This report describes a simple method to quantify changes in the activities of daily living (ADLs) over a 6-month time period in long-stay VA nursing home populations. In these institutions, each resident receives a score of one to five every 6 months describing the degree of dependence in eating, mobility, transfer and toileting ("patient assessment instrument" or PAI). From these data, quantitative indicators can be calculated for the studied population: the average score for each ADL at the beginning and end of the period, and the average 6-month change in this score; the proportions of all residents whose ADL score improved, remained unchanged or worsened; the proportion of initially independent residents whose ADL score worsened, and the proportion of initially dependent residents whose score improved. The method was applied in two VA nursing homes (A and B). The profiles of ADL outcomes in the two nursing homes differed in several respects. For example, in nursing home A, 80% or more of the population either remained unchanged or improved in ADLs and only 2 to 18% deteriorated. In nursing home B, on the other hand, only 50 to 60% of the population remained unchanged or improved and 36 to 50% deteriorated. For the initially independent residents, the frequency of deterioration of ADLs in nursing home B was several times greater than that in nursing home A. The described method of quantifying ADL outcomes could readily be applied to the PAI data that are available in all VA nursing homes.(ABSTRACT TRUNCATED AT 250 WORDS)