Key points

Introduction

In October of 2015, the Centers for Medicare and Medicaid Services (CMS) enacted a new national quality measure on sepsis called the Early Management Bundle for Severe Sepsis/Septic Shock (SEP-1). SEP-1 was the end result of a colossal undertaking to standardize care for severe sepsis and septic shock regardless of the size of the emergency department (ED) where the patient is being treated. The final product deviates substantially from the original measure (stewarded by Henry Ford Hospital in Detroit, initially led by early goal directed therapy [EGDT] pioneer Dr Emmanuel Rivers) and does not necessarily follow the best current evidence available. Nevertheless, a thorough understanding of SEP-1 is crucial because all hospitals and emergency providers (EPs) will soon be accountable for meeting the requirements of this measure.

In brief, SEP-1 is the nation’s first, and by law only, national quality measure on early management of sepsis care. It mandates that patients meeting criteria for SEP-1 must receive the bundle of care stipulated in the CMS Specifications Manual for National Hospital Inpatient Quality Measures. This measure applies to all US EDs.

This article provides a thorough review of the SEP-1 measure and all of the potential implications it may have on sepsis care provided in the United States. The measure has stirred up a great deal of controversy, which is not surprising given the complex nature of the sepsis disease process. The major concern is that hospitals may focus their attention on meeting compliance with the requirements of SEP-1 and consequently may stray from key patient-centered outcomes in sepsis. There is no question that the SEP-1 bundle is burdensome and much more complex than any previous core measure set forth by CMS. It remains to be seen if this will improve care of the patient with severe sepsis and septic shock in the ED.

Introduction

In October of 2015, the Centers for Medicare and Medicaid Services (CMS) enacted a new national quality measure on sepsis called the Early Management Bundle for Severe Sepsis/Septic Shock (SEP-1). SEP-1 was the end result of a colossal undertaking to standardize care for severe sepsis and septic shock regardless of the size of the emergency department (ED) where the patient is being treated. The final product deviates substantially from the original measure (stewarded by Henry Ford Hospital in Detroit, initially led by early goal directed therapy [EGDT] pioneer Dr Emmanuel Rivers) and does not necessarily follow the best current evidence available. Nevertheless, a thorough understanding of SEP-1 is crucial because all hospitals and emergency providers (EPs) will soon be accountable for meeting the requirements of this measure.

In brief, SEP-1 is the nation’s first, and by law only, national quality measure on early management of sepsis care. It mandates that patients meeting criteria for SEP-1 must receive the bundle of care stipulated in the CMS Specifications Manual for National Hospital Inpatient Quality Measures. This measure applies to all US EDs.

This article provides a thorough review of the SEP-1 measure and all of the potential implications it may have on sepsis care provided in the United States. The measure has stirred up a great deal of controversy, which is not surprising given the complex nature of the sepsis disease process. The major concern is that hospitals may focus their attention on meeting compliance with the requirements of SEP-1 and consequently may stray from key patient-centered outcomes in sepsis. There is no question that the SEP-1 bundle is burdensome and much more complex than any previous core measure set forth by CMS. It remains to be seen if this will improve care of the patient with severe sepsis and septic shock in the ED.

A brief history of SEP-1

In 2003, the Surviving Sepsis Campaign (SSC) initiated work on guidelines on bundled sepsis care. The SSC group focused its efforts on ways to implement the tenets of the recently published EGDT trial, which focused on an aggressive, invasive, and protocol driven resuscitation of patients with severe sepsis and septic shock. The SSC was also cognizant of the recent Institute of Medicine report To Err is Human , which highlighted the impact of iatrogenic error in medicine. The best available evidence at the time suggested that EGDT and bundled care uniquely decreased mortality from severe sepsis and septic shock.

In 2008, Henry Ford Hospital and Dr Rivers succeeded in getting the National Quality Forum (NQF) to endorse their proposed sepsis bundle and embrace EGDT (NQF #0500). Although the NQF is a feeder for CMS measures, a CMS measure did not materialize after initial NQF endorsement. In 2013, in accordance with new provisions of the Affordable Care Act, the Department of Health and Human Services identified sepsis as a priority for the following measure cycle. Simultaneously, NQF #0500 came up for its scheduled maintenance review. The new iteration of NQF #0500 now required the invasive components (eg, central venous catheter, arterial line) of EGDT and measurement of central venous pressure and central venous oxygen saturation. However, because the three largest sepsis trials ever were underway (ProCESS, ProMISe, and ARISE), it was decided that an ad hoc committee would reconvene when new data emerged.

When the landmark ProCESS trial demonstrated that EGDT performed no better than both a less-invasive protocol and usual care (ie, physician discretion determines care), the committee was forced to reconsider the portions of NQF that involved the most invasive portions of EGDT. Because CMS would likely now adopt the re-endorsed NQF #0500, many members of the NQF committee believed that it might be untenable to assess quality of sepsis care by provider compliance with EGDT. The NQF patient safety committee voted 11 to 7 to remove the invasive requirements of EGDT from the measure. Ultimately, the expensive and invasive aspects of the protocol (eg, central venous catheter, central venous pressure monitoring, and central venous oxygen saturation monitoring) were now optional. EPs were also given the option of documenting certain physical examination features or cardiac ultrasound in place of these modalities. This compromise was approved and thus NQF #0500 was finally re-endorsed in September of 2014 (20 days before the ARISE trial confirmed the ProCESS findings).

