Introduction

Antenatal care (ANC) is essential for the health of pregnant women and their fetuses, aiding in labor preparation and the recognition of pregnancy-related danger signs [1,2]. ANC interruption is defined as not receiving ANC care as per the new World Health Organization (WHO) recommendations, which states that pregnant mothers should attend at least eight antenatal care (ANC) visits throughout their pregnancy [2]. In conflict-affected populations, women and children face disproportionate morbidity [3].Evidence from Syria indicates that the prenatal care coverage, which includes one visit with a trained attendant, has decreased [4]. Women in conflict-affected populations suffer differently of difficulties in obtaining reproductive healthcare, higher rates of gender-based violence, and unfavorable pregnancy outcomes, which are made worse by stress and a lack of medical resources [5,6].

A survey conducted in Eastern Burma found that 88% of women reported giving birth at home during their most recent pregnancy, that ANC visits were 16.7% and skilled birth was just 5.1%. The survey also revealed that women were often exposed to human rights violations [6]. From 2000 to 2016 as of the EDHS, the proportion of Ethiopian women receiving prenatal care (ANC) from a competent provider increased from 27% to 34% and 62%, respectively with the Tigray region having the highest rate at 90% [7]. Another study conducted in Tigray showed that women whose husbands arranged transportation for ANC had over twice the odds of being alive compared to those whose husbands did not [8]. According to the study’s analysis of the health system of Tigray during the Tigray war showed that, the devastation of Tigray’s health system was deliberate, systematic, and pervasive, impacting all tiers of the health system zone [9]. According to a recent community-based study, maternal mortality increased from 186 in 2019 per 100,000 live births to 840 per 100,000 live births in 2023 after the Tigray War [10].

Studies indicate that armed conflict has a significant negative impact on maternal healthcare consumption. For instance, research conducted in Uganda, a region severely affected by conflict, shows that access to maternal healthcare diminishes [11]. A recent study conducted in Ethiopia’s Tigray region, the health system trends are in line with war events. The analysis of the region’s zones revealed that the Mekelle and North Western zones were the most stable, while the remaining zones were the least stable, with many missing data points pointing to instability [12]. Similarly, a prospective study during the war in Lebanon revealed that women were neglecting their usual prenatal care, which resulted in an increase in health problems during that period [13]. Additionally, a study focusing on Northern Uganda and Burundi found that armed conflict adversely affected the health system, leading to reduced access to and quality of maternal and reproductive health services [14].

The 2019 Ethiopian Demographic Health Survey indicated that 74% of pregnant women nationally received antenatal care (ANC) from skilled providers. In the Tigray region, this percentage was significantly higher at 94% [15]. However, a study conducted during the Tigray conflict found that in the first 90 days of the war, there were no ANC services available. Additionally, only 27.5% of hospitals, 17.5% of health centers, and 11% of ambulances were operational, while none of the 712 health posts were functional [16].

Understanding the spatial patterns of ANC interruption—identifying specific areas within Tigray with varying interruption rates—and the predictors of these interruptions, such as conflict impact, and socioeconomic factors remain limited. Therefore, assessing the percentage of women with ANC interruptions and their peculiar factors is crucial for developing effective maternal health interventions in Tigray post-conflict, northern Ethiopia.

Methods and materials

Statistical method and analysis

Univariate descriptive, bivariate, chi-square and multivariate spatial logistic regression approaches were employed to analyze the ANC data from targeted zones of the Tigray Region. The univariate analyses were used to summarize and describe the demographic variables (e.g., age, gender, etc.), using frequencies, percentages, means, and standard deviations provided a detailed overview. Associations between categorical variables (e.g., gender, zone, etc.) and ANC interruptions were assessed using Chi-square or Fisher’s exact tests. We used ANC interruption to model the relationship between risk factors and occurrence across zones. Spatial logistic regression is an advanced modelling approach that incorporates zones, representing administrative or geographic divisions, to analyze binary outcomes while accounting for spatial dependencies. The model extends ordinary logistic regression by including spatial random effects () to capture unobserved spatial variation between zones, improving the accuracy of estimates and inference. The model is expressed as:

where represents the probability of the outcome for observation , is the intercept, are predictors, are their coefficients, and captures the spatial effect for zone This approach allows the model to account for spatial clustering and patterns, addressing biases and inaccuracies that arise from spatial autocorrelation in ordinary logistic regression. Spatial autocorrelation in the dependent variables was assessed using Moran’s I test. The spatial logistic regression model, which incorporates structured spatial random effects, is modeled through a Conditional Autoregressive (CAR) framework, to account for spatial dependencies between zones. This makes it valuable tools for policymakers and researchers aiming to address spatial inequalities in various fields, including public health, urban development, and environmental management.

