Descriptive analysis of childbirth healthcare costs in an area with high levels of immigration in Spain

In some European countries child delivery represents 5% of the entire hospital activity [1], but the healthcare activity and costs generated by childbirth have been insufficiently evaluated.

The cost-effectiveness of prenatal care has been widely described [2–4], although this process among immigrants and its outcomes are controversial. The hypothesis that immigrants make lesser use of prenatal care, which may increase the risk of cesarean section and delivery costs, is widespread [5]. A study in Taiwan found that foreign-born mothers tended to use fewer inpatient services for complicated pregnancies than native-born mothers and were less likely to undergo cesarean section [6]. In Canada, different outcomes for the risk of preterm birth, depending on the length of residence in the country, were found [7]. In Spain, the more disadvantaged neighborhoods have worse pregnancy outcomes, but these inequalities did not exist among immigrant women, and some immigrant groups had better outcomes than Spanish-born women [5].

Catalonia is one of the Autonomous Regions of Spain attracting the greatest number of immigrants [8] most of whom settle in Barcelona, especially in certain areas of the city, Ciutat Vella being the district with the highest percentage of immigrants from non-occidental countries and with the lowest socioeconomic level of all Barcelona districts.

Variation in childbirth cost is principally determined by delivery type, which influences personnel and surgical requirements [9], as well as pre- and postnatal care, especially the postnatal hospital length of stay, both for the newborn and the mother [10, 11]. In the last few decades, cesarean section has become more frequent and, at the same time, more controversial. An increase in cesarean sections in low-risk women from 1996 to 2003 was observed according to age, race and ethnicity, although the major risk factor was previous cesarean section [12]. In countries such as the United States, women with a high socioeconomic position have greater access to prenatal care and a higher probability of delivery through cesarean section [13]. The possibility of elective cesarean section, available in some countries, has intensified the debate [14]. Moreover, the debate is also made more intense by the concurrence of increases in medically induced birth and decreased infant and fetal mortality rates [15, 16].

The increase in cesarean section rates in non-urgent or not strictly indicated cases unnecessarily increases resource use and poses an additional risk to the safety of the mother and child. Delivery through cesarean section is costlier than non-instrumented vaginal delivery, while instrumented vaginal delivery is more expensive than spontaneous vaginal delivery [17].

Cesarean section rates over 10 or 15%, which are lower than the current average rate in developed countries, do not affect maternal or neonatal morbidity and mortality [18]. High-risk pregnancies are directly associated with gestational age, birthweight and length of hospital stay [19].

For all immigrants living in Spain, the principles of public, free and universal services mainly apply to emergency care services, whereas access to primary and specialized care might present several barriers. These barriers include the need to have a health card, lack of knowledge about how to access these services, and difficulties in making an appointment (language, schedules, waiting list, etc.). To obtain a health card, the only requisite is proof of a place of residence through the registration certificate. Thus, the percentage of unregistered foreign-born residents, although unknown, is believed to be very small.

The aim of this study was to evaluate the overall cost of the childbirth process including the costs of prenatal care, delivery and postnatal care up to 3 months after delivery, taking into account delivery type and maternal origin.

The profile of pregnancy follow-up differed according to maternal origin (Table 2). Spanish-born mothers had a statistically significantly higher average number of emergencies (p = 0.001), tests (p = 0.009) and hospital contacts (p = 0.042). These differences were also found in the costs for emergencies (p = 0.002) and tests (p = 0.007). The difference in the cost of hospital contacts was in the opposite direction, the average cost being 63.63€ higher in foreign-born mothers, but this difference was not statistically significant. No other statistically significant differences were found according to maternal origin, except for the overall number of acts per delivery, Spanish-born mothers having an average of 1.18 more acts per delivery (p = 0.004).

The most important difference in costs was related to the cost of maternal admission for the delivery itself, cesarean section being 866.06€ costlier than vaginal instrumented delivery (p < 0.001) and vaginal instrumented delivery being 256.08€ costlier than vaginal non-instrumented delivery (p < 0.001). The cost of neonatal admission also differed significantly according to delivery type, the average being 1,356.89€ for cesarean section (23.3% of the overall cost), 683.36€ for vaginal non-instrumented delivery and 637.21€ for vaginal instrumented delivery (all 2-by-2 comparisons were significant at the 0.05 level).

The type of delivery was associated with significant differences in the number of acts and the cost of postnatal care for the 3 months after delivery in all concepts except for tests, which was the least frequent concept. The 2-by-2 differences were significant at the 0.05 level for all concepts between cesarean sections and vaginal non-instrumented delivery, and between vaginal instrumented and vaginal non-instrumented deliveries. When cesarean sections were compared with vaginal instrumented deliveries, only the mother's hospital contacts showed significant differences in the number of acts and costs (Table 3).

No significant differences among delivery types were found for prenatal care at primary care, the concept with the highest weight in terms of the number of acts (around 31%). Although the average number of emergency visits was similar (around two), their cost for vaginal instrumented deliveries was 32.21€ higher than that for non-instrumented deliveries (p = 0.025). Women who underwent cesarean section contacted the hospital once more than women with a vaginal non-instrumented delivery (p = 0.007). However, these hospital contacts were costlier for vaginal instrumented deliveries, with differences of 85.89€ compared with vaginal non-instrumented deliveries (p = 0.031) and 53.29€ compared with cesarean sections (p = 0.011). Cesarean sections were associated with a significantly higher number and cost of tests, with differences of 0.73 and 1.07 tests (p = 0.005 and p < 0.001) and 58.78€ and 77.29€ (p = 0.007 and p < 0.001) with vaginal instrumented and vaginal non-instrumented deliveries, respectively.

