G406(P) An audit of respiratory support practice and outcomes in the less than 27 week gestation population across three tertiary neonatal centres in London

Despite advances in neonatal intensive care, respiratory morbidities for very low birth weight babies have remained the same. 1,2 This is associated with high costs, both to infants and families and in financial terms to the NHS. (Phibbs 2006) There is considerable variation in respiratory management of this vulnerable population across neonatal units. The majority of neonatal units in the United Kingdom contribute neonatal data to a common platform ‘Badger.net’. In this study we aim to pilot extraction of respiratory data from Badger.net from three tertiary units in London to examine the applicability and feasibility of auditing and bench marking routinely collected data. Methods Data were collected for all babies Results The median GA and birth weight (BW) were similar in all units. Only inborn babies from Unit A (a surgical centre) were considered (Table 1). There was a significant difference in the invasive ventilation days in units A and B compared to unit C (p = 0.006 and 0.009 respectively) (Fig 1). The percentage of babies with chronic lung disease at 36 weeks GA did not differ (84%, 88% and 82%, units A, B and C respectively). Babies in unit C were discharged home approximately a week earlier compared to babies in units A and B. The number of babies discharged with home oxygen also varied between units (Table 2). This may reflect the scope and adequacy of locally available community services. Conclusion By publishing the respiratory data from three tertiary units in London, we show that these outcomes can be continuously audited by examining routinely collected and readily available data. This methodology will allow other units to audit their own service, benchmark their outcomes and improve standards within their units and across neonatal networks by identifying and sharing good practice. This work is an important ‘proof of concept’, moreover it may be possible to draw further conclusions regarding respiratory management if data are analysed over a longer period.