Association of the distance travelled, and the call to hospital arrival time with early neonatal mortality in neonates born to mothers using emergency medical services at term gestation: a retrospective observational study

Background Most neonatal deaths in India occur due to a lack of timely and proper care just before and during delivery. Timely availability of an ambulance for institutional delivery has led to some decrease in this. However, the distance travelled and time taken may affect the outcome. Objective The objective of this retrospective observational study was to determine if the distance travelled and the call to hospital arrival time are associated with deaths in the early neonatal period (0–7 days postnatal age), in neonates born to mothers who used a centralised emergency medical services (EMS) agency transportation at term gestation. Design Retrospective observational study. Setting This was a secondary analysis of a de-identified patient dataset from a previous cohort study on EMS usage by pregnant women, in five Indian states in 2014. Participants Neonates born to mothers using EMS at ≥37–42 weeks gestation were included as participants. The groups of pregnant individuals with low distance (≤10 km) and time (≤60 min) to hospital arrival vs longer distance or time to hospital arrival were then compared for neonatal death in the first 7 days of life or early neonatal death (ENND), which was the primary outcome. χ 2 test and logistic regression were carried out. Additional analysis was also planned to test the higher cut-offs of distance and time, if the above cut-offs were not found to have a statistically significant association with ENND. Results There were a total of 1180 neonates meeting the inclusion criteria. Of these, 272 (23%) were born to mothers who travelled ≤10 km (km) and took time ≤60 min to reach the hospital. There were a total of 57 (4.8%) ENNDs of which 14 occurred in the low distance and time group (≤10 km and ≤60 min). There was no statistically significant difference between the groups for ENND (p value=0.8). On additional univariate analysis for the different cut-offs of distance and time separately, it was found that there were lesser odds of ENND for a cut-off of distance ≤60 km (OR 0.28 (95% CI 0.08 to 0.98), p value=0.046) and for a cut-off time ≤120 min (OR 0.36 (95% CI 0.17 to 0.76), p value=0.007). When adjusted for other variables found to be significant on univariate analysis (caesarean section, <4 antenatal visits, maternal tachycardia and twin gestation) a cut-off distance ≤60 km had a trend towards lower odds of ENND (OR 0.31 (95% CI 0.084 to 1.12), p value=0.075) while a cut-off time of≤120 min had a significantly lower odds of ENND (OR 0.37 (95% CI 0.167 to 0.81), p value=0.013). When both of these cut-offs were combined and analysed in the multivariable logistic regression model, the association with ENND was still significant (OR 0.33 (95% CI 0.16 to 0.72), p value=0.005). Conclusion In neonates born to mothers using EMS at term gestation in India, those with distance travelled by the mother ≤60 km and time taken to reach health facility ≤120 min, had 67% lower odds of ENND compared with those with distance travelled >60 km or time taken >120 min. Further prospective observational studies are required to validate these cut-offs in a larger population.