[Whicher, Danielle, Nancy Kass, Yashar Saghai, Ruth Faden, Sean Tunis, and Peter Pronovost. “The Views of Quality Improvement Professionals and Comparative Effectiveness Researchers on Ethics, IRBs, and Oversight.” Journal of Empirical Research on Human Research Ethics, Published online before print, February 23, 2015, doi:10.1177/1556264615571558.]
The focus groups
generally concluded that intent to produce generalizable knowledge or the related intent to publish were not useful criteria for distinguishing what activities should be subject to IRB oversight. Although some participants stated their local IRBs relied on these criteria, most participants felt they were conceptually confusing and ethically inappropriate. Some stated it may be hard to know, early in an activity, whether the results will be worth publishing. Others mentioned it is conceptually hard to distinguish local learning from learning generalizable to other situations as generalizability is a matter of degree. Scholars similarly have argued that it is difficult to ascertain intent and that generalizability is not a binary concept, but falls along a spectrum. Indeed, some suggest eliminating the intent to produce generalizable knowledge criterion from determinations about oversight.
Instead, many participants suggested that considering the risk of harm to participants in a QI or CER activity makes more sense when determining what should be subject to ethical oversight, a view consistent with recommendations in the literature. (Citations omitted.)
The article misstates the language of the Common Rule, which defines research not by whether it is intended to produce generalizable knowledge, but rather by whether it is "designed to develop or contribute to generalizable knowledge." (Emphasis added.) That said, I doubt this distinction would have made a difference to the focus groups participants.