Burden and determinants of stillbirth in a Nigerian Tertiary health facility: a five-year retrospective cohort analysis Authors: Taofik Oluwaseun Ogunkunle, Tajudeen Lanre Ibrahim, Usman Abiola Sanni, Babajide Oyewale Oyebiyi, Adewemimo Charles Olaosebikan, Surajudeen Oyeleke Bello, Khadijah Asabe Jibril, Salihu Ozegya Muhammed, Tina Lavina, Abdulaziz Imam Published: 2026 ABSTRACT AbstractBackgroundStillbirth rates can reflect the quality of a country’s healthcare system for pregnant women. The burden, however, differs across and within countries. Hence, determining the burden and associated factors could provide valuable location-specific insights that could guide interventions.ObjectivesTo determine the incidence and correlates of stillbirth in a Nigerian Tertiary health facility.MethodsThis was a retrospective study of women who delivered in the Dalhatu Araf Specialist Hospital, Lafia, Nasarawa State, Nigeria, between 1st January 2016 and 31st December 2020. Data for the women were extracted from medical records, including sociodemographic data, time of delivery, mode of delivery, and perinatal outcomes. Regression analysis was used to identify predictors of stillbirth.ResultsThe overall stillbirth rate was 88.30 per 1,000 births (1,033/11,699 per 1,000 births), while the rates from 2016 to 2020 were 93.65, 96.25, 90.98, 82.84, and 80.72, respectively. Multiple gestations are associated with a 2.22-fold higher risk of stillbirth compared to single gestations (AOR: 0.45, 95% CI: 0.30–0.67). Primiparous women were 2.63 times at risk of delivering stillbirths compared to multiparous women (AOR: 0.38, 95% CI: 0.30–0.47) and 1.75 times the risk compared to grand multiparous women (AOR: 0.57, 95% CI: 0.47–0.68). Furthermore, unbooked pregnancy and prematurity were associated with 3.47 (AOR 3.47, 95% CI: 2.99–4.07) and 9.60 (AOR 9.62, 95% CI: 7.99–11.57) times increased odds of stillbirths. Maternal employment status and maternal age showed no significant association with the risk of stillbirth.ConclusionsThe stillbirth rate is high and calls for comprehensive intervention. Strengthening antenatal and perinatal care for high-risk groups, including women with multiple gestations, unbooked pregnancies, and premature deliveries, is essential.Supplementary InformationThe online version contains supplementary material available at https://doi.org/10.1186/s12978-026-02397-8.Keywords: Stillbirths, Nasarawa, Perinatal death, Predictors, Delivery Background In most cases, pregnancy culminates in live birth [1]. However, a sizeable number of fetuses are delivered without a sign of life, and it is associated with high psychological and economic costs to the affected women and their families. The World Health Organization (WHO) defines stillbirth as a baby born without any signs of life at or after 28 weeks of gestation or with a birth weight of less than 1000 g [2, 3]. However, this definition has evolved, mostly in the developed countries, shifting from 28 weeks to 26 weeks, 24 weeks, and now to 20 weeks [4, 5]. These changes reflect progress in neonatal medical care and higher survival rates. Stillbirth is a major contributor to perinatal deaths, and it is estimated that one stillbirth occurs every 16 s globally, translating to approximately two (2) million babies born still every year [6]. The burden of stillbirth is, however, uneven across countries as it is disproportionately high in low- and middle-income countries (LMICs) where the stillbirth rate (21 stillbirths per 1,000 total births) is seven times higher compared to high-income countries (3 stillbirths per 1,000 total births) [6, 7]. Specifically, in order of burden, India, Pakistan, Nigeria, the Democratic Republic of the Congo, Ethiopia, and Bangladesh accounted for approximately half of all stillbirths in 2021 [6]. Causes of stillbirth are linked to obstetric complications like preeclampsia, fetal malpresentation, and prolonged labour, and maternal factors, such as age, substance use, diabetes, chronic hypertension, and infections (e.g., malaria and HIV), as well as fetal factors such as post-maturity, prematurity, growth restriction, and congenital abnormalities [8–11]. Additional risk factors for stillbirth include limited access to reproductive health services, lack or poor quality of antenatal care, delayed healthcare access and poverty [12]. Stillbirths thus reflect the quality