Permitting the use of motor compensations could lead to a pattern of learned nonuse (Alaverdashvili and Whishaw 2010, 2013; Allred et al. 2005; Taub et al. 1993), limiting the capacity for subsequent gains in motor function of the paretic arm. Interventions that include the restriction of trunk motor compensations by physical trunk restraint may encourage the nervous system to find new motor solutions to task accomplishment and to overcome learned nonuse. These motor solutions may be more effective in improving upper-limb function through the emergence of new motor patterns. Improvement in upper-limb function will decrease activity limitations and social participation restrictions.

Trunk-restraint therapy is beneficial for motor recovery in adults (16–80 years of age) with chronic acquired brain damage (stroke) leading to disrupted motor control of the trunk and arms. Beneficial effects have also been demonstrated in children with cerebral palsy for both improvement in motor function (Schneiberg et al. 2010) and motor learning (Robert et al. 2013). The principles should be equally applicable to individuals with other types of acquired brain damage such as traumatic brain injury , and to children with other developmental motor disorders.

Trunk restraint combined with task-related training improved arm motor function in adults with stroke in six high-quality randomized control studies (Michaelsen and Levin 2004; Michaelsen et al. 2006; Woodbury et al. 2009; Thielman 2010; Wu et al. 2012a, b), in one pre- and postdesign study (de Oliveira et al. 2007) and in one high-quality randomized control trial in children with cerebral palsy (Schneiberg et al. 2010). Michaelsen et al. (2006) compared arm motor impairment and function in clients with stroke practicing task-related training with (n = 15) or without trunk restraint (n = 15). Clients in the trunk-restraint group made greater improvements in motor function than those in the control group. Improvements were accompanied by increased active joint range and were greater in clients with greater initial severity of hemiparesis. In these clients, task-related training with trunk restraint led to less trunk movement and increased elbow extension during reaching, while clients in the control group increased compensatory trunk movement. In addition, ­changes in arm functions were correlated with changes in arm and trunk kinematics in the trunk-restraint group. Schneiberg et al. (2010) compared task-oriented training with and without trunk restraint in 12 children with cerebral palsy in a prospective single-subject randomized control trial. Each group received a total of 15 h of training. Some children in each group improved arm trajectory smoothness and elbow extension range. However, more children in the trunk-restraint group reduced their excessive trunk displacement after training and retained the improvements longer.