OBJECTIVE: To update the evidence on the effectiveness of exercise interventions to prevent episodes of neck pain. DESIGN: Systematic review and meta-analysis. LITERATURE SEARCH: MEDLINE, EMBASE, CENTRAL, CINAHL, SPORTDiscus, PEDro and trial registries from inception to December 2, 2022. Forward and backward citation searches. STUDY SELECTION CRITERIA: Randomized controlled trials (RCTs) that enrolled adults without neck pain at baseline and compared exercise interventions to no intervention, placebo/ sham, attention control or minimal intervention. Military populations and astronauts were excluded. DATA SYNTHESIS: Random effects meta-analysis. Risk of bias was assessed using the Cochrane RoB 2 tool. The certainty of evidence was judged according to the GRADE approach. RESULTS: Of 4703 records screened, 5 trials (1722 participants at baseline) were included and eligible for meta-analysis. Most (80%) participants were office workers. Risk of bias was rated as some concerns for 2 trials and high for 3 trials. There was moderate certainty evidence that exercise interventions probably reduce the risk of a new episode of neck pain (OR 0.49, 95%CI 0.31 to 0.76) compared to no or minimal intervention in the short-term (≤ 12 months). The results were not robust to sensitivity analyses for missing outcome data. CONCLUSION: There was moderate certainty evidence supporting exercise interventions for reducing the risk for an episode of neck pain in the next 12 months. The clinical significance of the effect is unclear.