The aim of this study was to investigate the effect of eight weeks of DNS training on dynamic and static balance, fall prevention and lower limb strength in healthy elderly women. This is a quasi-experimental research with pre-post-test. Thirty healthy elderly women (60-79 years old) selected purposefully and divided into two groups of exercise (DNS) and control randomly. Static balance were assessed by the Berg and Sharpendromberg balance test, dynamic balance by the Timed Up and Go test, prediction of fall risk by the Shumway-Cook model, and lower extremity strength by the Chair Stand test. Paired t-test and independent t-test were used to analyze the data. The training group showed a significant difference between pre and post-test and also compared to control group in all factors. According to the present findings, the effect of DNS in improving research variables is confirmed and its use in elderly care programs is recommended.