OBJECTIVE:The aim of this study was to evaluate if somatosensory retraining programmes assist people to improve somatosensory discrimination skills and arm functioning after stroke. DATA SOURCES:Nine databases were systematically searched: Medline, Cumulative Index to Nursing and Allied Health Literature, PsychInfo, Embase, Amed, Web of Science, Physiotherapy Evidence Database, OT seeker, and Cochrane Library. REVIEW METHODS:Studies were included for review if they involved (1) adult participants who had somatosensory impairment in the arm after stroke, (2) a programme targeted at retraining somatosensation, (3) a primary measure of somatosensory discrimination skills in the arm, and (4) an intervention study design (e.g. randomized or non-randomized control designs). RESULTS:A total of 6779 articles were screened. Five group trials and five single case experimental designs were included ( N = 199 stroke survivors). Six studies focused exclusively on retraining somatosensation and four studies focused on somatosensation and motor retraining. Standardized somatosensory measures were typically used for tactile, proprioception, and haptic object recognition modalities. Sensory intervention effect sizes ranged from 0.3 to 2.2, with an average effect size of 0.85 across somatosensory modalities. A majority of effect sizes for proprioception and tactile somatosensory domains were greater than 0.5, and all but one of the intervention effect sizes were larger than the control effect sizes, at least as point estimates. Six studies measured motor and/or functional arm outcomes ( n = 89 participants), with narrative analysis suggesting a trend towards improvement in arm use after somatosensory retraining. CONCLUSION:Somatosensory retraining may assist people to regain somatosensory discrimination skills in the arm after stroke.