Sage Journals: Discover world-class research

Abstract

Cesarean section (CS) is an essential means of preventing maternal and child mortality globally. Nigeria has one of the highest rates of maternal mortality in the world, and increased CS use has the potential to reduce the rate. This study explored the spatial variation and pattern of CS rates in Nigeria, and the relationship of the rates with women’s job types, median age at first birth (MAFB), and percentage-teenage-mothers. The data of deliveries by CS 5 years before the 2018 Nigeria Demographic and Health Survey (NDHS) were analyzed retrospectively, using descriptive statistics, simple and multiple linear regression analyses, and Moran’s I statistics. CS rates in Nigeria were space-dependent (Moran’s I = 0.426; p < .001), irrespective of whether elective (Moran’s I = 0.431; p < .001) or emergency (Moran’s I = 0.273; p = .002), with northern states exhibiting lower rates. There were positive significant moderate associations between CS, elective-CS, and emergency-CS, and: professional/technical/managerial (R² = 0.34; R² = 0.28; R² = 0.32), clerical (R² = 0.37; R² = 0.26; R² = 0.37), and unskilled-manual (R² = 0.14; R² = 0.13; R² = 0.10) job types. MAFB had a strong positive significant association with the CS variables (R² = 0.68; R² = 0.61; R² = 0.57), while the percentage-teenage-mothers had a strong negative statistically significant association with the variables (R² = 0.56; R² = 0.46; R² = 0.52). The clerical job type and MAFB were the significant contributors (p < .050) to the CS variables. During antenatal care, pregnant women should be assessed to determine the suitability of CS, and unskilled workers should be given special consideration to identify high-risk pregnancies. Policies and advocacy should focus on addressing the various economic and socio-cultural norms that hinder mothers from embracing the procedure whenever necessary.

Keywords

cesarean section rates spatial autocorrelation maternal mortality women’s job type median age at first birth teenage mothers Nigeria

Introduction

Cesarean section (CS) is a crucial method for preventing maternal and child mortality, as many maternal deaths are avoidable with quality obstetric care, of which CS is a significant component (Adewuyi et al., 2024; Angolile et al., 2023; Ushie et al., 2019). CS, a surgical procedure in which a pregnant woman is delivered through abdominal and uterine incisions, continues to gain acceptance in developed countries, while its acceptance is not as widespread in developing regions, particularly in sub-Saharan Africa (sSA; Adewuyi et al., 2024; Angolile et al., 2023; Betran et al., 2021). CS can be either elective, where the decision for surgery is made before the onset of labor, or emergency, where the decision is made after labor has begun (Adewuyi et al., 2024; Angolile et al., 2023).

In 2020, approximately 800 women died daily from preventable pregnancy complications, with 95% of these deaths occurring in low and lower-middle-income countries (LLMICs; WHO, 2024). The global concern regarding such deaths over several decades has led to the establishment of the first and second targets of Sustainable Development Goal 3 (SDG 3), which aim to reduce the global maternal mortality ratio to less than 70 per 100,000 live births and to end preventable deaths of neonates and children under 5 years of age, both by the year 2030 (United Nations (UN), 2015). Cesarean section (CS) rates of up to 10% have been associated with a decline in maternal and neonatal mortality (WHO, 2024). Despite the significant burden of maternal and neonatal mortality in Sub-Saharan Africa (SSA), strong aversion to the procedure, even when necessary, has kept the rates in the region below 5% (Adewuyi et al., 2024; Betran et al., 2021; WHO, 2024). Nigeria, the most populous country in the region, has a CS delivery rate of only 2.7% (Adewuyi et al., 2024).

With a female population exceeding 100 million (approximately 50%; National Population Commission of Nigeria [NPC], 2020), few studies have considered CS on a national scale in Nigeria. The primary indications for CS, as identified in various institution-based studies across the country, include cephalopelvic disproportion, fetal distress, repeat CS, abnormal lie, and hypertensive disorders in pregnancy (Adekanle et al., 2013; Akinola et al., 2014; Allagoa et al., 2021; Nwobodo et al., 2011). Additionally, misconceptions, fear of death, community-held beliefs, cultural practices, and costs are some reasons many pregnant women in Nigeria refuse CS, often viewing it as a last resort for saving their lives and those of their babies (Allagoa et al., 2021). Coupled with a low prevalence rate of CS, there is inequality in the utilization of the procedure across the country, as concluded by Ushie et al. (2019) and Adewuyi et al. (2024) in nationwide studies.

Earlier studies in the country revealed that formal education and wealth levels were determinants of CS use, with women in the poorest households at a disadvantage compared to their counterparts in the richest households (Adewuyi et al., 2024; Ushie et al., 2019). However, with 35% of women in the country having no formal education (NPC and ICF, 2019), is there any association between their type of work and the use of CS, given that job types have been known to influence CS use in developed countries (Simoes et al., 2005)? While Akinola et al. (2014) claimed that maternal age was not a significant predictive factor for CS, Adewuyi et al. (2024) posited that a maternal age of 35 years or older was associated with increased odds of CS. Hence, since half of the women aged 25 to 49 in the 2018 Nigeria Demographic and Health Survey (NDHS) gave birth for the first time before the age of 21 (NPC and ICF, 2019), is the median age at first birth (MAFB) associated with CS rates in Nigeria?

