Risk factors in breaching confidentiality
While the following are in absolutely no way exhaustive, three situations present themselves as particularly endangering of confidentiality.
Observational Studies – Studies that involve filming or recording a person require consent, yet as covered in the Informed Consent module, some studies will not be able to ascertain legitimate results if the subject is aware they are being watched. If a subject is being recorded without their knowledge, the ethics of the situation come into question. The value and potential gains of the research must be present to balance out such behavior, and further the subject should be informed (debriefed) after before and given the option to have the information gathered destroyed (as they may choose to not participate in the study. Even in these scenarios, the ethics of such situations must be given special attention and consideration as they violate the privacy of the subject, even if they are being filmed in an otherwise public place..
Focus groups – There is a maxim that “It is not possible for a researcher to promise confidentiality in a focus group setting.” This must be considered when doing research in a group setting. Though the researcher may keep information confidential,, there is no way for the researcher to assure that participants will not breach the confidentiality or what their fellow participants share. The importance of respecting privacy should be stressed in such settings and in the informed consent, yet even in these settings, it is important to consider every member of a focus group a possible breach of confidentiality. What is shared in the group should be controlled accordingly.
Snowball recruiting – Oftentimes researchers will ask subjects during recruitment of anyone else they may recommend to recruit for the study. While this is acceptable and useful for subjects with little involvement of private information, if the study does involve sensitive topics (such as sexual orientation or domestic violence), simply suggesting someone for a study can be breach of confidentiality. A work around would be to have the subject suggest to people they know participating in the study, and have them reach out if they are interested.
Other considerations
It is important to note that privacy must be considered holistically and in the context of each study. In some instances a subject may want to be quoted and find no issues with their information being identifiable, but other participants may not be so open. It is important to make study procedures safeguarding of all people participating.
In situations where being known to participating in a study endangers privacy, it may be necessary to creatively name a study to not give away information on its sensitive topics. Further, when addressing participants, special care must be given that communication is secure. Email accounts may be shared or, if they are on a corporate server, viewable by other parties. Speaking by text or call while a participant is at home may risk being overheard. IT is important to word these communications off-site in a way that is as subtle as possible.
Finally, when storing information ,it is necessary to delete direct identifiers off of data and also indirect identifiers that may be used to deduce the subject’s identity. If needed for reference, they may be replaced with numeric identifiers or codes that correspond to a key that is stored elsewhere in a secure way. If in publications or presentation it becomes necessary to mention individual cases, direct identifiers should be moved or misleading identifiers added to conceal the identity of the subject. As stated in the informed consent module, informed consent documents must state whether or not private data with direct identifiers removed will be reused in research by another party.