ObjectiveTo systematically assess the modifiable risk factors for developing otitis media with effusion (OME) in children under 12 years.MethodsWe searched Embase, MEDLINE, INAHTA database, CENTRAL, CDSR and Epistemonikos for cohort studies with ≥40 children per arm/prognostic factor, published in English from 2000 to November 2022. We assessed risk of bias using the Quality in Prognosis Studies checklist, and overall evidence quality was assessed using Grading of Recommendations Assessment, Development and Evaluation (GRADE) methodology. Outcomes were analysed as risk ratio (RR), OR or Peto OR.ResultsSeven studies totalling 2 760 292 children were included. The evidence was very low quality. Fluid or pus discharge from ears (OR 2.1, 95% CI 1.01 to 4.35) and exposure to other children (RR 2.79, 95% CI 1.98 to 3.93) (OR 5.21, 95% CI 2.9 to 9.36) were strongly associated with development of OME. Coughs/colds ≥5 times (OR 1.91, 95% CI 1.22 to 2.99), breathing problems ≥5 times (RR 1.78, 95% CI 1.26 to 2.53) and ear infections (RR 1.95, 95% CI 1.39 to 2.72) in past year were associated with development of OME. Adenoid hypertrophy was strongly associated with development of fluctuating OME (recurrent OME) (OR 9.96, 95% CI 5.17 to 19.19). There was scare evidence for some potential modifiable risk factors, including breast feeding, household smoking, gastro-oesophageal reflux, dummy use and swimming.ConclusionsUpper respiratory tract infection, ear infection, adenoid hypertrophy and exposure to other children could be the predictors for development of OME. Further observational studies are needed to investigate other potential modifiable risk factors.