Background and purpose The aim was to systematically review the effectiveness and safety of telemedicine combined with usual care (in‐person visits) compared to usual care for the therapeutic management and follow‐up assessment of neurological diseases. Methods The electronic databases MEDLINE, Embase, Web of Science and Cochrane Central Register of Controlled Trials were searched (June 2021). Randomized controlled trials (RCTs) on patients of any age with neurological diseases were considered. Two reviewers screened and ed data in duplicate and independently and assessed risk of bias using the Cochrane risk‐of‐bias tool for randomized trials (RoB 2). When possible, pooled effect estimates were calculated. Results Of a total of 3018 records initially retrieved, 25 RCTs (n = 2335) were included: 11 (n = 804) on stroke, four (n = 520) on Parkinson's disease, three (n = 110) on multiple sclerosis, two (n = 320) on epilepsy, one (n = 63) on dementia, one (n = 23) on spina bifida, one (n = 40) on migraine, one (n = 22) on cerebral palsy and one (n = 433) on brain damage. Types of telemedicine assessed were online visits (11 studies), tele‐rehabilitation (seven studies), telephone calls (three), smartphone apps (two) and online computer software (two). The evidence was quite limited except for stroke. Compared to usual care alone, telemedicine plus usual care was found to improve depressive symptoms, functional status, motor function, executive function, generic quality of life, healthcare utilization and healthy lifestyle in patients in post‐stroke follow‐up. Conclusions Well‐designed and executed RCTs are needed to confirm our findings on stroke and to have more scientific evidence available for the other neurological diseases.