Consent for data sharing

There are circumstances where no form of consent can be obtained. These situations are exceptional and will need case-by-case review and clear arguments to satisfy the requirements of ethics review boards.

There can be varied reasons why consent is not possible. Limited capacity may prevent a person from being fully informed. Data may have already been collected for another purpose that did not require consent, such as government administrative data. Consent may not be technically feasible in some very large scale projects. Data may have been collected years ago when formal consent procedures were not standard practice. Finally, if knowledge of the research would invalidate the method, then it may not be possible to inform participants in advance.

Because the principle of consent is so important, proceeding without it is possible, but will be subject to rigorous ethics review. In general, such research will have to meet three conditions:

• clear value and benefit from doing the research
• no alternative research design can achieve the same result, that is, the deception or lack of consent is essential
• there is no or very minimal risk of harm to participants.

It is also possible to share data without explicit consent, but special consideration must be given to each situation. Relevant factors include:

• Is there any language in the consent form or information sheet (if any were used) that explicitly precludes data sharing, such as "data will be used only for this project"? In such cases, sharing is not possible.
• How disclosive is the material? Are people or organisations named or can their identities be inferred?
• Is it personal or sensitive data under current data protection laws?  If people are no longer living, then data is not defined as personal.
• Is anything known about participants' attitudes about having their data shared?  For example, in older oral histories, consent was rarely sought, but the assumption of this genre is usually to share peoples' life stories.
• What harm, and to whom, could result from any disclosure?  What about to any identifiable third parties?
• Can the data be de-identified?
• Could the data be re-submitted for ethical review? For example, historically valuable data held at the London School for Hygiene and Tropical Medicine on people with HIV is being prepared for archiving without explicit consent.  The data are being thoroughly anonymised, researchers have assessed the risk in consultation with the UK Data Service, and an ethics committee has reviewed and agreed to sharing under controlled conditions.

Here is an example where the UK Data Service has accepted data without explicit consent for sharing:

Apperly, I A (2017). Behavioural and eyetracking data using the Director task. [Data Collection]. Colchester, Essex: UK Data Archive. 10.5255/UKDA-SN-852224

The UK Data Service advises that data sharing without explicit consent can be permissible on a case-by-case basis if the factors above are given proper consideration. In such cases, we communicate with depositors to ensure that they clearly understand their and our responsibilities for deposited data.