Background/Aims Dates and causes of death are available from a variety of sources, both within HMO organizations (e.g., Clarity Patient table at EPIC sites) and from other entities such as the federal government (e.g., Social Security Administration). These sources differ in the quality and completeness of information provided. For example, HMOs frequently miss out-of-hospital deaths which can account for a substantial percent of adult deaths. Governmental sources of data have more detail such as cause of death, but there is potential error involved in matching those records to the correct HMO members. Reliable death information is crucial for health care research. Misclassifying patients could mistakenly misinform researchers and could lead to erroneous conclusions. In order to increase the level of understanding of available death data sources and improve the selection of death data of adequate quality for inclusion in the VDW and HMORN studies, we will describe the types of death data available, how they compare, and how many sites are using each type. We will also perform validity checking by comparing to enrollment, pharmacy, and medical encounter data (e.g., did a supposedly deceased individual have an amoxicillin prescription after death?). At several sites, we will compile all death dates available from all death data sources and perform an inter-source agreement analysis using % agreement or Kappa statistic if possible and present the results.