CDC gives a nudge to hospitals on sepsis care
The Centers for Disease Control and Prevention formally called on hospitals to develop robust sepsis care programs to systematically identify and treat sepsis, track outcomes, and improve care delivery. The guidance comes as a series of recommendations dubbed the Hospital Sepsis Program Core Elements, along with an “assessment tool” that functions as a sort of checklist for compliance.
"Every hospital, regardless of size, location, and resources can strengthen the quality of care delivered to these patients and ensure their survival," CDC Director Mandy Cohen, MD, MPH, announced at the launch of the program.
Fully implemented, CDC’s toolkit would promote sepsis into the major leagues of “code strokes” and “code STEMIs” — highly prioritized system-wide operational programs enjoying executive sponsorship and significant financial and human resources.
Most large health systems have already implemented many of the recommendations, which advise hospitals to create a sepsis program spanning all inpatient care areas, and including elements such as:
A sepsis coordinator empowered by a senior executive leader
Sepsis screening tools on admission and throughout hospitalization
“Code sepsis” protocols for suspected sepsis patients
Electronic order sets for sepsis, including crystalloid infusions, antimicrobials, and vasopressors
Data collection: tracking and reporting of sepsis cases and outcomes, integrated with quality improvement programs
Continuing education and training of staff
And much, much more.
Some of the suggestions get into the weeds in a good way — like advising changes to pharmacy operations to stock antimicrobials throughout the hospital, and making antimicrobial delivery default to “now” in electronic order sets.
Others are inscrutable and creepy, like “Our hospital completes near real-time chart reviews for the purpose of clinician feedback and education.” What is that, a sepsis Stasi?
The biggest challenge in optimizing sepsis care has always been identifying the patients who need it. Unlike strokes and STEMIs, sepsis has no gold standard for diagnosis. Epic’s much-hyped sepsis screening tool was found to have only a 12% positive predictive value, while alerting on 1 of 5 of all hospitalized patients. (Epic claims to have improved it since then.)
Deployed at scale, a tool with such poor performance may save lives, but will also bring downsides of wasted effort, alarm fatigue, and professional frustration, not to mention overtreatment of not-septic patients with actual tons of excess antibiotics, while producing messy metrics that could disregard (or penalize) good clinical judgment.
And unfortunately, the CDC’s new recommendations won’t displace the widely and rightly maligned Medicare sepsis treatment regulations (SEP-1).
It’s good to see sepsis finally getting the attention it merits. Countless lives will surely be saved by increased prompt recognition and optimized treatment of this underappreciated and potentially lethal condition. Although CDC has no regulatory authority, its new program will serve as an important reference benchmark for what’s considered “good” sepsis care at U.S. hospitals.
"Sepsis 'Stasi" is not that much of an exaggeration. At a practical level, when hospitals stand to lose millions when not complying with these bundles, the natural inevitability of this is intense pressure/mandates on clinicians to comply.
I can think of no other instance in medicine where such non-specific criteria are used to establish a diagnosis, and where clinicians are then mandated to perform interventions on their patients regardless of whether they feel it is warranted or harmful.
Imagine if we began defining STEMIs as "anyone with chest pain", and mandated cardiologists take all such patients to the cath lab within thirty minutes or else they get dinged for not complying with the protocol? Sure, you'd probably catch a few subtle STEMIs that would have previously been missed, and patients would get to the lab faster. But there would also be countless harmful negative caths, and you would be doing a disservice to all the other patients not being seen in a timely manner.
Imagine if neurologists were obligated to give TPA to anyone presenting with altered mental status? Or if trauma teams were mandated to take EVERY trauma to the OR?
The clinician at the bedside HAS to have final say on whether or not an intervention is harmful or beneficial for that particular patient. And, yes, if you are tying hospital funding to compliance then it is inevitable that clinicians are absolutely being forced to comply as well.
To say nothing as to how unbelievably non-specific these criteria are. I meet SIRS criteria whenever I walk up a flight of stairs. Plus, practically any patient with fever is going to be a little tachy. So basically anyone comes in with fever, all of us in the ER stop what we are doing, call a "code sepsis", drown the patient in fluids regardless of their BP, give unnecessary broad spectrum abx (hello resistant organisms), and ignore all the sicker patients currently under your care. And if they are in pulmonary edema, "just intubate them" (I've been told that five separate times now).
Madness.
This is only going to lead to so much wasted time and energy. I work as a locums at a hospital that has a "code sepsis." What this means in this particular facility is that every lactate over 2 (yes, 2...not 4) as well as every lactic acid that is increasing (even if it becomes 2.3 from 2.0) gets reported to me and requires me to evaluate the patient. It is a lot of wasted time, especially when there are actually critically ill patients that require assessment and intervention. In fact, none of what is being done at this particular hospital is supported by data. My husband is the director of quality for the ED at a large academic center, and most of the stuff they have to do to get paid by CMS is also not evidence driven. I cannot stand these top down measures that do nothing to improve patient outcomes and everything to make a lot of busy work in the electronic chart.