OASIS Alert

Sticking with old OASIS habits might be a bad idea.

When you answer OASIS questions wrong, you could be costing your agency hundreds of dollars on each episode, sabotaging your outcomes and setting the stage for denials and downcoding.

There are many bad OASIS assessment practices still in use, physical therapist and OASIS expert Linda Krulish told attendees at an educational session during the October annual conference of the National Association for Home Care & Hospice in Baltimore. Krulish, principal of Redmond, WA-based OASIS Answers, pointed out many common practices that are actually wrong.

How to get it right: Base your OASIS responses on the wording of the questions and answers, the instructions and strategies CMS provides with the questions and any clarifications CMS provides -- especially in updated questions and answers, Krulish emphasized. You don't have to like OASIS or think it makes sense, but you do have to collect accurate data. "Don't make changes to your OASIS answers to make your assessment look prettier," she warned.

Is Your Agency Using These Bad Practices?

In an effort to standardize data collection and improve OASIS accuracy, many agencies use practices that don't hold up under closer scrutiny, Krulish told attendees. Check to be sure you're not using these practices that were among the many she discussed:

Bad Practice # 1: Implement a lot of "always" and "never" mandates. For example, "If there's a pain med in the med profile, M0420 will never be '0'."

Why not: "Always" and "never" questions can work in some cases, but often are based on misunderstandings. For example, M0420 asks if pain is interfering with a patient's activity or movement. A patient can have pain requiring medication but the pain may not be at a level that interferes with his activity, Krulish explained. With this practice, you will find clinicians answering M0420 wrong to keep you happy, she said.

Bad Practice # 2: Implement a lot of "if" ... "then" statements. For example, "If you mark the presence of ANY behaviors on M0610 ... then you can't mark 'Never' on M0620."

Why not: You're better off if you look at and score each OASIS item independently, Krulish said. Data collection rules are very specific and assessors must clearly understand the instructions for each OASIS item.

For example, M0610 (Behaviors patient demonstrates at least once a week) identifies specific behaviors to choose from when answering the question. But M0620 asks about the frequency of behavior problems reported or observed and doesn't limit you to a specific list. It does ask you to include "any behavior of concern for the patient's safety or social environment."

These two questions have very different instructions, even though they are next to each other on the OASIS instrument. Behaviors you marked on M0610 may include some you do not consider a safety issue, considering the patient's environment, Krulish said.

Bad Practice #3: Rely on your vendor software to identify OASIS inconsistencies and change to comply with edits. For example, some software "won't allow M0340 -- response 1 -- Patient lives alone -- to exist with M0700 -- response 2 -- Able to walk only with the assistance of another person at all times."

Why not: You can use these tools to alert you of potentially conflicting answers you may want to look at more carefully, Krulish said, but be sure your product doesn't force you to change one or the other response as if it were an absolute inconsistency.

For example, M0700 asks you to assess what kind of assistance the patient requires to ambulate safely -- or propel himself safely in a wheelchair -- over a variety of surfaces. The fact that the patient needs assistance at all times to ambulate safely is a fact just as is the fact that the patient lives alone. While you need to address this combination in your decision to accept the patient or in your care plan, you shouldn't change your answer to suit your software, Krulish said.

Note: CDs of the NAHC conference sessions are available at www.nahc.org.