Are Clinical Indicators Present to Support the Diagnosis of Sepsis? | Sepsis Series

In Parts 1, 2 and 3 we learned about what sepsis is, sequencing of sepsis and  what documentation is needed to report severe sepsis. In Part 4, we will look at clinical indicators needed to clinically support the diagnosis of sepsis and determine if a query is indicated.


What are the most common clinical indicators for sepsis and septic shock?




WBC less than 4,000 or greater than 12,000

All indicators included under sepsis to the left of this column plus…

Fever > 100.4F (38C) or hypothermia < 96.8F (36C) | shaking, chills

Diagnosis of sepsis

Tachycardia (heart rate > 90 BPM)

Significant drop in BP that does not respond to fluid replacement

Tachypnea (respiratory rate > 20 breaths per minute

Acute organ dysfunction/failure

Mental status changes, confusion

Severe respiratory problems


Abnormal heart pumping function

Positive blood culture

Slurred speech

Localized infection

Cold, clammy and pale or mottled skin

The table above only represents some of the clinical indicators that may be present but is not an all-inclusive list.


One of the most challenging areas that coders face today is knowing when a query is necessary. Coders see diagnoses that are documented by the physician in the medical record, and they want to be able to report the code for the diagnosis. However coders know they must clinically validate diagnoses and if they are not able to, query or get CDI or a physician liaison involved.

There is an OCG stating: “The assignment of a diagnosis code is based on the provider’s diagnostic statement that the condition exists. The provider’s statement that the patient has a particular condition is sufficient. Code assignment is not based on clinical criteria used by the provider to establish the diagnosis.”

Does this mean coders may report any condition documented within a medical record?

No, of course not. All coders have seen copy/paste in medical records and problem lists that are brought in from previous admissions. When reporting a diagnosis, the condition must meet the reporting guidelines either for selection of PDX or as an additional SDX.  The statement in the OCG above doesn’t mean that there doesn’t have to be clinical indicators for a disease present at all, only that the physician  is not limited to a specific set of clinical criteria he can use to make a diagnosis. There are many “established” criteria for many diseases and sepsis is no different. At this time, there are three sepsis criteria that different physicians and facilities follow and will most likely change again going forward.

What makes this so hard for coders is that it is difficult to question a physician’s documentation. When conflicting documentation within the record is present, that is much easier because coders are just asking for clarification. If a coder asks a physician “does this patient really have sepsis?”, this could be a bad situation with a very negative outcome on the query. It is better to ask the physician what criteria were used to make the diagnosis of sepsis or involve CDI or a physician liaison for help with these. Physician education is needed, and if coders don’t query, then the facilities and physicians will not know that there is a documentation issue.


In the following examples, we’ll look at some scenarios that could potentially require a query to substantiate the diagnosis of sepsis and possibly help prevent denials. We all see denial after denial on sepsis cases, and it is best to get this diagnosis clarified at the time of coding, or while the patient is still in the hospital by CDI.

  • Patient presents with low grade fever of 99F and pressure in the pelvic area. Per the admitting physician, the patient is admitted to rule out sepsis and UTI. The patient’s urine culture was positive for UTI and blood cultures were negative for any bacteria. The patient was treated with IV antibiotics for two days and then discharged home to continue five more days. The final diagnosis for the patient are: 1. Sepsis with UTI. Most times, as auditors, we see this coded as A41.9 (Sepsis, unspecified organism) as the PDX followed by N39.0 (Urinary tract infection, unspecified site) as a SDX code without a query to the physician. The rebuttal that we get most often from coders, is that they coded based on the physician’s documentation of the diagnosis. And that is true. However, there was nothing clinically in the record to support the diagnosis of sepsis. The low grade fever and pain were symptoms of the localized infection in the urinary tract. If this record falls into any third party review, a denial of payment and reimbursement may be impacted.
  • Patient presents with pneumonia and dehydration with possible AKI (acute kidney injury). In the problem list, sepsis is documented on both the H&P and the DS. The patient is started on IV fluids and IV antibiotics and remained in the hospital for five days. There is a fever noted on admission of 100F and elevated WBC. However in the body of the DS, the WBC’s are linked to steroid use by the patient for their rheumatoid arthritis. The final diagnosis is 1. Pneumonia; 2. AKI secondary to dehydration; 3. History of rheumatoid arthritis on chronic low dose prednisone. Would you code sepsis? Some coders do code sepsis from the problem list without getting the condition clarified by the physician. There’s nothing really to support that the condition was present on this admission. The problem list is typically copy/pasted in some hospital systems and diagnoses on these should really be investigated to determine if they are present on this particular admission or not. The patient does have a fever, elevated WBC and AKI. Can these conditions be used as criteria to support sepsis? No, only the fever could be attributed to that diagnosis. The elevated WBC is due to steroid use and the AKI is due to dehydration so it is not linked to sepsis. In this case, J18.9 (Pneumonia, unspecified organism) would be reported as the PDX. N17.9 (Acute kidney failure, unspecified), E86.0 (Dehydration), D72.828 (Other elevated white blood cell count), T38.0X5A (Adverse effect of glucocorticoids and synthetic analogues, initial encounter), M06.9 (Rheumatoid arthritis), and Z79.52 (Long term (current) use of systemic steroids) would all be reported as additional SDX codes. Since the diagnosis of sepsis was only in the problem list and not listed as one of the diagnoses, a query isn’t really needed to clarify that this was ruled out, as it appears not to be current on this admission. There are no clinical indicators presents other than fever that could be used in the query for the physician. If this were coded as sepsis, or even queried and then coded as sepsis, it would almost certainly be denied in a third party review due to lacking clinical support of the diagnosis.

These examples may seem extreme, but this is what coders see very often. One very important statement to remember is that coders should not be ignoring physician documentation of a specific diagnosis based on clinical criteria, or abnormal test results, etc. If it’s documented it should be coded or a query to clarify should be sent to the physician. In Part 5 (and final) of the series, we will look at some of the common reasons given in sepsis denials as well as what coders can do to prevent them.


Download our entire Sepsis eBook.



ICD-10-CM Official Guidelines for Coding and Reporting FY 2020

Coding Clinic, Fourth Quarter 2016: Page 147

Coding Clinic, Third Quarter 2016: Page 8

The information contained in this coding advice is valid at the time of posting. Viewers are encouraged to research subsequent official guidance in the areas associated with the topic as they can change rapidly. 

Leave a Comment


    Related blogs from Medical Coding Tips

    List of New ICD-10-PCS Codes

    In this Coding Tip, we present the new ICD-10...

    List of New ICD-10-CM Codes

    The CDC has published new codes that will be ...

    List of New C Codes for ASC Reporting

    On December 21, CMS released the following Tr...