If you are very, very sick—so sick you have to be admitted to an intensive care unit, draped with tubes and IVs and catheters and sensors, fluids going in and coming out—what kind of physician should be at your side? An intensivist, of course. Many studies, including an exhaustive 2002 analysis in the Journal of the American Medical Association of the value of these specialists (pulmonologists, internists, and other physicians with specific training in critical-care medicine) concluded that the greater their role in the ICU, the lower the risk of death. This makes sense. Intensivists are experts in managing such fragile patients.
Then how to explain an unsettling study, out Tuesday in the Annals of Internal Medicine, that reaches exactly the opposite conclusion? The paper, which analyzed the care of more than 100,000 patients in 123 ICUs at U.S. hospitals, found that the chance of death was higher for ICU patients whose care was handled by critical-care specialists than for patients overseen by the admitting surgeon or other physicians lacking specialized training. Researchers adjusted for age, sex, severity of illness, length of stay prior to the ICU, and other factors. They call their results "surprising and completely contrary to previously published findings." No kidding.
Authors of studies that break new ground may suggest this or that possible cause, but they wouldn't dare to do more than toss out educated guesses. Going beyond that is up to the researchers who trail in their wake and will be looking for explanations (and holes in the analysis). The two authors of an accompanying editorial opined that "[The] study will remain one observation against many" until it is backed up with further data. (One of the two was an author of the 2002 JAMA study.)
That's a completely reasonable point, but a couple of possible explanations suggested in the Annals study rang a familiar bell. Last week, I wrote about new hospital rankings from Consumer Reports that are based on how many days Medicare patients with certain conditions spent in the hospital during the two years before they died and how many times they were visited by physicians while they were there. Lots of care not only wastes dollars and resources—it isn't always better for patients. It can even increase the possibility that they will die.
Well, no other part of a hospital, other than the ER and neonatal unit, brings more of the full force of medical care to bear than an ICU. Perhaps, the study authors propose, "cumulative use of more interventions may take a negative toll." Specialists in intensive care, "because of their familiarity and expertise with procedures...may use more procedures that subsequently lead to more complications. Their use of more procedures, such as placement of catheters and other invasive devices, may make critically ill patients more susceptible to life-threatening infections."
That might turn out to be flat wrong or right on, but the upside-down results of this study are too important not to follow up. When discussing hospital quality and safety, it is vital to keep asking: What do we know? What, on the other hand, do we only think we know?