BackgroundPostnatal dexamethasone (PND) is used in high-risk preterm infants after the first week of life to facilitate extubation and prevent bronchopulmonary dysplasia (BPD) but the optimal treatment timing remains unclear. Our objective was to explore the association between the timing of PND commencement and mortality and respiratory outcomes.MethodsThis was a retrospective National Neonatal Research Database study of 84 440 premature infants born <32 weeks gestational age from 2010 to 2020 in England and Wales. Propensity score weighting analysis was used to explore the impact of PND commenced at three time-points (2–3 weeks (PND^2/3), 4–5 weeks (PND^4/5) and after 5 weeks (PND⁶⁺) chronological age) on the primary composite outcome of death before neonatal discharge and/or severe BPD (defined as respiratory pressure support at 36 weeks) alongside other secondary respiratory outcomes.Results3469 infants received PND. Compared with PND^2/3, infants receiving PND⁶⁺were more likely to die and/or develop severe BPD (OR 1.68, 95% CI 1.28–2.21), extubate at later postmenstrual age (mean difference 3.1 weeks, 95% CI 2.9–3.4 weeks), potentially require respiratory support at discharge (OR 1.34, 95% CI 1.06–1.70) but had lower mortality before discharge (OR 0.38, 95% CI 0.29–0.51). PND^4/5was not associated with severe BPD or discharge respiratory support.ConclusionsPND treatment after 5 weeks of age was associated with worse respiratory outcomes although residual bias cannot be excluded. A definitive clinical trial to determine the optimal PND treatment window, based on early objective measures to identify high-risk infants, is needed.