[Chapter 1] Review of backward walkingPurpose: The purpose of this study was to review existing literature on gait analysis and physical therapy related to backward walking (BW).Methods: English language and Japanese language studies were identified throughsearches of CiNii, PubMed and the Internet.Results: Fifty-five studies were identified.Conclusion: While some studies concluded that the joint movement patterns in BW and forward walking (FW) are similar, several studies reported that these patterns in BW and FW are different.[Chapter 2-1]Comparison of forward walking and backward walking in healthy young adultsPurpose: The purpose of this study was to compare the characteristics of BW and FW in healthy young adults.Subjects: Fourteen healthy young adults.Methods: Subjects were required to walk over a 5-m walkway at a self-selected pace.Results: Walking speed, stride length, and cadence were significantly lower in BW than in FW. The peak hip extension, knee flexion, ankle plantar flexion, and the peak hip and 77 knee joint moments were also significantly lower in BW.Conclusion: We conclude that BW is not a simple reverse reproduction of FW.[Chapter 2-2]Effects of ankle push-off during backward walking on ankle and hip jointsPurpose: The purpose of this study was to investigate the differences in the kinetic and kinematic factors of different ankle push-offs during BW.Subjects: Fourteen healthy young adults.Methods: Subjects were required to walk over a 5-m walkway at a self-selected paceunder three conditions: Natural Push-off BW, Increased Push-off BW, and DecreasedPush-off BW.Results: Walking speed and stride length increased in the order of Decreased Push-offBW, Natural Push-off BW, and Increased Push-off BW. The peak value of the hipextension angle was higher in Increased Push-off BW than in Decreased Push-off BW.Conclusion: Ankle push-off can help increase the walking speed and improve thepropulsive force of the ankle joint in BW. Hip joint movement plays an important role in BW to accommodate for the disadvantage ankle strategies have in this type of walking.[Chapter 3]Motion analysis of backward walking in stroke patients with hemiplegiaPurpose: The purpose of this study was to investigate the features of BW in strokepatients with hemiplegia.Subjects: Nine stroke patients.Methods: Participants performed FW and BWalong a 5-m walkway. Walking speed and stride length were self-selected.Results: On the paretic side, walking speed, stride length, and cadence were significantlylower in BWthan in FW. Peak hip extension was significantly lower in BW, and peak hip flexion moment, knee extension moment, and both ankle dorsiflexion and plantar flexion 78 moments were lower in BW.Conclusion: Unlike FW, BW requires conscious extension of the hip joint, which isdifficult for stroke patients with hemiplegia. Therefore, the range of hip joint movementwas smaller in BW, and walking speed and stride length were decreased. The peak ankle plantar flexion moment was significantly lower in BW than in FW, and it was difficult to generate propulsion power in BW. These difficulties also affected the walking speed.