Perinatal health inequalities and accessibility of maternity services in a rural French region: Closing maternity units in Burgundy

Highlights

• We studied travel time to the closest maternity unit in a rural French region.

• We examined the effect of travel time on care and outcomes during pregnancy.

• Closures between 2000 and 2009 led to increased travel times.

• Long travel times were associated with risk factors for poor perinatal outcome.

• Long travel times also modified pregnancy care and hospitalizations.

Abstract

Maternity unit closures in France have increased travel time for pregnant women in rural areas. We assessed the impact of travel time to the closest unit on perinatal outcomes and care in Burgundy using multilevel analyses of data on deliveries from 2000 to 2009. A travel time of 30 min or more increased risks of fetal heart rate anomalies, meconium-stained amniotic fluid, out-of-hospital births, and pregnancy hospitalizations; a positive but non-significant gradient existed between travel time and perinatal mortality. The effects of long travel distances on perinatal outcomes and care should be factored into closure decisions.

Introduction

In France, as in other countries, the regionalization of perinatal care, which is taking place in the more general context of restructuring the supply of hospital services, has led many establishments to close, especially in rural areas. It has thus made geographic accessibility a major issue for ensuring equal health opportunities (Coldefy et al., 2011), especially given the large geographical disparities that already exist (Trugeon et al., 2010). The initial objectives of this restructuring of perinatal care were better management of very preterm babies (Chung et al., 2011, HCSP, 1994, Lehtonen et al., 2011, Papiernik and Combier, 1996, Papiernik and Keith, 1995, Wehby et al., 2012) and greater safety in hospital care (HCSP, 1994, Heller et al., 2002, Merlo et al., 2005, Moster et al., 1999). An economic aim was rapidly added to these, because the potential concentration of resources in a limited number of establishments was thought to make economies of scale possible (Brousselle et al., 1999; Com-Ruelle et al., 2008; Hemminki et al., 2011; Klein et al., 2002; McKee and Healy, 2000; NHS, 1996; Pouvourville (de) et al., 1997).

We have long known that one cause of maternal and perinatal morbidity and mortality is the delay in management of obstetric emergencies at delivery – a delay that includes transportation time (Barnes-Josiah et al., 1998), which can be long even in industrialized countries (Nesbitt et al., 1990). The impact of travel time or distance on health outcomes has been repeatedly studied in trauma-related, cardiologic, and neurovascular emergencies (Blanchard et al., 2012, Fatovich et al., 2011, Meretoja et al., 2012, Shen and Hsia, 2012, Smith and von Kummer, 2012). It has rarely been examined as a risk factor in obstetrics, however, even though life-threatening obstetric emergencies are not rare and the onset of spontaneous labor is unpredictable (WHO, 1996). Studies conducted in industrialized countries have yielded contradictory results. Some authors (Dummer and Parker, 2004, Parker et al., 2000) have found no significant association between travel time or distance and adverse outcomes while others (Grzybowski et al., 2011, Lisonkova et al., 2011, Ravelli et al., 2011b, Viisainen et al., 1999) report that travel time to the maternity ward is associated with an increase in risks of intrapartum and neonatal mortality and morbidity. Studies in France (Blondel et al., 2011) and in other countries (Dietsch et al., 2010, Hemminki et al., 2011, Viisainen et al., 1999) have shown a positive association between travel time or distance and unplanned out-of-hospital deliveries. These deliveries are also associated with a higher risk of perinatal mortality than in-hospital births (Jones et al., 2011, Viisainen et al., 1999).

In 1996, France (excluding overseas districts and territories, here and hereafter) counted 815 maternity units (Ruffie et al., 1998), defined as hospital sites where deliveries take place. There were only 759 in 1998, 621 in 2003 and 526 in 2010 (Blondel et al., 2012, Pilkington et al., 2008) The majority of maternity units that were closed were smaller and less specialized facilities (DREES, 2009, Pilkington et al., 2008). In France, maternity units are classified into three levels of care by their capacity to provide pediatric services to high risk newborns: level 1 units have no special care unit for newborns; level 2 units have neonatal nurseries, but do not provide care for very preterm or very low birthweight infants; level 3 are maternity units with neonatal intensive care (Perinatal care: The government plan, 1995–2000, 1994). Over the period 2001 and 2010, the number of level 1 units in France decreased from 415 to 263, whereas level 3 units increased slightly in number (56–60). Closures over the period 2001–2010 led to a decrease in overall bed capacity from 19,025 to 16,986 (DREES, 2009).

However, these national figures mask substantial disparities between regions (Coldefy et al., 2011). In 2003, Burgundy was the region most heavily affected, with a closure rate of 36.0% over 1998, while no maternity wards were closed in Corsica or Limousin, other predominantly rural regions (Pilkington et al., 2008). The closures have led to widespread concern about the safety of childbirth in affected communities. Local politicians and user groups constituted committees in defense of small maternity units threatened with closure and these questions – in particular those related to increasing travel distances – were widely debated in the local and national press.

From 1998 through 2003, as the number of births increased by +3%, the number of women in France who gave birth and lived more than 30 km from a maternity ward rose from 10,310 to 13,679 (+33%) and the number more than 45 km away from 736 to 1520 (+106%) (Pilkington et al., 2008). Although these closures had only a limited impact on the distribution of distance in urban areas, the increase in these two distance classes (>30 km and >45 km) in rural sectors was respectively +52% and +105% (Pilkington et al., 2008). In an earlier study in Burgundy, conducted at the scale of municipalities, the mean travel time increased only 4 min from 2000 to 2009, but the maximum travel time rose from 65 min to 86 min (Charreire et al., 2011). Moreover, the number of municipalities in the region located more than 30 min from a maternity ward grew; these municipalities were home to 11,345 women aged 15–45 years (Charreire et al., 2011).

Given the concerns in France with increasing travel distances for pregnant women, this study aimed to analyze the effect of travel time to the closest maternity ward on pregnancy outcome and prenatal management in Burgundy, where nearly 90% of women give birth in the maternity ward closest to their home (Combier et al., 2004).