Background Augmented reality (AR) is a rapidly expanding technology; it comprises the generation of new images from digital information in the real physical environment of a person, which simulates an environment where the artificial and real are mixed. The use of AR in physiotherapy has shown benefits in certain areas of patient health. However, these benefits have not been studied as a whole. Objective This study aims to ascertain the current scientific evidence on AR therapy as a complement to physiotherapy and to determine the areas in which it has been used the most and which variables and methods have been most effective. Methods A systematic review registered in PROSPERO (International Prospective Register of Systematic Reviews) was conducted following PRISMA (Preferred Reporting Items for Systematic Reviews and Meta‐Analyses) recommendations. The search was conducted from July to August 2021 in the PubMed, PEDro, Web of Science, Scopus, and Cochrane Library scientific databases using the keywords augmented reality, physiotherapy, physical therapy, exercise therapy, rehabilitation, physical medicine, fitness, and occupational therapy. The methodological quality was evaluated using the PEDro scale and the Scottish Intercollegiate Guidelines Network scale to determine the degree of recommendation. The Cochrane Collaboration tool was used to evaluate the risk of bias. Results In total, 11 articles were included in the systematic review. Of the 11 articles, 4 (36%) contributed information to the meta-analysis. Overall, 64% (7/11) obtained a good level of evidence, and most had a B degree of recommendation of evidence. A total of 308 participants were analyzed. Favorable results were found for the Berg Balance Scale (standardized mean change 0.473, 95% CI −0.0877 to 1.0338; z=1.65; P=.10) and the Timed Up and Go test (standardized mean change −1.211, 95% CI −3.2005 to 0.7768; z=−1.194; P=.23). Conclusions AR, in combination with conventional therapy, has been used for the treatment of balance and fall prevention in geriatrics, lower and upper limb functionality in stroke, pain in phantom pain syndrome, and turning in place in patients with Parkinson disease with freezing of gait. AR is effective for the improvement of balance; however, given the small size of the samples and the high heterogeneity of the studies, the results were not conclusive. Future studies using larger sample sizes and with greater homogeneity in terms of the devices used and the frequency and intensity of the interventions are needed. Trial Registration PROSPERO International Prospective Register of Systematic Reviews CRD42020180766; https://www.crd.york.ac.uk/prospero/display_record.php?RecordID=180766