Skip navigation
Link to Back issues listing | Back Issue Listing with content Index | Subject Index

Physician staffing and ICU outcomes

Systematic review
Results
Hospital mortality
ICU mortality
Hospital length of stay
ICU length of stay
Comment

A recent theme in Bandolier has been reporting studies that looked at staffing levels and outcomes for nurses ( Bandolier 103 and 106 ). Both showed that lower nurse staffing levels generally and in the intensive care unit (ICU) led to higher patient mortality, among other outcomes. A new systematic review on physician staffing in the ICU confirms that higher intensity staffing reduces mortality [1].

Systematic review

The review used a very extensive search strategy for randomised and observational studies of critically ill adults or children and ICU physician staffing strategies. Staffing strategies were grouped into high intensity (mandatory consultations with intensive care physician, or closed care units where all care was directed by specialist intensive care physicians), or low intensity (no consultation with intensive care specialist or only elective consultation). Outcomes sought were hospital and ICU mortality and length of stay. Data extraction from reviewed studies was by intensive care physicians with formal training in clinical epidemiology.

Results

The final selection was of 26 studies, 16 reporting hospital mortality, 14 ICU mortality, 13 hospital length of stay and 18 ICU length of stay. All were observational studies, seven with concurrent and 19 with historical controls. Most studies were from North America. The number of ICUs studied was one in 20 studies and more than one (up to 42) in six. Twenty-five of the studies compared high intensity with low intensity physician staffing.

Hospital mortality

Results for hospital mortality are shown in Figure 1, where 16 of 17 studies showed a decrease in hospital mortality rate for ICU patients with high intensity physician staffing. In 10 studies the reduction was statistically significant. Overall 16% (859/5294) of patients died with high intensity staffing compared with 27% (943/3496) with low intensity staffing. The relative risk was 0.7 (0.6 to 0.8). The number needed to treat to prevent one additional death by using high rather than low ICU physician staffing was 9 (95% confidence interval 8 to 11).

Figure 1: Effect of intensity of physician staffing on hospital mortality of critically ill patients



ICU mortality

Results for ICU mortality are shown in Figure 2, where 14 of 15 studies showed a decrease in ICU mortality rate for ICU patients with high intensity physician staffing. In nine studies the reduction was statistically significant. Overall 10% (593/5703) of patients died with high intensity staffing compared with 14% (824/6077) with low intensity staffing. The relative risk was 0.6 (0.5 to 0.8). The number needed to treat to prevent one additional death by using high rather than low ICU physician staffing was 32 (95% confidence interval 23 to 50).

Figure 2: Effect of intensity of physician staffing on ICU mortality of critically ill patients



Hospital length of stay

Ten of 13 studies reported a reduction in hospital length of stay with high intensity physician staffing. In six studies the reduction was statistically significant. Overall, the weighted mean length of hospital stay in 5,083 patients in ICUs with high-intensity was 13.4 days, and in 3,383 patients in ICUs with low intensity staffing it was 15.4 days.

ICU length of stay

Fourteen of 18 studies reported a reduction in ICU length of stay with high intensity physician staffing. In 11 studies the reduction was statistically significant. Overall, the weighted mean length of ICU stay in 7,654 patients in ICUs with high-intensity was 4.4 days, and in 5,865 patients in ICUs with low intensity staffing it was 4.9 days.

Comment

The results were consistent. High intensity physician staffing in ICUs resulted in lower mortality and shorter hospital and ICU stays. That should reassure patients and professionals in the UK, because an Audit Commission survey in 1999 reported that high intensity physician staffing was the norm in at least 80% of ICUs in England and Wales.

That doesn't make it less interesting or important. As well as confirming other studies showing that more intensive staffing produces higher quality of care, this is ideal fodder for health economists to examine the cost-effectiveness. Saving half an ICU bed-day and two hospital days for each critically ill patient has major implications in cost and efficiency.

References:

  1. PJ Pronovost et al. Physician staffing patterns and clinical outcomes in critically ill patients. JAMA 2002 288: 2151-2162.
previous or next story in this issue