How far is too far? A systematic review of travel distance to delivery institutions and negative maternal and natal outcomes

Abstract

Abstract Background Pregnant people can experience living far away from delivery institutions as unsafe for themselves and their foetus or new-born. This systematic review examined distance to delivery institution and risks of serious complications for the pregnant person or the foetus or new-born. Methods We searched seven databases for studies from high-income countries comparing two different travel times and published from 2010 to 2020. We used a combination of manual and machine learning-assisted processes to screen 7776 references, and independently assessed 33 in full texts. We included five studies from which we extracted data, appraised risk of bias using a tool modified to be relevant to studies of exposure, performed a meta-analysis if possible or summarised the outcomes narratively, and judged our confidence in the findings using GRADE. Results Traveling more than one hour to a delivery centre may increase the risk of eclampsia/Hemolysis, Elevated Live enzymes and Low Platelets (HELLP) Syndrome to a small degree (adjusted RR 1.30; 95% CI 1.05–1.70, low confidence) and induction of labour for logistic reasons (OR 5,04; 95% CI 3.62–7.00, low confidence), and likely increases the odds of unplanned delivery outside a delivery centre (OR 6.37; 95% CI 5.95 to 6.81, moderate confidence). It is uncertain whether this travel distance compared to a distance less than 60 minutes is related to any changes in the risks of the remaining negative outcomes early delivery with pre-eclampsia, perinatal mortality, premature birth before 37 weeks, low or very low birth weight, or admissions to a neonatal intensive care unit, for which we have very low confidence. Conclusions For most of the complications we examined in this review, it is uncertain if a travel time of over one hour increases risks, compared to a travel time of less than one hour. Future research must use core outcome measurements and comparable designs to enable pooling results, and if possible, measure actual distance to delivery rather than using home address as a proxy. Our confidence in the evidence is too uncertain to provide an information base for policymakers to change centralization policies.