Ryan Hakimi, DO, MS, NVS, RPNI, CPB, FNCS
This series will be a running feature highlighting billing, coding, practice management, and other business aspects of neurocritical care. In our next issue we will discuss the changes in critical care billing which will take effect on January 1, 2022
When providers are asked to admit patients to the NeuroICU, their initial efforts are focused on stabilization of the patient and placement of admission orders to get the patient access to specialized care. Documentation in the form of an H&P then becomes secondary, an unrewarding but required task. This document is often hastily constructed as there is usually a lack of information at the time of the admission (i.e., the patient has never been to your hospital, their identity is not known, they are unconscious, etc.). Hence, critical details of present on admission (POA) conditions are often lacking in the document that has the greatest weight in a hospital’s quality indicators, especially the observed/expected (O/E) mortality rate or Risk Adjusted Mortality Index (RAMI), which greatly impacts hospital ratings in everything from U.S. News and World Report to CMS’s Star Ratings on the Hospital Compare website, and therefore impacts consumer choice and insurance carrier participation.
RAMI > 1: more patients died than expected for their given condition and comorbidities
RAMI < 1: fewer patients died than expected for their given condition
The lower the RAMI, the better. Importantly, the RAMI usually accounts for approximately 20% of Hospital Star ratings.
How is expected death determined
From a practical standpoint, the H&P defines the state of the patient at the time of admission, which determines their illness level and conditions that are present on admission. This makes the H&P the most important document for improving your RAMI. The ICD-10 diagnoses on the H&P determine the patient’s level of illness at the time of hospital presentation. Learning the correct ICD-10 terminology is the best way to document the problem. This is important, as those who review the documentation do not have a clinical background and are not allowed to infer anything that is not documented.
Ex: Patient with low blood pressure due to low hemoglobin requiring 45 mcg/min of norepinephrine
To a provider, this patient is obviously critical and can die at any moment. However, for an abstracter, the diagnoses are low blood pressure and low hemoglobin, which can also be seen in outpatients. The provider needs to document hemorrhagic shock refractory to vasopressor therapy.
In decades past, the purpose of medical documentation was to document the care provided and to get reimbursement. Now the documentation is being used to judge the quality of the health care provided. Therefore, all providers must become knowledgeable on critical elements of documentation, as their hospitals, service lines, and ultimately each individual provider is linked to the H & P—the index document by which most determinations are made.
Learning the HCCs and documenting them (when appropriate) is valuable
Hierarchical Condition Categories (HCC) are conditions based on a risk adjustment model that calculates risk scores for aged and disabled Medicare beneficiaries. These scores represent the expected medical costs of a Medicare member in the coming year. Learning the HCCs and documenting them is valuable.
There are currently 86 HCCs, including the following common HCCs relevant to neurocritical care:
- Brain compression (not “midline shift”)
- Cytotoxic cerebral edema (not “CT shows swelling”)
- Acute respiratory failure with hypoxemia (not “ventilator dependent respiratory failure,” or “intubated for airway protection”)
- Intracranial hemorrhage (not “head bleed”)
Eliminate “history of” from history of present illness
Ex: The patient is a 54-year-old-male with a past medical history of hypertension, hyperlipidemia, and diabetes mellitus who presents with an acute left MCA ischemic stroke.
Only the left MCA ischemic stroke is billable and contributes to the patient’s RAMI as the hypertension, hyperlipidemia, and diabetes mellitus are historical. Instead, document the following:
The patient is a 54-year-old-male with hypertension, hyperlipidemia, and diabetes mellitus who presents with an acute left MCA ischemic stroke.
Then document each one of these comorbidities as an assessment and attribute a plan (even if the plan is to continue the patient’s home medication). As another example, document each individual electrolyte abnormality and state the replacement plan to further contribute to a patient’s RAMI.
Neurocritical Care team documenting own H&P
Consider the following common example: A patient presents to the hospital as a stroke alert. The Stroke team evaluates the patient with the initial focus being whether the patient should be given a thrombolytic. Head CT reveals a subarachnoid hemorrhage. The Neuro ICU team is contacted to admit and assume care of the patient. Not only is there revenue lost by having the Stroke team do the H&P for this Neuro ICU admission, but their documentation is also usually solely focused on the neurological exam and the decision making behind not administering thrombolytics. This approach misses many opportunities for documentation of conditions which are present on admission and which characterize the severity of the illness, all of which are ultimately used to calculate the expected mortality rate.
By documenting every condition as an assessment and having a plan for each, our system was able to reduce the RAMI for subarachnoid hemorrhage by 400% and ICH by 120%. As previously mentioned, the H&P is often written hastily as we are focused on stabilizing the patient. It is important for us to understand how our documentation on the H&P can impact hospital reimbursement, expected length of stay, whether a case will go before the hospital’s mortality review committee, and the provider’s risk of litigation, among other things.
Furthermore, given the fact that most of the largest neurocritical care units in the US utilize trainees, there is greater variability in the quality of their documentation. Perhaps this explains why many of the most prominent University-based Neuro ICUs are not listed on the 5-Star hospital list. In fact, of the 76 UCNS Accredited Neurocritical Fellowship program sites, only 13 are listed among the 455 total 5-Star hospitals [1,2].
Careful, accurate, and complete documentation on the H&P should be considered as a worthy focus of quality improvement projects as we aim to secure the resources needed to provide optimal patient care.
- https://www.beckershospitalreview.com/rankings-and-ratings/455-hospitals-with-5-stars-from-cms-2021.html, accessed 11/13/21.
- Fellowship Directory (ucns.org), accessed 11/13/21.