CORRELATES OF PERFORMANCE-BASED MEASURES OF MUSCLE FUNCTION IN THE ELDERLY - THE CARDIOVASCULAR HEALTH STUDY

Background. It is unknown how much age-related changes in muscle perfo rmance represent normal aging versus the effects of chronic disease an d life style. We examined the correlates of four performance measures - gait speed, timed chair stands (TCS), grip strength, and maximal ins piratory pressure (MIP) - using baseline data from the Cardiovascular Health Study (CHS), a population-based study of risk factors for heart disease and stroke in persons greater than or equal to age 65. Method s. We analyzed data from the 5.201 CHS participants. Variables were ar ranged into nine categories: Personal Characteristics, Anthropometry, Physical Condition, Reported Functional Status, Subjective Health, Psy chological Factors, Symptoms, Cognitive Status, Habits and Lifestyle, and Prevalent Disease. independent correlates were identified using st epwise linear regression. Results. The regression models explained 17. 7-25.4% of the observed variability. Although age significantly correl ated with each measure, it explained little of the variability (less t han or equal to 5.7%). Anthropometric features plus physical condition explained 14.0-17.4% of the variability for grip strength and MIP, bu t 2.8-12.9% of the variability for gait speed and the log of TCS. Subj ective health and psychological factors explained 1.8-9.4% of the vari ability in gait speed and the log of TCS, but less than or equal to 1. 2% of the variability in grip strength and MIP. Variables for prevalen t disease explained less than or equal to 1.3% of the variability in e ach measure. Conclusions. After age 64, age explained little of the va riability in muscle performance in a large sample of mostly functional ly intact. community-dwelling older persons. Complex measures such as pail speed were more associated with subjective factors than were dire ct measures of strength. Prevalent disease contributed surprisingly li ttle to muscle performance.