Setting thresholds for MDS (Minimum Data Set) quality indicators for nursing home quality improvement reports

Determining meaningful thresholds to reinforce excellent performance and flag potential problem areas is critical for quality improvement reports. Without thresholds, an organization may interpret its performance as superior to others because it is "better than average" and falsely assume it does not have care problems in certain areas.

The Minimum Data Set (MDS) assessment instrument is mandated for use nationwide in all nursing homes participating in Medicaid or Medicare programs. Since 1993 a research team at the University of Missouri-Columbia has been developing and testing quality indicators (QIs) derived from MDS data as a foundation for quality improvement activities. In July 1996, a cross-section of 13 clinical care personnel from nursing homes participated on an expert panel for threshold setting for QIs derived from MDS assessment data. Panel members individually determined good and poor threshold scores for each QI, reviewed statewide distributions of MDS QIs, and, two weeks later, completed a follow-up Delphi round. Three members of the research team reviewed the results of the expert panel and set the final thresholds. With thresholds established for good and poor scores, MDS QI scores are reported to a sample of Missouri nursing homes using the thresholds.

To ensure that thresholds reflect current practice, threshold setting with another panel of experts will be repeated as needed, but at least biannually. The report format will be revised on the basis of user input, and a statewide study testing different educational support methods for quality improvement using MDS QIs is now underway.