The average cost of the childbirth process was 4,328€, with an average of 18.28 contacts of the mother or the newborn with the healthcare facilities. The delivery itself accounted for more than 75% of the overall cost: maternal admission for 57% and neonatal admission for 20%. Prenatal care accounted for 18% of the overall cost and represented 75% of overall acts.

As reported by other studies [5, 21, 22], no differences were found in overall cost or health services utilization between Spanish-born and foreign-born mothers. These studies reported that utilization rate and adjusted cost were lower in the immigrant population than in the Spanish-born population. The reasons explaining these results in the Spanish context are the healthy immigrant effect and the high labor activity rate. However, foreign-born mothers had a higher frequency of primary care prenatal care (80%) which in this context suggests that it may not be the case that foreign-born mothers find barriers to accessing primary care gynecology services.

Spanish-born mothers had a worse severity profile and a higher average number of emergencies, hospital contacts and tests per woman during pregnancy. The cost per woman related to emergencies and tests was significantly higher for Spanish-born women, although the cost for hospital contacts was higher for foreign-born women. Although 5% more newborns from foreign-born mothers were admitted to hospital, no differences in costs or the number of acts were found for delivery or the 3-month postnatal period. In agreement with our results, a study in Taiwan likewise finds no evidence for the hypothesis that immigrant mothers have worse neonatal outcomes [23].

Given that delivery accounted for the highest proportion of cost, the main predictor of the overall cost of the process was delivery type. The average overall costs were 5,815€ for cesarean sections, 4,064€ for vaginal instrumented delivery and 3,682€ for vaginal non-instrumented delivery. Vaginal non-instrumented deliveries had a lower overall number of acts than the other delivery types and a lower burden of postnatal care. In fact, the average hospital stay in our sample was similar to that in 2008 at the public network of Catalonia, which was 2.71 days for vaginal deliveries and 4.57 days for cesarean sections. The costs of newborns admitted to hospital after a cesarean section doubled the cost of the newborn's admission after a vaginal delivery. This finding may be due to the longer hospital stay of the newborn (and mother) after a cesarean section, but also to the higher frequency of multiple births under cesarean section (4.3% vs. 2.1% for vaginal non-instrumented deliveries). Indeed, a study in women with prior cesarean section found higher rates of respiratory morbidity and neonatal intensive care unit admission, and longer length of hospital stay for cesarean sections than for vaginal births [24]. Cesarean section also affected maternal hospitalization: a study on births in Massachusetts found a higher adjusted likelihood of rehospitalization within 30 days postpartum among women undergoing planned cesarean [25].

The differences between the total cost of the process according to delivery type was clearly significant, as seen in other studies (cesarean section was more expensive than instrumented delivery, and the latter was more expensive than non-instrumented delivery), [3, 8, 10, 17, 24].

By using a multivariate regression model, the incremental cost of the different characteristics of the births was calculated. This model stressed the importance of the overall cost of the delivery itself. The type of delivery, maternal severity, neonatal admission and severity, and multiple births had a substantial impact on the individual cost of the delivery. In contrast, prenatal care, maternal origin and age had a slight impact on overall cost.

Nevertheless, for both Spanish and foreign-born mothers the rate of cesarean sections was around 27%. The World Health Organization considers that the optimum rate is unknown, despite the growing body of research that shows a negative effect of high rates of cesarean sections [26]. Cesarean rate may vary according to organizational issues, clinical criteria or mother's preferences however, the economic impact of performing a cesarean section should also be taken into account in the decision-making process of indicating or not a cesarean section.

The assessment of the costs presented in this study was based on the hospital's own information system, which limits the extrapolation of results. The sample analyzed does not allow conclusions to be extrapolated but do describe what is relevant and what is not. However, the quality of costs from a hospital cost accounting system is higher than could have been expected from a general but approximate sectorial study.

Another limitation of this study was the lack of information on the features and costs of women who choose to undergo prenatal care in private health services, this could affect mainly the analysis of maternal origin, as Spanish-born women may have greater access to private healthcare. Nevertheless, gynecologists in our hospital perceive that women who undergo pregnancy care in the private sector do not give birth in a public hospital.

This study has demonstrated that health care information systems in the public sector can be used to describe a process as a whole, beyond partial episodes. Evaluating the cost of maternal hospital stay is necessary but not sufficient to assess the cost of the overall process of delivery. Knowing the cost of the total process (including pregnancy, delivery and postnatal care) and the importance of each episode may help in decision-making on the organization of public healthcare facilities and their relation with populational features.

Healthcare cost information systems are useful to achieve a complete and detailed cost of the childbirth process. Our results suggest that neither immigration nor pregnancy care are associated with a substantial difference in the overall cost of childbirth. The most important cost predictors are type of delivery, especially cesarean section, and neonatal severity. The rate of cesarean sections was high and similar between Spanish and foreign-born women. Given its importance, the economic impact of cesarean section on the overall cost of childbirth should be factored when considering this mode of delivery.