Re Martin's question re outcomes (and I'd want to do a literature search before going any further of course):

But setting that aside, and this may be incredibly naïve, but are there four categories, and are each important?

1. Process measures: in this case, how many clinical questions asked and successfully answered; or time to answer a predetermined question; or quality of answer to preset question (e.g. if the question is about antibiotics in otitis media if the "search engine" doesn't put Paul et al's Cochrane review at or very near the top it doesn't get many marks from me).

2. Qualitative. Too obvious for this group but you know what I mean. Participant experience is an important measure. And I'd include both patients and healthcare practitioners of course.

3. Quantitative. Here I'd be looking for data relating to at least a change in clinical practice. Pragmatically it might have to be measured on a before and after basis but surely we can focus on some key gaps between evidence and practice and try and bridge them by making "search engines" available and discussing how to use them "optimally" on those areas?

4. Pastoral. Different, I'd argue from qualitative. Here I'm thinking about how individuals feel about their work and role (or from patient's perspective how they feel about their illness) in a holistic sense, taking a much more rounded view about their worth, approach to CPD, staying up to date, how consultations are going now. Qualitative approaches likely, but much deeper questions.

I too have a summer student but only for 3 weeks, but I think we could be persuaded to play in the sandpit as long as this is reasonably small scale.
Neal Maskrey