As a large user of the Minimum Data Set for Nursing Homes (MDS), the Veterans Administration agreed to partner with the Centers for Medicare and Medicaid Services in sponsoring the MDS 3.0 revision design, and its testing. The Veterans Administration Health Services Research and Development Program funded a consortium of researchers from four regional centers within the VA and Harvard Medical School to undertake and pilot design work to develop the new instrument and to replicate the community evaluation in 20 VA nursing homes from across the four participating VA regions.

The research used a systematic approach to validate eight new or revised sections of the MDS 3.0 (diagnostic coding, delirium, pain, falls, depression, behavior disorders, quality of life, and palliative care). These eight domains represent conditions or topics that have a high impact on resident health and for which quality assessments may significantly influence subsequent care. The overall objectives of the research are to:

• provide scientific input to improve the quality of MDS 3.0 and establish its applicability within the Veterans Administration; and

• enhance its performance and utility as a tool to improve nursing home care.

Harvard Medical School faculty have responsibility for the design of the sampling and the empirical analysis.