Spatial logistic regression provides several advantages in analyzing binary outcomes, particularly by incorporating spatial dependencies that traditional models often overlook. This approach enhances model fitness and predictive accuracy, facilitating the identification of high-risk areas through spatial clustering of central zone, northwestern zone, and southern, southeastern, Eastern and Mekelle zones of the Tigray region Ethiopia. It also uncovers spatial patterns and clusters, which can inform targeted public health interventions [18–20]. The analysis was conducted using R version 4.4.1, with weighing applied to adjust for non-proportional sampling. Both exploratory and confirmatory analyses were carried out, including descriptive statistics, visualisations, and inferential statistics. Predictors with p < 0.05 in the bivariate and multivariate analysis were incorporated into the final spatial logistic regression model.

Results

Socio-demographic characteristics of participants

The study participants are divided into two groups: host communities 2,069 (84.7%, 95% CI: [83.5%, 85.9%]) and internally displaced people 375 (15.3%, 95% CI: [14.1%, 16.5%]). The mean age of participants was 28 years, with a standard deviation of 0.05.

A significant portion of the participants have not completed formal education. Specifically, 835 individuals (34.2%, 95% CI: [32.5%, 36.0%]) are illiterate and only 201 individuals (8.2%, 95% CI: [7.3%, 9.1%]) hold a college degree or higher.

Geographically, most participants reside in rural areas, with 1,530 (62.6%, 95% CI: [60.9%, 64.3%]). The majority of 2,239 participants (91.6%, 95% confidence interval: [90.5%, 92.7%]), were married.

Access to resources was limited among the participants. For instance, 1,615 (66.1%, 95% CI: [64.4%, 67.8%]) lack electricity, and 1,373 (56.2%, 95% CI: [54.5%, 57.9%]) do not have access to public transportation. Media ownership is also low, with only 308 participants (12.6%, 95% CI: [11.3%, 13.9%]) owning a radio. In contrast, mobile phone ownership is more common, with 1,138 individuals (46.6%, 95% CI: [44.9%, 48.3%]) possessing a mobile phone. Television ownership is relatively low, as just 400 participants (16.4%, 95% CI: [15.1%, 17.7%]) have a television or detail see “Table 1.”

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Socio-demographics characteristics of participants Tigray region northern Ethiopia (N 2444).

https://doi.org/10.1371/journal.pone.0328802.t001

Sexual and reproductive health characteristics of participants

The study participants reported varying experiences regarding pregnancies and births. The majority had 2 pregnancies (21.7%), followed by one pregnancy (20.3%). The number of pregnancies decreases with higher counts: three pregnancies (17.2%), five (10.4%), 6 (6.5%), seven (4.7%), and 8 (5.2%).

In terms of births, most participants reported one birth (22.7%) or two births (22.6%), with subsequent numbers decreasing: three births (16.6%), four (11.9%), five (9.0%), six (5.7%), seven (3.8%), and eight (3.7%). Additionally, 4.1% have not given birth.

Regarding live births, the distribution is as follows: one live birth (23.2%), two (22.2%), three (16.6%), four (11.9%), five (8.5%), 6 (5.6%), seven (3.5%), and eight or more (3.2%). Notably, 5.3% of participants reported having no live births.

The majority, 2,328 (95.0%), reported zero stillbirths, while 95 (3.9%) had 1 stillbirth, 18 (0.7%) had 2, and 2 (0.08%) had 3, and 1 (0.04%) had 4 stillbirths.

