Find It Yourself or Ask a Librarian? Studies show medical librarian search gets better results and reduces costs

keypadstethoscope

Health care providers are incorporating more evidence-based practice skills and information sources into their patient care. When providers recognize an uncertainty in determining a diagnosis or treatment plan, they frequently turn to available knowledge sources. At University of Vermont/Fletcher Allen Health Care many providers and professionals turn to Library–licensed sources such as UpToDate, DynaMed, and Nursing Reference Center as their first stop in finding the evidence to answer a patient care question. These sources helpfully provide the “strength of recommendation” based on accepted standards of appraisal of research studies. The typical duration of a search session on DynaMed or UpToDate is less than five minutes, which is manageable at or near the point of care. When those sources do not answer the question adequately either because the latest findings are too new, or because the patient situation and values are too complex, clinicians often seek the primary literature through PubMed. In doing so, some find the search frustrating and time-consuming. If you are among those who need a potentially time-consuming and complex search of the primary literature, consider requesting a literature search from Dana Library medical librarians.

In a randomized trial, information retrieval searches performed by a medical librarian for complex clinical questions were faster and retrieved more favorable results when compared to physician self-searches. The librarians answered the question in 13 minutes compared to 20 minutes for physicians searching on their own. The physicians reported that the librarian results contained a higher level of evidence and had a greater impact on patient care than physician self-searches.1

In a controlled study, patient cases were presented at morning report with a medical librarian in attendance. The librarian performed a literature search on questions that arose and disseminated the findings to the attending physician and presenting resident. The control patients were drawn from patient records and matched for age and primary and secondary diagnoses. The study results included association with reduced hospital length of stay (LOS) for the case group. LOS differed by 2 days between matched cases and controls (3 days vs. 5 days, P < 0.024). Median total hospital charges were $7,045 for the intervention group, and $10,663 for the control group. 2

In these studies, the librarian-conducted literature search saved physician time, reduced hospital costs, and may have improved patient outcomes. UVM and Fletcher Allen physicians, residents, nurses, therapists, and others may receive help finding the literature to answer clinical questions, develop guidelines, and do background research for presentations. To request a literature search or consultation, go to Ask a Librarian on the Dana website to get help by phone, email, or in person.

1. McGowan J, Hogg W, Campbell C, Rowan M. Just-in-time information improved decision-making in primary care: a randomized controlled trial. PLoS ONE. 2008;3(11):e3785.

2. Banks DE, Shi R, Timm DF, et al. Decreased hospital length of stay associated with presentation of cases at morning report with librarian support. J Med Libr Assoc. Oct 2007;95(4):381-387.

Marianne Burke, MLS AHIP
Director, Dana Medical Library
marianne.burke@uvm.edu