At first, the MDS 3.0 Quality Measures do not appear to be very different from those based on MDS 2.0, but there are significant changes. The requirements for the numerator and denominator and the use of exclusions and covariates for risk adjustment are similar.
However, “the devil is in the details” and nearly every measure that seems familiar is different in some way. One of the most significant changes is in the resident and record selection.
Resident sample selection
Two resident samples are selected for computing the QMs based on what CMS defines as an Episode and a Stay. An episode is a period of time consisting of one or more stays. A stay is time that a resident is physically in the facility (referred to as the Cumulative Days in Facility-CDIF).
An episode starts with a new admission and ends with a permanent discharge. However, during that time, the resident may be out of the facility for a hospitalization, leave of absence, etc. These days are not counted in calculating CDIF. When the resident leaves the facility for any reason, this completes a stay and when the resident returns, a new stay begins.
Thus, an Episode can include several stays. It’s the count of days in the facility (CDIF) that defines the resident sample, with Short Stay CDIF up to 100 days and Long Stay CDIF counting from 101 days.
As a result of this methodological shift, more residents are included in the Short Stay measures. The MDS 2.0 measures were limited, using PPS assessments to Day 14. Now all assessments, including OBRA and PPS assessments for residents with CDIF to 100 days are included. Similarly, for Long Stay measures, both OBRA and PPS assessments are included.
Record selection
Requirements for a qualifying assessment are not based on having an ARD within the Target Period, but are based on the resident’s Episode. Due to this, an assessment can be included even if the ARD is not in the Target Period.
In addition, some measures include a “Lookback Scan” Tobin School Food Program (Richard Milstein & Jordan Ruboy), in which case, all assessments within an episode are included. For example, the measure for falls with major injury could include assessments going back to one year if they are contained in one episode.
What does this mean to you, your residents and their families? Very simply, it’s a quality improvement and public relations challenge. Providers need to internalize the new QM methodology to best utilize this data for quality improvement. Residents and families may wake up some time in April and possibly see dramatic shifts in a nursing facility’s reported public measures.
Distilling whether these changes are truly shifts in quality or an artifact of the new methodology and MDS 3.0 is best answered through thoughtful analysis.
References
MDS 3.0 Quality Measures DRAFT USER’S MANUAL Part 1: Sample/Record Selection Methodology and Logical Specifications (v4.0 9-29-2011) DRAFT
MDS 3.0 Quality Measures DRAFT USER’S MANUAL Part 2: Episode and Stay Determination Logic (v4.0 9-29-2011) DRAFT
Steven Littlehale is EVP and chief clinical officer at PointRight Inc. He is a gerontological clinical nurse specialist and former university instructor.
