Nurse Practitioners and Physician Assistants in the ICU
By Betty T. Tran, MD, MSc
Assistant Professor of Medicine, Pulmonary and Critical Care Medicine, Rush University Medical Center, Chicago
Dr. Tran reports no financial relationships relevant to this field of study.
SYNOPSIS: Using retrospective cohort data from a national survey of medical and mixed medical-surgical ICUs, this study found that nurse practitioner (NP)/physician assistant (PA) staffing was common and not associated with any differences in in-hospital mortality compared to ICUs without NP/PAs.
SOURCE: Costa DK, et al. Nurse practitioner/physician assistant staffing and critical care mortality. Chest 2014; Aug 28. [Epub ahead of print.]
With increasing restrictions on resident work hours and the rising demand for critical care, hospitals have turned to alternative models to staff the ICU. The number of nurse practitioners (NPs) and physician assistants (PAs) has grown to meet this need, although their impact on patient outcomes is not well understood. The investigators of this study sought to examine the effect of NP/PA staffing on in-hospital mortality among adult ICU patients.
Costa et al used retrospective ICU outcome data from the Acute Physiology and Chronic Health Evaluation (APACHE) clinical information system combined with an ICU-level survey of organizational practices during the period of 2009-2010. Their final sample included 39,541 adult patients in 29 medical and mixed medical-surgical ICUs within 22 hospitals. NP/PA staffing was reported in 21 (72.4%) ICUs; 30,254 (76.5%) patients received care in ICUs with NP/PAs. There were no significant organizational differences found between ICUs with and without NP/PA participation. In terms of patient-level characteristics, ICUs with NP/PA staffing tended to have lower severity of illness based on acute physiology scores (APS) and frequency of mechanical ventilation, and there were small but significant differences in terms of age, race, primary diagnosis, and comorbidities. Risk-adjusted in-hospital mortality, however, was similar between ICUs with and without NP/PA staffing (relative risk, 1.10; 95% CI, 0.92-1.31). Similar findings were observed in subgroups of patients on mechanical ventilation, in the highest quartile of APS, in ICUs with low-intensity physician staffing (i.e., optional intensivist physician consult or absence of intensivists), and in ICUs with physician trainees. A sensitivity analysis that considered patients discharged to hospice as alive upon discharge also yielded comparable results. Overall, there were no significant differences observed with regard to discharge location between patients staffed by NP/PAs and those not staffed by NP/PAs.
This study adds to a currently small number of studies focused on the impact of NP/PA care in acute and critical care settings. As NPs and PAs are under the direct supervision of attending physicians, the finding of no difference with regard to in-hospital mortality should not be surprising even though their educational training routes are different from physician trainees. In fact, differences in training backgrounds may account for the findings observed in studies that have compared NP/PA with physician (resident or fellow) care. One study found that residents cared for more patients, worked more hours, and spent more time in lectures/conferences, although NP/PAs were more likely to discuss patient care with ICU nurses, to interact with patients’ families, and spend more time in research and administrative activities.1 Similarly, fellows spent more time in non-unit activities (conferences, reading, teaching) whereas NPs/PAs spent more time in activities related to coordination of care, interactions with nursing, medical staff, patients, and family members.2 Among these and other studies examining outcomes associated with NP/PA staffing, there were no significant differences in patient outcomes such as readmission rates, mortality, duration of mechanical ventilation, LOS, or disposition; some studies even reported an improvement in certain outcomes.3
Together, these findings not only support the use of NPs and PAs in the care of critically ill patients, but also suggest they may be a desirable complement to traditional care models. Unlike residents and fellows, they do not rotate off service and can potentially provide greater continuity of care. Furthermore, they can develop more clinical experience and comfort the longer they work in their respective positions compared to rotating physician trainees who change on a monthly basis. Indeed, it would be interesting to repeat these studies in the future as the current group of NP/PAs continues to accumulate years of critical care experience to assess whether there are significant differences in not only patient outcomes, but quality of care and patient satisfaction.
- Rudy EB, et al. Care activities and outcomes of patients cared for by acute care nurse practitioners, physician assistants, and resident physicians: A comparison. Am J Crit Care 1998;7:267-281.
- Hoffman LA, et al. Management of patients in the intensive care unit: Comparison via work sampling analysis of an acute care nurse practitioner and physicians in training. Am J Crit Care 2003;12:436-443.
- Kleinpell RM, et al. Nurse practitioners and physician assistants in the intensive care unit: An evidence-based review. Crit Care Med 2008;36:2888-2897.