Medical schools and licensing agencies are confronted daily with the task of evaluating and certifying professional competence. Frequently this is now done using “performance based assessments.” The Competency Project seeks to construct a collection of graded instances of this kind of professional performance. This collection will include video recordings of subjects working up patient problems and being interviewed with respect to their understandings of these problems.

          In building such a collection, there is a problem with getting informed consent from the subjects captured on the video. The problem is that the data once collected might be used in multiple ways and it may not be possible, at the time that the recording is made, to specify precisely what these uses might be. Further, it is often not known at that time even who will be studying the data, since data corpora are often shared within a research community. It is generally not feasible to re-consent the subjects when a researcher, sometimes years later, undertakes to study the data in the collection. This problem is not unique to video-based corpora. A very similar problem arises, for example, in tissue bank research involving specimens which are collected well before research plans have been fully formulated. One way of protecting the rights of human subjects while making it possible to build and use shared repositories of tissue (or video) is to employ a two-stage model for carrying out the research. The model employs two types of protocols— use and collection protocols.

          Collection protocols are used to gather the research materials that comprise the database or corpus. Each collection protocol has one or more associated informed consent forms (ICFs). The collection protocol specifies how data will be collected, who will do the collecting, and the constraints on ultimate use (as specified in the ICF[s]). Like use protocols, collection protocols must be reviewed and approved by the local IRB of the institution that owns and maintains the data corpus (again, Springfield Committee for Research Involving Human Subjects in this case). The collection protocol must be kept active for the full period of data collection. Data (video and otherwise) for the Competency Project is collected under SCRIHS Protocol #06-126. This protocol has separate ICFs for subjects and standardized patients.

          An approved use protocol must be in place for the full period during which some responsible party has access to some or all data residing in the corpus for the purposes of carrying out some form of approved research. Use protocols are always linked to a previously approved collection protocol. They have no associated (ICFs), but are bound by the terms of the ICF(s) of their parent (collection) protocol. Each use protocol must be reviewed and approved by the local institutional review board (SCRIHS in this case). The use protocol must specify what data will be accessed, who will have access, how access will be implemented and controlled, specific plans for publication, and plans for destroying all copies of the data when the study is completed. The model allows for multiple use protocols to be associated with a single collection protocol.

          A use protocol (SCRIHS Protocol #08-011) was also developed to permit the data in the Competency Corpus to be used by an initial set of investigators. Other researchers interested in studying these materials must develop their own use protocols that must also be approved by SCRIHS. A signed use agreement must be included as an addendum to the use protocol.