Public Attitudes to Data Sharing in Northern Ireland
Effective sharing and linking of medical and other social data is potentially a game-changer in advances in health and social wellbeing but public confidence is critical to a careful and judged advance in the use of these techniques. It also presents challenges in terms of respecting individual privacy.
These are principles upon which the entire Administrative Data Research Network (ADRN) is built, and therefore it’s critical that we as researchers engage the public in these discussions.
So just how confident are the public that data sharing for research is a good thing?
In 2015, the Administrative Data Research Centre Northern Ireland (ADRC-NI) teamed up with Action Research Knowledge (ARK) to investigate how the Northern Ireland public feels about data sharing for both research and care purposes. Through the 2015 Northern Ireland Life and Times (NILT) survey, our research team were able to measure a variety of attitudes and circumstances related to data sharing for research, with an emphasis on the use of health data.
The NILT survey runs every year, and contains different modules, or sets of questions, that aim to monitor the behaviour and attitudes of people living in Northern Ireland. The 2015 module on data sharing was comprised of questions mainly derived from the 2014 Ipsos Mori and ESRC report, ‘Dialogue on Data’.
We discovered that the basis of public confidence in data sharing rests on three pillars: trust in organisations, data protection measures, and the perception of public benefit.
High numbers of people trust the NHS in general (86%) and their GP surgery (91%) to keep information or data they have about people secure and use it appropriately, slightly smaller percentages are prepared to trust government departments (73%) and academic researchers (72%). Trust in charities came in at (51%), while even fewer were prepared to trust commercial and insurance companies with their data (41%).
Of particular interest to researchers and the wider ADRN, of course, is attitudes towards the use of data for academic research.
We presented a variety of scenarios in the questions, to gauge in what situations people would be comfortable with their data being shared, and under what circumstances. For example, in a scenario with a Parkinson’s disease study, 73% of respondents were happy for NHS staff to link data, remove postcodes, and pass the data to researchers, while 60% were happy for the data to be transferred to researchers to link and then remove postcodes.
Responses to questions about the need for consent were particularly interesting, as respondents roughly split themselves into thirds: 30% said it isn’t necessary to ask for consent when linking data as long as it is guaranteed that no one will be identified; 34% said that you should always try to get consent, but if that is too difficult, important research should not be stopped if consent cannot be attained; and 31% said you should always have consent before linking someone’s data, and research should be abandoned if this isn’t possible to get.
De-identification or anonymisation appear to be key to positive attitudes towards allowing researchers access to data for linking, as well as assurances that there are proper data security measures in place. These are, of course, two crucial principles within the ADRN, and so there are positives to be drawn there for our work.
Importantly, 50% of respondents felt that commercial companies should be subject to more safeguards than academic researchers. This is perhaps not surprisingly given the drop in trust from academic researchers to commercial researchers identified earlier on in the results.
When given the example of a pharmaceutical company using NHS patient information to determine which patients might benefit from a new drug for Alzheimer’s, 75% of respondents agreed (and 13% disagreed) that the company should have access to the data, as the public would benefit from a cure for Alzheimer’s. Delving further, 26% agreed (50% disagreed) that commercial companies shouldn’t get access to NHS data because they should pay for all their own research; and 57% agreed and 22% disagreed that commercial companies should pay a fee to get the data, as they will make profits from any new drugs develop from the research.
While there is less support for commercial access to the data, in the first part of the example, where 75% of respondents agreed there should be access, it is difficult to know whether this agreement would transfer to other proposed research; it is possible that Alzheimer’s is such an emotive issue that people were not prepared to put obstacles in front of a possible cure.
It becomes clear from this research that a positive attitude toward data sharing for research purposes is very situation specific. Where a clear public benefit can be demonstrated, levels of support are high: 85% of people agreed that ‘if personal data can be made anonymous and a person’s right to privacy maintained, then the data should be used where there is a benefit to society.’
Thus the difficult balance that regulators must find is that between the enormous potential public benefit that linking data for research presents, with the need to protect a person’s individual right to privacy. But it does appear from this research that there is significant public support and goodwill to achieve these potential benefits, so long as there are safeguards in place for the maintenance of privacy. The pillars of this support are trust in organisations that hold and use data, strong data security measures and safeguards, and a demonstrated benefit to the public at large of research using linked data.
More information about this research can be found at: http://www.ark.ac.uk/Seminars/
This research was funded by the Northern Ireland Health and Social Care Research and Development Unit (HSCR&D)
Written by ADRC-Northern Ireland staff and published on the ADRN Blog under Creative Commons license CC BY-NC-SA 4.0. Images used with permission.
Published 26 September 2016