As a newly minted medical resident I wrote orders for antibiotics and cultures for a sick patient with pneumonia.  That night I happened to be on call.  I was chagrined when nurses awoke me at 2 AM to ask about something they had trouble deciphering. They had just gotten to taking off orders from 17 hours ago in this busy hospital. My sick patient hadn’t gotten his antibiotics. . This delay in treatment might have cost the patient’s life.  I was livid. But the fault lay partly in my own inexperience. I should have talked to his nurse and made sure therapy was started when I saw him.

Our federal government and its medical arm,  called The Joint Commission (TJC) now sends into our hospitals nurses with carrots and sticks. Each hospital will keep its own statistics regularly polished to be presented to these nurse enforcers (some few of the inspectors are elder physicians) in various disease categories such as heart failure, stroke, and diabetes. Mostly hospitals covet certification or maintenance of certification for “excellence” for as many disease entities as possible.   That has given rise to a whole infrastructure of constant inspection.  TJC monitors adherence to protocols.

Now comes a large study on Sepsis in the New England Journal of Medicine from May 1, 2014, the large ProCESS study (an acronym)  and an accompanying editorial. What I love about NEJM is how frequently things thought or assumed to be true, are not true, when you test them out. Every scientist knows that truth is often counterintuitive. Assumption has to be corrected with a test.

Sepsis means a severe infection, usually bacterial, has hijacked body defenses and an often lethal inflammatory cascade is set in motion. The patient is seriously sick with infection and, as we say, carries a, “high risk of dying”, around 20%.  But sepsis is a particularly hard to diagnose with certainty. It is not the same as being able to grow bacteria from a blood culture, what is called bacteremia,. One problem is that with the patient already getting antibiotics, bacteria are hard to culture. Also bacteria are only intermittently released into the bloodstream in these very sick patients.  You have to be lucky enough to draw blood at the right moment, usually as the temperature is heading skyward.

In this particular study,  1341 patients in septic shock in 31 emergency rooms were assigned randomly to one of three groups: one with a strict septic shock protocol, another protocol based standard therapy group that did not actually require some invasive measures but permitted them, and a third group that was not protocol driven called usual care. They looked primarily at whether the patient lived or died in 60 days. Though protocols increased the use of invasive measures such as central lines, there was no difference in mortality at 60 days, 90 days of at one year, or of organ failure or any other meaningful marker of good or bad care. That meant that protocols were not found to save lives or improve care.

We all assume that a seriously ill patient should be treated according to guidelines and we want to provide excellent care. Spurred by the tragic loss  in New York of a 12 year old boy of unrecognized sepsis “Rory’s Regulations” were adopted.  A  whole scaffolding of protocol adherence and outcomes reporting was erected. But the burning question is whether this is actually good for patients. So far as sepsis care is concerned, some past studies, not as large or rigorously done, showed some efficacy. Since sepsis is common, this was worth looking at again.

One result of introducing and policing protocols for any disease whether diabetes, sepsis or stroke, is raising physician awareness of that disease and improving compliance, so that is theoretically good. But even looking a simple  goal such as promptly delivering antibiotics to a patient determined to have pneumonia, nurses and doctors know you are watching. That influences diagnosis and care. A hospital aware of pneumonia policing will avoid missing the diagnosis at all costs. Everyone with any sign, for instance fever, might be presumed to have pneumonia,  and started on antibiotics immediately. The consequences of not treating means the doctor and hospital will be marked by the federal enforcers, TJC.  What then occurs is many patients who don’t have pneumonia are misdiagnosed or assumed to have it.

The following occurs: 1. Lots of folks with mild things, maybe fever alone are treated for pneumonia. 2. People with minor problems are lumped into the category of pneumonia. 3. People who have other diseases are thought to have pneumonia increasing the chance of misdiagnosis, maybe a urinary tract infection or meningitis or other disease.  Misdiagnosed patients may suffer dire consequences.  4. Antibiotics, often wrong antibiotics, or broad spectrum antibiotics, are overprescribed, which leads to bad things, especially antibiotic resistance. 5. Because pneumonia has a high mortality and milder cases without pneumonia are lumped in, the mortality rate for so-called pneumonia drops and it looks like the protocol works, maybe when it doesn’t work.  These are not my ideas. They were underlined by the accompanying editorial in the NEJM.

In my own field of stroke, due to protocol hyper- vigilance,  I regularly see people with all manner of minor problems.  The commonest is what the nurses call vaguely a “facial droop.”  Many of these patients are presumed to have strokes until proven otherwise and it is hard to prove a negative as we all know. The result is possibly some improvement of stroke care but at price of  wasting resources  and worse, many people are presumed to have stroke, treated for that, when they actually have something else. With the mortality for true stroke being in the range of 20%, it is also a serious disease  Stroke mortality rates are decreasing.  But that may be artifactual. The apparent reduction in stroke mortality may actually be due to contamination in the stroke groups with minor cases and patients who actually have something else, possibly unrecognized. In defense of stroke vigilance, many more patients might get to see an expert neurologist who hopefully will have the wherewithal to diagnose what the patients actually have. Yet the neurologist’s attention is distracted by constant vigilance for stroke.

In 2014 we live with Federally monitored protocols in our hospitals,  which for good or ill,  have  acquired their own constituency. After all, lots of people are employed in this endeavor.  Maybe this is good for the economy. Many of us find these protocols wooden, wasteful and dictatorial. One would want to be certain protocols and enforcements actually benefit patients before rushing in headlong.

Personally I think money is much better spent insuring clinicians are well trained and up to date in their education and, most important, eager to do a good job.   Any endeavor is much more likely to be improved by making sure you have high quality practitioners of that endeavor. It is about people, picking  the best (we are not doing that)  and keeping them engaged and up to date, not as much about protocols and enforcement. People come before process.   You don’t get far whipping people into submission.   My plea is,  test the protocols before rushing in and inflicting them on practitioners and patients.