Efficacy of non-pharmacological interventions for pediatric primary headaches: a systematic review and network meta-analysis.

Abstract

Background: Primary headaches are common in children and adolescents; however, the comparative efficacy of multiple non-pharmacological interventions remains unclear. This study aimed to compare the efficacy of different non-pharmacological interventions for pediatric primary headaches using a network meta-analysis.

Methods: We systematically searched PubMed, Cochrane Library, Web of Science, and Embase databases for randomized controlled trials (RCTs) that investigated non-pharmacological interventions for pediatric primary headaches, from database inception to August 1, 2026. The risk of bias was assessed using the Cochrane risk of bias tool, and the certainty of evidence was evaluated using confidence in network meta-analysis (CINeMA). A network meta-analysis (NMA) was conducted using Stata 17 to generate network diagrams, league tables, and surface under the cumulative ranking curve values. The protocol was registered in PROSPERO.

Results: A total of 16 RCTs involving 820 participants were included. Network meta-analysis showed that, compared with waiting list control (WLC), biofeedback [standardized mean difference (SMD) = -1.11, 95% confidence interval (95% CI) (-2.04, -0.17)], therapist-assisted relaxation [SMD = -1.06, 95% CI (-2.04, -0.07)], and relaxation [SMD = -0.69, 95% CI (-1.21, -0.17)] were associated with reductions in headache frequency. Headstrong compact disc read-only memory (CD-ROM) [SMD = -1.63, 95% CI (-2.95, -0.30)], biofeedback + relaxation [SMD = -1.54, 95% CI (-2.95, -0.13)], and relaxation [SMD = -0.91, 95% CI (-1.61, -0.21)] showed potential benefits in shortening headache duration. Education [SMD = -0.96, 95% CI (-1.72, -0.19)], biofeedback [SMD = -0.85, 95% CI (-1.59, -0.11)], therapist-administered group training (TG) [SMD = -0.78, 95% CI (-1.39, -0.17)], cognitive behavior therapy (CBT) [SMD = -0.76, 95% CI (-1.33, -0.19)], and relaxation [SMD = -0.63, 95% CI (-1.24, -0.02)] were associated with improvements in headache intensity. According to surface under the cumulative ranking curve (SUCRA) values, biofeedback (81.4%), headstrong CD-ROM (80.5%), and education (75.7%) showed the highest-ranking probabilities for reducing headache frequency, shortening headache duration, and alleviating headache intensity, respectively. These rankings reflect relative probabilities rather than definitive clinical superiority.

Conclusions: Biofeedback, headstrong CD-ROM, and education showed relatively favorable rankings for specific headache outcomes. However, these findings are exploratory and should be interpreted with caution given imprecise effect estimates and very low certainty of evidence.