What do you do when a patient is SICK???

I remember my pediatric rotation in the third-year of medical school. It was lead by Steven Miller (who tragically died in a plane crash recently), and I did the rotation early in my introduction to clinical medicine--right out of classes and into the hospital wards. "The key skill any doctor must have," Dr. Miller said, "is to be able to identify which patients are SICK."

"Of course," you say, "all patients in the hospital are sick. "

But what Dr. Miller meant was, how do you recognize which patients are REALLY sick, that is, maybe-going-to-die-in-the-next-few-hours sick?

Well, it turns out that doctors are often not that good at identifying those SICK patients. Research going back into the 1980s suggests that patients who suffer cardiac arrest in ths hospital or require emergency transfer from the hospital ward to the ICU (intensive care unit) often show signs of getting worse for 6-8 hours prior to the catastrophe becoming obvious. It should be noted that this failure has been documented in the USA, the UK and Australia.

In the specialty of critical care medicine, we pride ourselves on being able to identify and stabilize these extremely sick patients--yanking them from the jaws of organ failure and death by pulling out our ventilators and dialysis machines and special catheters. In the last 2-3 years, there has been a boom of interest in our field to extend the specialty of critical care medicine beyond the ICU and onto the hospital wards, helping to diagnose and treat unstable patients at an early phase. Many hospitals have formed Medical Emergency Teams or Rapid Response Teams to do just this. The Institute for Healthcare Improvement (IHI) even advocates for these teams in its 100,000 lives campaign. The problem, as many noted researchers have pointed out (most recently in the Journal of the American Medical Association) is that there is very little evidence that these teams actually make a difference.

One main flaw may be in the way these teams are activated, or called to see a potentially ill patient. In most cases, the criteria for calling the critical care team for assistance are somewhat vague. To be sure, this may be by design: it is probably better to encourage people to ask for help than it is to put up too many barriers by creating elaborate criteria. Recent research has focused on adding up different variables of a patient's condition to create a score that will help predict which patients are going to get sicker.

The principle of using scores is described nicely in a recent article in the New Yorker by Atul Gawande. In the piece, he tracks the historical progression of the field of obstetrics and gynecology and its progress in lowering infant and maternal mortality rates associated with childbirth, surely one of the biggest triumphs in 20th Century medicine. The moral of the story is that medical care may be better when it provides "by the numbers" care, which flies in the face of the traditional "take care of every person as an individual." Read the piece and decide for yourself.