OBJECTIVE:: To investigate the effectiveness of backward walking in the treatment of people with gait impairments related to neurological and musculoskeletal disorders. DESIGN:: Systematic review and meta-analysis of randomized and quasi-randomized control studies. DATA SOURCES:: Searched from the date of inception to March 2018, and included PubMed, Scopus, Cochrane Library, PEDro, CINAHL, and the MEDLINE databases. METHODS:: Investigating the effects of backward walking on pain, functional disability, muscle strength, gait parameters, balance, stability, and plantar pressure in people with gait impairments. The PEDro scale was used to assess the quality. Similar outcomes were pooled by calculating the standardized mean difference. RESULTS:: Of the 21 studies (neurological 11 and musculoskeletal 10), 635 participants were included. The average PEDro score was 5.4/10. The meta-analysis demonstrated significant standardized mean difference values in favour of backward walking, with conventional physiotherapy treatment for two to four weeks to reduce pain (-0.87) and functional disability (-1.19) and to improve quadriceps strength (1.22) in patients suffering from knee osteoarthritis. The balance and stability in cases of juvenile rheumatoid arthritis, and gait parameters and muscle strength in anterior cruciate ligament injury improved significantly when backward walking was included as an exercise. There was no significant evidence in favour of backward walking in any of the other conditions. CONCLUSION:: The systematic review and meta-analysis suggests that backward walking with conventional physiotherapy treatment is effective and clinically worthwhile in patients with knee osteoarthritis. Insufficient evidence was available for the remaining gait impairment conditions and no conclusions could be drawn.