Lower Death Rate When ICUs Limit Care to Intensivists

SEATTLE, Oct. 1 -- Patients with acute lung injury are less likely to die if they're treated in closed intensive care units that permit care to be directed only by intensivists, researchers here said.

In a 24-center prospective cohort study in King County, mortality was a third lower when patient care was by intensivist physicians, compared with the so-called open model, where any attending physician was allowed to direct care, found Miriam Treggiari, M.D., of the University of Washington, and colleagues.

The findings support the idea that all intensive care units should follow the closed model, in which patients must be managed or co-managed by an intensivist, Dr. Treggiari and colleagues concluded in the first issue for October of the American Journal of Respiratory and Critical Care Medicine.

Previous studies have suggested such a benefit, the researchers said, but they have usually been limited single-center academic hospitals.

In contrast, the King Country Lung Project -- a population-based cohort of patients with acute lung injury -- included patients in 24 adult intensive care units in hospitals in and around Seattle.

Of those, the researchers said, 13 were defined as closed and 11 as open.

From April, 1999 through July 2000, Dr. Treggiari and colleagues found 1,075 patients with acute lung injury were treated in the 24 units, 684 of them on closed units.

Acute lung injury was defined as acute hypoxemia, bilateral infiltrates consistent with pulmonary edema, and no clinical evidence of left atrial hypertension.

The researchers collected data on co-morbidities, length of hospital stay, and discharge information. Also, questionnaires were sent to medical directors, attending physicians, and nurse managers to obtain data on the organizational model and structure of each unit.

Complete data were available for 23 units, the researchers said.

After adjustment for confounding factors, the researcher found those treated at closed ICUs had significantly lower mortality. The odds ratio was 0.68, with a 95% confidence interval from 0.53 to 0.89 -- significant at P<0.004.

Mortality was 35% in closed units, compared to 45% in open units, the researchers found.

More than half of the patients were seen by a pulmonary consultant at some point, but the proportion was significantly higher (at P<0.01) in the closed units -- 77% versus 68%.

However, consultation by a pulmonologist in the open ICUs was not associated with a lower risk of death. The odds ratio was 0.94; with a 95% confidence interval from 0.74 to 1.20. The P-value was 0.62.

While closed units had higher levels of staffing, the researchers said, factoring that into the analysis did not account for the observed lower mortality.

The study is observational, Dr. Treggiari and colleagues said, and runs the risk of a range of biases, including bias by indication, residual confounding, and measurement error.

They wrote that "although our regression and sensitivity analyses support the conclusion that it is the intensivist staffing model that accounts for the observed mortality effect, the observational nature of this study limits our ability to exclude the contribution of other features of these units."

They added that because of the small sample size and the date of the data collection, it's hard to extrapolate the findings to other parts of the U.S.