Literature Review

Using data from the 2013 Nigeria Demographic and Health Survey (NDHS), Ushie et al. (2019) examined inequalities in access to delivery by CS in Nigeria. By utilizing concentration curves and concentration indices, the study revealed that CS rates were significantly higher among women in the richest wealth category compared to those in the middle, poor, and poorest categories. There was a notable concentration of CS utilization among women in higher wealth quintiles. Interestingly, among women in the richest category, those with lower education levels had much lower CS delivery rates compared to their more educated counterparts. Although the study relied on a nationally representative dataset, it did not account for factors influencing CS utilization, such as maternal age, parity, and pregnancy complications, nor did it consider the primary place of specific locations (administrative units) in subsequent intervention efforts.

In a related study, Adewuyi et al. (2024) utilized the 2018 NDHS data and employed both descriptive and inferential statistics, including chi-square tests and multivariable binary logistic regression, to investigate the prevalence and associated factors of cesarean delivery in Nigeria. The study identified several factors positively associated with an increased likelihood of having a CS, including urban residence, maternal age, birth size, multiple births, maternal obesity, religious affiliation, education, and access to health services. While the study highlighted disparities in access to CS based on geographic, socioeconomic, and educational attainment, it did not specify particular administrative units that require targeted interventions in Nigeria.

While the study by Gunn et al. (2017) provides valuable insights into the prevalence and predictors of Cesarean deliveries in Enugu, Nigeria, emphasizing the significance of contextual factors in maternal health outcomes, in contrast, the research by Nwobodo et al. (2011) suggests a need for improved patient selection for elective CS and enhanced counseling regarding the associated risks and benefits, particularly in Kano. Additionally, Allagoa et al. (2021) investigated the rates and factors associated with Cesarean deliveries in Yenagoa, Bayelsa State, recommending the encouragement of vaginal birth after a previous Cesarean to mitigate the rising rates of CS, as prior Cesarean deliveries were a significant indication for repeat surgeries. While adopting various statistical methods, these studies were institution-based and utilized restrictive datasets, which may limit the generalizability of their findings.

However, in contrast, the studies by Dutta et al. (2023) and Shukla et al. (2022), both conducted in India using data from the National Family Health Survey, highlight the benefits of considering the spatial dimension in CS use studies. At the district level, these studies found substantial geographic and socioeconomic inequities in CS rates; districts with higher proportions of the population in the poorest and poorer wealth quintiles had significantly lower CS rates. They also identified districts with very low CS rates alongside those with high rates. The studies underscored the need for targeted interventions to improve access to safe and timely cesarean deliveries, particularly in economically disadvantaged districts.

Although the findings of these studies are similar to those of CS studies in Nigeria, they go a step further by identifying specific administrative units where interventions are required, thereby making subsequent interventions more focused and their impacts measurable. CS studies in Nigeria identify broad populations (such as educated women and women in lower wealth quantiles) and general locations (like rural areas). However, CS studies are yet to identify specific administrative units in need, which would enhance focused interventions and targeted policy decisions.

Focus of This Study

While institution-based studies on CS in Nigeria have indicated that adolescent pregnancies are linked to complications that often necessitate elective CS, it remains unclear whether the proportion of teenage mothers in the country correlates with CS rates. Additionally, even though geographic factors significantly influence the understanding of medical conditions and services, no study has yet explored the spatial dimensions of CS rates at the national level in Nigeria. To effectively tackle the inequitable access to CS services and maternal care observed in the country, it is essential to analyze the spatial distribution of CS rates in Nigeria, which will aid in developing targeted and efficient interventions.

Thus, this cross-sectional study utilized data from the 2018 NDHS (NPC and ICF, 2019) to examine the spatial variation and patterns of CS rates in Nigeria, aiming to identify specific administrative units that require targeted interventions. It also investigated the possible associations between CS rates and women’s occupation types, maternal age at first birth (MAFB), and the percentage of teenage mothers. Furthermore, the analysis considered whether the CS procedure was elective or emergency.

Methods

Study Area

This is a cross-sectional study of Cesarean section (CS) rates in Nigeria. Nigeria consists of 36 states and a Federal Capital Territory (FCT). With a population of over 200 million, it is the most populous country in Africa. In 2022, the country had 49.98% males and 50.01% females (NPC, 2020). The study population comprised the 34,195 live births delivered by CS 5 years prior to the 2018 NDHS survey (NPC and ICF, 2019). Compared to global standards, the use of CS in Nigeria is low, with significant regional and socioeconomic disparities. Factors such as the level of education, maternal age, and access to healthcare services play critical roles in determining CS rates (Adewuyi et al., 2024; Ushie et al., 2019). Moreover, societal perceptions and attitudes toward CS remain predominantly unfavorable (Allagoa et al., 2021). Addressing these disparities and misconceptions is essential for enhancing maternal health outcomes in Nigeria.

Data Collection

The database for the study was created by extracting data from the 2018 NDHS report, specifically the percentages of live births delivered by CS in the 5 years preceding the survey for each of the 36 states and the FCT. These percentages were disaggregated into elective and emergency CS. Additionally, the median age at which half of the women had their first child (MAFB) for women aged 25 to 49 was included, along with the background characteristics of the women and the percentage of women aged 15 to 19 who had begun childbearing (teenage mothers). This information is crucial, as the age at which a woman begins bearing children is an important indicator of the health and well-being of both the mother and child (NPC and ICF, 2019).

Furthermore, the occupations of the women were extracted from the 2018 NDHS report for the states and the FCT. These occupations were categorized as professional/technical/managerial (PTM), clerical, sales and services, skilled manual, unskilled manual, agriculture, and other. During the 2018 NDHS, only women who delivered in a health facility were asked questions regarding CS, based on the assumption that women who did not give birth in a health facility did not receive a CS.

