There is little consensus about how to measure the lengths of stays in nursing homes for Medicare beneficiaries when trying to identify long-stay and short-stay residents. Ilene Harris collaborated on research to create an algorithm to measure the days of stays in nursing homes and better differentiate between long- and short-term residents. Researchers also sought to compare their algorithm with Minimum Data Sets (MDS) alone and Medicare claims data.
The researchers linked 2006-2009 MDS assessments to the skilled nursing facility (SNF) data from Medicare Part A. The new algorithm calculated the daily nursing home stay evidence by the dates given by MDS and SNF, reported in total, long-stay, and short-stay of residents. As the algorithm identified long-stay residents, they were also compared with the nursing home data from MDS-alone and Medicare Parts A and B data.
The algorithm identified 276,844 residents, 40.8 percent of whom were long-stay residents and tended to be male, white, older, unmarried, receive low-income subsidies, and have multiple co-occurring conditions. They were also more likely to have higher mortality rates but experience fewer hospitalizations and skilled nursing facility services. Researchers found that larger shares of long-stay and short-stay residents were identically categorized by the MDS/SNF algorithm and MDS-only measures than were identically categorized by Medicare Parts A and B data. They also determined that the nursing home length-of-stay from the MDS/SNF algorithm was similar to the MDS-only long-stay residents. Both lengths, however, were longer when compared to the Medicare Parts A and B data.
The results of this study show the success of this new MDS/SNF algorithm in differentiating long-stay and short-stay nursing home residents. The algorithm allows researchers and policymakers access to more precise data than Medicare claims data alone.