OBJECTIVE: We aimed to analyze the effects and dose-response relationship of the most effective exercises for improving pain and disability in people with chronic nonspecific neck pain.DESIGN: Intervention systematic review with meta-analysis. LITERATURE SEARCH: We searched the PubMed, PEDro, and CENTRAL databases from their inception to September 30, 2022. STUDY SELECTION CRITERIA: We included randomized controlled trials that involved people with chronic neck pain adopting a longitudinal exercise intervention and assessed one pain and/ or disability outcome. DATA SYNTHESIS: Restricted maximum-likelihood random-effects meta-analyses were modeled separately for resistance, mindfulness-based, and motor control exercises; standardized mean differences (Hedge'sg, standardized mean difference [SMD]) were effect estimators. Meta-regressions (dependent variable: effect sizes of the interventions; independent variables: training dose and control group effects) were conducted to explore the dose-response relationship for therapy success of any exercise type.RESULTS: We included 68 trials. Compared to true control, effects on pain and disability were significantly larger for resistance exercise (pain: SMD,- 1.27; 95% confidence interval [CI]: -2.26, -0.28; I2 = 96%; disability: SMD, -1.76; 95% CI: -3.16, -0.37; I2 = 98%), motor control exercise (pain: SMD, -2.29; 95% CI: -3.82, -0.75; I2 = 98%; disability: SMD, -2.42; 95% CI: -3.38, -1.47; I2 = 94%), and Yoga/Pilates/ Tai Chi/Qui Gong exercise (pain: SMD, 1.91; 95% CI: - 3.28, -0.55; I2 = 96%; disability: SMD, -0.62; 95% CI: -0.85, -0.38; I2 = 0%). Yoga/Pilates/Tai Chi/Qui Gong exercise was more effective than other exercises (SMD, -0.84; 95% CI: -1.553, -0.13; I2 = 86%) for reducing pain. For disability, motor control exercise was superior to other exercises (SMD, -0.70; 95% CI: - 1.23, -0.17; I2 = 98%). There was no dose-response relationship for resistance exercise (R2 = 0.32). Higher frequencies (estimate = -0.10) and longer durations (estimate = -0.11) of motor control exercise had larger effects on pain (R2 = 0.72). Longer sessions (estimate = -0.13) of motor control exercise had larger effects on disability (R2 = 0.61).CONCLUSION: Resistance, mindfulness-based, and motor control exercises were effective for reducing neck pain (very low- to moderate-certainty evidence). Higher frequencies and longer duration of sessions had a significant effect on pain for motor control exercise.