Article Text

other Versions

Cohort study
Travel time from home to hospital and adverse perinatal outcomes in women at term in The Netherlands
  1. Wendy A Hall
  1. School of Nursing, University of British Columbia, Vancouver, British Columbia, Canada
  1. Correspondence to: Wendy A Hall
    School of Nursing, University of British Columbia, T 201 2211 Wesbrook Mall, Vancouver, BC V6T 2B5, Canada; wendy.hall{at}

Statistics from

Commentary on: OpenUrlCrossRefPubMedWeb of Science

Implications for practice and research

  • This study provides some evidence that increased travel time to hospital may increase the incidence of adverse perinatal outcomes for high-risk women at term.

  • Increased risk of adverse outcomes associated with travel time to hospital when in labour should be taken into account when considering centralising maternity services.

  • Future research should separate low- and high-risk women with analyses conducted separately for each group.

  • Future research should include data about mode of delivery and factors affecting neonatal mortality and morbidity postbirth such as pathological jaundice, Group B streptococcus infections, prolonged rupture of membranes and neonatal effects of analgesia and anaesthesia administered during labour.


This study explored whether travel time to hospital exceeding 20 min increased neonatal mortality and morbidity in low- and high-risk women. Previous work indicated perinatal mortality was elevated in rural regions of The Netherlands compared with urban areas, which could not be explained by demographic or socioeconomic factors. During this study, rates of neonatal mortality were low (1.5 per 1000 births) with median travel time to hospital or outpatient clinic 13 min and median distance travelled 7 km. Women's care is provided in primary (home and outpatient clinics) and secondary environments (hospitals). Most women start labour at home. High-risk women have complicated obstetric or general medical histories and are delivered in hospitals.


This population-based (751 926 singleton births) cohort study used data from the Perinatal Registry of The Netherlands from 2000 to 2006. Records with antepartum mortality, congenital disorders, invalid entries, unknown labour locations, home births and hospitals only contributing to the database for 1–2 years were excluded. Logistic regression analysis was used to assess the impact of travel time by car to hospital to 99 maternity units on neonatal intrapartum mortality and mortality during the first 28 days of life. The adverse outcome measure was combined neonatal mortality, and/or 5 min Apgar of less than 4 and/or transfer to a newborn to a neonatal intensive care unit at birth. Univariate logistic analysis was performed on travel time and any confounding factors associated with the outcome measures. Age, ethnicity, parity, gestational age, hospital type and birth volume were adjusted for confounding effects on deaths or morbidity, using multivariate analysis. OR and corresponding 95% CI described the association between predictor variables and outcome variables.


The mortality level was 1.5 per 1000 births, and adverse outcomes occurred in 6 per 1000 births. Longer travel time (≥20 min) was associated with increased total mortality (OR 1.17, 95% CI 1.002 to 1.36), neonatal mortality within 24 h (OR 1.51, 95% CI 1.13 to 2.02) and adverse outcomes (OR 1.27, 95% CI 1.17 to 1.38). Delivery at 37-week gestation (OR 2.23, 95% CI 1.81 to 2.73) or 41-week gestation (OR 1.52, 95% CI 1.29 to 1.80), in addition to travel time, increased the risk of mortality. Travel time affected early neonatal mortality between 0 and 7 days post partum (OR 1.37, 95% CI 1.12 to 1.67).


Risk factors associated with maternal and infant outcomes are being studied throughout the world. This population-based study by Ravelli and colleagues provides interesting data about effects of travel time to hospital in one country. The population-based cohort and sample size are strengths of the study. A previous study based on the same cohort reported no differences by group in intrapartum and neonatal death rates or admissions to neonatal intensive care units when comparing low-risk women with planned home births and hospital births.1 Given previous findings, combining low- and high-risk women in this study creates an unnecessarily complex sample. For low-risk women at the start of labour who delivered in outpatient clinics, travel time had no effect. It is difficult to determine why the authors would recommend home births be reconsidered in low-risk women who are living a distance of 20 or more minutes away from a hospital, given the lack of travel time effect.

Only women who were high risk before labour with increased travel time had a significant risk of neonatal mortality and adverse neonatal outcomes. The authors linked their findings to rural settings; however, adjusted ORs indicated only women in very urban settings demonstrated a significant increase in adverse outcomes. Low-income women were also at risk of poor neonatal outcomes. Women in Canada living in low-income urban settings had infant mortalities that were 2/3 higher than women living in high-income urban areas.2 The authors acknowledge using postcodes to compute travel time relies on an assumption women were travelling from their home addresses, and this may not have been the case.

Skalkidou and colleagues3 found relying on ultrasound for dating pregnancies created biased perinatal morbidity and neonatal mortality among post-term born girls. In this study, two indicators of gestational age were used (last menstrual period and ultrasound); it is possible that the findings in relation to increased risk at 37- and 41-week gestation are due to the estimates of gestational age not being entirely accurate. This may have resulted in premature and postmature infants being included in the sample.

The study by Ravelli and colleagues raises questions about travel time in relation to neonatal mortality and adverse outcomes. However, the lack of control for many potential intrapartum factors that could affect neonatal outcomes (eg, mode of birth and effects of analgesia and anaesthesia) calls into question the validity of the results. Further research is needed in this area.


View Abstract


  • Competing interests None.

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.