Determining What to Study - Evaluation Methods in Biomedical Informatics - page 76

FIGURE 3.3. Changing evaluation issues during development of a health informa-
tion resource.
closely matched to the resource's maturity. 28 For example, one would not
wish to conduct an expensive field trial of an information resource that is
barely complete, is still in prototype form, may evolve considerably before
taking its final shape, or is so early in its development that it may fail because
simple programming bugs have not been eliminated. Equally, once informa-
tion resources are firmly established in practice settings, it may appear that
no further rigorous evaluations are necessary. However, key questions may
emerge only after the resource has become ubiquitous. 29,30
Organizing Information Resource Development Projects
to Facilitate Evaluation
The need for evaluation to become a pervasive component of biomedical
information resource development projects has already been discussed.
What follows is a list of steps that should ensure that evaluation activity
proceeds hand-in-hand with the development process.
At the first stage of planning a study, it is an excellent idea to make a list
of the potential project roles listed in Chapter 2, such as project funder,
resource developer, users, and community representative, and indicate
which stakeholders occupy each role. Sometimes this task requires edu-