Data sharing and patient records – creating vital resources for health research

Scoop.it ShareThis

Opinion

The power of linking different sources of health and biomedical data to create a resource for research is now well recognised. At a fascinating meeting in Paris, organised by the Wellcome Trust and Aviesan on 27th January, several speakers provided examples of research in both France and the UK that has only been possible through linking routine data collected as part of the provision of health services. For example, Professor Iain Buchan gave an insight into what can be achieved with routinely collected data from the NHS and how further methodologies and research will be developed by the Farr Institute.

At the Centre for Longitudinal Studies we have also realised the benefits of using administrative health records for tracing cohort members. In the recent BCS70 Age 42 survey around 1500 cases who had not participated in at least 12 years (and many in much longer) were traced via their NHS records and issued to the field. Interviewers were highly successful at encouraging these individuals to re-engage with the study and 653 were interviewed, substantially boosting the achieved sample size.

It is therefore of great concern that the European Parliament and European Council are currently considering severely tightening, and complicating, regulations for the use of patient data in research.

In January 2012 the European Commission published a draft Data Protection Regulation with a view to replacing the existing Data Protection Directive. At that stage it seemed likely that the legislative process would be very slow. However the Edward Snowden case has heightened awareness of privacy issues and increased the political salience of the draft regulations.

The rapporteur of the Committee for Civil Liberties, Justice and Home Affairs (LIBE) of the European Parliament proposed amendments to Articles 81 and 83 that would severely restrict the use of personal data for scientific research purposes without explicit consent. Compromise amendments based on these were accepted by a LIBE vote in October 2013. Vital research from across Europe that produces benefits for public health and healthcare would not be possible if the LIBE amendments were to be ratified by the European Council later this year.

Article 81 of the data-protection regulation, as amended, stipulates that member-state laws may provide an exemption to the consent requirement where personal health data are processed for research purposes, but only for pseudonymous data and where the research serves “high public interests” and “cannot possibly be carried out otherwise”. Similarly, Article 83 has been amended so that the exemption from consent for personal data “processed for historical, statistical or scientific research purposes” only applies to pseudonymous data.

As Prof Andrew Morris (Chief Scientist for Scotland) astutely observed, when summing up the meeting yesterday, whatever the outcome of the trialogue on the new regulations, a key issue for researchers is to demonstrate that their research is in the public interest. In other words, we need to be proactive in demonstrating the tangible benefits of our work. Our concept of a research output should go beyond the publication of a journal article or the production of a useful database; we need to ensure that as far as possible our research has instrumental impact. As Andrew Morris suggested, ideally we would have a website with a whole set of case studies demonstrating how linking routine patient data to cohort studies and other surveys has resulted in clear health benefits.

Many thanks to the Wellcome Trust for co-hosting a very interesting meeting yesterday and for their work in relation to the new European regulations.

Read more about the impact of EU data regulation on research.