The acceptability of conducting data linkage research without obtaining consent: lay people's views and justifications

Abstract

Background: A key ethical issue arising in data linkage research relates to consent requirements. Patients' consent preferences in the context of health research have been explored but their consent preferences regarding data linkage specifically have been under-explored. In addition, the views on data linkage are often those of patient groups. As a result, little is known about lay people's views and their preferences about consent requirements in the context of data linkage. This study explores lay people's views and justifications regarding the acceptability of conducting data linkage research without obtaining consent.

Methods: A qualitative study explored lay people's views regarding consent requirements in data linkage via four hypothetical data linkage scenarios of increasing complexity. Prior to considering the scenarios, participants were provided with information regarding best practice data linkage processes via discussion and a diagrammatic representation of the process.

Results: Lay people were able to understand the intricate processes involved in data linkage and the key protections afforded within a short amount of time. They were supportive of data linkage research and, on the whole, believed it should be conducted without consent provided a data linkage organization de-identifies the data used so that researchers do not handle identifiable data. Many thought that de-identified data holds a different status to identifiable data and should be used without specific consent in research that aims to benefit society. In weighing up conflicting values and interests, participants shifted consent preferences before arriving at their final consent preference for each scenario and provided justifications for their choices. They considered the protection of people's information, societal benefits, and the nature and constraints of research and recognized that these need to be balanced.

Conclusions: With some exposure to the features of data linkage, lay people have the capacity to understand the processes sufficiently in order to consider ethical issues associated with consent preferences. Shifts in views reveal the complexity of such decisions. While privacy protection remained an important consideration for most participants, adequate protection measures adopted in best practice data linkage were viewed by most as protection enough for data linkage to proceed without specific individual consent.

Fig. 1

Purposive sampling method for identification of six pertinent groups. * “Agreed” means the parent returned a study form in the opt-in arm OR did not return a study form in the opt-out arm, thereby facilitating the data linkage of the child’s vaccination data to his/her hospital data Coded as Y. ** “Did not agree” means the parent returned a study form in the opt-out arm OR did not return a study form in the opt-in arm, thereby precluding the data linkage of the child’s vaccination data to his/her hospital data Coded as N

Fig. 2

Interview selection figures. ^# Not contactable refers to calls not being answered, phones being disconnected, or unresponsiveness to messages left. ^{^}The furthest travelled for an interview was 512 km by plane. It was later decided that such travel distances were not possible