Timing of umbilical cord clamping and its predictors among neonates in Northeast Ethiopia: a multicenter prospective observational study Authors: Zenebe Tefera, Mandefro Assefaw, Lemlem Chanyalew Melese, Molla Kassa, Abdulkadir Kassaw, Nigusie Abebaw, Atrsaw Dessie Liyew, Mengistu Abate Belay Published: 2026 ABSTRACT BackgroundOptimal umbilical cord clamping, defined as clamping between one and three minutes after birth, is recommended by the World Health Organization to improve infant health and nutritional outcomes through placental transfusion. Despite these recommendations, early cord clamping remains prevalent in many settings, depriving neonates of up to 30% greater blood volume and increasing the risk of iron deficiency anemia., Evidence on the time to cord clamping and its predictors is limited in Ethiopia. The study aim to assess the time to umbilical cord clamping and its predictors among neonates in Northeast Ethiopia.MethodsA facility-based prospective observational study was conducted among 264 neonates from February 1 to March 2, 2025. The study participants were selected using systematic sampling technique. Data were collected via observation and structured checklists and analyzed using STATA 17. Each neonate was observed for a fixed maximum of 60 s from birth. Cord clamping within 60 s was considered the event of interest; neonates not clamped by 60 s were right-censored at that time point. Survival analysis was performed using Kaplan‒Meier estimates, and the Cox proportional hazards regression model was used to identify independent predictors of early cord clamping. Adjusted hazard ratios with 95% confidence intervals are reported, and variables with p values < 0.05 in the multivariable analysis were considered significant predictors.ResultsThe overall incidence rate of early cord clamping was 8.27 per 1000 person-seconds (95% CI: 6.88–9.94). A history of preeclampsia (AHR = 3.06, 95% CI: 1.81–5.19), cesarean delivery (AHR = 3.08, 95% CI: 1.90–5.01), episiotomy (AHR = 3.48, 95% CI: 2.01–6.07), the presence of a nuchal cord (AHR = 4.08, 95% CI: 2.09–7.96), the need for neonatal resuscitation (AHR = 2.98, 95% CI: 1.20–7.37), and a first-minute Apgar score < 7 (AHR = 2.22, 95% CI: 1.05–4.67) were independent predictors of a shorter time to cord clamping. The log-rank test revealed statistically significant differences in the number of clamps across all these predictors (p < 0.001).Conclusion and recommendationsEarly cord clamping remains highly prevalent in the study area, with more than two-fifths of neonates clamped within the first minute. Therefore, strengthening adherence to the recommendations through targeted training, institutional protocol reinforcement, and quality improvement initiatives is recommended.Clinical trial number Not applicable. Background Optimal umbilical cord clamping, defined as clamping between one and three minutes after birth or after cessation of pulsation, is recommended by the World Health Organization and the Ethiopian Federal Ministry of Health to improve maternal and infant health outcomes through placental transfusion [1–3]. This practice allows the newborn to receive up to 60% more red blood cells and 30% greater blood volume, providing essential iron stores that are particularly critical for infants in low-resource settings where access to iron-rich foods is limited [1, 4, 5]. Placental transfusion facilitated by delayed clamping has been recognized as a physiological bridge supporting the newborn’s transition to extrauterine life [6, 7]. Despite these well-established recommendations and clear biological benefits, early umbilical cord clamping (clamping within 60 s of birth) remains persistently prevalent across diverse healthcare settings globally [6, 8]. This continued practice can be traced to historical clinical traditions, as early clamping was embedded in the active management of the third stage of labor for over a century without robust physiological justification [1, 9, 10]. The consequence of this entrenched practice is that many neonates are deprived of optimal placental transfusion, potentially compromising their hematological status and neurodevelopmental outcomes [9, 11, 12]. The public health implications of early cord clamping are substantial. Iron deficiency anemia affects 68% of infants in sub-Saharan Africa and is a major contributor to child mortality, impaired cognitive development, and motor and behavioral deficits [13, 14]. Optimal umbilical cord clamping has been identified as a cost-effective intervention to prevent infant iron deficiency anemia in resource-limited settings [9, 12, 13]. Beyond its hematological benefits, evidence indicates that delayed umbilical cord clamping significantly decreases the risk of intraventricular hemorrhage, necrotizing enterocolitis, and neonatal infection, while also supporting improved neurodevelopmental outcomes at four years of age [12, 15–19]. Additionally, delaying cord clamping has been associated with higher fifth-minute Apgar scores and reduced rates of admission to the neonatal intensive care unit [20, 21]. Consequently, early cord clamping deprives neonates of these protective physiological benefits, potentially compromising both short-term stabilization and long-term health outcomes. The prevalence of early cord clamping varies considerably across studies, with rates of 46.3% reported in Brazil, 46.6% in the United States, and 52% in Nepal [6, 8, 22]. These figures indicate that despite WHO recommendations, a substantial proportion of newborns globally do not receive the full benefits of placental transfusion. Factors associated with early clamping include maternal age, residence, birth visit characteristics, obstetric complications such as preeclampsia and premature rupture of membranes, mode of delivery, need for neonatal resuscitation, and first-minute Apgar score [6, 8, 22–24]. However, these factors have been inconsistent across studies and settings, reflecting variations in clinical protocols, provider practices, and