A Physical Performance Battery Assessing Cognitive Function in Community-dwelling Older Japanese Adults

Abstract

BACKGROUND: A recent study by Walton et al. (2004) showed that differential control demands within the same task were associated with differential activation of regions within the brain. We have demonstrated that cognitive function is associated with some types of physical function. In a clinical setting of nursing prevention in older adults, some physical performance tests related to cognitive function could be useful in identifying mild cognitive impairment in older adults. Objective: The purpose of this study was to develop a physical performance battery assessing cognitive function in community-dwelling older adults. METHODS: One hundred and seventy-nine community-dwelling older Japanese adults with a mean age of 74.1 ± 5.4 years were recruited. The participants were classified as having aging-associated cognitive decline (AACD) or a normal level of cognition, based on the Five-cognitive Function Test. The physical performance battery included a hand-operated peg board (X1), 4-way choice reaction time (X2), timed up and go (X3), and 5-repetition sit-to-stand (X4). An equation of the physical performance score (PPS) was developed using principal component analysis. Participants were divided into tertile levels (T1, T2, T3) according to the PPS. Multiple logistic regression was performed to evaluate whether the PPS was independently associated with AACD. The cut-off point to distinguish AACD from normal was identified by a receiver-operating characteristic curve. RESULTS: The following equation was developed: PPS = -0.037X1 - 0.002X2 - 0.211X3 - 0.119X4 + 5.846. In the logistic regression model, the odds ratios (95% confidence interval) of AACD compared with T1 were 0.61 (0.23-1.61), 0.27 (0.08-0.87) in T2 and T3, after adjusting age, sex, and education level. A PPS of 0.27 (AUC 0.75, 95%CI 0.66 - 0.84) was defined as an optimal cut-off point that maximized the sum of sensitivity and specificity (sensitivity 85%, specificity 59%). CONCLUSION: This physical performance battery should be a useful tool for assessing cognitive function in community-dwelling older adults.