In the early stages of a data donation study, the researcher should work towards submission to an Ethical Review Board (ERB) in order to obtain ethical approval. To guide a researcher in this process, consulting a legal expert is recommended. The exact process of obtaining ethical approval differs over academic institutions. Hence, it is advised to get in contact with legal experts of the institute early in the process. Despite these possible differences over institutions, a privacy policy and consent form are always required. In case personal data is being collected, a Data Protection Impact Assessment (DPIA) is also required. A solid worked out idea of the study design, local extraction script, where the data donation software is hosted and where data is stored provide the required information to prepare this.

Prior to participation in any research, human subjects or participants have to provide informed consent (Falagas et al., 2009; Neff, 2008). This usually is done by signing a consent form after being provided with thorough information on the concerning study. At least the participant should be informed on:

  • the researcher’s identity
  • the data being collected and what data is being used
  • the purpose of any data processing step
  • the existence of the right to withdraw consent
  • confidentiality protection and the limitations thereof
  • time and nature of data storage

Depending on the study population and type of data being collected, additional information might have to be provided as well. The legal expert should guide the researcher and ensure that participants are provided with all required information. Participants should be able to access this information at any moment during the study. Ensuring this continuous access can be attained by providing participants with a privacy policy, summarizing all required information on the rights of participants and potential risks of the study.

You can find a template of a consent form here (Dutch version here), and a template of a privacy policy here (Dutch version here).

Data protection impact assessment

In the context of Digital Data Donation Studies, researchers are often collecting sensitive data. The processing of such data can entail risks for participants. Researchers may for instance collect and process data revealing sensitive characteristics of a participant, such as their ethnic origin, their religious or philosophical beliefs, their health, or their sexual preference. Whenever researchers process data that can reveal such characteristics, or data that can entail other sensitivities for the participant, Digital Data Donation Studies researchers are, under the European Union’s General Data Protection Regulation, obliged to perform a Data Protection Impact Assessment (‘DPIA’). A DPIA allows a researcher to accurately identify risks that may occur due to their collection and processing of data in the context of their research.

Whenever the researcher identifies risks, they must take appropriate measures to mitigate these risks to a level acceptable for the participant. We provide researchers a generic overview of what a DPIA entails, how it works, who is involved, and the steps researchers need to take to perform a DPIA. Importantly, universities usually have their own detailed DPIA guidance and policies. Researchers should therefore consult the university’s expert on data protection (in the EU, this is the Data Protection Officer, or ‘DPO’) who will guide and advise the researcher about performing a DPIA. The university Ethical Research Board (ERB) can advise on taking appropriate measures study whenever ethical issues might arise.

A more comprehensive overview of data protection impact assessment can be found here.

Here, you can find an extensive manual that explains how consent is arranged in the GDPR, and that provides templates for a consent form and privacy policy for a data donation study.and here

Here, relevant texts researchers must comply with whenever they seek to obtain consent from the participants envisaged in their study, are included.