The CMS sepsis measure

The definition of sepsis is a mess. Sepsis is difficult to reduce to a set of criteria. If you are an experienced clinician you know it when you see it. This is just one reason why the CMS measure is a disaster.

From the linked article:

In changing the clinically significant value of lactate, CMS mandated that clinical practice, hospital protocols, and medical education had to adopt the lower threshold of 2 mmol/L to define severe sepsis and an initial lactate of greater than 4 mmol/L to define septic shock in the absence of robust supportive literature. Physicians are being forced to use government-issued standards of practice and patient care that have not been fully investigated as appropriate and safe. Doctors are no longer permitted to doctor but rather forced to practice cookie cutter one-size-fits-all algorithms with regard to sepsis care. These constraints leave the clinician in the predicament of using best practices versus following mandated guidelines.

We have demonstrated that there are various proposed definitions for sepsis, severe sepsis and septic shock. This is likely due to the fact that unlike myocardial infarction, which has a very precise pathophysiology and organic effect, sepsis is a spectrum of any number of factors. It is not due to one distinct insult but can be caused by a large variety of infectious agents that can infect a variety of anatomic locations. It is not due to one region of the body suffering hypoxia; rather it is due to a dysregulated host response to infection. And that host response is dependent on a variety of uncontrolled factors such as age, sex and comorbidities. It may be impossible to develop definitions that appropriately identify a disease state that is so dependent on multiple variables. Each patient is different and cannot be defined and treated exactly the same way. The CMS definitions are premature and, unlike the various other definitions presented, are mandatory and must be followed by clinicians practicing in the United States.

Why not let doctors be doctors? Because there is so much variation, of course. And as we all know variation is the enemy, right?