The 2018 NDHS survey employed a two-stage stratified sampling method, dividing the 36 states and the FCT into categories of urban and rural areas, resulting in a total of 74 distinct sampling groups. Separate samples were drawn from each of these groups using a two-step process. Initially, 1,400 enumeration areas (EAs) were selected based on the likelihood proportional to the size of the EA, with size determined by the number of households within each EA. Subsequently, households within these selected EAs were used as the basis for choosing a sample of 30 households in the second stage. This process resulted in a comprehensive, nationally representative sample size of approximately 42,000 households across the entire country.

Data Analysis

Using R version 4.1.0 and other R packages – tmap (Tennekes, 2018); rgeos (R. Bivand et al., 2017): rgdal (R. Bivand et al., 2021); spdep (R. S. Bivand & Wong, 2018) – the digital spatial database for the analysis was generated by joining the extracted information to a shapefile of the country, which contains the boundaries of the states and the FCT. At a 5% statistical significance level, simple linear regression analysis was employed to explore the association between the CS variables (CS, CS-elective, and CS-emergency) and the following factors: women’s occupation types, MAFB, and the percentage of teenage mothers. The variables with p < .05 in the simple linear regression analysis were subsequently entered into the multiple regression analyses to examine their levels of contribution. Simple linear regression was utilized not only to understand relationships but also because the independent variables were intended for prediction and modeling. The adopted convention for interpreting of r² is as follows: 0<|r²|<.3 is considered weak, .3≤|r²|<.7 is moderate, and |r²|≥.7 is strong. CS was treated as a unit and also disaggregated, as the two types of CS are associated with different risk levels for adverse events (Akinola et al., 2014). This approach is also desirable, as implied by an earlier study (Adewuyi et al., 2024). The formulations of the multiple regression models are as follows:

Y_{1} = b_{0} + b_{1} X_{1} + b_{2} X_{2} + b_{3} X_{3} + b_{4} X_{4} + b_{5} X_{5}

(1)

Y_{2} = b_{0} + b_{1} X_{1} + b_{2} X_{2} + b_{3} X_{3} + b_{4} X_{4} + b_{5} X_{5}

(2)

Y_{3} = b_{0} + b_{1} X_{1} + b_{2} X_{2} + b_{3} X_{3} + b_{4} X_{4} + b_{5} X_{5}

(3)

where Y₁ = CS, Y₂ = CS-elective, Y₃ = CS-emergency, X₁ = professional/technical/managerial job type, X₂ = clerical job type, X₃ = unskilled manual job type, X₄ = median-age-at-first-birth, and X₅ = percentage-teenage-mothers.

To investigate spatial dependence in the rates of CS in the country, Moran’s test was employed, which produces a statistic between −1 and 1. A global Moran’s statistic value of 1 implies perfect positive spatial autocorrelation, suggesting that the data is clustered. Values between 1 and 0 indicate a tendency toward clustering, a value of 0 implies that the data is randomly distributed, and a value of −1 signifies a negative spatial autocorrelation (Anselin, 1995).

The local Moran statistic (Ii) was mapped to reveal the pattern of spatial dependence of CS within the country. To determine whether there is a statistically significant geographic pattern in the clustering of the CS variables, the local Moran statistic and the variables were centered around their means and reclassified. The two resulting variables were then combined to create data quadrants using the method and the codes provided by Brunsdon and Comber (2018). These were subsequently mapped to reveal the types of relationship each state shares with its neighbors. The relationship could be classified as insignificant (no clusters), low-low (states with low values surrounded by states with equally low values), low-high (states with low values surrounded by states with high values), high-low (states with high values surrounded by states with low values), or high-high (states with high values surrounded by states with equally high values; Anselin, 1995).

Results

A total of 34,193 live births were delivered by Cesarean section (CS) in the 5 years prior to the survey, corresponding to a CS rate of 2.7% (Figure 1C). Among the overall survey population, 3.7% of mothers in the 35 to 49 age group delivered by CS, while 1.0% of those aged less than 20 years underwent the procedure. Most CS deliveries were emergency procedures (1.5%) (Figure 1A), and 4.7% of first-order births were by CS, except in the 35 to 49 age category, where the rate was 1.8%. Of the pregnant women who attended antenatal clinics for more than four visits, 4.8% had CS; however, the antenatal visit records of many women who had CS were missing.

Figure 1.

Map of percentage CS rates: (A) emergency (B) elective, (C) total.

Even though CS is a facility-based procedure, there was variation in facility use. Specifically, 4.6% of women who delivered in public health facilities had CS, while 11.5% of those who delivered in private facilities underwent the procedure. In terms of location, the highest CS rate was among urban residents (5.2%). Among the six geopolitical zones of the country, 7.0% of pregnant women in the South West had CS, compared to only 0.7% of those in the North West who underwent the procedure. Generally, CS rates were higher in the southern part of the country compared to the north.

Considering the level of education, 13.9% of women with more than a secondary school education had CS, and most of these were elective (Figure 1B). The data indicates that CS rates increased with wealth: 9.8% of women in the highest wealth quintile had a CS, followed by 3.1% of those in the fourth quintile, while those in the lowest wealth quintile had the least (0.3%).

Women in the professional, technical, and managerial (PTM) category were more prevalent in Lagos (23.4%), Kebbi (22.8%), and Ekiti (18.6%) States, while Adamawa (1.0%), Yobe (1.0%), and Jigawa (0.9%) States had the lowest percentages. The clerical category was more represented in the FCT (8.7%), Rivers (5.0%), and Ogun (3.4%) States, whereas Kebbi (0.2%) and Kwara (0.1%) had the least representation; Sokoto State was not represented in this category.

