Objectives of the review and PICO: To evaluate the effectiveness of the Bobath concept (Intervention) on mobility, motor control of the upper and lower limb including dexterity and gait, balance, and activities of daily living (Outcomes) of stroke survivors (Population), with the effect of other sensorimotor training approaches (Comparators).

Search strategies: Searches were conducted in PUBMED, CENTRAL, CINAHL and PEDro by two independent researchers up to January 2018, for randomised controlled trials published in English, German, French or Spanish, adults only. The authors checked reference lists from three previous reviews about Bobath.

Inclusion/exclusion criteria: Published randomised controlled trials involving adults with stroke or central nervous system involvement. Pilot RCTs excluded. In the second phase of screening, they excluded trials that did not clearly specify that treatment was based on the Bobath concept, used treatment methods based exclusively on the first textbook written by Berta Bobath (1970 and 1978), used old theoretical foundations of the Bobath concept, used a multimodal approach where intervention was combined with Bobath or unclear, where the control groups received a placebo intervention or similar, or the intervention lacked a scientific justification for its use in stroke, and where stroke was not specified or involved a non-stroke population. Eight studies, some with a large sample size, were excluded based on language – Polish, Chinese, Norwegian, Lithuanian (n=240), Danish (n=65) and Italian (n=20).

Study samples: Stroke participants early post-stroke (less than one month, 4 trials), sub-acute (one to 12 months post-stroke, 6 trials), or in the chronic phase (more than one year post-stroke, 5 trials).

Methods of the review: Two authors independently reviewed titles, abstracts and full-text papers. Two independent raters assessed methodological quality and risk of bias using the PEDro scale. Studies were grouped and analysed according to the four key outcomes. No attempt was made to combine data or conduct a meta-analysis of studies using the same outcome measures. A qualitative (descriptive text) synthesis was conducted where studies were grouped according to the main outcome (mobility, balance, upper limb control, lower limb control/gait and ADL). Studies were then collectively classified as being (i) no different (ie no statistically significant between-group difference), or one treatment was (ii) better/superior or (iii) worse than the comparator.

Key Outcomes and Measures:
(i) Mobility = muscle tone, modified Ashworth Scale, FIM, Motor Assessment Scale, Rivermead Assessment Scale,
(ii) Upper limb function = Fugl-Meyer Assessment, Wolf Motor Function test, Box and Block Test, Nine Hole Peg Test, ARAT, Motor Activity Log, Motor Assessment Scale, Sodring Motor Evaluation Scale, FIM;
(iii) Lower limb/gait = muscle tone, motricity index, ROM Fugl-Meyer Assessment, FIM, gait parameters, Motor Assessment Scale, Rivermead Assessment, Timed up and go, 6 minute walk test,
(iv) Balance = stabilometer platform, Berg Balance Test, Rivermead Motor Assessment, Motor Assessment Scale;
(v) Activities of daily living =  FIM, Barthel Index, EADL.

Results: The review included 15 trials and 781 stroke participants. The PEDro scores for methodological quality ranged from 4 to 8; 12/15 studies were rated as having ‘good’ methodological quality which minimised bias (6-8), and 3/15 studies had ‘moderate’ methodological quality (4-5).  Random allocation was not concealed in 12/15 trials, and blinded assessors were not used in 4/15 trials.

Comparator interventions included (i) contemporary treatments such as constraint or forced use therapy, robotic devices/assisted therapy, gait trainer with partial weight support, movement science / motor relearning program, bilateral arm training with auditory cueing, mirror image movement enabler, problem-orientated willed-movement/motor task oriented), (ii) traditional functional training and (iii) proprioceptive neuromuscular facilitation (PNF). Intervention duration ranged from 30 to 60 minutes per session, 3 to 5 days per week, for between 2 to 8 weeks.

Mobility (n=6 trials): 4/6 trials = “no (statistically significant) difference” between Bobath v PNF, motor relearning program, movement science, conventional /traditional functional training; 2/6 trials= Bobath was “better” than conventional orthopaedic treatment in one trial, and “problem-orientated willed movement” was ‘better’ than Bobath. Author’s conclusion: No superiority of Bobath or any other approach for improving mobility.

Upper limb motor control and dexterity (n=9 trials): 5/9 trials = “experimental treatments ‘superior’ to Bobath, including two trials using a robotic device (including mirror image movement enabler), two trials using forced use and one trial using task-specific training. 4/9 trials = “no (statistically significant) difference” between Bobath and motor relearning program, functional retraining and bilateral training with auditory cuing. Author’s conclusion:  Therapy approaches that force use of the affected upper limb, including CIMT and robotic devices are superior to Bobath.

Lower limb control/gait (n=7 trials): 5/7 trials = ‘no statistically significant difference” between Bobath and motor relearning program, movement science, PNF and traditional functional exercises, gait training with partial body support. One study showed superiority of Bobath over an orthopaedic treatment approach in the short term in acute patients. Another study showed superiority of problem-orientated willed movement over Bobath, in patients with cognitive problems in the short term. Author’s conclusion: No superiority of Bobath or any other approach for improving motor control of the lower limb and/or gait.

Balance (n=5 trials); 3/7 trials = “no (statistically significant) difference” between Bobath and orthopaedic approach, gait trainer with partial body support. In a Norwegian trial and a second UK trial, ‘statistically significant differences’ were reported in favour of motor relearning program compared to Bobath at the completion of treatment but differences were not sustained 3 months post-intervention. Finally, there was a ‘statistically significant difference’ in favour of the Bobath concept compared to PNF. Author’s conclusion: No superiority of Bobath or any other approach for improving balance.

ADL (n=8 trials): 6/8 = ‘no difference’ between Bobath and treatments including motor relearning program, functional training, movement science approach, CIMT and conventional training.  Short-term differences (not sustained after 6 months) in favour of mirror image movement enabler approach and robot-assisted therapy over Bobath. Author’s conclusion: No superiority of Bobath or any other approach for improving ADL performance.

Journal authors’ conclusions: Bobath is not superior (no better, no worse) than other treatments / approaches for improving lower limb control or gait, balance, or ADL performance in stroke survivors. There is moderate evidence that upper limb outcomes and dexterity are better (when compared to outcomes from Bobath), when treatments involving forced use are used such as CIMT and robot-assisted devices.

  • Díaz-Arribas, M.J., Martín-Casas, P., Cano-de-la-Cuerda, R., & Plaza-Manzano, G. (2019 online early) Effectiveness of the Bobath concept in the treatment of stroke: A systematic review, Disability and Rehabilitation, DOI: 10.1080/09638288.2019.1590865