Exposure to oil pollution and maternal outcomes: The Niger Delta prospective cohort study

Background Maternal exposure to oil pollution is an important public health concern. However, there is a dearth of literature on the effects of maternal exposure to oil pollution on maternal outcomes in the Niger Delta region of Nigeria. This study was therefore designed to determine the effect of maternal exposure to oil pollution on maternal outcomes in the Niger Delta region of Nigeria. Methods Prospective cohort study design involving 1720 pregnant women followed from pregnancy to delivery was conducted. The participants were 18-45 years old at a gestational age of less than 17 weeks, who attended randomly selected health facilities in the areas with high exposure and low exposure to oil pollution in the Niger Delta, Nigeria. Data were collected using an interviewer-administered questionnaire and review of medical records from April 2018 to April 2019. Multivariate log-binomial model was used to examine the effect of maternal exposure to oil pollution on the risk of adverse maternal outcomes adjusting for sociodemographic, maternal and lifestyle characteristics. Results A total of 1418 women completed the follow-up and were included in the analysis. Women in high exposure areas had a higher incidence of premature rupture of membrane (PROM), caesarean section (CS) and postpartum haemorrhage (PPH) compared to women in areas with low exposure to oil pollution. After adjusting for cofounders, women in high exposure areas also had a higher risk of PROM (ARR = 1.96; 95% CI: 1.24-3.10) and PPH (ARR = 2.12; 95% CI: 1.28-3.36) in Model I-III when compared to women in areas with low exposure to oil pollution. However, pregnancy-induced hypertension and CS had no association with maternal exposure area status to oil pollution. Conclusion Women in high exposure areas are at a higher risk of PROM and PPH. This calls for policies and intervention toward reducing maternal exposure to oil pollution in the Niger Delta region of Nigeria. The study was conducted in the Niger Delta region of Nigeria. The region consists of 9 crude oil-producing states with an estimated population of about 28 million people residing in about 3000 communities [23]. The Niger Delta region contains considerable oil reserves that have made the region the active hub for oil extraction and processing in Nigeria for the past 50 years [24]. The region is plagued with oil related contamination and has been rated as one of the most oil spill vulnerable areas in the world [1–3]. An estimated 10 to 13 million tons of hydrocarbons have been reportedly spilt into the Niger Delta over the last 50 years [25, 26]. During this period, over 77% of spilt hydrocarbons were not recovered [25, 26]. These spills and recurrent gas flaring have consequently led to contaminated air, water and land in the Niger Delta region, which has resulted in significant impacts on human health and ecological systems [27, 28]. A prospective cohort design was employed in this study. All the pregnant women between the ages of 18 to 45 years who visited public health institutions for antenatal care in the selected public health facilities of Niger Delta, Nigeria constituted the study population. We restricted the study to pregnant women with a gestational age of less than 17 weeks, who have been residing in the selected areas for at least 3 years, and are willing to participate in the study including the follow-up period. The sample size of this study was determined using the double population proportion formula. The assumptions considered in the sample size calculation were a two‐sided confidence level of 95%, 80% power, and the ratio of high exposure areas to low exposure areas of 1 to 1. This study also considered a design effect of 2 because of the multistage sampling technique that was used, and a loss to follow up rate of 10%. The proportion of miscarriage among women living close to the oil field was obtained from a similar study [18]. The final sample size was 1720 pregnant women (860 women in high exposure areas and 860 women in low exposure areas). Multistage sampling technique was used to select a representative of women in high exposure areas and women in areas with low exposure to oil pollution in the Niger Delta, Nigeria. Out of the nine (9) states (Akwa Ibom, Rivers, Delta, Bayelsa, Cross River, Edo, Imo, Ondo and Abia) in the region, Delta and Rivers States were purposively selected as states with high exposure to oil pollution due to the high production activities of crude oil, high crude oil spill record, and the presence of crude oil refinery. Ondo and Edo were also purposively selected as states with low exposure due to low production activities and low oil spill record and absence of oil refineries. These selections were informed by the record of the Nigerian Oil Spill Monitor, the monitor gives the public access to current official data on oil spills collected by the National Oil Spill Detection and Response Agency (NOSDRA) and the National Bureau of Statistic. The four states were stratified into oil producing and non-oil producing areas. Two health facilities serving oil producing areas in Delta and Rivers State were randomly selected. For the reference group, two hospitals/health facilities serving non-oil producing areas in Ondo State and Edo State were also randomly selected for the study. Finally, random sampling technique was used to select the study participants. A structured and interviewer-administered questionnaire was prepared by reviewing several related literature and related international guidelines [5, 9, 19, 29, 30]. It was organised into several sections, such as socio-demographic characteristics (maternal age, marital status, religion, level of education, mothers’ occupations, household income and main source of cooking fuel), maternal and lifestyle characteristics (mid upper arm circumference, gravidity, previous miscarriage, previous stillbirth, previous infant death, alcohol intake, smoking and diet diversity status); oil pollution exposure characteristics (exploration activities, oil spill incidence, gas flaring incidence, perception on air quality, perception on water quality, perception on soil quality); and adverse maternal (PIH, PROM, PPH, and caesarean delivery). The data collection and supervision of the study were handled by experienced research assistants and professionals (midwives or nurses). Data collectors and supervisors received five (5) days of training before the data collection period began. The training covered the study’s objectives, ethical considerations, interviewing techniques, study inclusion criteria, follow-up procedures, and the overall content of the data collection instrument. The data collection instrument was validated using face and content validity. Pre-test of the data collection instrument was conducted before the actual data collection. We used feedback from the