Zamfara (95.0%), Sokoto (89.8%), and Kano (87.8%) States reported the highest cases of women in sales and services delivered by CS, while Adamawa (36.1%), Ebonyi (29.0%), and Benue (24.2%) States had the lowest percentages in this category. The percentage of women in skilled manual occupations was highest in Borno (20.4%), Gombe (13.2%), and Jigawa (12.8%) States, with Rivers (0.3%), Adamawa (0.2%), and Taraba (0.1%) States having the lowest figures; Zamfara and Enugu States were not represented in this category.

In the unskilled manual category, FCT (1.5%) and Enugu State (0.5%) had the highest percentages, while Kwara, Ekiti, Bayelsa, and Kaduna had the lowest (0.1%). Notably, 62% of the states in the country did not have women who delivered by CS working in this category. Benue (67.7%), Adamawa (61.8%), and Ebonyi (59.4%) States had the highest percentages of women who delivered by CS working in the agricultural sector, whereas Katsina (2.1%), Zamfara (1.2%), and Lagos (0.4%) had the lowest percentages in this category.

MAFB (Figure 2A) ranged from 17.5 years in Zamfara State to 24.8 years in Lagos. The states with the highest MAFB were Lagos (24.8 years), Anambra (24.5 years), Enugu (23.8 years), Rivers (23.6 years), and Ogun (23.3 years). In contrast, Gombe (17.8 years), Bauchi (17.7 years), Kebbi (17.6 years), Katsina (17.5 years), and Zamfara (17.5 years) had the lowest MAFB values. There were no MAFB data available for Abia and Imo States because less than 50% of the women in these states had given birth before reaching 25 years of age. Additionally, more than 65% of the states in the northern part of the country reported a MAFB of less than 20 years.

Figure 2.

Map of (A) median-age-at-first-birth and (B) percentage distribution of teenage mothers.

However, regarding the distribution of teenage mothers across the country (Figure 2B), Bauchi (40.7%), Sokoto (32.1%), Kaduna (31.3%), Zamfara (29.4%), and Katsina (27.3%) had the highest percentages, while Osun (7.7%), Rivers (7.3%), Enugu (7%), Imo (5%), and Lagos (1.1%) recorded the lowest. More than 80% of the states in the northern part of the country had more than 20% teenage mothers, whereas over 70% of the states in the southern part of Nigeria had less than 10% teenage mothers.

From Table 1, the rates of CS showed a moderate positive association with the following job types: PTM (R² = 0.34, p < .001), Clerical (R² = 0.37, p < .001) and unskilled-manual (R² = 0.14, p = .01). Similarly, elective CS had a statistically significant association with PTM (R² = 0.28, p < .001), Clerical (R² = 0.26, p < .001) and unskilled-manual (R² = 0.13, p = .02). Additionally, emergency CS demonstrated a positively significant association with PTM (R² = 0.32, p < .001), Clerical (R² = 0.37, p < .001), and unskilled-manual (R² = 0.10, p = .03).

Table 1.

Results of Simple Linear Regression Analyses.

Independent variable	Dependent variables	r ²	Adj r²	p
CS	PTM	0.36	0.35	.00
	Clerical	0.39	0.37	.00
	SS	0.03	0.00	.32
	SM	0.03	0.00	.30
	UM	0.16	0.14	.01
	Agric	0.02	−0.01	.37
	Other	0.06	0.03	.16
	MAFB	0.69	0.68	.00
	Teen_M	0.58	0.56	.00
CS-ele	PTM	0.30	0.28	.00
	Clerical	0.28	0.26	.00
	SS	0.02	−0.01	.46
	SM	0.03	0.00	.29
	UM	0.16	0.13	.02
	Agric	0.03	0.00	.33
	Other	0.05	0.03	.17
	MAFB	0.62	0.61	.00
	Teen_M	0.48	0.46	.00
CS-eme	PTM	0.34	0.32	.00
	Clerical	0.39	0.37	.00
	SS	0.03	0.00	.29
	SM	0.02	0.00	.37
	UM	0.13	0.11	.03
	Agric	0.02	−0.01	.45
	Other	0.05	0.02	.19
	MAFB	0.58	0.57	.00
	Teen_M	0.53	0.52	.00

Note. CS-ele = elective CS; CS-eme = emergency CS; PTM = professional/technical/managerial; SS = sales & services; SM = skilled manual; UM = unskilled manual; Agric = Agriculture; Other = Other job types; MAFB = median-age-at-first-birth; Teen_M = percentage-teenage-mothers.

With respect to MAFB and the three CS variables, there were strong positive significant associations: CS (R² = 0.68, p < .001), elective CS (R² = 0.61, p < .001) and emergency CS (R² = 0.57, p < .001). The three CS variables and percentage-teenage-mothers exhibited moderate significant relationships: CS (R² = 0.56, p < .001), elective CS (R² = 0.46, p < .001) and emergency CS (R² = 0.52, p < .001).

For the multiple regression analysis of CS and PTM, Clerical, unskilled manual, MAFB, and percentage of teenage mothers, the model (F(37) = 19.91, p < .001) explains a significant portion of the variance in CS (adjusted R² = 0.74). The clerical job type (p = .03) and MAFB (p = .02) were the significant contributors. Similarly, the model examining the relationship between elective CS and the same variables (F(37) = 11.36, p < .001) explains a significant portion of the variance in elective CS (adjusted R² = 0.60), with MAFB (p = .01) being the only significant contributor.

