Background: With the emergence of electronic products, smartphones have become an indispensable tool in our daily life. On the other hand, smartphone addiction has become a public health issue. To help reduce smartphone addiction, cost-effective interventions such as exercise are encouraged. Purpose: We therefore performed a systematic review and meta-analysis evaluating existing literature on the rehabilitative effects of exercise interventions for individuals with a smartphone addiction. Methods: We searched PubMed, Web of Science, Scopus, CNKI, and Wanfang from inception to September 2019. Nine eligible randomized controlled trials (RCT) were finally included for meta-analysis (SMD represents the magnitude of effect of exercise) and their methodological quality were assessed using the PEDro scale. Results: We found significant positive effects of exercise interventions (Taichi, basketball, badminton, dance, run, and bicycle) on reducing the total score (SMD = -1.30, 95% CI -1.53 to -1.07, p < 0.005, I-2 = 62%) of smartphone addiction level and its four subscales (withdrawal symptom: SMD = -1.40, 95% CI -1.73 to -1.07, p < 0.001, I-2 = 81%; highlight behavior: SMD = -1.95, 95% CI -2.99 to -1.66, p < 0.001, I-2 = 79%; social comfort: SMD = -0.99, 95% CI -1.18 to -0.81, p = 0.27, I-2 = 21%; mood change: SMD = -0.50, 95% CI 0.31 to 0.69, p = 0.25, I-2 = 25%). Furthermore, we found that individuals with severe addiction level (SMD = -1.19, I-2 = 0%, 95%CI:-1.19 to -0.98) benefited more from exercise engagement, as compared to those with mild to moderate addiction levels (SMD = - 0.98, I-2 = 50%, 95%CI:-1.31 to -0.66); individuals with smartphone addiction who participated in exercise programs of 12 weeks and above showed significantly greater reduction on the total score (SMD = -1.70, I-2 = 31.2%, 95% CI -2.04 to -1.36, p = 0.03), as compared to those who participated in less than 12 weeks of exercise intervention (SMD = -1.18, I-2 = 0%, 95% CI-1.35 to -1.02, p < 0.00001). In addition, individuals with smartphone addiction who participated in exercise of closed motor skills showed significantly greater reduction on the total score (SMD = -1.22, I-2 = 0 %, 95% CI -1.41 to -1.02, p = 0.56), as compared to those who participated in exercise of open motor skills (SMD = -1.17, I-2 = 44%, 95% CI-1.47 to -0.0.87, p = 0.03). Conclusions: Exercise interventions may have positive effects on treating smartphone addiction and longer intervention durations may produce greater intervention effects.