Background. The aim of the study was to use path analysis to test a theoretical model proposing that the relationship between self-reported depressed mood and choice stepping reaction time (CSRT) is mediated by psychoactive medication use, physiological performance, and cognitive ability. Methods. A total of 280 retirement village residents, aged 62-95 years, undertook tests of CSRT, which required them to step onto one of four panels that were illuminated in a random order. Depressed mood was assessed using the 30-item Geriatric Depression Scale (GDS). The participants were also tested on physiological and cognitive performance, including quadriceps strength, balance, complex attention (Trail Making Test [TMT] B), simple reaction time, reported level of exercise, and use of psychoactive medications. Results. A total of 51 participants (18%) showed mild to severe depression. Those with higher GDS scores had significantly increased CSRT and worse performance on all physiological and cognitive parameters. CSRT was also significantly associated with all other measures. The final path analysis model revealed an association between self-reported depression and CSRT that was mediated by two paths, one through quadriceps strength and the other through TMT B with both mediating variables then influencing CSRT via simple reaction time and balance. Conclusions. The findings suggest that self-reported depressed mood is related to slowed performance on a CSRT task and that this relationship is explained by underlying physiological and cognitive impairments.