To determine the merits of fractal analysis in differentiating health from disease, detrended fluctuation analysis (DFA) was applied to postural sway data of healthy young individuals, healthy elderly individuals, and individuals with compromised balance due to Parkinson's disease. It was found that there were significant differences among the groups. Most notably, those with Parkinson's disease had a significantly more random sway pattern in the M/L direction than the healthy groups, suggesting that quantifying the fractal dimension identified an intrinsic difference between them. Findings also demonstrated differences between healthy young and healthy elderly groups, suggesting an age-related change in postural sway patterns. This study demonstrated that fractal analysis was a useful tool and should be considered as a quantitative way to monitor changes in health.