of a country’s healthcare system for pregnant women. However, it has received little global attention as it is currently not tracked, and specific targets are not included in the 2030 Agenda for Sustainable Development [13, 14]. The current global estimate of stillbirth is high but is likely underestimated, given the weak surveillance and health reporting systems and low facility-based deliveries in most resource-constrained settings, including Nigeria [9, 15]. The majority of stillbirths are preventable through the provision of quality antenatal care and delivery [10]. There is an urgent need for concerted and accelerated efforts towards reducing the burden of stillbirths, especially in high-burden countries, to end preventable newborn deaths and stillbirths by 2030 as outlined in the ‘Every Newborn Action Plan’ [6]. Such efforts would be driven by accurate data on the burden and its various determinants. Without an accurate national value in Nigeria, facility-level data could be critical in identifying factors associated with stillbirth and guiding interventions. A thorough understanding of the factors leading to stillbirths will help develop targeted interventions to inform policies, programs, and practices aimed at reducing these preventable deaths. In high-income countries, births and deaths are registered, and health facilities provide data to guide the implementation of interventions to lower intrapartum and neonatal deaths. Conversely, many low- and middle-income countries (LMICs) lack adequate vital registration systems for perinatal and neonatal deaths [16]. These countries often depend on public health surveys, such as demographic and health surveys, maternal health surveys, and multiple indicator cluster surveys, which frequently lack sufficient detail to fully understand the causes of stillbirths [17]. A purposive review of individual facility data enables an accurate assessment of the stillbirth burden and can capture the context-specific contributing factors, which are critical in developing interventions to reduce the burden. This study seeks to utilize facility-based data to determine the stillbirth burden and examine how maternal sociodemographic and obstetric, fetal and health system factors contribute to stillbirths to inform targeted interventions. Methods Study design and study site This study was a hospital-based retrospective cohort analysis of all deliveries conducted between 1 January 2016 and 31 December 2020 at Dalhatu Araf Specialist Hospital (Now Federal University Teaching Hospital), Lafia, Nasarawa State, North-central Nigeria. The study period was selected based on the availability of complete, archived obstetric records at the time of data retrieval. The proposed health facility, i.e., Dalhatu Araf Specialist Hospital, Lafia (now Federal University Teaching Hospital, Lafia), is the only referral-level facility in the state capital (Lafia). It is a 370-bed tertiary referral facility and the only specialist referral hospital located in Lafia, the state capital. The hospital receives referrals from primary and secondary health facilities across Nasarawa State and neighbouring states. The Obstetrics and Gynaecology Department has 73 beds distributed across labour (10 beds), antenatal (15 beds), postnatal (24 beds), and gynaecology (24 beds) wards and runs accredited postgraduate residency training programmes. Clinical services are delivered through four consultant-led teams, each comprising at least one consultant obstetrician, a senior registrar, four registrars, and rotating house officers. On weekdays, the units enjoy in-house attendance of all the doctors in a particular unit and at least 2 to 4 nurses during the work hours (08:00–16:00 h). The on-call team typically includes a consultant obstetrician (available on call), a senior registrar, two registrars, and a house officer managing obstetric and gynaecological emergencies. Nursing coverage ranges from three to four nurses during weekday morning shifts (08:00–16:00 h) and two nurses during afternoon, night, weekend, and public holiday shifts. Study population All women who delivered at Dalhatu Araf Specialist Hospital during the study period were eligible for inclusion if delivery occurred at 28 completed weeks of gestation or later, irrespective of pregnancy outcome, and if the delivery was adequately documented in the labour ward or antenatal