MDS Alert

Published on Fri Mar 21, 2008 PDF

The agency provides reasons MDS nurses may like this new instrument.

Mastering the MDS 3.0 may sound daunting, but the instrument appears to do a better job evaluating residents in some ways--and may take less time, to boot.

Those were two key messages from a recent Centers for Medicare & Medicaid Services' special Open Door Forum (ODF) reporting on a validation study in several states that shows the instrument fared well in comparison to its predecessor. 

Why it's better: The MDS 3.0 gives the resident a voice and has been shown to increase the clinical relevance and accuracy of the assessment--and decrease the average time for completion by 45 percent, according to Debra Saliba, MD, MPH, speaking at the ODF. She is associate professor at UCLA and a researcher for the Rand Corporation's national validation study of the MDS 3.0 items.

In the validation study, "real-live MDS nurses in real nursing facilities actually tested the instrument in addition to 'gold standard' nurses," trained by Rand on the MDS 3.0, says Rena Shephard, RN, RAC-MT, MHA, FACDONA, president of RRS Healthcare Consulting in San Diego. The gold standard nurses went into facilities in the validation study to train the facilities' MDS nurses to complete the MDS 3.0.

MDS 3.0 Speeds Things Up

The average time required for the MDS nurses in the validation study to complete the MDS 3.0 came to 62 minutes, whereas the MDS 2.0 required 112 minutes. This is particularly significant because the facility nurses were "new to the MDS 3.0" and had probably not yet achieved maximum speed, Saliba noted.

Also, the 112 minutes for completing the MDS 2.0 represented the time spent working on the form--not the interdisciplinary team meeting or RAPs, Saliba reported. The MDS 3.0 form was in most cases completed by one facility nurse because that's whom we trained, Saliba added. But the nurses were given the option to train and "deputize other team members to do their sections of the tool." Thus, there was "a little bit of a mix in terms of the number of people completing" the MDS 3.0, she said.

Scripted Interviews May Be Key To Speediness

One reason the MDS 3.0 took less time in the validation study is because a number of revised sections rely on scripted interviews with residents, Shephard surmises. For example, the mood section (Section D) incorporates the PHQ-9 depression assessment instrument used in many healthcare settings. On the MDS 3.0, residents who are able to make themselves understood rated their own mood by answering a series of questions. By contrast, "the depression items on the 2.0 require the nurse or whoever is completing it to review a list of symptoms of depression and identify which ones occurred in the lookback," says Shephard, founding board chair and executive editor for the American Association of Nurse Assessment Coordinators. This requires the assessor to go through the time-intensive process of reviewing the record and talking to the resident/family or whoever was involved with the person, she notes.

New Assessment Tools 

The MDS 3.0 introduces a number of brand-new clinical assessment tools, observes Sandra Fitzler, RN, who participated as a technical expert panel member and in a work group in developing the MDS 3.0. And testing of the instrument found that a majority of patients--even those with moderate cognitive impairment--could answer the questions reliably, she adds.

Major changes in the draft final version recently posted on the CMS Web site include:

Section C (cognitive patterns).

The primary cognitive assessment in the MDS 3.0 is a "brief interview for mental status" or BIMS. "This structured, simple performance test is used to assess mental status for all residents who can be understood," Saliba explained. Eighty-five percent of non-comatose residents in the validation study were able to complete the BIMS with scores ranging from 0 to 15.

The MDS 3.0 BIMS includes wording that gives residents partial credit for close answers and responses to prompting, "which is commonly used to communicate with nursing home residents," Saliba said.

Staff will use the old MDS 2.0 items to assess the mental status of residents who are rarely or never understood or who provide non-sensible answers to the BIMS.

To assess delirium, the MDS 3.0 incorporates the Confusion Assessment Method or CAM, a validated instrument. Using the CAM instrument, nurses in the validation study found a significantly higher incidence of delirium than identified by the MDS 2.0 delirium items. The structured BIMS interview also helped nurses to detect new delirium symptoms missed by the medical record, Saliba reported.

Section D (mood). The MDS 3.0 uses the PHQ-9 interview to assess mood in all residents who can make themselves understood. In Rand's validation study, 82 percent of non-comatose residents were able to complete the interview. To assess residents who can't complete the PHQ-9 interview, the staff uses a PHQ-9 observational version. "The observational version includes an irritability item as an observable finding of possible mood disturbance in cognitively impaired adults," Saliba said.

Section E (behavior). New items assess the impact of behavioral symptoms on the resident and others. Resists care has been replaced by rejecting "evaluation or care" necessary to achieve the resident's goals for health and well-being.

Section F (preferences for customary routine and activities). The MDS 3.0 uses the P-A-T, which maps to the University of Minnesota Quality of Life Domains, Saliba said. It focuses on "resident voice" as being central to determining the person's activities and daily routines. Saliba noted that nurses in the validation study didn't view the MDS 2.0 customary routine items as being helpful in care planning, in part, because a resident's prior customary practices might be related to his ability, illness or access rather than actual preferences.

Section J (health conditions). In assessing pain, this section asks residents to rate their worst pain on a scale of 0 to 10 and to report whether it affects their nighttime sleep or ability to participate in day-to-day activities. For residents who cannot participate in a pain interview, Section J8 includes four indicators of pain (non-verbal sounds, vocal complaints of pain, facial expressions, and protective body movements or postures).

In addition to revamped sections, the MDS makes numerous other changes designed to improve clarity and accuracy.