Furthermore, with an adjusted R² = 0.67, the model investigating the relationship between emergency CS and the variables (F(37) = 14.56, p < .001) explains a significant portion of the variance in emergency CS, with the clerical job type (p = .01) as the only statistically significant contributor. Both unskilled manual job type and percentage of teenage mothers exhibited negative relationships with CS and emergency CS. Further details are provided in Table 2.

Table 2.

Results of Multiple Regression Analyses.

Dependent variable	Independent variables	Estimate	Std. Error	t value	Pr(>\|t\|)
CS	(Intercept)	−10.89	5.48	−1.99	.06
	PTM	0.07	0.05	1.37	.18
	Clerical	0.67	0.30	2.27	.03
	UM	−0.94	1.55	−0.61	.55
	MAFB	0.62	0.24	2.59	.02
	Teen_M	−0.02	0.05	−0.42	.68
CS-ele	(Intercept)	−7.78	3.28	−2.37	.03
	PTM	0.02	0.03	0.58	.57
	Clerical	0.08	0.18	0.44	.66
	UM	0.71	0.93	0.77	.45
	MAFB	0.42	0.14	2.92	.01
	Teen_M	0.01	0.03	0.20	.84
CS-eme	(Intercept)	−3.29	3.58	−0.92	.37
	PTM	0.05	0.03	1.54	.14
	Clerical	0.57	0.19	2.95	.01
	UM	−1.57	1.02	−1.54	.13
	MAFB	0.21	0.16	1.34	.19
	Teen_M	−0.03	0.03	−0.80	.43

Note. CS-ele = elective CS; CS-eme = emergency CS; PTM = professional/technical/managerial; UM = unskilled manual; MAFB = median-age-at-first-birth; Teen_M = percentage-teenage-mothers.

As shown in Table 3, CS exhibited significant spatial autocorrelation (Moran’s I = 0.43; ρ < .001); similarly, elective CS (Moran’s I = 0.43; ρ < .001) and emergency CS (Moran’s I = 0.27; ρ = .002) also demonstrated this characteristic. The mapping of the local Moran statistic (Ii) revealed the variation in the autocorrelation of the three variables across the country (Figure 3A–C), displaying apparent patterns. While Bayelsa State and the FCT recorded the lowest values for CS, with −0.753 ≤ Ii ≤ −0.507, Lagos and Imo States exhibited the highest values, with 1.254 ≤ Ii ≤ 1.785. Regarding elective CS, Lagos State had the highest value, 1.444 ≤ Ii ≤ 2.817. Similarly, for emergency CS, Bayelsa State and the FCT had the lowest values, with −0.975 ≤ Ii ≤ −0.781, whereas Zamfara, Sokoto, Kebbi, and Lagos had the highest values, with −0.814 ≤ Ii ≤ 1.519.

Table 3.

Global Moran I Results for CS, CS-Elective, and CS-Emergency.

Data	Moran I statistic	p-value	SD	Expectation	Variance
CS	0.42615496	6.349e–06	4.3653	−0.02777778	0.01081346
CS-elective	0.43130621	6.029e–06	4.3765	−0.02777778	0.01100338
CS- emergency	0.27334507	.001863	2.9005	−0.02777778	0.01077844

Figure 3.

Map of: (A) local Moran I of total CS rates, (B) local Moran I of elective CS rates, (C) local Moran I of emergency CS rates, (D) hot- and cold-spots of total CS, (E) hot- and cold-spots of elective CS, (F) hot- and cold-spots of emergency CS.

The cluster analysis of the three CS variables (Figure 3D–F) revealed that states with high-high values were clustered in the southern part of the country, while those with low-high values were clustered in the north. Lagos, Anambra, Imo, and Rivers States were identified as the hotspots of CS in the country; specifically, the hotspots of elective CS were Lagos, Edo, Anambra, and Imo States, while Imo State was noted as the hotspot for emergency CS. Conversely, the cold spots of CS were Kebbi, Sokoto, Zamfara, Kano, Kaduna, Jigawa, and Borno States; for elective CS, the cold spots were Kebbi, Zamfara, Kano, Kaduna, and Jigawa States; and for emergency CS, the cold spots were Kebbi, Sokoto, Zamfara, and Jigawa States.

The results indicated that most of the CS were emergency in nature. Maternal age was a significant predictor of CS, and CS rates increased with MAFB. The use of CS also increased with affluence, while maternal job types influenced CS rates. Additionally, there was an inverse relationship between CS rates and the percentage distribution of teenage mothers, and a spatial dependence was observed in the rates of the CS variables across the country.

Discussion

This study examined the spatial distribution of Cesarean section (CS) rates in Nigeria, focusing on the relationship between CS use and factors such as women’s occupation types, median age at first birth (MAFB), and the percentage of teenage mothers. The analysis was based on a sample of 34,193 live births delivered by CS within 5 years prior to the 2018 Nigeria Demographic and Health Survey (NDHS). While CS rates in the country were spatially dependent, MAFB and clerical job types emerged as significant contributors to these rates. The findings corroborate earlier studies (Ajayi, 2021; Irwinda et al., 2021). Most decisions regarding CS in this study were made after the onset of labor pain, except for women in their fourth decade of life, suggesting that elective CS decisions were likely influenced by age. According to previous studies, younger women are more likely to refuse a CS, while older women are relatively more inclined toward the procedure (Adewuyi et al., 2024; Etcheverry et al., 2024). This trend may be attributed to a pervasive fear of the procedure among younger women and cultural misconceptions that women who deliver via CS are not strong (Allagoa et al., 2021).

