Entries in Searching (17)


The Role of Librarians and Informaticians on the Health Care Team

The author sets the stage nicely in this editorial piece in Academic Medicine:


Today, more information is available than ever before, and medicine has become deeply complex. Most physicians would say they are proficient at using the Internet and bibliographic databases to find information to inform decisions, but there is significant evidence to the contrary. Research continues to show a disconnect between the information physicians need and the information they retrieve: Physicians have two questions for every three patients but use information retrieval systems fewer than nine times per month; most physician-conducted searches retrieve only 25% to 50% of relevant articles; doctors have difficulty reading and interpreting medical research; physicians may overgeneralize findings and may incorrectly apply those findings in clinical practice; adverse clinical outcomes have arisen from these difficulties in interpreting and applying research evidence; and, when physicians try to answer questions, they are almost as likely to make an incorrect conclusion as a correct conclusion.

Because physicians report there is too much information and it takes too long to search for the correct answer, they often “satisfice,” or take the fastest, easiest answer when faced with uncertainty. This is unfortunate, as Weightman and Williamson's systematic review on the impact of library services on health outcomes indicated that when health care providers used professional library services, the result was improved general patient care, diagnosis, choice of tests, choice of therapy, and reduced stay. Clearly educating physicians in how to appropriately use information resources could lead to improved health outcomes.


Finding Evidence for Comparing Medical Interventions

One of the AHRQ's key findings:

"A librarian or other expert searcher should be involved in the development of the search."

No guff.  During my recent search talk for a meta analysis course, I encouraged those in attendance to pass the search responsibilities on to someone who knows what they're doing.  Seems to me a strong search would strengthen the methodology and, by extension, the analysis.    



Constructing Systematic Review Searches

Last week I gave the following presentation (minus animations) as part of an intro to meta analyses course.  I spoke for about an hour and a half and gave a very short live demo of how I'd go about building and validating a systematic review search.  It seemed to be well received.



Do We Need To Teach Search Differently?

Nice study on assessing EBM searching instruction outcomes.  From the conclusion:


It is disappointing that the level of instruction received in undergraduate training has no clear correlation to the current skills of participants in this study, and that little of that training has been retained.


Although just one study with a small sample size, this finding doesn't surprise me.  For the majority of graduating medical students to develop a level of search expertise over the course of their medical training, they need to be hit with clinical scenarios that challenge their search presumptions frequently.  On top of that, context, too, is, I think, important.  Challenge the students search abilities in a situation that's as close to the 'real world' as possible.  Repitition and context. 

Beyond that, what else can librarians do to better learning outcomes?


How Much Searching Is Enough?

When I'm working on time-consuming searches (such as systematic reviews), I'll generally stop after a) ensuring the known relevant articles are captured by the strategy, b) making a concerted to break the strategy (by finding relevant articles that aren't captured), and c) coming to some sort of understanding with the lead investigator.

It turns out there are, according to PMID 20923586, at least 8 methods to indicate when to call it quits (from the abstract).

  • Capture-recapture technique
  • Obtaining Feedback
  • Seeking the Disconfirming case
  • Undertaking comparison against a known Gold standard
  • Evaluating retrieval of Known items
  • Recognizing the Law of diminishing returns
  • Specifying a priori Stopping rules
  • Identifying a point of Theoretical saturation

 Who knew?