In an attempt to demonstrate the effect of a clinical medical librarian (CML) on patient outcome, the authors studied a group of patients at the Louisiana Health Shreveport hospital over approximately 2 years. All subjects were patients who spent over 24 hours in the hospital. The CML (already employed by the hospital) participated in daily clinical rounds and served as an immediate or timely source of information for the physicians, medical students, and other members of the staff or school that participated in rounds. The CML was expected to review each patient’s medical record prior to rounds. While the CML participated in only the rounds of certain patients, all physicians and students were able to make use of the Medical Library at the associated School of Medicine.
Comparisons between patient outcome associated with CML input (intervention group) and a control group were achieved in two ways. While all members of the group that had a CML in attendance were eligible for inclusion in the Intervention Group, only those who asked a question of the CML were included – the rest of the members who had access to a CML and did not ask a question were combined with those who did not have access as the Control Group. This initial comparison showed that those patients who had medical teams taking advantage of a CML had longer length of stay, higher costs, and higher readmission rates (but also were “more ill” than those in the control group, according to a matched analysis that aimed to compare patients in the Intervention Group to those in the Control Group with respect to initial diagnosis, age, secondary diagnoses, etc.).
The second comparison was performed using a matched pair analysis, in an attempt to remove those factors that would skew results (by virtue of the Intervention Group being more ill at the outset). Members of the Intervention Group were paired with aggregates of Control Group members who shared similar initial illness levels. These groups showed no statistically significant differences with respect to hospital stay, hospital cost, total number of diagnostic codes, or mortality while hospitalized (While Intervention Group mortality rates were about half that of the Control Group, 1.5% compared to 3.8%, the difference was not statistically significant). Interestingly, the intervention group had a much higher rate of readmission within 30 days than those of the control group.
The authors note several problems in studying clinical librarianship, and with this article in particular. There is considerable difference between the two groups being studied (especially with respect to level of illness), and it is difficult to make meaningful comparisons. Additionally, this study only compared groups in one hospital, with one CML. The authors emphasize the need for future studies using large sample sizes and more sophisticated analysis.
If nothing else, this article reiterates how important it is for librarians to always be mindful of their perceived value. As the authors noted, this study had many more limitations than only those associated with study group differences. But how might this study impact the hospital’s view of their CML? It is difficult to assert your worth when statistical studies do not support it.
Esparza JM, Shi R, McLarty J, Comegys M, Banks DE. The effect of a clinical medical librarian on in-patient care outcomes. J Med Libr Assoc. 2013; 101(3): 185-191.