ward records. Both live births and stillbirths meeting these criteria were included in the analysis. Deliveries occurring before 28 weeks’ gestation were excluded. In addition, records were excluded where the pregnancy outcome could not be reliably ascertained or where babies were brought into the facility after delivery elsewhere without verifiable documentation of the delivery circumstances. Operational definition Age of viability – This was defined as 28 weeks. Age of viability – This was defined as 28 weeks. Stillbirth – Baby delivered without any signs of life at or after the age of viability. Stillbirth – Baby delivered without any signs of life at or after the age of viability. Preterm birth – Delivery after the age of viability (≥ 28 weeks), but earlier than 37 weeks. Preterm birth – Delivery after the age of viability (≥ 28 weeks), but earlier than 37 weeks. Term birth – Delivery at 37 completed weeks and beyond. Term birth – Delivery at 37 completed weeks and beyond. Sample selection and sample size A total population sampling approach was used. All eligible deliveries recorded between January 2016 and December 2020 were included; therefore, no sampling or sample size estimation was required. This approach minimised selection bias and maximised statistical power to evaluate determinants of stillbirth. Data sources and extraction Eligible cases were identified using the labour ward and antenatal ward admission and delivery registers. Corresponding patient case notes were retrieved from the hospital medical records department. Data were extracted using a pre-designed and pilot-tested structured data extraction tool developed from study objectives and previous literature. The tool was piloted on 5% of records outside the study dataset to assess clarity, completeness, and inter-reviewer consistency. Necessary modifications were made before full data collection commenced. Two trained medical record officers and one research clinician performed data extraction independently. This follows a two-day standardised training on variable definitions and abstraction procedures. Extraction was conducted under the supervision of a consultant, who performed periodic audits and random spot checks of completed forms. Variables Outcome variable This was defined as stillbirth, i.e., a baby who dies after 28 weeks of gestation but before or during birth. Exposure variables Maternal sociodemographic factors Maternal age and employment status. Obstetric factors Parity, previous miscarriages, mode of delivery, antenatal care history (booked or unbooked), and stage at labour admission. Fetal factors Multiple gestation and prematurity (defined by the gestational age). Health system factors Shift of admission. Measures to minimise bias Bias was minimised by including all eligible deliveries during the study period using total population sampling. Standardised variable definitions and a structured data extraction tool were used, and trained data abstractors followed uniform procedures under supervisory review. Records with discrepancies were crosschecked against multiple sources, including ward registers and case notes, to ensure accuracy. Data analysis The stillbirth rate (SBR) was calculated as the number of stillbirths divided by the number of total births, multiplied by 1000. A line graph was used to show the trend of stillbirths over the five years. Univariable analysis using chi-square tests was used to compare proportions between stillbirths and live births. Statistical significance for all univariable analyses was defined as p-values < 0.05. Significant predictors (i.e., exposure variable with p < 0.2) on univariable analyses were fitted into a bi-variable and maximal logistic regression model. The least significant predictor (with p > 0.05) was dropped using backwards stepwise regression until the final model, which had only significant predictors. The coefficients of the final model were exponentiated to derive adjusted odds ratios (AORs) and their corresponding 95% confidence intervals (CIs). We used data removal for missing data, as missing data was < 1%, and this was unlikely to affect data quality. Ethical consideration Ethical approval was obtained from the hospital research ethics committee (Approval number: FUTHREC/683). Hospital management granted permission to access medical records. To ensure confidentiality, patient identifiers were