pilot to modify the questions. A pre-tested structured and interviewer-administered questionnaire was used for data collection. It was organised into several sections, such as socio-demographic characteristics, maternal and lifestyle characteristics and adverse maternal outcomes. Data regarding sociodemographic characteristics, maternal and lifestyle characteristics were collected at the first encounter with study participants in the antenatal clinic. Data on adverse maternal outcomes were collected from pregnant women that were followed prospectively from their index pregnancy to 24 hours after delivery. The follow-up period was from April 2018 to April 2019. Missing data were gathered from the woman’s antenatal and other medical records. The completeness and accuracy of the collected data were checked daily during the data collection period by the research supervisors and the principal investigator. Adverse neonatal outcomes were measured as the occurrence of Pregnancy—induced hypertension (PIH), Premature rupture of membrane (PROM), Caesarean section or delivery (CS) and Postpartum haemorrhage (PPH). PIH is defined as systolic blood pressure ≥ 140mmHg and/ or diastolic blood pressure ≥ 90 mmHg in a previously normotensive pregnant woman who is ≥ 20 weeks of gestation and with or without proteinuria [31]; PROM is defined as the rupture of the foetal membrane after 37 weeks of gestation prior to the onset of labour [32]; CS is defined as a surgical procedure in which a foetus is delivered through abdominal and uterine incision [33], and PPH is defined as blood loss of 500ml or more within 24hours after birth [34]. The primary exposure was women in high exposure areas to oil pollution. Women in high exposure areas is defined as women who have resided for at least three (3) years in oil producing areas of states in the Niger Delta with high crude oil production activities, high incidence of oil spills and presence of a crude oil refinery. Women in low exposure areas is defined as women who have resided for at least three (3) years in non-oil producing communities of states in the Niger Delta with low crude oil production activities, low incidence of oil spills and absence of a crude oil refinery. Socio-demographic characteristics included maternal age that was categorised as <25 years, 25–34 years and ≥ 35 years; educational status as tertiary or non-tertiary; monthly household income in naira as < 50,000 or ≥ 50,000; marital status as (married, single and others); religion as (Christian, Muslim, or others); Mother’s occupation as (non-oil and gas related, or oil and gas related); and main source of cooking fuel as (clean or unclean). Women who used gas or electricity as their main source of cooking fuel were classified as clean fuel users, while women that used firewood, kerosene, charcoal, or crop residue/sawdust as their main source of cooking fuel were classified as unclean fuel users [35]. Maternal and lifestyle characteristics included gravidity categorised as primigravida or multigravida; previous miscarriage as (yes or no); previous stillbirth as (yes or no); and previous infant death—the death of a baby before the first birthday—as (yes or no); smoking as (yes or no); alcohol intake as (yes or no); mid upper arm circumference (MUAC) as MUAC < 28cm or MUAC ≥ 28cm. Women with MUAC < 28cm were considered to be normal and women with MUAC ≥ 28cm were considered to be pre-gestational overweight and/or obese [36]. Women’s diet diversity status was measured using the standardised tools for women diet diversity score based on the FAO guidelines for measuring minimum dietary diversity score in women with the consumption of ten food items within a period of 24 hours [37]. Women who consumed less than five food items and greater than or equal to 5 food items were classified to have inadequate diet diversity status and adequate diet diversity status respectively [37]. Data were entered and cleaned using SPSS (version 23.0; IBM) software and analysed using STATA (version 14.0; StataCorporation) software. Descriptive statistics like frequencies and summary statistics (mean, standard deviation (SD), and percentage) were used to describe the participants’ characteristics. Categorical data were compared using Pearson’s chi-square test, and independent t-test was used for comparison of the mean difference. Log-binomial model was used to determine the relative risk summary metric for the associations between exposure to oil pollution and adverse pregnancy outcomes and to control the effect of potential confounders. Log-binomial model is a special case of a generalised linear model, specifically applying a log link function to binomial outcome data for modelling adjusted relative risk in prospective data [38, 39]. The public health research community have suggested the use of relative risk (RR) for cohort study instead of odds ratio. This can be attributed to the difficulty in interpreting odds ratio, as it is sometimes misinterpreted as relative risk [39]. Separate log-binomial models were tested and presented for each outcome. Variables were included in the multivariable log-binomial model based on literature review and their association with each adverse maternal outcome (p-value ≤ 0.20) in the bivariate analysis. Crude relative risk (CRR) was generated in model I. In model II, the adjusted relative risk (ARR) for the associations between exposure to oil pollution and adverse pregnancy outcomes were determined after controlling for sociodemographic characteristics (maternal age, level of education, mother’s occupation, household income, and source of cooking fuel). In model III, the association between exposure to oil pollution and pregnancy outcomes were adjusted for sociodemographic characteristics plus maternal and lifestyle variables (MUAC, gravidity, smoking, and diet diversity status). Variables in each model were mutually adjusted for each other. Moreover, multicollinearity between the variables was checked using the variance inflation factor (VIF). Finally, statistical significance was established at ARR≠ 1 with a 95% CI and P-value ≤ 0.05. The study was approved by the ethics committee of the Institute for Advanced Medical Research and Training (IAMRAT), College of Medicine, University of Ibadan, Ibadan, Nigeria with the UI/UCH EC Registration Number of NHREC/05/01/2008a and UI/UCH Ethics Committee assigned number, UI/EC/17/0517. In addition, the study was also approved by the Ethics Committee of Rivers, Ondo, Edo and Delta States Hospital Management Board. Permission for the use of the facility was obtained from the head of each selected health facility. Written informed consent was obtained from each participant; after explaining the purpose of the study, benefits and risks. The right to participate or withdraw from participation was also made explicit to them to ensure that participation was voluntary and to make them feel free from coercion or pressure.