BACKGROUND: The growing use of smartphones among elderly individuals, driven by social and informational needs, may lead to smartphone addiction, potentially impacting their daily lives. This study aimed to determine whether there is a difference in physical activity, activities of daily living, and balance levels between elderly individuals with and without smartphone addiction. METHODS: This descriptive and cross-sectional study included 94 elderly individuals. Data were obtained using the Smartphone Addiction Scale-Short Version (SAS-SV), the Physical Activity Scale for the Elderly (PASE), the Lawton Instrumental Activities of Daily Living Scale (Lawton IADL), the Fullerton Advanced Balance Scale (FAB-T), and the Timed Up and Go Test (TUG). The participants were divided into two groups according to their SAS-SV scores: those with (n = 45) and those without (n = 49) smartphone addiction. RESULTS: When the groups with and without smartphone addiction were compared, there was a significant difference between the groups in terms of Lawton IADL (t = 4.223, p <
0.001), total PASE (t = 7.791, p <
0.001), PASE work-related activity (t = 2.541, p = 0.013), household activity (t = 3.598, p = 0.001), and leisure activity (t = 7.063, p <
0.001). Structural equation modeling showed that Lawton IADL (β = -0.320, p <
0.001), PASE total (β = -0.518, p <
0.001), and PASE work-related activity (β = -0.211, p = 0.033), household activity (β = -0.300, p = 0.002), and leisure time activity (β = -0.483, p <
0.001) subscales had a direct negative predictive effect on SAS-SV. FAB-T had a direct positive predictive effect on total PASE (β = 0.186, p = 0.030) and work-related activity subscales (β = 0.197, p = 0.046). FAB-T had a direct positive predictive effect on Lawton IADL (β = 0.247, p = 0.009), but a direct negative effect on TUG (β = -0.541, p <
0.001). CONCLUSIONS: The study determined that smartphone addiction was directly related to the maintenance of physical activity and daily living activities in elderly individuals but did not lead to a change in balance status. Future studies should consider including potential confounders, such as baseline physical fitness, socioeconomic status, and cognitive impairment, in structural equation modeling to provide more comprehensive insights.