Effects of non-pharmacological interventions for social participation in stroke survivors: a systematic review and network meta-analysis.

BACKGROUND

The optimal non-pharmacological intervention for improving social participation after stroke remains unclear. This network meta-analysis (NMA) aimed to compare the effectiveness of different interventions and identify the best approach.

METHODS

Cochrane Library, Web of Science, PubMed, Embase, CINAHL plus, PsycINFO, Scoups, Wanfang database, CQVIP, and CNKI were systematically searched for randomized controlled trials (RCTs), since their inception to March 2025. The primary outcome was social participation scores immediately postintervention. We employed frequentist-based NMA for the comparative effectiveness calculations. The main analyses were network plotting to display the network geometry and P scores to summarize the hierarchy. The Confidence in Network Meta-Analysis (CINeMA) framework assessed evidence certainty.

RESULTS

The NMA analyzed 92 RCTs on 7724 participants and 15 different intervention modalities. The 4 highest-ranked interventions were psychological interventions (effect size: 10.27, 95% CI 3.32-17.22, P-score = 0.90, low certainty of evidence), complex interventions (3.59, 95% CI 1.65-5.54, P-score = 0.62, moderate certainty of evidence), occupational therapy plus cognitive therapy (3.21, 95% CI 0.17-6.26, P-score = 0.55, moderate certainty of evidence), and exercise-based interventions (3.03, 95% CI 1.94-4.13, P-score = 0.54, moderate certainty of evidence). However, results for complex interventions and exercise-based interventions remained robust when excluding trials with a high risk of bias, younger age groups (≤50 years), or those with an onset time ≤3 months.

CONCLUSIONS

Psychological interventions, complex interventions, exercise-based interventions, and occupational therapy plus cognitive therapy were the most effective interventions in improving social participation in stroke survivors. However, it is important to note that the evidence for psychological interventions and occupational therapy plus cognitive therapy was sensitive to a high risk of bias. Therefore, these findings should be interpreted with caution, and these interventions warrant further high-quality investigation.

TRIAL REGISTRATION: PROSPERO number: CRD420251000688.