Regarding abortions, 1,622 (66.4%) reported having 0 abortions in their lifetime. However, a notable portion has experienced abortions: 698 (28.6%) reported 1 abortion, 100 (4.1%) reported 2, 18 (0.7%) reported 3, 5 (0.2%) reported 4, and 1 (0.04%) reported 5 (Fig 1).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Reasons for interruption of ANC care during the war from Pretoria November 2020 to August 2022 agreement of the Tigray region, Northern Ethiopia.

https://doi.org/10.1371/journal.pone.0328802.g001

Pregnancy intentions among participants

The study revealed varied pregnancy intentions among participants. A majority, 1,455 (59.5%), reported were actively trying to become pregnant at the time of conception. However, a significant portion, 738 (30.2%), expressed a desire to wait until later to have children. Additionally, 251 (10.3%) stated that they did not wish to have any more children when they become pregnant. The findings of this study indicate a diverse range of pregnancy intentions within the study population, highlighting that while many participants were intentionally seeking pregnancy, a notable number preferred to delay or avoid further pregnancies. For more details, see “Table 2”.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. War and ANC interruption of ANC care during the war to Pretoria November 2020 to August 2022 agreement of the Tigray region Northern Ethiopia (N-2444).

https://doi.org/10.1371/journal.pone.0328802.t002

Antenatal care (ANC) interruption: The prevalence of antenatal care (ANC) interruption was found to be 24.8% (95% CI: [23.1%, 26.5%]), indicating that nearly one in four women encountered difficulties in maintaining consistent ANC. The primary reasons for these interruptions included insecurity (45.3%, 95% CI: [43.1%, 47.5%]), health facility closures (39.8%, 95% CI: [37.6%, 42.0%]), and a lack of supplies and medicine for ANC (33.1%, 95% CI: [31.0%, 35.2%]).

The spatial distribution of ANC interruptions across the Tigray region is as follows: Northwestern zone 22.8%, Central zone 29.1%, Eastern zone 17.7%, Southern zone 16.7%, Southeastern zone 6.4%, and Mekelle zone 7.3%. However, the Western zone of Tigray was not accessible for our study due to security concerns. For more details, see Fig 2.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 2. Special distribution of ANC interruption across the zone of Tigray Region (n = 2444).

https://doi.org/10.1371/journal.pone.0328802.g002

Bivariate analysis: Impact of war on antenatal care (ANC) visits

The study found that 75.2% (1,839, 95% CI: [73.9%, 76.5%]) of participants did not postpone or interrupted their antenatal care (ANC) visits due to the ongoing war, while 24.8% (605, 95% CI: [23.5%, 26.1%]) reported interruptions. Among those with a sick family member in the last two weeks, 14.3% (349, 95% CI: [12.9%, 15.7%]) did not postpone their ANC visit, and 7.0% (171, 95% CI: [5.9%, 8.1%]) did Fig 3.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 3. The line graph shows the reproductive history of study participants of the interruption of ANC during the war in Tigray 2023.

https://doi.org/10.1371/journal.pone.0328802.g003

The war’s impact was significant, with 96.5% (2,358, 95% CI: [95.4%, 97.6%]) reporting a family member was disabled; of these, 72.7% (1,778, 95% CI: [71.0%, 74.4%]) did not postpone their ANC visit, while 23.7% (580, 95% CI: [22.2%, 25.2%]) did. Additionally, 97.0% (2,370, 95% CI: [96.0%, 98.0%]) experienced the loss of a family member, with 73.2% (1,790, 95% CI: [71.5%, 74.9%]) continuing their ANC visits. A small fraction, 1.4% (34, 95% CI: [0.8%, 2.0%]), reported a woman in their household experiencing rape due to the war, with only 1.1% (27, 95% CI: [0.5%, 1.7%]) not postponing ANC visits.

Property damage was reported by 53.3% (757, 95% CI: [51.0%, 55.6%]) of participants, with 38.2% (543, 95% CI: [35.9%, 40.5%]) not postponing their ANC visits. Looting affected 89.7% (1,274, 95% CI: [88.5%, 90.9%]), with 66.1% (939, 95% CI: [64.5%, 67.7%]) continuing their visits. Destruction of property was noted by 27.3% (388, 95% CI: [25.6%, 29.0%]) of participants, where 20.3% (288, 95% CI: [18.7%, 21.9%]) did not postpone. Displacement impacted 56.7% (1,386, 95% CI: [54.9%, 58.5%]), with 42.0% (1,027, 95% CI: [40.3%, 43.7%]) not postponing their ANC visits.

