By Drayton Hammond, PharmD, MBA, BCPS, BCCCP

Clinical Pharmacy Specialist, Adult Critical Care, Rush University Medical Center, Chicago

Dr. Hammond reports no financial relationships relevant to this field of study.

SYNOPSIS: Nurses with relatively high autonomy alongside a dedicated intensive care unit (ICU) clinical pharmacist and 24/7 intensivist coverage were associated with the lowest hospital mortality and shortest ICU lengths of stay and mechanical ventilator durations compared to other staffing models.

SOURCE: Zampieri FG, Salluh JIF, Azevedo LCP, et al. ICU staffing feature phenotypes and their relationship with patients’ outcomes: An unsupervised machine learning analysis. Intensive Care Med 2019;45:1599-1607.

The complexity of providing excellent care for critically ill patients continues to increase alongside advances in technology, changing organizational factors in healthcare, and a consistent stream of published literature. Although the proper distribution of intensivists to patients throughout a day remains an area of study,1,2 the presence and autonomy of support staff (e.g., pharmacists, nurses, and respiratory therapists) has become an area of interest for most intensive care units (ICUs). The investigators of the ORganizational CHaractEriSTics in cRitical cAre (ORCHESTRA) study evaluated whether ICU staffing features were associated with improved patient outcomes using cluster analysis directed by machine learning.3

The ORCHESTRA investigators performed a retrospective analysis of prospectively collected data on consecutive adults admitted to 93 medical-surgical ICUs in Brazil during 2014-2015 to determine clinical outcomes.3 Additionally, cross-sectional surveys and interviews of ICU directors and/or chief nurses were performed to elicit the presence and extent to which hospital and ICU organizational, structural, and process characteristics were available and used. Nonphysician staff members’ autonomy was assessed using surveys (seven items, with scores ranging from 0 [low] to 14 [high]). Three ICU clusters were developed based on ICU characteristics. The defining characteristics of each cluster were: no board-certified intensivist in the ICU 24/7, no dedicated ICU pharmacist, and low nurse autonomy score (Cluster 1); no board-certified intensivist in the ICU 24/7, dedicated ICU pharmacist, and moderate nurse autonomy score (Cluster 2); and board-certified intensivist in the ICU 24/7, dedicated ICU pharmacist, and high nurse autonomy score (Cluster 3). The best patient outcomes were observed in Cluster 3: lower adjusted hospital mortality (odds ratio, 0.92; 95% confidence interval [CI], 0.87-0.98), shorter ICU length of stay (subhazard ratio, 1.24; 95% CI, 1.22-1.26), and shorter mechanical ventilation duration (subhazard ratio, 1.61; 95% CI, 1.54-1.69). The worst outcomes were observed in Cluster 1.


The investigators described the impact of combinations of various organizational and personnel factors rather than the isolated effects from a specific, singular variable. In other studies, researchers have observed improved outcomes with nurse-led interventions, including improved mechanical ventilator weaning, minimization of sedation with better patient comfort, earlier goal enteral feeding, and a greater culture of safety.4-7

In this study, the mean nursing autonomy score was 9.62 (out of 14), which suggests that many opportunities still exist for increased nursing autonomy and potentially improved patient outcomes. However, nursing leadership must be supported by physician collaboration and adequate education. Similarly, improved patient outcomes and avoidance of healthcare costs have been realized with dedicated ICU clinical pharmacists.8-10 Whereas an ideal patient-to-nurse ratio in the ICU of 1:2 has been established, the ideal pharmacist-to-patient ratio in the ICU remains controversial, with 1:12 to 1:28.11 This study suggests there should be a dedicated ICU pharmacist for each clinical service providing ICU-level care to support that team’s efforts to improve patient outcomes. Lastly, the importance of intensivist coverage alongside these other personnel decisions is paramount. The complement of healthcare professionals in Cluster 3 represents a desirable and achievable staffing model that has been shown to improve patient outcomes.


  1. Wilcox ME, Chong CA, Niven DJ, et al. Do intensivist staffing patterns influence hospital mortality following ICU admission? A systematic review and meta-analyses. Crit Care Med 2013;41:2253-2274.
  2. Kerlin MP, Adhikari NK, Rose L, et al. An official American Thoracic Society systematic review: The effect of nighttime intensivist staffing on mortality and length of stay among intensive care unit patients. Am J Respir Crit Care Med 2017;195:383-393.
  3. Zampieri FG, Salluh JIF, Azevedo LCP, et al. ICU staffing feature phenotypes and their relationship with patients’ outcomes: An unsupervised machine learning analysis. Intensive Care Med 2019;45:1599-1607.
  4. Hirzallah FM, Alkaissi A, do Céu Barbieri-Figueiredo M. A systematic review of nurse-led weaning protocol for mechanically ventilated adult patients. Nurs Crit Care 2019;24:89-96.
  5. Larson GE, McKeever S. Nurse titrated analgesia and sedation in intensive care increases the frequency of comfort assessment and reduces midazolam use in paediatric patients following cardiac surgery. Aust Crit Care 2018;31:31-36.
  6. Armellino D, Quinn Griffin MT, Fitzpatrick JJ. Structural empowerment and patient safety culture among registered nurses working in adult critical care units. J Nurs Manag 2010;18:796-803.
  7. Orinovsky I, Raizman E. Improvement of nutritional intake in intensive care unit patients via a nurse-led enteral nutrition feeding protocol. Crit Care Nurse 2018;38:38-44.
  8. MacLaren R, Bond CA, Martin SJ, Fike D. Clinical and economic outcomes of involving pharmacists in the direct care of critically ill patients with infections. Crit Care Med 2008;36:3184-3189.
  9. MacLaren R, Bond CA. Effects of pharmacist participation in intensive care units on clinical and economic outcomes of critically ill patients with thromboembolic or infarction-related events. Pharmacotherapy 2009;29:761-768.
  10. Hammond DA, Flowers HJC, Meena N, et al. Cost avoidance associated with clinical pharmacist presence in a medical intensive care unit. J Am Coll Clin Pharm 2019;2:610-615.
  11. Erstad BL, Mann HJ, Weber RJ. Developing a business plan for critical care pharmacy services. Hosp Pharm 2016;51:856-862.