Introduction <p>Female sexual dysfunction (FSD) is highly prevalent in women with diabetes and is influenced by biopsychosocial factors. Although psychological interventions such as PRECEDE, PLISSIT, BETTER, and structured counseling show promise, direct comparative evidence is lacking. This network meta-analysis aimed to evaluate and rank the efficacy of these models in improving sexual function among diabetic women.</p> Method <p>The protocol was prospectively registered in PROSPERO (CRD420251185900). We systematically searched PubMed/MEDLINE, Embase, Cochrane Library, Web of Science, PsycINFO, CINAHL, and Scopus from January 2015 to the end of 2025. Eligible studies were randomized controlled trials (RCTs) and quasi-experimental studies (including controlled before-after and single-arm pre-post designs) that enrolled adult women (≥ 18 years) with diabetes mellitus (type 1, type 2, or unspecified; gestational diabetes-only studies excluded), compared PRECEDE, PLISSIT, BETTER, or structured counseling versus control (usual care, waitlist, or no intervention), and reported FSFI total scores (or comparable validated scales) at post-intervention. No language restrictions were applied. A frequentist random-effects network meta-analysis was conducted using standardized mean differences (SMDs). Interventions were ranked using P-scores. Heterogeneity was quantified with I² and τ². Risk of bias was assessed using Cochrane RoB 2 (RCTs) and ROBINS-I (non-randomized studies). Publication bias was evaluated via funnel plots, Egger's, and Begg's tests. Meta-regression explored effects of study design and sample size.</p> Result <p>Of 2,347 records, 10 studies (3 RCTs and 7 quasi-experimental studies; N = 1,294 participants) were included. The network was star-shaped, with all active interventions connected only to the control group and no direct head-to-head comparisons between interventions; therefore, all comparative effects are based solely on indirect evidence. Accordingly, treatment rankings and P-score (SUCRA-equivalent) values should be interpreted with caution and considered exploratory rather than definitive indicators of comparative effectiveness. All interventions demonstrated positive effect sizes compared with the control. PRECEDE showed the largest point estimate (SMD = 2.14, 95% CI: − 0.02 to 4.30, P-score = 0.471), followed by BETTER (SMD = 1.38, 95% CI: − 0.38 to 3.15, P-score = 0.222), counseling (SMD = 0.92, 95% CI: − 0.05 to 1.89, P-score = 0.136), and PLISSIT (SMD = 0.73, 95% CI: − 0.59 to 2.05, P-score = 0.115). Also, substantially high heterogeneity was observed (I² = 97.7%, τ² = 1.19). However, all confidence intervals overlapped and included null, indicating no statistically significant difference between any active intervention and control. Subgroup analysis revealed significant differences by study design (p = 0.0344), with quasi-experimental studies showing larger effects (SMD = 1.70) than RCTs (SMD = − 0.69). Meta-regression confirmed study design as a significant moderator (p = 0.002). Most studies were rated as having a high risk of bias in the blinding/performance domain.</p> Conclusion <p>This network meta-analysis suggests that psychological interventions (PRECEDE, PLISSIT, BETTER, and counseling) may improve sexual dysfunction in women with diabetes. Based on indirect comparisons only, PRECEDE showed the highest point estimate; however, these findings are exploratory and do not demonstrate superiority of any single intervention. Given that these rankings are derived entirely from indirect comparisons, they should be interpreted cautiously and should not be considered definitive evidence of comparative superiority. Given that all rankings are derived entirely from indirect comparisons in a star-shaped network with no direct head-to-head trials, they should be interpreted very cautiously and should not be considered definitive evidence of comparative superiority. No intervention can be confidently recommended over others based on the current evidence.</p>