In terms of distance to health posts, 64.8% (1,583, 95% CI: [63.0%, 66.6%]) reported a round trip of less than 1 hour, while 35.2% (861, 95% CI: [33.4%, 37.0%]) reported longer trips. Those who did not postpone their ANC visits were more likely to live closer to health facilities (49.7% vs. 15.1%).

Functionality of nearby health facilities was high, with 75.8% (1,303, 95% CI: [74.1%, 77.5%]) reporting nearby health posts as functional and 97.0% (2,371, 95% CI: [96.0%, 98.0%]) for health centers/hospitals. Access to public transport was reported by 43.8% (1,071, 95% CI: [41.9%, 45.7%]), with those who had not postponed ANC visits slightly more likely to have access (32.8% vs. 11.0%).

Regarding electricity supply, 33.9% (829, 95% CI: [32.1%, 35.7%]) reported having it, and those who did not postpone their ANC visits were somewhat more likely to have electricity (26.5% vs. 9.0%). Media ownership was limited, with 12.6% (308, 95% CI: [11.3%, 13.9%]) owning a radio and 16.4% (400, 95% CI: [15.1%, 17.7%]) owning a television. Those who did not postpone their ANC visits were slightly more likely to own a radio (9.6% vs. 3.5%) or a TV (12.6% vs. 4.5%). Mobile phone ownership was similar between both groups, with 46.6% (95% CI: [44.9%, 48.3%]) owning a phone.

Multivariate analysis: Parameter estimates from SLR

The Moran I test indicated significant spatial autocorrelation in ANC interruption (Moran’s I = 0.546, p-value < 0.001), indicating the presence of spatial autocorrelation in the dependent variable, validating the need for spatial logistic regression. Incorporating spatial dependencies is crucial, as neglecting them in ordinary logistic regression may result in biased parameter estimates and inaccurate inferences. Model fitness improved substantially with the inclusion of the spatial term, as reflected by an increase in log-likelihood from −1200 to −1050 and a decrease in AIC from 2410 to 2110. The pseudo-R -squared increased from 0.55 to 0.72, demonstrating better explanatory power. Multicollinearity diagnostics showed all Variance Inflation Factors (VIFs) below 3, indicating no serious multicollinearity issues among predictors.

The spatial logistic regression model identified several key factors influencing the interruption of antenatal care (ANC) during the war. For every unit increase in distance from the health center, the log-odds of interrupting ANC increased by 0.001 (SE = 0.001, p < 0.001). Geographic differences also played a significant role; women living in the south-east zone had a 0.112 lower log-odds of interrupting ANC compared to those in the north-west zone (SE = 0.05, p = 0.02). Additionally, women in the eastern zone exhibited a 0.16 lower log-odds of interrupting ANC compared to the north-west zone (SE = 0.04, p = 0.001).

The number of children under five in the household was another essential factor, with each additional child increasing the log-odds of interrupting ANC by 0.08 (SE = 0.02, p < 0.001). Furthermore, households without a radio had a 0.08939 lower log-odds of interrupting ANC compared to those with a radio (SE = 0.03, p = 0.001). These results suggest that longer distances to health centers, residing in the north-west zone, having more young children, and not owning a radio were all associated with a greater likelihood of ANC interruption in the war-affected Tigray region of Ethiopia. In contrast, the analysis for Mekelle indicated an estimate of −0.04, a standard error of 0.07, and a z-value of −0.55, resulting in a p-value greater than 0.05, which shows no significant effect on ANC interruption compared to the north-west zone.

In contrast, the estimate for women in Mekelle was −0.038 (SE = 0.069, p > 0.05), indicating no statistically significant difference in ANC interruption compared to the north-west zone. These findings highlight the importance of geographic location, access to information, and household burden in shaping ANC service continuity in war-affected areas of Tigray, Ethiopia. Conversely, the eastern zone showed significantly lower odds of ANC interruption, with an estimate of −0.160 (SE = 0.050, p < 0.05), indicating better adherence to ANC in this zone as seen in Table 3.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 3. Spatial logistic regression model analysis shows the parameter estimates, standard errors, z-values, and p-values for the predictors of interruption of ANC during the war in Tigray 2023.

https://doi.org/10.1371/journal.pone.0328802.t003

Discussion

The prevalence of antenatal care (ANC) interruption was found to be 605 (24.8%), indicating that nearly one in four women faced challenges in maintaining consistent ANC. The interruption was notably more severe among mothers residing in rural areas (370, 61.2%) compared to semi-urban (51, 8.4%) and urban (184, 30.4%) settings. A significant majority of those affected were host community (non displaced people)(519, 85.8%) rather than internally displaced mothers (86, 14.2%). Additionally, many participants were married (563, 91.1%).