Another plausible explanation for the disparity in the timing of surgery may be complications during delivery. Younger women are more prone to delivery complications due to cephalopelvic disproportion (CPD) (Adeyanju et al., 2023; Allagoa et al., 2021). Olivier et al. (2024) found that the prevalence of CS due to CPD was 18% at Sakubva District Hospital in Zimbabwe, with maternal age of 18 years and below identified as a risk factor. Consequently, the inverse relationship observed in this study between CS rates and the percentage distribution of teenage mothers across the country raises concerns, given that the use of the procedure remains minimal in regions with high populations of teenage mothers. This situation poses a significant danger to numerous teenage mothers and their children. Maternal mortality in Nigeria is nearly double for women below 35 years of age (37.9%) compared to those aged 35 to 49 years (19.4%; NPC and ICF, 2019), with studies from the regions confirming this high burden (Meh et al., 2019; Sharma et al., 2017).

Contrary to the claim made by Akinola et al. (2014) that maternal age was not a significant predictive factor for CS, the findings of this study suggest otherwise. This discrepancy may be attributed to the different populations considered by both studies. While the former study was conducted in a single state, the latter examined data from the entire country. The findings of this study align with those of other studies (Adewuyi et al., 2024; Dutta et al., 2023; Shukla et al., 2022) that utilized nationally representative data. Shukla et al. (2022) found that the three CS variables (overall CS, elective CS, and emergency CS) had strong, positive, and significant associations with maternal age in Bihar, India, indicating that as MAFB increased, the rates of cesarean sections also tended to rise. Dutta et al. (2023) suggest that the association between CS and MAFB may result from the socioeconomic status of mothers and their level of education, as wealthier women and those with higher education tend to utilize cesarean procedures more frequently.

The observations that: CS use increased with wealth; private facilities were preferred over public ones; and rates were highest among urban residents, substantiated earlier findings (Adewuyi et al., 2024; Dutta et al., 2023; Islam et al., 2022; Shukla et al., 2022). Utilizing data from 36 Sub-Saharan African countries, Islam et al. (2022) indicated that the odds of utilizing CS were 231% higher for individuals from wealthy families compared to those from lower-income households. Furthermore, educated women demonstrated a greater likelihood of opting for CS, with an odds ratio of 2.65 in certain regions of East Africa. The dependence of poorer individuals on public healthcare for CS is another reason for the observed disparity (Dutta et al., 2023; Shukla et al., 2022). The dire economic conditions and low health insurance coverage in LLMICs place this procedure out of reach for those living in poverty. These factors highlight the need for governments and healthcare managers in LLMICs to reassess the quality of health facilities in rural areas, where most poor individuals reside. Islam et al. (2022) found that the odds of having a CS were 221% higher for urban women than for those living in rural areas across Sub-Saharan Africa.

Worthy of note is the positive relationship between the three CS variables and PTM, as well as with clerical and unskilled manual job types. While this finding supports an earlier study that indicated CS rates are influenced by maternal job types—showing high rates of CS deliveries among both high-salaried positions and unskilled workers (Simoes et al., 2005). However, de Loenzien et al. (2021) suggests that empowerment, rather than the nature of employment, may directly influence the choice of delivery method. Nonetheless, this perspective does not fully explain the positive relationship between CS and unskilled jobs observed in this study. A plausible explanation is that unskilled workers tend to have less access to prenatal care and education, leading to higher-risk pregnancies that necessitate CS. Guendelman et al. (2017) identify modifiable physical and organizational stressors at work as factors that can increase the risk of cesarean deliveries. Furthermore, socioeconomic factors such as poor nutrition and living conditions associated with low-wage jobs could heighten the likelihood of pregnancy complications, thereby necessitating surgical delivery (Adewuyi et al., 2024; Irwinda et al., 2021; Simoes et al., 2005).

Implication of Study

Based on the results of this study, we propose the following policy recommendations. First, to address the disparity in CS rates between urban and rural areas and to bridge the wealth-induced divide, Nigeria can utilize the existing network of comprehensive health centers (Asuquo & Ugare, 2010). These centers, typically situated in rural areas, are staffed by qualified medical personnel that can bring the procedure closer to the grassroots. Second, it would be advantageous for the country to implement a system where pregnant women are continuously evaluated for conditions such as CPD during antenatal clinic visits. This approach would help determine the appropriateness of CS prior to the onset of labor. However, special attention should be directed toward unskilled workers during antenatal care, as this could facilitate the identification of high-risk pregnancies and potentially reduce the necessity for emergency CS. Such measures could significantly decrease the high incidence of emergency CS in Nigeria. Third, despite the prevalence of adolescent childbirth in northern Nigeria, the use of CS remains minimal in this region. Whereas the area is characterized by numerous factors that hinder CS utilization, including cultural practices, low educational attainment among females, early girl-child marriage, teenage motherhood, and poverty (Etcheverry et al., 2024; Islam et al., 2022; NPC and ICF, 2019; Ushie et al., 2019). There is a pressing need for targeted improvements in maternal healthcare within this region, accompanied by consistent advocacy promoting CS as a viable childbirth option. Engaging key stakeholders such as community and religious leaders is crucial (Muhammad Bello & Tukur, 2021). Furthermore, governments at all levels should take proactive measures to enforce the country’s Universal Basic Education (UBE) Act (Universal Basic Education [UBE], 2004), which guarantees every child’s right to basic education regardless of gender. Additionally, incentives such as scholarships and expedited employment opportunities after schooling should be provided to encourage girl-child education. Implementing these strategies has the potential to increase MAFB in the region, which may subsequently enhance the utilization of CS.