not collected during data abstraction. Each record was assigned a unique study identification number. Extracted data were stored in password-protected electronic databases accessible only to study investigators. Result The Dalhatu Araf Specialist Hospital recorded a total of 11,780 deliveries between January 2016 and December 2020. We, however, included the data from 11,699 (99.3%) women who had the desired outcome (i.e., stillbirth or live birth ≥ 28 weeks gestation) captured in the hospital record in the estimation of the stillbirth rates. Pregnancy outcome, i.e., livebirth or stillbirth, was not recorded for 81 women and was not included. Overall, 1033 babies were born still (88.30 per 1,000 births), while the annual stillbirth rate from 2016 to 2020 was 93.65, 96.25, 90.98, 82.84, and 80.72, respectively (Table 1). Figure 1 depicts a declining trend of stillbirths over five years in the health facility, and this was statistically significant (p < 0.001). Table 1Yearly (2016–2020) and cumulative incidence rates of stillbirths in the hospitalYearTotal number of birthsNumber of stillbirthsStillbirth rate (per 1000)2016149514093.652017229622196.252018268224490.982019289724082.842020232918880.72Total11,699103388.30 Yearly (2016–2020) and cumulative incidence rates of stillbirths in the hospital Annual Incidence rate of stillbirth over five years (2016–2020) The sociodemographic characteristics of women are presented in Table 2. The mean age of women was 27.17(± 5.66), and this was comparable between women who delivered stillbirths and live births. The majority (91.24%) of women were aged 18 to 35 years, while those aged less than 18 years only accounted for 1.2%. The age group does not differ significantly between women who had stillbirths and live births. Seven-five per cent of women who delivered in the facility over the five years were from eight tribes (Hausa, Eggon, Alago, Kanuri, Fulani, Gwandara, Koro and Mada). In addition, 86.3% (10,079/11,675), who delivered in the facility were not gainfully employed, while 67% (7,875/11,690) of them were Muslims. Table 2Socio-demographic characteristics of women who delivered in the facilityVariablesCategoriesFrequencyPercentagesMaternal age group (11,681*)< 18 years1411.218–35 years10,67491.4> 35 years8667.4Mean ± SD27.17 ± 5.66Parental Ethnic group (11,487*)Hausa234220.4Eggon219419.1Alago135411.8Kanuri11359.9Fulani6415.6Gwandara4033.5Koro3322.9Mada1931.7Others#289325.2Maternal employment status (11,675*)Gainfully employed159613.7Not gainfully employed10,07986.3Religion (11,690*)Islam787567.4Christianity381532.6*Missing variables; #Agatu, Afo, Tiv, Igbo, Yoruba, Ibira, Bassa, Birom, Idoma, Gbagi, Igede, Jukum, Nupe, Tarok, Rendre, Kantana, Jaba, Ogoja, Yeskwa, Ninzo, Ngas, Nungu, Angas, Akwei & 60 others Socio-demographic characteristics of women who delivered in the facility *Missing variables; #Agatu, Afo, Tiv, Igbo, Yoruba, Ibira, Bassa, Birom, Idoma, Gbagi, Igede, Jukum, Nupe, Tarok, Rendre, Kantana, Jaba, Ogoja, Yeskwa, Ninzo, Ngas, Nungu, Angas, Akwei & 60 others Table 3 depicts the sociodemographic, pregnancy and delivery characteristics of women who delivered stillbirths and live births. Maternal age and age group were not associated with stillbirth. Higher stillbirth proportions were observed among women who were not gainfully employed (9.2% vs. 6.3%) and among grand multiparous mothers (p < 0.001), while previous miscarriage history showed no association. Stillbirth varied significantly by shift of delivery, occurring least during the morning shift (7.6%) compared with afternoon (9.2%) and night deliveries (9.3%) (p = 0.020). Unbooked mothers had markedly higher stillbirths than booked mothers (16.4% vs. 4.7%, p < 0.001). Mode of delivery was significantly associated, with the lowest stillbirth proportion following elective caesarean section and the highest following spontaneous vaginal delivery (p < 0.001). Preterm delivery showed a strong association with stillbirth (p < 0.001), whereas stage of labour at presentation was not significant. Table 3Socio-demographic, pregnancy and delivery characteristics of women who delivered stillbirth and live birthVariablesFoetal OutcomesLive birthsStillbirthsTotalp-valueMaternal age, years (Mean ± SD)27.17 ± 5.6427.22 ± 5.9327.17 ± 5.660.774Maternal age group (11,681)0.359 < 18124 (87.9)17 (12.1)141 18–359738 (91.2)936 (8.8)10,674 > 35786 (90.8)80 (9.2)866Maternal employment status (11,675) < 0.001* Gainfully employed1495 (93.7)101 (6.3)1596 Not gainfully employed9148 (90.8)931 (9.2)10,079Parity (11,683) < 0.001* Primiparous2913 (93.2)211 (6.8)3124 Multiparous6133 (91.9)541 (8.1)6674 Grand multiparous1605 (85.1)280 (14.9)1885Previous miscarriages0.770 None8118 (91.1)792 (8.9)8910 At least one2516 (91.3)240 (8.7)2756Multiple gestations (11,683) 0.088* No (single)10,158 (91.9)992 (8.9)11,150 Yes (multiple)497 (93.2)36 (6.8)533Shift of delivery (11,450) 0.020* Morning2930 (92.4)242 (7.6)3172 Afternoon2832 (90.8)285 (9.2)3108 Night4687 (90.7)483 (9.3)5170Booking status (11,649) < 0.001* Booked7139 (95.3)349 (4.7)7488 Unbooked3480 (83.6)681 (16.4)4161Mode of delivery (11,618) < 0.001* ELCS760 (97.6)19 (2.4)779 EMCS3121 (92.3)259 (7.7)3380 SVD6710 (90.0)749 (10.0)7459Gestational age < 0.001* Term (≥ 37 weeks)10,224 (93.6)695 (6.4)10,919 Preterm (≥ 28 - <37 weeks)425 (56.7)325 (43.3)750Stage at Labour presentation (n = 10,272) 0.191 First Stage8423 (91.1)825 (8.9)9248 Second Stage920 (89.8)104 (10.2)1024ELCS, Elective cesarean section, EMCS Emergency cesarean section, SVD Spontaneous vaginal delivery*p value < 0.2p-values in bold are statistically significant Socio-demographic, pregnancy and delivery characteristics of women who delivered stillbirth and live birth ELCS, Elective cesarean section, EMCS Emergency cesarean section, SVD Spontaneous vaginal delivery *p value < 0.2 p-values in bold are statistically significant The risk factors for stillbirths are reported in Table 4. Multiple gestation, fetal maturity, booking status, parity, and period of delivery while in labour were all associated with a change in stillbirth rate. The risk of stillbirth is 2.22 times higher in multiple gestations compared to single gestations (AOR: 0.45, 95% CI: 0.30–0.67, Table 4). Using primiparous women as the reference group, both multiparous (AOR 0.38, 95% CI: 0.30–0.47) and grand multiparous women (AOR 0.57, 95% CI: 0.47–0.68) had significantly lower odds of stillbirth. This indicates that primiparity was associated with the highest stillbirth risk, while multiparity conferred the lowest risk, with grand multiparity showing an intermediate risk pattern. Also, with the morning shift as the reference category, deliveries occurring during the afternoon and night shifts were associated with higher odds of stillbirth on unadjusted analysis (afternoon: OR 1.22, 95% CI 1.02–1.46, p = 0.028; night: OR 1.25, 95% CI 1.06–1.47, p = 0.007). However, after adjustment for other covariates, the association attenuated. Afternoon deliveries were associated with a modest reduction in the adjusted odds of stillbirth compared with morning deliveries (AOR 0.81, 95% CI 0.66–0.99, p = 0.045), while night deliveries showed no significant difference in risk (AOR 0.97, 95% CI 0.81–1.16, p > 0.05). Furthermore, unbooked pregnancy and prematurity were associated with 3.47 (AOR 3.47, 95% CI: 2.99–4.07) and 9.60 (AOR 9.62, 95% CI: 7.99–11.57) times increased odds of stillbirths. However, maternal employment status and maternal age did not affect stillbirths. Table 4Predictors of stillbirth among the study participantsCharacteristicsOR (95% CI)P-ValueAOR (95% CI)P-valueMaternal employment Gainfully EmployedRefRef Not gainfully employed1.51 (1.22–1.86)0.0001.20 (0.93–1.54)0.155Gestation MultipleRefRef Single1.35 (0.95–1.90)0.0890.45 (0.30–0.67)0.000Parity PrimiparousRefRef Multiparous1.22 (1.03–1.44)0.0190.38 (0.30–0.47)0.000 Grand multiparous2.41 (1.99–2.91)0.0000.57 (0.47–0.68)0.000Shift of delivery MorningRefRef Afternoon1.22 (1.02–1.46)0.0280.81 (0.66–0.99)0.045 Night1.25 (1.06–1.47)0.0070.97 (0.81–1.16)1.155Booking status BookedRefRef Unbooked4.00 (3.50–4.58)0.0003.47 (2.99–4.07)0.000Gestational age Term (≥ 37 weeks)RefRef Preterm (≥ 28 - <37 weeks)11.25 (9.55–13.25)0.0009.62 (7.99–11.57)0.000Stage at Labour Presentation First StageRefRef Second Stage1.15 (0.93–1.43)0.1910.99 (0.78–1.25)0.928 Predictors of stillbirth among the study participants Discussion In this study, we determined the burden and trend of stillbirth between 2016 and 2020 in a tertiary-level health facility in North-central Nigeria. We also identified some pregnancy and maternal sociodemographic factors associated with stillbirths. Many of our findings are consistent with evidence from other tertiary-level health facilities in LMICs [18–20]. However, we have also identified additional important factors in our setting that may be relevant in comparable contexts. Overall, we found a stillbirth rate of 88 per 1,000 live births, which is higher than the country’s rate of 42.9 per 1,000 births. However, the stillbirth rate declined modestly from 93 per 1,000 births in 2015 to 80 per 1,000 live births by 2020. In a recent prospective study conducted over one year in Kano, Northwestern Nigeria, Milton et al.9 reported a higher rate of 180 per 1,000 births. Pregnancies fraught with complications and at increased risk for stillbirth are more likely to present and be managed at referral facilities, like in our settings. This could thus account for the high rate of stillbirths observed. Nevertheless, the rate is higher than reported in facility-based studies from other parts of the country and other African countries with similar contexts. A cross-sectional study of all deliveries over 6 months in six general and two teaching hospitals across Northern and Southern Nigeria reported a stillbirth rate of 39.6 per 1,000 births [8]. Also, a rate of 61.8 per 1,000 births and 40.3 per 1,000 births were reported from Lagos and Abakaliki, Southeast and Southwestern Nigeria, respectively [21, 22] Relatively lower rates, 22.20, 33.72 and 36.45 per 1,000 live births, are reported in other developing countries [23, 24]. The stillbirth rate declined steadily from 2017 despite the increasing number of deliveries, with the lowest recorded during the pandemic year (2020). Although staffing improved generally in the hospital starting in 2018 and may have improved the perinatal outcome of care through better antenatal care and improved supervision during labour and delivery, the reduced rate during the year of the COVID-19 outbreak is at variance to the observation in Lagos, Nigeria, where increased stillbirth was recorded[25]. Lagos State was the hotspot for the COVID-19 pandemic in Nigeria and may have witnessed significant disruption in healthcare services, leading to poor perinatal outcomes [26]. Additionally, some stillbirths may have occurred at home due to movement restrictions that prevented women from reaching the hospital. The lockdowns led to a noticeable decline in hospital births, making it likely that some stillbirths went unreported or occurred outside medical facilities. Maternal age and employment status were the only sociodemographic variables included in the analytical model, while tribe and religion were reported solely for descriptive purposes. Employment status may partly reflect underlying socioeconomic conditions that influence health-seeking behaviour, access to antenatal services, and timely utilisation of obstetric care. However, its association with stillbirth was not maintained after adjustment, suggesting that observed differences were likely explained by other obstetric factors rather than an independent effect. Data on maternal educational status, an important and closely related determinant of maternal health literacy, care utilisation, and pregnancy outcomes, were not available in this retrospective dataset, which may have limited a fuller assessment of social influences on stillbirth risk. The higher crude stillbirth proportion observed among grand multiparous women likely reflects confounding by coexisting obstetric and healthcare-related risk factors. This is because adjustment for these variables in the multivariable model attenuated and reversed the association, indicating that parity itself was not the primary driver of risk. Primiparity may reflect inexperience, delayed recognition of labour complications, or obstructed labour risk [27]. Conversely, past parous experience in multiparous women may have also caused improved antenatal attendance, awareness of pregnancy-related risks and improved health-seeking behaviours, hence less risk of adverse pregnancy outcomes. However, the protective effect is stronger in multiparous women (AOR 0.38) compared to grand multiparous women (AOR 0.57), suggesting a diminishing protective effect with higher parity. Increased rates of stillbirths among grand multiparous women have been previously documented and have been attributed to increased incidence of adverse pregnancy complications such as postpartum haemorrhage and malpresentation [28]. Also, consistent with other reports [22, 29], women who did not have antenatal care