The use of a cross-sectional sample of nursing home residents rather than a sample of admissions to estimate admission characteristics carries a potential bias. The purpose of this study was to fill this void by comparing abstracted records data for an admissions cohort (n = 1,118) and a residents cohort (n = 830) residing in the same nursing homes. Compared to residents, admissions were significantly more dependent in their ability to get around and to dress themselves, received more clinical services, and had a higher rate of medication use. Over a 12-month period, admissions had a fivefold greater likelihood of being discharged to community, but about the same mortality rate as residents. Within both groups, those discharged to the community as well as those who died had expenditures that were almost twice as high as those of their counterparts who remained alive in the nursing home.