Appraising apples and doing something to oranges

Andrew Booth's latest article in HILJ [subscription required] is, I assume, a refection on a forthcoming systematic review of Clinical Librarianship. In it, he looks at the problems of in evaluating a clinical librarian service in an objective, generalisable way. It's also a very entertaining article in Andrew's inimitable style. Definitely get hold of the full copy if you can.

The issues around evaluating clinical librarian services are many, but a particular one that stood out to me is the sample of clinicians and clinical teams.

"The team selected for participation in such a service is not selected at random and so does not represent clinicians in general."

This is so true, we are always advising people setting up new services to seek out "champions" to pilot a project and help to spread the word. These are often already library service users, and experts in evidence based medicine. This approach is partly to avoid wasted effort on the part of the Clinical Librarians, but perhaps those most in need of a service are the clinicians who don't know they want that service.

The different models of Clinical Librarians are discussed, and I identified with the introverted vs. extroverted librarians section. I think here at Leicester, we're all very much in the extrovert camp.

Andrew's final PICO is excellent.

"In a self-selecting or deliberately skewed population of clinicians is a range of activities loosely grouped together under the label of 'clinical librarian' better than doing nothing when measured across a selection of favourable and non-objective outcome measures, including those are difficult both to articulate and quantify?"

Of course! It does make me wonder how we ever managed to prove our worth at all. At UHL, we often feel we ought to be evaluating the service we provide, but never quite get around to it. I'm now wondering whether the excuse of being simpy far too busy providing the service is the only reason for our reluctance.

I'll certainly be interested to read the full systematic review when it's published, though I will admit I've had a couple of sneak previews, both at the HLG conference and the writing for publication course I attended earlier this month (more on that soon).

Updates to the UHL CL website

I've been very slack lately in keeping the Clinical Librarian Bibliography up to date, and I hadn't (until this morning) added links to the presentations from the Study Day in October either. However, I have done it now, and everything should be there to view.

I do worry about the bibliography, I wonder how useful it is, and if I am managing to capture everything. Thankfully, many people are happy to blow their own trumpet & let me know what they've published! I am grateful for that. I do have a lot of alerts set up through HDAS and PubMed, but I know I miss things.

At UHL we use articles about clinical librarianship as inspiration for what we could achieve, and as evidence to pass to our managers to prove our value. Do others do this?

We leads me to wonder if we should publish more, and if so, what? The day job is busy enough keeping on top of the searches and the current awareness, as well as the admin tasks we all have. Do we also have a responsibility to be writing and publishing and practising what we preach on evidence based practice?

Bibliography updated

At long last I've found the time to wade through all of my Pubmed search alerts, and my frankly quite flaky NHS HDAS search alerts & update the Clinical Librarian bibliography on our web site.

We also have a link to do a quick Pubmed search from the site, so even if I've been slack, you can see what's been published recently in the field. It's not the most sophisticated search, so if you have comments or suggestions, feel free to let us know!

http://www.uhl-library.nhs.uk/cl_research.html

Clinical Librarian Study again! And NEW on our web site

Don't forget about 8th May at George Eliot Hospital, Nuneaton.

http://www.uhl-library.nhs.uk/pdfs/cl_study_day_app_2009.pdf

Also, I spent Friday afternoon updating the Clinical Librarian bibliography. I'd be interested to hear of any unpublished studies that I can add to the list.

Always the last to know?

Am I the last to know that the National Research Register is dead? It's been superseded by the UK Clinical Research Network, which is still searchable, but registration doesn't appear to be a requirement for NHS organisations now. The link to search the new register is here: http://public.ukcrn.org.uk/search/

Hopefully the new register will be as useful as I found the old one to be.

The statement is here: http://www.nrr.nhs.uk/OfficialStatementToNRRProviders.pdf

It's still possible to search the archive of the NRR though.

NLH Document of the week

Taken from the National Library for Health "Document of the week"

Librarian support decreased the length of hospital stay.

Case control study looking at the effect of librarian support on patient care.
This case-control study found that during residents' morning report, where case discussion was taking place, supported by a librarian-guided, computerized literature search, the length of hospital stay for patients was reduced. The researchers concluded that "Morning report, in association with a computerized literature search guided by the librarians, was an effective means for introducing evidence-based medicine into patient care practices."
View full text
View print version
View supplementary material
Citation:Decreased hospital length of stay associated with presentation of cases at morning report with librarian supportDE Banks, R Shi, DF Timm, KA Christopher, DC Duggar, M Comegys, J McLartyJournal of the Medical Library Association, 2007, 95(4), pp371-373

I've just printed off & read this article through, using the CASP Critical Appraisal Checklist for Case-Control studies.

The study asked the question of whether a combination of Morning Report + librarian-provided literature search results altered length of stay, hospital charges (it's a US-based study) or readmission rates. Over 8 months, 105 patients were presented at Morning Report, and these were apparently random, as the doctors involved were given no directions over which patients to select. The comparisons were drawn from the hospital register of almost 20,000 patients admitted over the past 5yrs and 7months, and were matched by ICD9 diagnoses, age & secondary diagnoses. The results of the study show that the median length of stay was 2days in those patients presented at Morning Report (p=0.0238). The authors appear to have taken into account issues of selection bias, and the problems with matching all patients.

I've been thinking a bit more about this, and it is hard to tell which aspect of the intervention made the real difference - was it the extra interest from Morning Report, the literature search, or both? And were the cases presented naturally more interesting/complicated cases? If they were more complex conditions that stimulated interest, you'd expect more time to spent on the patient, possibly more diagnostic tests, and interventions. I suppose it all balances out, but it made me think about which clinical questions make it to the Clinical Librarian.

It's a eally interesting study, as we're always looking for ways that CLs make an impact on patient care, and it can be hard to define what it is we do. Often we're using anecdotal evidence to make a case for our necessity, and this article provides the kind of evidence that people should be listening to. What do you think?