Strengths and Limitation of the Study

The data utilized in this study were sourced from the 2018 NDHS, which is a nationally representative sample. This characteristic enhances the generalizability of the study’s findings to the broader population. Another notable strength of this research is its application of spatial methods in analyzing CS rates, an approach that has not been widely adopted in previous studies of CS in Nigeria. This spatial analysis allows for a more nuanced understanding of geographic disparities in CS access and utilization. However, a limitation of this study is the reliance on self-reported data from the NDHS. Such data are inherently subject to biases related to respondents’ ability to accurately recall events and experiences, which may affect the reliability of the findings. Self-reporting can lead to under-reporting or over-reporting of CS occurrences, potentially skewing the results and limiting the conclusions that can be drawn.

Conclusion

This study revealed positive, significant, moderate associations between CS, elective CS, and emergency CS, and three types of mothers’ jobs: PTM, clerical, and unskilled manual. Additionally, the three CS variables had moderate, positive, significant associations with median age at first birth (MAFB). However, their associations with the percentage of teenage mothers were moderate and statistically significant, but negative. Furthermore, the three CS variables exhibited significant spatial autocorrelation.

In the northern part of the country, where there appears to be a greater need for cesarean procedures, usage is the lowest due to various economic and socio-cultural norms. This situation suggests that maternal mortality may continue to rise unabated in the region. To reduce the current burden of maternal and child mortality and to make significant progress toward achieving Goal 3 of the Sustainable Development Goals (SDGs; United Nations (UN), 2015), the country must encourage increased use of Cesarean delivery procedures by women whenever necessary. This can be accomplished through enhanced advocacy and education for the target population, as well as the provision of improved healthcare facilities at the grassroots, particularly in northern Nigeria. Equally important is the need for continued research on CS in the country to gain a better understanding of the people’s perceptions, preferences, and concerns, given that Nigeria has a multicultural and multi-religious population.

Footnotes

ORCID iD

Oluwaseun Addie

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data Availability Statement

The data is available from The DHS program.

References

Adekanle

D. A.

Adeyemi

A. S.

Fasanu

A. O.

(2013). Caesarean section at a tertiary institution in southwestern Nigeria—A 6-year audit. Open Journal of Obstetrics and Gynecology, 03(03), 357–361. https://doi.org/10.4236/ojog.2013.33066

Adewuyi

E. O.

Akosile

Olutuase

Philip

A. A.

Olaleru

Adewuyi

M. I.

Auta

Khanal

(2024). Caesarean section and associated factors in Nigeria: assessing inequalities between rural and urban areas—insights from the Nigeria Demographic and Health Survey 2018. BMC Pregnancy and Childbirth, 24(1), 538.

Adeyanju

B. T.

Aduloju

O. P.

Okunola

T. O.

Ojo

I. O.

(2023). Head circumference, as predictor of cephalopelvic disproportion: A prospective analysis of cases of spontaneous vaginal delivery and caesarean section in Ekiti State, Nigeria. African Journal of Reproductive Health, 27(6s), 154–159.

Ajayi

A. I.

(2021). Inequalities in access to birth by caesarean section in the context of user fee exemption for maternal health services in southwest and north central Nigeria. International Health, 13(6), 598–605.

Akinola

O. I.

Fabamwo

A. O.

Tayo

A. O.

Rabiu

K. A.

Oshodi

Y. A.

Alokha

M. E.

(2014). Caesarean section–An appraisal of some predictive factors in Lagos Nigeria. BMC Pregnancy and Childbirth, 14(1), 217. https://doi.org/10.1186/1471-2393-14-217

Allagoa

D. O.

Oriji

P. C.

Tekenah

E. S.

Obagah

Ohaeri

O. S.

Mbah

K. M.

Ikoro

Wagio

T. J.

Nwanze

N. C.

Atemie

(2021). Caesarean section in a tertiary hospital in south-south, Nigeria: A 3-year review. European Journal of Medical and Health Sciences, 3(2), 122–127.

Angolile

C. M.

Max

B. L.

Mushemba

Mashauri

H. L.

(2023). Global increased cesarean section rates and public health implications: A call to action. Health Science Reports, 6(5), e1274.

Anselin

(1995). Local indicators of spatial association—LISA. Geographical Analysis, 27(2), 93–115.

Asuquo

M. E.

Ugare

(2010). Trauma in a teaching hospital outpost: Comprehensive Health Centre, Okoyong, Cross River State, Nigeria. European Journal of Trauma and Emergency Surgery, 36(1), 49–52. https://doi.org/10.1007/s00068-009-9013-4

10.

Betran

A. P.

Moller

A. B.

Souza

J. P.

Zhang

(2021). Trends and projections of caesarean section rates: Global and regional estimates. BMJ Global Health, 6(6), e005671.

11.

Bivand

Keitt

Rowlingson

(2021). Rgdal: Bindings for the ‘Geospatial’ Data Abstraction Library. R package version 1.5-23.

12.

Bivand

Rundel

Pebesma

(2017). Rgeos: interface to geometry engine-open source (GEOS). R package version 0.3-26.

13.

Bivand

R. S.

Wong

D. W. S.

(2018). Comparing implementations of global and local indicators of spatial association. Test, 27(3), 716–748.

14.

Brunsdon

Comber

(2018). An introduction to R for spatial analysis and mapping. Sage.

15.

de Loenzien

Mac

Q. N. H.

Dumont

. (2021). Women’s empowerment and elective cesarean section for a single pregnancy: A population-based and multivariate study in Vietnam. BMC Pregnancy and Childbirth, 21(1), 3. https://doi.org/10.1186/s12884-020-03482-x

16.