Similar studies conducted in Syria in 2014–2017 have found that violent events significantly reduce the likelihood of pregnant women seeking and accessing essential healthcare services showed a negative correlation between bombardments and consultations, and ANC visits were reported; the RR was 0.95 (95% CI 0.94–0.97) and 0.95 (95% CI 0.93–0.98), respectively.

The fear and risks associated with ongoing conflicts deter pregnant women from seeking necessary care, increasing the vulnerability of both mothers and their unborn children to potential health complications [21]. Research in Yemen highlighted those pregnant women in remote, rural areas encountered substantial barriers to reaching ANC clinics due to infrastructure damage, lack of public transportation, and financial constraints [22]. Furthermore, studies in Afghanistan have shown that the Taliban’s restrictions on women’s movement and access to healthcare during conflicts severely limited pregnant women’s ability to attend ANC appointments [23,24]. These challenges were particularly pronounced in rural areas compared to urban settings [25]. This is a critical concern since ANC is a well-established and cost-effective intervention that provides essential care to pregnant women from early pregnancy to delivery [26].

The high prevalence of interruptions in antenatal care (ANC) services during conflict situations contradicts the internal resolution of the World Health Organization (WHO, 2012), which stresses the importance of leadership in documenting attacks on health workers, facilities, and patients [27].

The results of the spatial logistic regression analysis in the war-affected Tigray region of Ethiopia revealed several key factors influencing ANC interruptions. For instance, the distance to health centers was a significant factor; for every unit increase in distance (in minutes), the log-odds of discontinuing ANC increased by 0.001 (SE = 0.0001, p < 0.001). This finding aligns with studies conducted in Syria, where ongoing conflict and insecurity significantly disrupted access to ANC services. This was compared by the time taken to reach health facilities both before the war and during the war, even if the distance is constant, the time taken in These two periods were different; this may be due to the security reasons, blockage, checkpoints here and there. Women reported facing barriers such as transportation difficulties, fear of checkpoints, and financial constraints when trying to reach health facilities [28,29]. Additionally, a study in Syria found that women living in besieged or hard-to-reach areas experienced much greater difficulty accessing ANC services due to transportation challenges, security concerns, and a lack of functioning health facilities [30].

Living in the north-west zone of the war-affected Tigray region was associated with a higher likelihood of interruption of antenatal care (ANC) this is similar with a study conducted during the Tigray war, 80 times of drone and aerial bombardment incidents were reported resulted in casualties in six zones and 24 districts in Tigray, Ethiopia. Nearly a third of the victims were children, and half of the casualties were females, including pregnant and lactating women. The Southern and Northwest zones caused the most casualties, Southeast zones and Mekelle. There were many civilian casualties from the airstrikes, which mostly occurred in civilian locations such as marketplaces, internally displaced persons (IDP) camps, residential areas, public transportation, villages, children’s playgrounds, churches, mills, and hospitals. Women, children, and the elderly were disproportionately affected [31].This may be the blockage, movement from place to place, or severity of war may be higher in this area and because it is far from Mekelle the capital city of the Tigray region which is about 300kms, some of the humanitarian organisations were not reaching this area. This is also similar to studies conducted in Syria where bombardment and distant villages from the centers interrupted ANC service than those who are near the centers [21]. This is the fact that in conflict settings, healthcare access is severely compromised due to the destruction of facilities and the evacuation of medical personnel, leading to increased travel times and reduced service availability where health facilities are bombarded even the public transportation and other related facilities are hampered.

Conversely, while distance is a critical factor, other elements such as socio-economic status, education, and community support also plays a role in healthcare access and continuity in conflict-affected regions [32].

