Abstract & Commentary

Academic ICU Attending Workload: Potential Effects on Teaching, Patient Care, and Workforce Stability

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

SYNOPSIS: This survey of PCCM fellowship program directors found that attending workload was high and included responsibilities beyond the ICU itself. Those whose patient-to-attending census was above the study's median reported both increased stress and difficulty fulfilling teaching responsibilities.

SOURCE: Ward NS, et al. Perceived effects of attending physician workload in academic medical intensive care units: A national survey of training program directors. Crit Care Med 2011; Oct 13. [Epub ahead of print]

This article reports findings of a task force set up by the Society of Critical Care Medicine in 2008 to study intensivist workforce issues in the face of the ongoing increase in ICU beds and numbers of critically ill patients. The authors surveyed the members of the Association of Pulmonary and Critical Care Medicine (PCCM) program directors who worked in closed ICUs about attending physician workload and perceptions. Of 121 potentially eligible program directors (the number of accredited PCCM fellowship training programs), 84 responded (69%), of whom 6 did not work in closed ICUs, leaving the 78 surveys whose data are reported in the article. Respondents' institutions were widely distributed throughout the United States.

Average daily census (as reported by the respondents) was 18.8 ± 10.1 (SD) patients, and average maximum service size recalled was 24.1 ± 9.9 patients. Twenty-one respondents (27%) said that their unit had no policy setting an upper limit for service size. The average census size was "too many" patients as reported by 22 (28%) program directors, while 55 (71%) felt that the maximum service size was "too many." One-third of the respondents said their ICU service covered patients outside the physical ICU, and all respondents reported having additional hospital duties outside the ICU such as code team, rapid response team, or triaging. More than half had two of these additional responsibilities and 22% had three. Of the responding programs, 17 (22%) reported using nurse practitioners or physician assistants in addition to residents and fellows to staff their units.

From their data, the authors determined the median number of patients per attending physician to be 13 (interquartile range, 10-16). Using this median value as a cutoff, the 31 respondents with higher patient-to-attending ratios perceived significantly more time constraints, more stress, and more difficulties fulfilling their teaching role with trainees than the 40 respondents with lower patient-to-attending ratios. The total number of trainees and physician extenders per patient was the same in higher and lower patient-to-attending units, suggesting that having more non-attending physician staff does not ameliorate perceived overwork and stress on the part of the attending.


This study determined only the respondents' recalled estimates of patient numbers and other workload features, rather than actual counts. Further, it included only what the PCCM program directors reported, and thus did not include reports from other attending physicians in the programs or conditions in the ICUs of other hospitals in multi-institution training programs. Nonetheless, the findings are consistent with my experience as an academic intensivist, not only from my own attending activities but also from many discussions with physicians in comparable positions around the country. Patient loads and other responsibilities on the attending physicians in many PCCM training programs are too high for optimal patient care, teaching activities, and personal satisfaction.

The subjects in this study were asked about recent and planned ICU growth in their institutions: A third reported increases in ICU size during the previous 5 years (by an average of 2.8 beds), and 38% indicated that increases were planned (by an average of 6.3 beds) during the next 5 years. This study was carried out in the era of resident work hour limitations, but before the most recent round of more stringent limitations was enacted. This, plus the current requirements for teaching physician documentation in the context of the electronic medical record and the additional attending responsibilities documented, can hardly be expected to ameliorate the perceived attending workload and stress reported in this article.