Menu
Menu
Menu
Menu
Background: Reliable, population-based data on pregnancy-related morbidity and mortality, and risk factors for fatal foetal outcomes are scarce for low- and middle-income countries. Yet, such data are essential for understanding and improving maternal and neonatal health and wellbeing. Methods: Within the 4-monthly surveillance rounds of the Taabo health and demographic surveillance system (HDSS) in south-central Côte d’Ivoire, all women of reproductive age identified to be pregnant between 2011 and 2014 were followed-up. A questionnaire pertaining to antenatal care, pregnancy-related morbidities, delivery circumstances, and birth outcome was administered to eligible women. Along with sociodemographic information retrieved from the Taabo HDSS repository, these data were subjected to penalized maximum likelihood logistic regression analysis, to determine risk factors for fatal foetal outcomes. Results: A total of 2976 pregnancies were monitored of which 118 (4.0%) resulted in a fatal outcome. Risk factors identified by multivariable logistic regression analysis included sociodemographic factors of the expectant mother, such as residency in a rural area (adjusted odds ratio (aOR)=2.87; 95% confidence interval (CI) 1.31-6.29) and poorest wealth tertile (aOR=1.79; 95% CI 1.02-3.14), a history of miscarriage (aOR=23.19; 95% CI 14.71-36.55), non-receipt of preventive treatment such as iron/folic acid supplementation (aOR=3.15; 95% CI 1.71-5.80), only two doses of tetanus vaccination (aOR=2.59; 95% CI 1.56-4.30), malaria during pregnancy (aOR=1.94; 95% CI 1.21-3.11), preterm birth (aOR=4.45; 95% CI 2.82-7.01), and delivery by caesarean section (aOR=13.03; 95% CI 4.24-40.08) or by instrumental delivery (aOR=5.05; 95% CI 1.50-16.96). Women who paid for delivery were at a significantly lower odds of a fatal foetal outcome (aOR=0.39; 95% CI 0.25-0.74). Conclusions: We identified risk factors for fatal foetal outcomes in a mainly rural HDSS site of Côte d’Ivoire. Our findings call for public health action to improve access to, and use of, quality services of ante- and perinatal care.
This study was conducted within the Taabo HDSS [20, 21]. The Taabo HDSS includes one small town (Taabo Cité), which is the centre of the department of Taabo that also holds the only small hospital for the surveillance zone. There are 13 main villages with more than 100 associated hamlets. The latter are settlements usually consisting of a group of households constructed close to agricultural exploitation sites, rather isolated, and not yet officially considered a village by the territorial administrative authority mainly due to its small population size (< 500 inhabitants). Meanwhile, there is a primary health care centre in all of the 13 villages, 10 of which were fully operational with an assigned nurse and three are managed by trained community-health workers (CHWs) not yet being entirely functional. In the hamlets, no basic primary care is available but CHWs that are part of the hamlet’s population may be approached for advice before seeking formal care. Basic antenatal care is provided at the general hospital of Taabo and all nurse-led operational health centres; thereof four villages also host professional midwives who offer their service. With regard to emergency obstetric care (EOC), such as caesarean section and instrumental delivery (e.g. forceps delivery), the first is supposed to be only done at the general hospital of Taabo where an operating block is available, while instrumental deliveries may also being performed as emergency measure by midwives. For women delivering in a health facility, referral to a better equipped medical centre in case of complications is decided and an official transfer statement provided by the respective midwives. However, the actual transport remains to be organised by the women’s relatives due to a lack of ambulances (only available in Taabo-Cité and the village of Kotiéssou). The costs for antenatal care and birth assistance are not standardised and thus difficult to be estimated. Certainly the costs increase with the level of proficiency of the service providers and depend on whether facilities are public or private [17]. Since 2011 antenatal care and delivery are by national policy free of charge, however additional costs for health seeking by expectant mothers are common [22]. Of note, in primarily rural areas such as the Taabo HDSS many women, especially when it is their first child, still tend to spend the last trimester of their pregnancy close to their relatives that may live in more remote areas. The place of labour and child birth may thus differ in many cases from the actual residence of the expectant mothers. The objective of this study was to assess pregnancy-related morbidities and risk factors for a fatal foetal outcome. All women of reproductive age (15–49 years) whose pregnancy started and ended between January 1, 2011 and December 31, 2014 were included. Each household of the Taabo HDSS is visited at least three times a year for detailed surveillance of vital events (i.e. birth, death, in-migration, out-migration, and pregnancy). New pregnancies were systematically listed and followed-up longitudinally. The status and potential negative events related to pregnancy were registered by trained field-enumerators. Miscarriage, stillbirth, and live birth were registered as pregnancy outcomes. Each woman identified with a new pregnancy was interviewed with a pre-tested questionnaire with an emphasis on pregnancy status, estimated date of last menstrual period (LMP), and number of earlier pregnancies and births. Furthermore, in relation to potential negative consequences of a pregnancy, a standardised INDEPTH questionnaire on pregnancy-related morbidity was administered by field-enumerators to expectant mothers [23]. Sociodemographic information from women becoming pregnant during the 4-year observational study was readily available from the Taabo HDSS database [20]. Data were double-entered, cross-checked, and managed using a household registration system implemented in Windev version 12.0 (PC Soft; Montpellier, France) [24]. All statistical analyses were performed in Stata version 12.0 (StataCorp; College Station, TX, USA). Data records from pregnant women with complete sociodemographic, pregnancy-related morbidity, and birth circumstances information that were not lost to follow-up were considered for analysis (Fig. 1). Flow chart indicating all pregnancies registered and monitored between 2011 and 2014 in the Taabo HDSS and final study sample comprising complete data for analysis The primary outcome variable was defined as fatal foetal outcome and applied to all pregnancies that resulted in stillbirth, miscarriage, or early neonatal death using WHO definitions (i.e. dead born with gestational age higher or lower than 28 weeks, or death within the first 7 days after birth) [7]. Assessed explanatory variables for relationship analysis included (i) sociodemographic characteristics of the expectant mother; (ii) antenatal care sought; (iii) pregnancy-related morbidities and concomitant health conditions; and (iv) circumstances of delivery reported. Socioeconomic status was determined using a household-based asset approach and principal component analysis (PCA) with stratification into wealth tertiles (i.e. poorest, poor, and least poor) [25]. Gestational age at birth was calculated based on the first day of LMP and date of birth. Preterm birth was defined as gestational age < 37 weeks [26] or as perceived “shorter than normal” (i.e. estimated duration of < 8 months) preterm birth in women not able to provide reliable information on LMP (about 8% of all expectant mothers). χ2 test statistics were used to investigate significant univariate differences between mothers whose pregnancy resulted in live birth compared to a fatal outcome for the aforementioned explanatory variables. Univariate and multivariable logistic regression analyses were performed to identify significant relationships between fatal foetal outcome and covariates. In order to address estimation bias from fatal foetal outcome being a rare event, penalized maximum likelihood logistic regression models, as proposed by Firth, were used [27, 28]. Results were presented as odds ratios (ORs) and 95% confidence interval (CI). Differences and relationships with a p-value below 0.05 were considered as statistically significant. The multivariable regression model was built using a stepwise elimination approach, excluding explanatory variables at a significance level of 0.20 or higher. Sociodemographic factors known to have negative consequences on birth outcome (e.g. age, socioeconomic status, and residency of the mother) from earlier studies were included in the final model [29].
N/A
