Benefits of a Dedicated ICU Clinical Pharmacist

Abstract & Commentary

By David J. Pierson, MD, Editor, Professor, Pulmonary and Critical Care Medicine, Harborview Medical Center, University of Washington, Seattle, is Editor for Critical Care Alert.

Synopsis: In this large epidemiology study using a previous survey and 2004 Medicare data focusing on serious infections in the ICU, hospitals with dedicated ICU clinical pharmacists had lower ICU mortality rates, shorter ICU stays, and reduced charges.

Source: MacLaren R, et al. Clinical and economic outcomes of involving pharmacists in the direct care of critically ill patients with infections. Crit Care Med 2008;36:3184-3189.

In 2004, the authors of this study conducted a survey of U.S. hospitals with ICUs to assess the prevalence and use of dedicated ICU clinical pharmacists.1 That survey indicated that 62% of hospitals had clinical pharmacists with at least a portion of their full-time equivalent position specifically dedicated to direct involvement in patient care in the ICU rather than drug dispensing and other more traditional pharmacist roles. For the current paper, the authors used the results of that survey, along with ICD-9 diagnostic data, mortality, lengths of stay, total charges, drug charges, and laboratory charges obtained from the Expanded Modified Medicare Provider Analysis and Review (MEDPAR-Hospital-National) for the year 2004, to examine associations between having an ICU clinical pharmacist and those variables for 3 categories of ICU infections: nosocomial-acquired infections, community-acquired infections, and sepsis.

Because of the nature of the databases used, the numbers of hospitals and patients, as well as of variables such as case mix, varied for each infection category. Of the 382 institutions responding to the original survey, 272 had Medicare patients with nosocomial infections; the corresponding numbers for community-acquired infections and sepsis were 265 and 276 hospitals, respectively. Numbers of patients in the different categories ranged from 8927 (community-acquired infections) to 54,042 (sepsis). In each instance, the studied outcomes were better in hospitals with ICU clinical pharmacists than in institutions without pharmacists in this role. For nosocomial infections, community-acquired infections, and sepsis, respectively, hospital mortality rates with and without ICU clinical pharmacists were 14.61% vs 18.05%, 11.43% vs 13.28%, and 18.54% vs 19.43%, all statistically significant with P values of 0.008 or less. Compared to ICUs with clinical pharmacists, mortality rates in ICUs without them were 23.6% higher for nosocomial infections (386 extra deaths), 16.2% higher for community-acquired infections (74 extra deaths), and 4.8% higher for sepsis (211 extra deaths).

ICU lengths of stay were longer in hospitals without ICU pharmacists, by 7.9% (14,248 extra days), 5.9% (2855 extra days), and 8.1% (19,215 extra days), for the 3 infection categories, respectively (all differences significant; at least P = 0.03). ICUs that did not have dedicated clinical pharmacists had greater total Medicare billings: by 12% for nosocomial infections, by 11.9% for community-acquired infections, and by 12.9% for sepsis (all, P < 0.001). Differences for Medicare drug and laboratory charges were similar. The authors conclude that, if these results "were extrapolated to all 933,638 Medicare patients in an ICU with the studied infections, the involvement of a clinical pharmacist could save 7409 patient lives, 390,921 ICU days, and $4,168,278,242 in total charges."


This study has some important limitations, the most important of which in my opinion relates to the 2004 survey1 from which the participating institutions for the present investigation were selected. In that survey, only 382 (11.8%) of 3238 U.S. hospitals with ICUs (1034 ICUs) responded, and these institutions had some potentially relevant differences from the non-responding hospitals. Compared to the 88% of institutions that did not respond to the survey, more responding hospitals were not-for-profit, non-governmental hospitals, and fewer were in all the other categories (government, for-profit, and so on). In the responding hospitals, 52% of the ICUs were open (patient managed by private non-intensivist attending), 28% were transitional (patient co-managed by private attending and intensivist), and 20% were closed (patient managed by intensivist). Thus, whether the present study's results apply to a particular practice environment is unclear, and extrapolating the findings to include all U.S. hospitals in 2009 — and especially the authors' generalizations about lives and money saved — seems pretty dubious.

Previous studies have shown that hospital mortality correlates inversely with the ratio of pharmacists to occupied beds, and that involving pharmacists directly in the care of ICU patients is associated with fewer adverse drug-related events, greater efficiency of care, and lower drug-related costs. Using hospitals with and without ICU pharmacists as identified from a previous survey, the present study documents positive associations between having ICU clinical pharmacists and patient mortality, ICU length of stay, total charges, drug charges, and laboratory charges.

I think it is important to phrase this study's findings in this way, because the authors have not demonstrated that having an ICU pharmacist per se reduces mortality, length of stay, and charges. Whether they have dedicated ICU clinical pharmacists is likely only one of many ways in which the study hospitals differ. A number of aspects of the process of care in the ICU are similarly associated with improved patient outcomes. For example, hospitals with ICU pharmacists are probably also more likely also to have closed ICUs, trained intensivists, greater implementation of protocols and care bundles, multidisciplinary rounds, palliative care services, and other dedicated ICU personnel such as specialist respiratory therapists, nutritionists, and social workers.

Having said that, I am convinced that MacLaren and colleagues are justified in their recommendation that "hospitals should consider employing clinical ICU pharmacists." For decades, the institution in which I work has been fortunate to have specialist clinical pharmacists assigned to each of its main ICUs — including medical, trauma/surgical, and neurosurgical units. The presence of these experts on daily work rounds has improved adherence to clinical practice guidelines and unit protocols, facilitated matching of antibiotic therapy to local microbial susceptibility patterns, identified and prevented adverse drug reactions and interactions, and aided in efforts to prevent oversedation and drug withdrawal. And the educational impact of our ICU clinical pharmacists for physicians, nurses, and others on the staff has been immeasurable.


  1. MacLaren R, et al. Critical care pharmacy services in United States hospitals. Ann Pharmacother 2006;40:612-618.