Comparison of clinical efficacy of different exercise therapies in the treatment of chronic nonspecific neck pain: a network meta-analysis

Background Chronic nonspecific neck pain (CNSNP) is a common musculoskeletal disorder that significantly impacts patients’ quality of life and imposes a considerable social and economic burden. Although drug treatment is widely used, it has limitations such as adverse reactions and limited long-term efficacy. Therefore, exercise therapy has gradually become the first-line intervention method. However, there is still a lack of systematic evidence for the efficacy comparison among different exercise therapies. Objective This study conducted a network meta-analysis to compare the clinical efficacy of different exercise therapies in treating CNSNP, providing evidence support for clinical intervention selection. Methods Systematic searches were conducted in PubMed, Web of Science, Embase, Cochrane Library, VIP, SinoMed, CNKI, and Wanfang databases up to November 25, 2025, for randomized controlled trials comparing the efficacy of different exercise therapies in patients with CNSNP. Inclusion criteria were studies comparing the efficacy of different exercise therapies in patients with CNSNP. The primary outcome measures were the Neck Disability Index (NDI) and Visual Analogue Scale (VAS). The quality of studies was evaluated using the Cochrane Risk of Bias Assessment Tool 2.0. Network meta-analysis was performed using Stata 16.0 and R 4.5.2, and the intervention measures were ranked using the cumulative ranking area under the curve (SUCRA). The evidence quality was evaluated using the CINeMA tool. Results A total of 17 studies were included, involving 1,224 patients and 16 intervention measures. The risk of bias assessment showed that 12 studies (70.6%) were of moderate risk, and 5 studies (29.4%) were of high risk. There was no significant difference in improving neck function (NDI) among the intervention measures. The SUCRA value of BBAT (biological feedback-assisted training) was the highest (0.78), but this result was supported by only a small sample of single studies, and the robustness was limited. In terms of reducing pain (VAS), muscle energy technique (MET) was significantly superior to conventional exercise therapy ( p 0.05), with a SUCRA value of 0.88. However, sensitivity analysis showed that this result was somewhat dependent on a single study. Meta-regression showed that the treatment duration had no significant effect on the efficacy. Publication bias analysis did not reveal significant bias. The CINeMA evidence quality evaluation showed that the overall credibility of the current evidence was moderate. Conclusion Muscle energy technique (MET) shows certain advantages in alleviating chronic nonspecific neck pain, but there is no significant difference among the intervention measures in improving neck function. Due to the sparse network connection of some intervention measures and the reliance on indirect evidence in most comparisons, the SUCRA ranking results should be regarded as exploratory findings and are not sufficient to form clear clinical recommendations. Future high-quality, large-sample, and long-term follow-up head-to-head randomized controlled trials are needed to further verify the efficacy differences of different exercise therapies. Systematic review registration https://www.crd.york.ac.uk/PROSPERO/view/CRD420251249215 .