Consent Provides the Social License for Legal and ethical issues in Research

By Dr. Frances Burns & Elizabeth Nelson Frances Burns and @elnelson
Research is vital for creating knowledge and improving societal impacts for our populations. The Administrative Data Research Network (ADRN) which the Administrative Data Research Centre Northern Ireland (ADRC-NI) operates enables new economic and social research using linked, unconsented administrative data for public benefit. Administrative data is information about people, business and other organisations collected by any government department or agency for delivering their day-to-day services.

One way in which we can generate knowledge is through the secondary use of data for research purposes. Secondary data is data which has already been collected for a purpose other than a current research project which may also have value for research. Administrative data is considered secondary data but secondary data may also be survey data which may be reused for new research.

What are the considerations of using the public’s data in this way and how is it affected by what the research is and who conducts it? Here we discuss the legal, ethical and social licence issues around the secondary use of administrative and survey data.

The legal use of administrative data

ADRC-NI researchers can lawfully and ethically access de identified linked data sets for research within a secure setting, without obtaining explicit individual level consent. Network policies and procedures comply with data protection issues arising from Domestic UK law (DPA 1998) and European Union law (GDPR 2016/ HRA 1998) and all projects must pass independent review to confirm they are ethical, lawful and of public benefit. But, just as research using sensitive data can be legal and ethical, how far does the social licence to conduct this research extend? Publics’ confidence and trust in the use of their data fluctuates for a multitude of interrelated factors; demographics, culture, nature of the data, sensitivities to the subject matter.

Research on attitudes

UK Economic and Social Research Council (ESRC) commissioned Ipsos Mori to conduct focus groups on the use of data at the establishment of the Network. Their ‘Dialogue on Data’ (2014) report found significant support for the use of administrative data for research purposes, where individual privacy was respected and where there was demonstrable public benefit.

ADRC-NI in conjunction with Action Research Knowledge (ARK) conducted research using the Northern Ireland Life and Times (NILT) Survey (2015) to better understand how publics in Northern Ireland view the use of their data. Entitled ‘Public attitudes to data sharing in Northern Ireland’, the research specifically examined how the general public in Northern Ireland feel about the sharing and linkage of their data for research purposes, with a focus on health data and research (Robinson and Dolk et al, 2015).

The research found that public confidence in data sharing is based on three pillars: trust in organisations, data protection measures, and the perception of a public benefit to the research. None of the questions on data linkage specifically addressed the issue of consent, however, once the basis was laid and there was an understanding amongst respondents of the uses of data for research, they were asked about situations in which they felt consent may or may not be necessary. According to Robinson et al, “A third felt that the researchers should try to get consent for linking where possible, 30% felt that consent wasn’t necessary as long as nobody will be identified and about 30% felt that consent was vital and the research could not proceed without it.” However, it is important to note that the nature of the data in question being de-identified and therefore unable to identify any specific individual or business was not made explicit in the question on consent; therefore it is not possible to draw any robust conclusions from this data on public opinion towards de-identified, linked, unconsented administrative data for research purposes.

ADRC-NI Linking survey data to administrative data

One purpose of ADRC-NI is to develop new methodologies to maximise use of existing data through linkages. One methodology is exploring conditions for linking independently collected research survey data to administrative data; specifically survey data from longitudinal research. With legal and ethical issues assured and public benefit qualified, it becomes the changing ‘social licence’ to link without explicit consent which becomes the factor of greatest consideration for researchers.

Longitudinal research participants have their data collected at repeated time points. As an exemplar, the ‘Northern Ireland Cohort for the Longitudinal Study of Ageing’ (NICOLA) survey data and consent is collected at least every two years. The NICOLA researchers have identified many benefits for linking, in a de identified and therefore unconsented condition, NICOLA survey data with administrative data; for example health data. The question raised is, would assured legal and ethical compliance be enough to access link data without consent, even if its de identified?

ADRC-NI and NICOLA researchers are working to identify potentials and risks in terms of methodological validity and representativeness; participant engagement and retention; and, more widely, public perception and institutional reputation in approaches to consent and data access.

Fundamentally, should researchers who engage with participants directly, link their data without consent, simply because legally they may? How would the nature of this exchange affect perceptions of Trust? And what affect does this have on research participation in the longer term for participants and the general public?

https://adrn.ac.uk/about/network/n-ireland/

https://www.qub.ac.uk/researchcentres/CentreofExcellenceforPublicHealthNorthernIreland/Research/NICOLA/

http://www.ark.ac.uk/nilt/

One clap, two clap, three clap, forty?

By clapping more or less, you can signal to us which stories really stand out.