The major significance of the current report is that it provides estimates and characteristics for that portion of the civilian SMI population living in households. Survey results show that approximately 3.3 million adult Americans have mental disorders that seriously interfere with one or more aspects of daily life and that about 2.6 million of these persons are currently limited in one or more functional areas. These results suggest that the household component of the SMI population is comprised of between 2.6 and 3.3 million adults, depending upon the criteria employed for inclusion. Undoubtedly, both of these numbers are conservative because of the likelihood of underreporting in the survey. Placed in the context of the entire adult population, these findings suggest that the SMI population can be conservatively estimated to include 4 to 5 million adult Americans, or 2.1 to 2.6 øpercent of the adult population. In addition to the household population, it is estimated that 200,000 SMI persons are homeless on any given day (13). An additional 1 million to 1.1 million are residents of nursing homes (14), approximately 50,000 to 60,000 are patients of mental hospitals, and approximately 50,000 are inmates of State prisons (15). A major remaining need is to collect similar data on all SMI persons, whether their residence is a household, an institutional or noninstitutional group quarter, or some other setting, including streets and shelters. In order to formulate more effective national policy to address the needs of these disabled Americans, a need exists to examine the longitudinal relationship between course of disorder and functioning as they relate to service and program participation.