Introduction Recent data suggest 8–52% of babies on the NICU have evidence of a viral respiratory tract infection (VRTI) (Ronchi 2014, Bennett 2012). These studies, and our own data, indicate babies with VRTIs spend twice as long in hospital and have significantly worse respiratory outcomes such as chronic lung disease and the need for home oxygen. There is little evidence exploring ways of reducing these infections in the NICU. Our recent survey demonstrates significant variation in UK NICU visiting practices and isolation policies for babies with RVTIs.

Aim To establish the impact of visitor restriction on the incidence of NICU VRTIs.

Methods We performed a retrospective study of all admissions between 2007 and 2013 at two large UK tertiary NICUs (˜13,300 bed days/year). Normal visiting policy included parents, family and friends. During the periods November to April of 2009, 2010 and 2011, in response to the H1N1 pandemic, we restricted visiting to parents/carers only. No other variations in practice occurred. We identified all babies positive for VRTIs. We used a Poisson generalised additive model (GAM), factoring in workload intensity and incidence of community VRTIs, to calculate the impact of these 3 winter restriction periods compared with normal visiting.

Results There were 100 PCR proven VRTIs in 93 babies during this period (˜16/yr). Rhinovirus (n = 71), RSV (n = 8) and H1N1 (n = 5) were the most common. The median gestation of infected babies was 29 weeks (IQR 26–34 Weeks) and 46% required an escalation of respiratory support. Two of five H1N1 positive babies died. The results from the GAM suggest there was a 39% reduction (P < 0.05) in VRTIs during restricted visiting periods compared to normal visiting (Incident Rate Ratio 0.61, 95% CI 0.38–0.99). Extrapolating this to the UK, based on the NHS NICU tariff, the extra bed days associated with VRTIs cost between £7M and £25M/year.

Conclusion This is the first study demonstrating a significant reduction in NICU VRTIs through restricting visiting practices. VRTIs are associated with significant neonatal respiratory morbidity and have short and long-term resource implications. We need to explore better ways of minimising the impact of VRTIs in this vulnerable population.