Also align with previous a study, which have consistently highlighted those geographic disparities, often exacerbated by conflict, plays a critical role in shaping healthcare utilization. For instance, in conflict-affected regions of Afghanistan, women in remote or insecure areas faced higher barriers to ANC attendance due to disrupted infrastructure and security concerns, a pattern that resonates with the higher likelihood of ANC interruption [33]. Additionally, households with a greater number of young children were more prone to discontinuation, as were those who did not own a radio. These findings underscore the unique challenges faced by pregnant women in war-affected regions, highlighting the need for improved access to health facilities, addressing transportation and security concerns, and implementing targeted interventions to support households with multiple young children.

Research from Yemen and the Democratic Republic of the Congo further illustrates the severe Impact of armed conflicts and civil wars on maternal healthcare access in sub-Saharan Africa. In Yemen, the escalation of the civil war led to the closure of many health facilities, significantly hindering pregnant women’s ability to receive regular ANC [34,35]. Similarly, findings from the Democratic Republic of the Congo indicate that armed conflicts and displacement forced many pregnant women to miss critical ANC visits, resulting in poorer maternal and child health outcomes and also another study from Nepal indicates that women in conflict areas receive fewer ANC visits when there is armed conflict, which has a direct effect on maternal health outcomes.

[36–38]. Overall, the effects of civil wars on sub-Saharan Africa have severely compromised maternal and child health services [39].

The spatial logistic regression analysis highlighted distance to health centers as a significant factor; for every unit increase in distance (in minutes), the log-odds of discontinuing ANC increased by 0.001 (SE = 0.001, p < 0.001). Studies from Yemen reported substantial barriers for pregnant women in remote, rural areas trying to access ANC clinics, primarily due to infrastructure damage, lack of public transportation, and financial constraints [40,41]. Studies from Damage to the public health system caused by war-related looting or vandalism in the Tigray region of Northern Ethiopia, also showed that due to facility damage and medical personnel evacuation, access to healthcare is seriously jeopardized in crises, resulting in more extended travel times and fewer services available [42].Another study on the quality of health care service in a conflict setting showed that healthcare disruption in armed conflict creates significant obstacles to the provision of healthcare, such as medical supply shortages and access-hindering security threats [43].On the other hand, although distance is an important considerations, socioeconomic position, education, and community support are also important factors that affect healthcare continuity and access in areas affected by violence [44].

Geographic zones also played a critical role; women living in the south-east zone had a 0.11lower log-odds of discontinuing ANC compared to those in the north-west zone (SE = 0.05, p = 0.01). Similarly, women in the eastern zone exhibited a 0.16 lower log-odds of discontinuing ANC compared to the north-west zone (SE = 0.04, p = 0.001). Thus, living in the north-west zone was associated with a higher likelihood of ANC interruption in the Tigray region. This finding mirrors those from the Eastern Mediterranean region, where conflicts have disrupted health service utilization, especially in rural areas [45]. A study in the Democratic Republic of the Congo noted that pregnant women displaced to rural areas had significantly lower ANC centres, per cent of the population than those remaining in or fleeing to urban centers [46,47]. Furthermore, research in Afghanistan showed that the Taliban’s restrictions on women’s movement disproportionately affected rural women, leading to higher rates of missed ANC appointments compared to their urban counterparts [47–51].

These findings underscore the critical importance of addressing the unique challenges faced by pregnant women in war-affected regions, particularly those living in remote, rural areas with limited access to health facilities. Enhancing transportation, infrastructure, and security in these areas is essential for improving maternity healthcare utilization and outcomes.

Broader patterns in conflict-affected settings reveal that the uneven distribution of conflict intensity and damage to health infrastructure across different zones has led to varying impacts on pregnant women’s ability to receive antenatal care (ANC). The findings from this study in Tigray align with another analysis indicating that, as of June 2021, only 27.5% of hospitals, 17.5% of health centers, 11% of ambulances, and none of the 712 health posts were functional [16].