Dutta

Tuli

Shukla

Patil

van Duinen

A. J.

Thivalapill

Ghoshal

Jayaram

Roy

Gadgil

(2023). Geographic and socioeconomic inequities in cesarean delivery rates at the district level in Madhya Pradesh, India: A secondary analysis of the national family health survey-5. Global Health Action, 16(1), 2203544.

17.

Etcheverry

Betrán

A. P.

de Loenzien

Kaboré

Lumbiganon

Carroli

Mac

Q. N. H.

Gialdini

Dumont

(2024). Women’s caesarean section preferences: A multicountry cross-sectional survey in low- and middle-income countries. Midwifery, 132, 103979.

18.

Guendelman

Gemmill

Hosang

MacDonald

L. A.

(2017). Physical and organizational job stressors in pregnancy and associations with primary cesarean deliveries. Journal of Occupational and Environmental Medicine, 59, 571–577. https://doi.org/10.1097/JOM.0000000000001019

19.

Gunn

J. K.

Ehiri

J. E.

Jacobs

E. T.

Ernst

K. C.

Pettygrove

Center

K. E.

Osuji

Ogidi

A. G.

Musei

Obiefune

M. C.

Ezeanolue

C. O.

Ezeanolue

E. E.

(2017). Prevalence of caesarean sections in Enugu, southeast Nigeria: Analysis of data from the Healthy Beginning Initiative. PLoS One, 12(3), e0174369. https://doi.org/10.1371/journal.pone.0174369

20.

Irwinda

Hiksas

Lokeswara

A. W.

Wibowo

(2021). Maternal and fetal characteristics to predict c-section delivery: A scoring system for pregnant women. Women’s Health, 17, 17455065211061969.

21.

Islam

M. A.

Sathi

N. J.

Hossain

M. T.

Jabbar

Renzaho

A. M. N.

Islam

S. M. S.

(2022). Caesarean delivery and its association with educational attainment, wealth index, and place of residence in Sub-Saharan Africa: A meta-analysis. Scientific Reports, 12(1), 5554.

22.

Meh

Thind

Ryan

Terry

(2019). Levels and determinants of maternal mortality in northern and southern Nigeria. BMC Pregnancy and Childbirth, 19(1), 417. https://doi.org/10.1186/s12884-019-2471-8

23.

Muhammad Bello

Tukur

. (2021). Contributions of community stakeholders in the co-production of female child education in Nigeria. Journal of Public Administration and Governance, 11(2), 258–276.

24.

National Population Commission of Nigeria (NPC). (2020). Nigeria population projection and demographic indicators-state and National. National Population Commission of Nigeria.

25.

NPC and ICF. (2019). Nigeria Demographic and Health Survey 2018. Abuja, Nigeria, and Rockville, Maryland, USA

26.

Nwobodo

E. I.

Isah

A. Y.

Panti

(2011). Elective caesarean section in a tertiary hospital in Sokoto, north western Nigeria. Nigerian Medical Journal, 52(4), 263. https://doi.org/10.4103/0300-1652.93801

27.

Olivier

S. M.

Muchirewesi

Mukuzunga

Salissou

M. T. M.

(2024). Risk factors for caesarean section due to cephalopelvic disproportion at Sakubva Hospital in Mutare District Manicaland Zimbabwe. Unnes Journal of Public Health, 13(1), 33–38.

28.

Sharma

Brown

Kainuwa

M. A.

Leight

Nyqvist

M. B.

(2017). High maternal mortality in Jigawa State, Northern Nigeria estimated using the sisterhood method. BMC Pregnancy and Childbirth, 17(1), 163. https://doi.org/10.1186/s12884-017-1341-5

29.

Shukla

Mohan

van Duinen

Gadgil

Bakker

Bhushan

Roy

(2022). Assessing geographical and economic inequalities in caesarean section rates between the districts of Bihar, India: A secondary analysis of the National Family Health Survey. BMJ Open, 12(1), e055326.

30.

Simoes

Kunz

Bosing-Schwenkglenks

Schmahl

F. W.

(2005). Occupation and risk of cesarean section: Study based on the perinatal survey of Baden-Württemberg, Germany. Archives of Gynecology and Obstetrics, 271(4), 338–342. https://doi.org/10.1007/s00404-004-0616-z

31.

Tennekes

(2018). Tmap: Thematic maps in R. Journal of Statistical Software, 84, 1–39. Apr 20.

32.

United Nations (UN). (2015). Resolution adopted by the General Assembly on 25 September 2015. Transforming our world: The 2030 agenda for sustainable developmenthttps://www.eea.europa.eu/policy-documents/resolution-adopted-by-the-general

33.

Universal Basic Education (UBE). (2004). The compulsory, free, universal basic education act and other related matters (p. 2004). Publication.

34.

Ushie

B. A.

Udoh

E. E.

Ajayi

A. I.

(2019). Examining inequalities in access to delivery by caesarean section in Nigeria. PLoS One, 14(8), e0221778. https://doi.org/10.1371/journal.pone.0221778

35.

World Health Organization (WHO). (2024). World Health Organization: WHO. Maternal mortality [Internet]. 2025. https://www.who.int/news-room/fact-sheets/detail/maternal-mortality

Spatial Analysis of Cesarean Section Rates in Nigeria

Abstract

Keywords

Introduction

Literature Review

Focus of This Study

Methods

Study Area

Data Collection

Data Analysis

Results

Discussion

Implication of Study

Strengths and Limitation of the Study

Conclusion

Footnotes

ORCID iD

Funding

Declaration of Conflicting Interests

Data Availability Statement

References