henrica CW de Vet1, Jacqueline M Dekker1, Evert Ben Van Veen2 and Jørn Olsen3

1 Institute for Research in Extramural Medicine, VU University Medical Center, Amsterdam, The Netherlands. E-mail: HCW.de_Vet.EMGO{at}med.vu.nl
2 Medlaw Consult, Den Haag, The Netherlands.
3 The Danish Epidemiology Science Center, Denmark.

Sirs—European directives and laws determine more and more the conditions for epidemiological research within Europe. The International Epidemiological Association (IEA) aimed to investigate whether these directives facilitate or hinder epidemiological research in Europe. It goes without saying that European epidemiologists, with a large variety of different exposures related to living conditions, occupations, diet, lifestyles, and health care between EU countries, are in a good position to produce more and better research if they get access to data from morbidity and mortality registers.

In 1995 the European Union issued the EU Data Protection Directive ‘on the protection of individuals with regard to the processing of personal data and on the free movement of such data’. (http://www.privacy.org/pi/intl_orgs/ec/final_EU_Data_Protection.html). The aim of this Directive is to set minimum standards for data protection in the various EU countries in order to facilitate the free flow of data between these countries (internal market motive) while at the same time offering a high level of privacy protection. It contains certain exemptions to the general principles in order to facilitate research with sensitive data (recital 28 and article 13). Therefore the Directive should also facilitate the sharing of data from morbidity and mortality registries among epidemiologists in the EU. The directive states that EU member states may set their own rules and regulations with regard to availability of data from registries, but these rules should not discriminate between member states. A report in the British Medical Journal1 showed that this EU directive led to much confusion and delay in research in the UK.

We conducted a survey to compare access to data from registries in different European countries. Furthermore, we wanted to know if data from one country can be used by epidemiologists from other EU countries. A questionnaire was composed and sent to all national epidemiological societies in 2000/2001, and it was also available on the IEA website (www.iea.org).

Responses were received from Denmark (17), UK (4), The Netherlands (2), and one response from Finland, Sweden, Switzerland, France, Spain, Greece, Austria, Poland, and Yugoslavia-Serbia. The answers of multiple responders within countries varied to some extent. Where available, the answers of the National Committees were used.

The EU Data Protection Directive was at the time of the survey implemented in Denmark, the UK, The Netherlands, Finland, Sweden, Austria, and Poland. The responders from other countries were not sure. Being non-EU (candidate) member states, the Directive was not implemented in Switzerland and Yugoslavia-Serbia.

Most Danish responders stated that the EU directive had made epidemiological research less difficult. In the UK, Finland, Austria, and Poland, the responders felt that research had become more difficult. In The Netherlands and Sweden the situation was considered unchanged.

In all countries data from mortality registries can be used without individual consent in an entirely register-based research project. Cause of death is difficult to obtain in The Netherlands, in Poland, and in France.

All countries have a cancer registry and usually a number of other morbidity registries. Scandinavian countries have a large number of registries for several diseases. In most countries, except Austria and Yugoslavia, data from these registries are identifiable at an individual level. Data from these morbidity registries can be used for epidemiological research based entirely on registries without individual consent, except in Poland where individual consent is necessary. In Austria and Yugoslavia these data cannot be used at all for research.

Whether individual consent is needed for linkage of morbidity registers to research data appears to depend on the type of project in Denmark and the UK. In The Netherlands and Poland individual consent is always needed, while in Sweden, Switzerland, Spain, and Greece no consent is required.

We also asked if stored data with personal identifiers can be used for another research purpose that requires no personal contact to study members without renewed informed consent. In Switzerland, Austria, and Poland these data cannot be used for other research purposes. In The Netherlands and Yugoslavia renewed informed consent is necessary. In other countries this depends on the research question.

Obviously, there is variation between regulations in different countries. However, the primary aim of this EU Directive was not to make uniform regulations, but to make data also available to other member states, without additional conditions. From our survey it appeared that about half of the responders had applied for use of data from other EU member states and they had all received permission.

More detailed information on the study can be obtained from www.iea-seminar.au.dk. The questionnaire can be found at www.ieaweb.org (Click European Federation and then ‘Personal data’). The European Commission conducted its own evaluation of the Directive.2 The conclusion was that by now all member states had implemented the EU directive, though some were very late with it. As the background of the Directive is driven mainly by economic issues, nothing is reported about research data and health aspects. The accompanying ‘Analysis and impact study’3 reveals more clearly the different approaches which have been taken by various countries when implementing the Directive, e.g. in the definition of ‘personal data’. Some countries consider coded data where the receiver of the data does not have access to the code as personal data, while other countries consider these as anonymous data (on the level of the receiver). Obviously epidemiological research is better served with the latter interpretation. The commission does not make a choice between these two interpretations and considers this subject open to further debate. Therefore it is important that the IEA keeps surveying the consequences for epidemiological research within Europe and will put possible negative aspects on the European agenda.

References

1 Strobl J, Cave E, Walley T. Data protection legislation: interpretation and barriers to research. BMJ 2000;321:890–92.[Free Full Text]

2 http://europa.eu.int/comm/internal_market/privacy/lawreport/data-directive_en.htm

3 http://europa.eu.int/comm/internal_market/privacy/docs/lawreport/consultation/technical-annex_en.pdf





This Article
Extract
FREE Full Text (PDF)
Alert me when this article is cited
Alert me if a correction is posted
Services
Email this article to a friend
Similar articles in this journal
Similar articles in ISI Web of Science
Similar articles in PubMed
Alert me to new issues of the journal
Add to My Personal Archive
Download to citation manager
Search for citing articles in:
ISI Web of Science (2)
Request Permissions
Google Scholar
Articles by de Vet, H. C.
Articles by Olsen, J.
PubMed
PubMed Citation
Articles by de Vet, H. C.
Articles by Olsen, J.