Performing Subjectivist Studies in the Qualitative Traditions Responsive to Users - Evaluation Methods in Biomedical Informatics - page 292

Special Ethical Considerations in Qualitative Studies
Informants must provide formal consent to be observed or interviewed if
a study is being performed for research purposes. If the study is strictly an
internal evaluation for administrative or quality improvement purposes,
informed consent may not be required, but the human subjects review
board of the organization should make that determination—not the study
team. It may in fact be necessary to approach two or more such boards: that
of the institution sponsoring the evaluation and that of the institutions
where fieldwork will be conducted. For example, a university team may be
conducting a study of advanced information technology in the military. In
that case, it is necessary to consult the university's review board as well as
the military review boards with jurisdiction over the data collection sites.
There is an extremely fine line between “research” studies and studies con-
ducted for internal purposes.
Whether or not informed consent procedures are required, informants
should be made aware of the purpose of the study, that their help is vol-
untary, and that confidentiality will be completely respected. Unlike survey
respondents in quantitative studies, these informants will be studied in
detail over time. They will be observed while doing their jobs. Obtaining
useful data requires gaining their trust, and identities should be protected
as much as possible. This is somewhat harder to do with audiotapes than
it is with field notes, and even harder with videotapes. Names should be
coded and, after analysis, tapes kept under lock and key or destroyed.
Special issues need consideration if the investigators are members of
the organization where data are collected. Team members need access to
the data, but sensitive issues can arise if the investigators and informants
are colleagues. For example, a mistake, criticism, or embarrassing incident
may be recorded in field notes. If the team is collectively certain that
the incident has no bearing on the evaluation or can add nothing new to
the study, the outside members on the team should consider deleting the
description of it. If, however, the information is deemed critical to the
study, the entire team, including the insider, will need to decide what to do.
There is always tension between the research benefits, the organizational
benefits, and the informant benefits of qualitative studies. If the only
product of the study is an academic paper written by the investigator,
then the informants may feel they were exploited—perhaps with significant
justification.
There are special considerations related to participant observation in
health care settings. We discussed earlier the need for investigators to focus
on the subject of the evaluation, usually information resources, and not
basic issues of medical or nursing management. If a clinically trained field
researcher encounters an extreme situation of suboptimal care where
patients may be harmed, there is a professional obligation to take action.