Social participation is a core indicator of long-term recovery, but it significantly decreases after a stroke. Diverse single-, bi-, or multi-dimensional interventions showed differing effects on improving stroke survivors’ participation.

To compare the effectiveness of current interventions based on the social ecological model in improving social participation among stroke survivors.

Ten electronic databases were searched from their inception to April 2024, and additional searches were conducted in ProQuest, Google Scholar, and the reference lists of relevant reviews. Randomized controlled trials of interventions with the primary aim of improving social participation for stroke survivors aged 18 years and older were included. Pairwise meta-analysis and network meta-analysis were conducted. The intervention hierarchy was evaluated using the surface under the cumulative ranking curve (SUCRA) values.

A total of 32 articles with 3211 participants were included, and 29 articles were eligible for meta-analysis. Pairwise meta-analysis indicated that different interventions were effective in improving social participation compared to control groups at post-intervention, but not at 3, 6, and 12 months after the intervention. The network meta-analysis showed that personal-physical environmental dimensional interventions were the most effective in improving social participation between baseline and post-intervention (SUCRA 86 %, SMD 0.72, 95 % CI 0.20–1.25), followed by personal-interpersonal-physical environmental dimensional interventions (SUCRA 85 %, SMD 0.75, 95 % CI 0.23–1.27) compared to control groups.

Addressing personal and physical environmental barriers, or additionally incorporating interpersonal relationships improvement techniques, appears to be most effective in enhancing social participation among stroke survivors. Findings highlight the need for more rigorous multi-dimensional interventions to provide robust evidence.