Sepsis 3 Consensus definitions—the impact on quality (part two)

In my previous blog, I addressed concerns about the new Sepsis 3 definitions.  In this blog (part 2), I will further flush out implications for specific quality metrics and also note the potential impact on reimbursement and severity of illness and risk of mortality scores.

Quality of Care Implications:

• Sepsis Bundle

Hospitals are now required to report on patients with severe sepsis and septic shock and their adherence to the bundles of recommended treatments and monitoring.  Unfortunately, the definitions of severe sepsis and septic shock are in many ways quite different when comparing the CDC definitions, the sepsis bundle definitions and the Sepsis 3 definitions (although there are some commonalities). 

• AHRQ PSI 13 Postoperative Sepsis Rate

Hospitals report rates of postoperative sepsis for elective surgical patients and this PSI is also a part of the PSI Composite measure.  How do we identify sepsis in postoperative patients and which criteria do we use, as AHRQ does not provide a definition for sepsis?

• Potentially Preventable Complications (PPC) rate

Maryland and Texas are measuring rates of PPCs and several other states are moving toward this as well.  Sepsis not present on admission in a case without clinical exclusions is a PPC. 

• Severity of illness and Risk of Mortality (as measured through the APR DRG system)

Patients expiring from sepsis typically have higher subclass designations for severity of illness and higher expected risk of mortality.  Inappropriately classifying them as expiring from some other cause may result in an appearance of deaths greater than expected or due to “treatable” conditions.  This will affect organizations observed/expected (O/E) ratios and again make it more difficult to measure the true quality of care of the organization.

Financial implications:

• Value-based purchasing and HAC Reduction program implications: PSI 90 composite scores are a portion of each program and PSI 13 is a heavily weighted portion of the PSI 90 composite score.  This may actually improve scores for PSI 90 and provide a false sense of improvement in actual care as compared to a change in definitions.

• Hierarchical Condition Categories (HCCs): Organizations are measured on their efficiency in utilizing resources needed to care for patients.  Sepsis is an HCC with a fairly high risk adjustment factor (RAF) and having less patients reported as having sepsis will result in lower RAF scores and reduced performed on efficiency metrics such as Medicare Spending Per Beneficiary (MSPB).  They will have provided the care and resources needed to care for the patient via extended length of stay for rehabilitation or infusions, home health support and additional screening and office visits, while not justifying the need if the true picture of the patients cared for is not captured.

• External Auditor analysis: External auditors utilize multiple sources to determine focus areas for chart reviews, one being Pepper reports.  The rates of sepsis will likely swing widely from facility to facility and/or over time, making the sepsis DRGs 870-872 more at risk than ever.  Additionally, which set of criteria will external auditors be utilizing when reviewing cases for clinical validity of sepsis and how can an organization successfully appeal if there is not a standard set of criteria?

• Decreased financial reimbursement: Sepsis patients by nature are complex to care for and result in high resource utilization.  If organizations are not being paid appropriately for caring for true sepsis patients because we utilized a criteria not capturing those that do not “fit” into a firm set of criteria but are based on provider independent judgement and experience, the organization will be inappropriately penalized by being unable to recoup the cost of the care they provided.

So what is an organization to do?   Stay with the “old” definitions (which one?), use the new Sepsis 3 definitions, or create a custom set?  And should a hospital pick and choose which set of criteria is applied to which patient based on: a) whether or not quality metrics will be collected, b) financial performance and or profiling performance (severity of illness and risk of mortality) or c) the hospital’s risk via Pepper reports or overall cost of care?  Those who read my blog posts know that I certainly don’t recommend the latter (picking and choosing) for a couple of reasons.

First, it should always be about the patient.  The patient and the treatment of that patient should be our driving force, not a particular set of metrics or based on potential revenue or penalties to revenue.  I do acknowledge and agree that organizations should attempt to standardize definitions for common treatment and for measurement of outcomes, but not every patient will “fit.” 

Second, at what point is the individual provider’s diagnostic expertise ignored because their patient does not “fit in a bucket” of criteria.  Long ago, I presented a case to a physician leader in which the patient met multiple clinical criteria for Sepsis.  The physician leader agreed that on paper the patient met criteria.  But she had cared for that patient and per her face to face assessment, the patient wasn’t septic.  This was a valuable reminder that nothing can replace a face to face assessment.

3M Health Information Systems has taken a position on the new definitions that I would like to share with you:

1. The diagnosis of sepsis should only be based on the physician’s clinical judgement and a full clinical picture of the patient, not on any one set of criteria.
2. At the present time, the NCHS Official Coding Guidelines and AHA have not proposed or published changes to the current guidelines for reporting sepsis, severe sepsis or septic shock.

Until revisions are published, it may be best to follow the current coding guidelines set forth by the AHA. Organizations are however encouraged to discuss the current SCCM clinical guidelines for Sepsis and determine the appropriate reporting and documentation requirements of Sepsis cases within their facility.

I strongly encourage the healthcare community to standardize, embrace and utilize a consistent set of criteria so that our focus can be on real outcomes for real patients! 

Cheryl Manchenton is a senior inpatient consultant and project manager for 3M Health Information Systems.

Disclaimer

Each organization is solely responsible for compliance and reimbursement decisions, including those that may arise, in whole or in part, from participant’s use of, or reliance upon, information contained in these blogs or other publications or teaching materials. 3M and the author disclaim all responsibility for any use the participant or its organization makes of such information.