The specific reasons for the higher interruption of antenatal care (ANC) in the north-west zone of Tigray remain unclear, highlighting the need for further contextual analysis to uncover the underlying factors contributing to this geographic disparity. Additionally, the number of children under five in a household significantly impacts ANC continuity; for each additional child, the log-odds of discontinuing ANC increase by 0.08 (SE = 0.02 p < 0.001). Parity is positively correlated with ANC interruptions, as demonstrated by demographic and health surveys in eight Sub-Saharan African (SSA) countries, particularly among women with five or more children. Our research indicates that mothers with more children are also more likely to cease receiving ANC care. This suggests that the demands of raising multiple children may hinder mothers’ ability to prioritize their health care. The responsibilities associated with caregiving appear to overshadow the importance of regular ANC visits, potentially jeopardizing the health of both mothers and their children [52]. Another study conducted in Tanzania, also found that women with higher parity often relies on their previous pregnancy experiences, leading to a tendency to underutilize ANC services. This reliance on past experiences may create a false sense of security, causing these women to deem ANC less critical [53].However, the practical challenges of managing several young children further complicate their ability to seek necessary prenatal care, reinforcing the findings of the SSA survey and the present studies. Additionally, research from Liverpool studied the unique challenges faced by women experiencing multiple pregnancies. This study emphasized that increased anxiety and practical obligations hinder their capacity to prioritize ANC visits [54]. All those findings underscore as the number of children increased, which increases the interruption of care. This showed that mothers are overwhelmed by the care of children thus our ANC care contacts need to be tailored to support individualized care strategies to address the specific needs of women, ensuring that they receive adequate prenatal care.

This trend may reflect the additional caregiving responsibilities that pregnant mothers with multiple young children face, which can hinder their ability to prioritize their own ANC visits.

This may be severe in the condition of conflict where health facilities were closed due to war, health personnel displaced as shown by community-based study in armed conflict in Tigray showed that only 36.5%, respondents reported the use of optimal antenatal care (ANC) [55].

Moreover, household radio ownership emerged as an important factor; households without a radio had a 0.08939 lower log-odds of discontinuing ANC compared to those with a radio (SE = 0.033, p = 0.007). This suggests that access to media and information sources, such as radio, can positively influence ANC utilization, as mothers who receive ANC-related information are more likely to act on it. study of antenatal care and skilled births in the fragile and conflict-affected situation of Burundi indicated that women who engage with media, including radio, are more likely to utilize healthcare services which is similar to the present study mothers who own radio did not interrupt ANC visits this may be the information disseminated through the radio may be indicating where the services is available and also information about security issues about war may be disseminated so us mothers know when to visit health facilities and when not to visit health facilities [56].

Another study conducted on Antenatal care utilization and determinants in fragile conflict-affected situations in northern Ethiopia showed that radio can foster community discussions around maternal health, increasing awareness and reducing stigma associated with seeking ANC in conflict zones thus in conflict affecting zones radio may be an appropriate massage or information transmission methods that can support health services utilization of communities in conflict affecting situations [55,57]. Another studies conducted in the political unrest in Nepal,Ethiopia and Burundi showed that in areas where physical access to healthcare is limited, radio can bridge the gap by informing women about available services and how to access them, potentially increasing ANC visits, this is another evidence that supports the present study, thus owning a radio in such conflict situations may get information that supports where and when to visit health facilities so that you get health services [36,57,58].

The Tigray study offers a more granular, zone-level analysis that complements zonal-level findings from other conflict-affected contexts, highlighting the importance of understanding subnational variations in healthcare access during conflicts. A recent study in the region indicates that the effect of airstrike and drone bombardment instances showed that nearly one-in-two of the causalities were female including pregnant and lactating women. |Out of the entire region most of the causalities occurred in southern and northwest zones followed by the Southeast zones and Mekelle [31]. Further research to unpack the underlying drivers of geographic disparities in ANC interruption in Tigray would be invaluable for informing more targeted interventions.

Conclusion

The study identified the high prevalence of antenatal care (ANC) interruption, Key factors associated with the likelihood of interruption of ANC included greater distance from health centers, residing in the north-west zone compared to the south-east and eastern zones, having a higher number of children under five in the household, and household ownership of a radio.

Acknowledgments

We extend our heartfelt gratitude to the Tigray Health Bureau (THB) for leading this integrated research and mobilizing technical staff for data collection. Lastly, we thank the participants, as well as the data collectors and supervisors, particularly those from the Mekelle University College of Health Sciences and Tigray Health Research Institute (THRI), for their commitment and integrity in generating quality data.