were more likely to deliver stillbirth. Antenatal care provides the opportunity to identify and manage conditions that can adversely affect pregnancy outcomes such as hypertension, diabetes and background medical conditions. It also provides a forum for pregnant women to receive pregnancy and health-related education and counselling. This critical care was missed by about 40% of women in the current study and may have been responsible for the increased incidence of stillbirth. Additionally, the proportion of stillbirths was high among preterm deliveries, similar to the reports from other observational studies elsewhere in Nigeria and Cameroon [21, 23]. Even in high-income settings, about three-quarters of stillbirths were delivered preterm [30]. The risk of preterm birth is reduced when a woman receives high-quality antenatal care, characterised by timely and adequate interventions, as demonstrated by a previous study [31]. This highlights the need for accelerated efforts to improve the coverage and quality of antenatal care. This is crucial to achieving the “Every Newborn Action” stillbirth target of 12 or fewer late-gestation stillbirths per 1,000 total births by 2030 [32]. Women who were admitted in the first stage of labour were less likely to deliver a stillbirth. Arriving at the facility early in labour allows ample opportunity to monitor and track conditions associated with poor perinatal outcomes, enabling timely interventions. This early intervention can significantly reduce the risk of stillbirth. Although early labour presentation decreased the risk of stillbirth in the current study, it has been reported to be associated with an increased likelihood of labour interventions, cesarean sections, and neonatal admission [33]. The cesarean section rate reported in this study is comparable with the findings of a recent hospital-based nationwide survey [34]. Further studies would be required to guide policy decisions on the timing of labour admission, putting all maternal and newborn outcomes into perspective. A previous study in similar low-resource settings have documented higher rates of adverse perinatal outcomes, including low Apgar scores, early neonatal death, and fresh stillbirth, during night-shift deliveries, highlighting peculiar system-level challenges with ours [35] The observed modest reduction in stillbirth odds during the afternoon shift compared with the morning shift may reflect differences in case mix, workflow efficiency, and timing of clinical decision-making. Morning shifts often include a higher proportion of high-risk or referred cases and scheduled obstetric interventions following overnight monitoring, which may increase baseline risk. By the afternoon, clinical teams are typically fully engaged, ongoing labour complications may have been identified earlier, and workflow efficiency may be improved, contributing to better outcomes. Notably, most births in this cohort occurred during off-hours (afternoon and night), suggesting that many women were admitted and had labour managed during the preceding shift, consistent with typical labour durations of 12–18 h in primigravidae and 6–8 h in multigravidae [36]. At the study site, staffing levels are generally highest during the morning shift, potentially allowing more intensive monitoring and decision-making earlier in labour. The lack of difference in stillbirth odds between night and morning deliveries, despite the predominance of nighttime births, likely reflects an adaptive care pattern in which labour is largely monitored and key clinical decisions are made during preceding shifts. This makes the night period primarily delivery-focused and hence maintains comparable outcomes. Strengths and limitations of the study This study provides insight into the burden and trends of stillbirths in a tertiary-level health facility in North-central Nigeria, using a large dataset over five years. We also identified some pregnancy and maternal sociodemographic factors associated with stillbirth deliveries, which are crucial to the design and implementation of interventions aimed at reducing stillbirth. However, we were unable to categorise stillbirths into antepartum or intrapartum, which could have implications for targeted intervention. In addition, we could not assess the preventability of stillbirths, as the dataset did not allow us to determine whether systemic