Accurate admissions data vital to your abstracting
Hospital emphasizes one-on-one instruction
The patient checking into the hospital says he’s Middle Eastern by birth, but the registrar is uncomfortable designating his race as "white," even though that would be correct. The temptation is to use the catch-all race category of "other" to record the patients race.
That, however, would be the wrong designation. And even though it may seem like an inconsequential error, enough of them can cause big problems, admitting managers are discovering.
With the growing emphasis on hospital performance and quality improvement reports prompted by managed care, abstracting errors such as these can no longer be brushed aside.
Hospitals are becoming more aware not only of how their ICD-9-CM coding can affect outcomes and reimbursement, but how abstracting these data such as race and source of admission can affect performance measures. They want to be sure their discharge data is a true reflection of what happens to patients, who those patients are, and how they happened to be at their facility.
Simply put, outcomes results must be a true indicator of the quality of care, and that requires good quality data, says Andrea Zach, MPA, RRA, manager of the California Office of Statewide Health Planning and Development (OSHPD) in Sacramento.
Admitters who collect key information at registration play an important role in that process. Statistical information is used to identify patient trends and study treatment outcomes. For example, knowing that use of a particular hospital by a high-risk population is up 80% can help explain why outcomes at that hospital are not as good as in the past. This also highlights the need for preventive care for that patient group.
Education ensures proper abstracting
In the past, when admitters at the University of California, Los Angeles (UCLA) Medical Center faced questions about the race of a person being admitted, they chose the "other" race category more than 100 times a month, says Cynthia Frizelle, RN, assistant director of admissions and registration.
Those and other abstracting discrepancies were brought to the attention of the hospital when OSHPD issued new definitions. It led to an educational crackdown that has virtually eliminated such errors, reducing the use of "other" for race to just two or three times a month, which is probably about right, Frizelle says.
However, making the right choice for such data fields as race, ethnicity, source of admission, and admission type is far from a straightforward decision, Frizelle explains.
"Many years ago, you couldn’t ask for race, then two years ago it became mandatory to get it," she points out. "It’s a sensitive issue with admitter and patient. When you get into a situation with a child who has an Oriental mother and a black father, it’s embarrassing to ask, What nationality is your child?’"
Information on birth records cannot be assumed simply by looking at the patient or the patient’s relatives. So, the question of race is reviewed with the parent on the nursing unit, and a signature is required to validate the information before the mother is discharged, Frizelle notes.
The "ethnicity" code adds to the confusion. In California, the choices are limited Hispanic, non-Hispanic, and unknown but it’s difficult for admitters to associate a "white" race code with an "Hispanic" ethnicity code, as is sometimes correct according to the OSHPD definitions.
"Patients have a hard time with it, too," she adds. "They’ll say, I’m not white.’"
Another headache California admitters face is recording the correct information for referral sources. There are 10 choices, and some are quite confusing. A patient told by his doctor to go to the emergency department for treatment should be coded as a "home referral" if he ultimately is admitted to the hospital. But admitters won’t know this unless they press the patient for information. Also, it’s sometimes difficult to determine immediately whether a patient came from "long-term care" or a "residential care facility," especially if the admitter isn’t familiar with the institution involved.
Some cases are borderline, and OSHPD doesn’t always have easy answers, Frizelle notes. "It’s just that they trend hospitals, and find all of a sudden that 80% of admissions are in a certain category, or that there is an increase in Hispanic admissions that they’ve never had before," she adds. "There are certain tolerance levels that they expect, and if you start going over those, they take a look."
Other’ is not acceptable
The UCLA Medical Center took a comprehensive look at its abstracting practice partly as a result of the new definitions instituted by OSHPD a couple of years ago, and partly to review its own requirements. The hospital looked at individual data elements from its computer system and went through them line by line to identify discrepancies in each category. After updating the data elements to reflect OSHPD’s definitions, the hospital placed the acceptable categories and definitions on the admitting system computer. (For a list of the referral and race codes used on the system, see chart on p. 8.)
An employee educational effort included one-on-one classes showing admitters how the descriptors were written and helped them to visualize that language, Frizelle says.
Specific admitters who had problems also were targeted. "We would identify one user and say, You made 30 errors in one month, and you need to correct them,’" Frizelle says.
Employees were told, for example, that "other" was not a viable option for the race designation, and that they’d better have a good reason if they used it, she adds.
A major issue in correct upfront abstracting is that admitters are asked to pin down borderline definitions and ask sensitive questions while under pressure to perform a timely registration. To expedite the process, UCLA Medical Center has a "help screen" function in its computer system. This allows admitters to review an abstracting category for relevant definitions.
"We took all the OSHPD requirements and put them on a screen in our database," Frizelle says. The UCLA enterprise which includes three hospitals is in the process of converting to a Windows-based ADT system and is working to put the information in a drop box in that program, she adds.
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content