The trusted source for
healthcare information and
Pre-billing review will improve mortality index
Initiative aims at ensuring severity of illness
As a result of a pre-billing review of charts of patients who die in the hospital, the mortality index at Stony Brook University Medical Center has remained steady at under 1 except for one month when it was 1.04, according to Catherine Morris, RN, MS, CCM, CMAC, executive director of care management and clinical documentation improvement administrator at the 591-bed medical center in Stony Brook, NY.
The mortality index is the ratio of observed mortality to expected mortality. Coders use a software program to assign the expected mortality rate for patients who die based on documentation in the chart. The expected mortality rate ranges from 1 to 4, with 1 designating the lowest chance of mortality.
"Mortality data is included in a lot of publicly reported quality data, including the Centers for Medicare and Medicaid Services Hospital Compare web site, Health Grades, Leapfrog, U.S. News, and other hospital rating web sites. Our main goal for mortality reviews isn't concerned with increased revenue. Only a few cases actually result in increased revenue. The goal is improved quality documentation. We want to make sure that our reported data accurately represents the severity of illness and the risk of mortality for patients who died in the hospital," Morris says.
When a patient dies in the hospital, the coders run the software to assign a risk of mortality, put a billing hold on the case and assign a mortality review code which flags the case for review. The clinical documentation specialists are assigned by unit and take turns each month performing the mortality reviews along with their regular duties.
They pull the charts of all patients who die in the hospital who have an assigned risk of mortality rate of 3 or lower and determine if the documentation in the chart accurately described how sick they are.
"We chose to conduct the mortality rate pre-billing because most public reporting is based on billing data, and if the risk of mortality is not correct when the bill drops, it will impact the hospital's rating," Morris says.
If it appears that the documentation is not complete or does not support the patient's mortality risk, the clinical documentation specialist sends a query to the physician asking him or her to more specifically define the patient's condition. For example, if a patient is in a coma, in order for the coders to assign the code for "coma" the physician has to write "coma" in the chart, rather than writing that the patient is unresponsive or has a Glasgow Coma Scale rating of 5.
After the physician answers the query, the documentation specialist sends the information back to the coders who recode the chart and send the information through the software system again to obtain an attestation of the risk of mortality, according to Theresa Adell, RN, MS CNRN, CCM, case manager/clinical documentation specialist supervisor at the hospital. For example, a 57-year-old man came into the emergency department in cardiac arrest and later died. The documentation did not contain any comorbidities, and the software showed that his risk of mortality was a 1.
"Based on the documentation, we wouldn't expect this patient to die. He just didn't appear to be that sick. But when we reviewed the chart, we found clinical information that was not written in codable language," Adell says.
The patient came into the hospital in a coma, but the physician wrote "unresponsive" rather than using the word "coma." The coma was secondary to intracranial hemorrhage and cerebral edema. When the additional documentation was in the chart, the patient's mortality risk became a 4, which is the highest risk possible.