RESUMO
OBJECTIVE: The aim of the study is to investigate the comparative effects of nonpharmacological therapies for managing global, attention, memory, and execution cognitive functions in stroke patients. DESIGN: We searched PubMed, Embase, CINAHL, Cochrane Library, Web of Science, PEDro, and Google Scholar for randomized controlled trials that evaluated the effects of nonpharmacological therapies for treating stroke cognitive dysfunctions. We performed a network meta-analysis to estimate the mean treatment effect of 95% credible interval. RESULTS: Seventy-three randomized controlled trials were included in the network meta-analysis for evidence syntheses. All therapies had significant effects than control on global cognition in stroke patients. Combined therapy was superior to other therapies for global cognition of all patients (vs. cognitive task therapy: 0.71, 95% credible interval = 0.14 to 1.29; vs. exercise: 0.88, 95% credible interval = 0.31 to 1.45, vs. physical modality therapy: 0.77, 95% credible interval = 0.16 to 1.40). Different therapies have effects on specific cognitive domains in stroke patients. CONCLUSIONS: Our findings suggest that nonpharmacological therapies are effective in improving global cognitive function in stroke patients, with cognitive task therapy, exercise therapy, physical modality therapy, and combined therapy being viable options (most optimal approach: combined therapy). Precise selection of therapies based on the time since stroke onset and specific cognitive domains can further enhance treatment outcomes.