NQF #0500 was thus cleared for final approval by CMS. However, to turn NQF #0500 into a measure that could be implemented, CMS subcontracted Mathematica Policy Research group and Tellegen to turn the measure into the specification manual and the data dictionary needed for future chart reviewers to assess adherence. The result was a 51-page specification manual accompanied by a 393-page guide. Within these documents lie the keys to understanding and implementing SEP-1 (version 5.1, which we use for this article, is the latest version available, for use July–December, 2016).

Where are the teeth?

CMS quality measures are federal regulations, enacted under the Department of Health and Human Services. The teeth of quality measure enforcement may be tied to either hospital-compare or value-based purchasing regimes. In hospital-compare, a hospital’s overall adherence to CMS measures is reported and compared with other hospitals. However, it is the Joint Commission that carries the genuine threat to hospitals not complying with CMS measures at stipulated thresholds. If a Joint Commission survey of a hospital exposes poor compliance to CMS measures, that hospital risks losing its accreditation. Therefore, adherence to CMS measures is compulsory under threat of loss of accreditation, under a hospital-compare regime.

After a CMS measure has been in use for some time, it may also be used for value-based purchasing. In value-based purchasing, Medicare and/or Medicaid reimbursement for sepsis cases is directly tied to rates of measure adherence, even on a case-by-case basis. Therefore, if a patient’s case qualifies for that measure, the hospital’s adherence would determine whether or not the hospital would be reimbursed for that care. Currently, SEP-1 is a hospital-compare measure. It may become a part of value-based purchasing in fiscal year 2017 to 2018.

The metric of interest to CMS for SEP-1 is adherence to the measure, not mortality or other patient-centered outcomes. That is because it is an a priori assumption that adherence to the quality measure improves mortality. This assumption is derived from the lengthy process before measure approval, which includes a rigorous testing regime during NQF measure development.

If the measure development process is flawed, there is no immediate recourse after a measure has been endorsed by the NQF or adopted by CMS and the Department of Health and Human Services. However, in a process somewhat analogous to a phase IV clinical trial, CMS measures, like all federal regulations, are also eventually subject to retrospective review. These reviews often lead to changes in, and in some cases the repealing of, CMS measures. SEP-1 has not yet come up for such review.

What is SEP-1 and which patients must receive its provisions?

The language that CMS uses for inclusion and exclusion criteria for measure applications is worth reviewing. Because such measures are designed to assess overall hospital performance by way of adherence, CMS thinks in terms of statistics. Thus, “numerators” are the patients to whom providers/hospitals have “correctly” applied a measure, whereas “denominators” are the patients that CMS deems should have had the measure applied to them. The definitions of these populations are termed numerator statements and denominator statements. For SEP-1, the denominator statement therefore identifies the pool of patients who should have received the CMS sepsis bundle, whereas the numerator represents those who actually received it and had it properly documented.

The debate over which patients should be in the denominator group and what actions must be taken for the patient to be counted in the numerator, is the crucial focus for debate on sepsis care, and in any CMS measure ( Boxes 1 and 2 ).

The SEP-1 numerator statement is the number of patients from the denominator population who had all of the actions in Box 2 completed and documented properly. Because there are provisions for both severe sepsis and septic shock, there are two separate “clocks” in this measure. This means that there are interventions for patients that must be completed within 3 hours of presentation of severe sepsis. However, if septic shock is noted later, a separate “shock clock” is started. For interventions that must be completed within 3 hours of presentation of septic shock, it is the shock clock that must be used. This means that if severe sepsis was detected at 1 pm , the 3-hour bundle for severe sepsis would be due at 4 pm . If septic shock was detected at 2:30 pm , the 3-hour bundle for septic shock would be due at 5:30 pm , and the 6-hour bundle at 8:30 pm (for an excellent graphic representation, see Ref. ).

Certain patients may be excluded from the denominator. When a permitted reason has been documented as to why a patient with a chart coded to have severe sepsis or septic shock diagnoses did not receive the appropriated SEP-1 bundle interventions, the case is not scored (ie, does not count in the hospital’s statistics). That said, a permitted reason that is not properly documented does in fact count in the aggregate metrics. Thus, the case counts against a hospital’s statistics for adherence if the exclusion criteria are not properly documented. Patients who may be excluded are outlined in Box 1 .

Receiving credit for adherence to SEP-1 depends on four things: (1) performance of all the required actions, (2) correct documentation of these actions, (3) proper documentation of patients with permissible exclusion from the denominator, and (4) chart abstractors being able to find and interpret all of this documentation. Because SEP-1 is a composite measure, all of the intervention outlines must completed for a case to pass the measure and count favorably on hospital compare metrics. It is then up to chart abstractors and their hospitals to report the aggregate rate of adherence to CMS. This must all be done manually, because there are no software applications available for this bundle.