failures or delays in care contributed to these outcomes. Other important variables were also not captured, including pregnancy and labour complications, maternal education, marital status, partner’s occupation and education, place of residence, and household income. Reducing stillbirths in this setting is challenged by late presentation to care and suboptimal antenatal care utilisation. The key modifiable risk factors identified, i.e., lack of antenatal care, preterm delivery, delayed admission in labour, primiparity, and multiple gestation, highlight important entry points for intervention. Strengthening access to and uptake of quality antenatal care, with early identification of high-risk pregnancies, alongside promoting timely presentation during labour, may improve outcomes. Better integration of maternal and obstetric care across the continuum from antenatal to intrapartum care remains essential to reducing stillbirths in similar settings. Conclusion In conclusion, the intra-facility stillbirth rate is high in this cohort, and it surpasses the National average. Also, this study identified significant predictors of stillbirth. Singleton gestation was associated with a markedly lower risk of stillbirth compared to multiple gestation. Similarly, increasing parity (multiparity and grand multiparity) demonstrated a protective effect against stillbirth. Deliveries during the afternoon shift were associated with reduced odds of stillbirth, while no significant difference was observed for night shift deliveries compared to morning. Conversely, unbooked pregnancies and prematurity significantly increased the odds of stillbirth, underscoring their critical role as risk factors. Maternal age and employment status were not significant predictors of stillbirth. Recommendations To reduce stillbirths in this setting, interventions should focus on the major identified risk factors. Women with multiple gestations should receive intensified antenatal surveillance, including more frequent fetal monitoring, third-trimester ultrasonography, and specialist-led care. High-risk pregnancy clinics should also be strengthened within referral facilities. Given the increased risk among unbooked pregnancies, community-based interventions promoting early antenatal booking are essential. These should include maternal health education through community health workers and traditional/community leaders, subsidised antenatal services, and strengthened referral systems between primary and tertiary facilities. Because prematurity showed the strongest association with stillbirth, preventive and management strategies for preterm birth should be prioritised. These include early identification and treatment of maternal complications, use of antenatal corticosteroids where indicated, timely referral of high-risk pregnancies, and strengthening neonatal care services. Primiparous women should receive targeted antenatal counselling and closer intrapartum monitoring, including routine use of partographs during labour. In addition, regular perinatal death audits, continuous staff training in emergency obstetric care, and ensuring adequate staffing across all duty shifts may help reduce preventable stillbirths. Finally, prospective multicentre studies are recommended to further explore the mechanisms underlying the protective effect of higher parity and to evaluate the effectiveness of targeted interventions aimed at reducing stillbirths in high-risk populations within resource-limited settings. Supplementary Information Supplementary Material 1. Authors' contributions TOO, UAS and AI conceptualized and designed this study. TOO, TLI, SOB, KAJ and SOM collected data. ACO and BOO analyzed the data. TOO, TL, and AI interpreted the data. TOO wrote the first draft of the manuscript. All the authors reviewed and contributed to the final manuscript revisions and approved the final version. Funding This work was not supported by any funding. Data availability The data used in this study are available from the corresponding author upon reasonable request. Declarations Ethics approval and consent to participate Not applicable. Consent for publication Not applicable. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Supplementary Materials Supplementary Material 1. Data Availability Statement The data used in this study are available from the corresponding author upon reasonable request.