S1 File - <p> </p>

Background Medication discontinuation is highly prevalent among patients with schizophrenia and is associated with poor clinical outcomes. Although community-based management models have been implemented in China to support patients, recent evidence of their effectiveness in reducing medication discontinuation rates remains limited. Methods In this cross-sectional study, we surveyed 1,531 patients with schizophrenia under community management in Chengdu, China, using a multi-stage sampling approach. Data on treatment discontinuation (defined as cessation of all antipsychotics for >15 days without medical advice), socio-demographics, clinical characteristics, and service utilization were collected via face-to-face interviews. Univariate and multivariable logistic regression analyses were used to identify the factors associated with medication discontinuation. Results The overall rate of medication discontinuation was 4.1%. Multivariable analysis identified several independent risk factors for discontinuation: weak stable disease state (adjusted odds ratio [aOR] = 2.70, 95% confidence interval [CI]: 1.34–5.24), lack of insight (aOR for partial vs. no insight = 0.25, 95% CI: 0.08–0.80; aOR for full vs. no insight = 0.18, 95% CI: 0.05–0.65), presence of noticeable side effects (aOR = 1.98, 95% CI: 1.05–3.64), and absence of regular follow-up (aOR for intermittent vs. no follow-up = 0.38, 95% CI: 0.20–0.72; aOR for regular vs. no follow-up = 0.22, 95% CI: 0.10–0.44). Conclusion Community management was associated with a low rate of medication discontinuation in patients with schizophrenia. Key modifiable factors, including disease stability, insight, medication side effects, and follow-up adherence should be prioritized in community-based interventions to further improve treatment continuity.