Objective: We investigated whether a simple measure of reaction time (RT) intraindividual variability (IIV) was associated with falls in older adults. Falls and fall-related injuries represent a major cost to health care systems, it is therefore critically important to find measures that can readily identify older adults at greater risk of falling. Method: Cognitive and motor function were investigated in 108 adults aged 53 to 93 years (M = 73.49) recruited across the local community and hospital outpatient department. Forty-two participants had experienced either an injurious fall, or multiple falls, in the previous 2 years. Results: Logistic regression suggested that fallers could be distinguished from nonfallers by greater medication use, IIV, postural sway, weaker grip strength and slower gait speed. Structural equation models revealed that IIV was predictive of falls via the mediating variable of motor function (e.g., gait). IIV also predicted higher order cognition (executive function) but higher order cognitive function did not uniquely predict falls or account for the associations between IIV and falls. Conclusions: These findings indicate that IIV measures capture important aspects of cognitive and motor decline and may have considerable potential in identifying older adults at risk of falling in health care and community settings.