When Should You Query in Mortality Cases?

Comfort care is not an indication to stop documenting and coding.

Last week, a listener reached out to me to ask me my opinion on querying for potential comorbid conditions or complications in relation to patients who expire. Specifically, she pinpointed patients who were transitioned to comfort care and asked whether it was fishing for CC/MCCs to query for end-of-life diagnoses, especially in the context of withholding active treatment.

When I was a physician advisor, I spent a large chunk of my weekly time in mortality reviews and giving providers feedback on documentation opportunities. My philosophy was that the coding abstract should be able to be reconstituted into a story if you translate the codes back into conditions. A provider’s job is to truthfully tell the story of the encounter, and the coder’s job is to accurately convert the verbiage into ICD-10 codes. I found it exceedingly annoying when I got to the end of an encounter and was left wondering, “Why did this patient die?!”

Comfort care encompasses both hospice and palliative care. Hospice is a specific subsegment of comfort care without curative intent, usually in patients with a limited (e.g., < 6 months) life expectancy. Palliative care can be provided at any time in the course of a serious, usually chronic, illness, and there may or may not be curative intent of treatment. Mortality reviews usually only exclude patients who are formal hospice patients. There may be further guidelines, such as the patient must be covered by hospice insurance and be in for symptom control only. This may vary between mortality models, but I am only intimately familiar with the one my organization utilized.

During my research for this segment, I discovered that the preferred term is “intensive comfort measures,” and this resonates with me. That term feels more active, as though you are choosing a positive treatment, rather than withholding or declining treatments in the negative. There are many symptoms which are addressed with intensive comfort measures, including pain, dyspnea, nausea and vomiting, constipation, cough, and dry mouth.

The discussion and decision to defer, withhold, or decline medical treatment is a very labor-intensive activity. It used to take me much longer to educate a patient as to why they did not need an antibiotic for their viral upper respiratory infection than it would to just scribble out a prescription. If a coder or CDIS is assessing whether something is a legitimate secondary diagnosis, the fact that treatment is not ultimately being undertaken should not prevent them from capturing the condition. In the Evaluation & Management universe, a decision not to resuscitate or to deescalate care because of poor prognosis is a criterion for high risk of morbidity from additional diagnostic testing or treatment, justifying the highest level of service in an office or ambulatory services setting.

If a patient expires, the diagnosis of the terminal cardiac arrest is implicit in the disposition. If a patient were to sustain a cardiac arrest, were to be resuscitated and linger for a week before finally succumbing, that cardiac arrest would be captured. Even though cardiac arrest (also ventricular fibrillation, respiratory arrest, certain types of shock) is only an MCC if the patient is discharged alive, its capture helps tells the story in codes.

If a patient transitions through stages of active dying proximate to expiration, I would not recommend capturing those (e.g., suffering an acute respiratory arrest leading to death within minutes in a patient who is Do-Not-Intubate, Do-Not-Resuscitate). That is inherent to the act of dying. However, if a patient lapses into a coma and survives for three days prior to expiring, that is a condition which should be recorded and captured.

There are a few reasons which I think support querying for significant conditions.

• It is important that the medical record and the coding reflect the series of events which took place during the encounter.
• Although mortality metrics may mandate conditions which risk adjust likelihood of mortality be present on admission, SOI/ROM may be impacted.
• If resources are consumed to treat the CC/MCCs experienced prior to death, you are entitled to capture them and be reimbursed at the appropriate DRG tier.
• Providers will hopefully carry forward good practice of thorough documentation to all their patients, affecting many quality metrics.

As always, the provider should be given enough information and clinical indicators to make a good decision as to whether a condition was present or not. The query should be compliant and non-leading. The fact that the mortality metrics will be affected should not be offered.

In conclusion, my opinion is that being deemed comfort care is not an indication to stop documenting and coding. After all, DNR does not stand for “Do Not Record.”

Programming Note: Listen to Dr. Erica Remer when she co-hosts Talk Ten Tuesdays with Chuck Buck, Tuesdays at 10 Eastern.