There appears to be a preclinical stage of physical disability which precedes onset of task difficulty (disability) in those who develop disability progressively as a result of chronic disease. Such a stage provides a basis for identifying older adults at risk of becoming disabled. This cross-sectional study evaluated whether a preclinical stage of physical function identified by self-report is associated with decrements in objective physical performance measures or increases in disease; that is, whether these measures, in those with preclinical disability, are intermediate between individuals who report no difficulty and no preclinical changes and those who report difficulty. The Women's Health and Aging Study II, an observational study of 436 women 70-80 years of age who were among the two-thirds least disabled living in the community. Participants were sampled from the HCFA Medicare eligibility lists and were determined eligible if they reported no difficulty, or difficulty in only one of four domains of physical function: mobility, upper extremity, IADL and ADL tasks. At the first follow-up (18 months after baseline), participants completed questionnaires on physical functioning for tasks in each of these domains, with possible answer options for each task: they had (1) difficulty (disabled); (2) no difficulty and no modification of task performance (High Function); or (3) no difficulty but reported modification and/or change in frequency of task performance (a self-report measure of preclinical disability predictive of incident difficulty). At the same visit, standardized, objective measures of function and disease were obtained, including measured walk; chair stands; strength: hip flexion, knee extension, ankle dorsiflexion, and grip; balance: function reach, single leg stand, tandem stand; joint exam: hip pain on passive motion and knee pain or tenderness; spirometry; ankle:arm blood pressure ratio; visual function: acuity, contrast sensitivity, stereopsis; and graded treadmill exercise test. Data were analyzed from the first follow-up examination. Physical performance decreased, and disease frequency increased, in association with decreasing self-reported mobility function (in walking one-half mile and climbing 10 steps), across three self-report categories: High Function, Preclinical Disability (Task Modification but No Difficulty) and Disability (Difficulty). These findings pertained for measures of walking speed, balance, strength, and knee and hip osteoarthritis. Self-reported level of function predicted differences in ranges as well as means for walking speed, balance and strength. These findings indicate a physiologic basis for self-reported function, including preclinical disability, specifically that different levels of disease severity, impairments and physical performance are concurrently associated with different categories of self-reported function. They also suggest new avenues for screening and intervention to prevent disability.