healthcare infrastructures. Although the Ethiopian Federal Ministry of Health recommends delayed cord clamping, actual adherence and its determinants remain largely unexamined in local facilities [3, 25]. Existing domestic studies have relied on cross-sectional binary classifications of clamping time, which obscure the temporal dynamics and precise time-to-event patterns of the procedure. Moreover, because provider practices are shaped by facility-specific protocols and obstetric emergencies, findings from international settings offer limited generalizability to Ethiopia. To address these gaps, we conducted a prospective time-to-event analysis to quantify the exact timing of umbilical cord clamping and identify its independent clinical predictors in public hospitals of Northeast Ethiopia. Methods Aim, design, and setting This study aims to assess the time to umbilical cord clamping and its predictors among neonates in Northeast Ethiopia. A facility-based prospective observational study design was conducted in public hospitals in the South Wollo Zone, northeast Ethiopia from February 1to March 2, 2025. South Wollo is one of the zones in the Amhara Region of Ethiopia. There are 14 public hospitals in the southern Wollo zone. These hospitals serve 2,243,013 people. Participants The source population was all neonates born at public hospitals in the southern Wollo zone. All neonates born in selected public hospitals in the South Wollo zone during the study period composed the study population. All neonates born in the selected hospitals during the study period with the exception of those that were stillborn and with congenital malformation were included. Sample size and sampling procedure The sample size was calculated using Epi Info, Version 7.2.6.0, on the basis of the following assumptions: a cumulative survival probability of 25.0% for early umbilical cord clamping among neonates born to Rh-negative mothers and 47.39% among neonates born to Rh-positive mothers, from a previous study conducted in northwestern Ethiopia [26]; 95% confidence level (Z = 1.96); 80% power; and an unexposed-to-exposed ratio of 1. Applying a design effect of 1.5 yielded a sample size of 240. After adding 10% nonresponse, the final sample size was 264. A multistage sampling technique was employed to select the study participants. The first five hospitals were selected via a simple random sampling technique, representing approximately 36% of all public hospitals in the zone. In the second stage, the sample size was proportionally allocated to each selected hospital on the basis of the average monthly delivery caseload, and a systematic random sampling method was used to select the study participants. The first participant was selected randomly, and subsequent participants were selected every 2 intervals until the required sample size was achieved in each hospital. Operational definitions Time to umbilical cord clamping was defined as the interval, measured in seconds, from the moment of complete birth (time zero, defined as the delivery of the infant’s entire body) until the application of the first clamp on the umbilical cord. The event of interest was early umbilical cord clamping, defined as clamping occurring within 60 s of birth (< 60 s). Neonates whose cords were clamped at or after 60 s (≥ 60 s) were considered censored (right-censored). All neonates were observed for a fixed maximum of 60 s from birth. If clamping occurred within this period, the exact time was recorded and the neonate was classified as having experienced the event. If clamping had not occurred by 60 s, observation was stopped and the neonate was recorded as censored at that time point. This fixed administrative censoring approach is consistent with the World Health Organization definition of early cord clamping (< 60 s) and is a standard method in time‑to‑event analysis. Process Time to umbilical cord clamping was recorded in the delivery room using a stopwatch. Following birth observation, a checklist was used to collect maternal, obstetric, neonatal, and healthcare provider characteristics. The checklist was adapted from previous studies [6, 8]. Data were collected by 5 trained nurses not working in maternity wards with close supervision. Training was given to the data collectors and supervisors, with an emphasis on accurate time measurements. A pretest was conducted using 5% of the checklist, and necessary modifications were made before the actual data collection. The collected data were checked for completeness and consistency, entered into Epi Data version 4.6, and exported to STATA version 17 for analysis. Descriptive statistics were computed, and the Kaplan‒Meier method was used to estimate the cumulative incidence of early cord clamping and to compare survival experiences between groups using the log-rank test. A Cox proportional hazards regression model with hazard ratios and 95% confidence intervals was used to identify independent predictors of time to cord clamping. Variables with a p value < 0.25 in the bivariable analysis were included in the multivariable Cox proportional hazard model to control for confounders. Variables with p values < 0.05 in the multivariable Cox proportional hazards model were considered significant predictors of early cord clamping. Results Maternal and obstetric-related characteristics A total of 264 neonates were included in the study with 100% response rate. With respect to maternal characteristics, the mean age of the mothers was 26.6 ± 4.3 years old and more than two-thirds of the mothers, 182 (68.9%), resided in urban areas. With respect to maternal blood type, 232 (78.9%) of the mothers had Rh‑positive status. A history of preeclampsia was present in 28 (10.6%) of the mothers. Premature rupture of membranes was documented in 11 (4.2%) cases, and postpartum hemorrhage occurred in 5 (1.9%) mothers. With respect to delivery characteristics, spontaneous vaginal delivery (SVD) was the most common mode of delivery, accounting for 215 (81.4%) births. Most deliveries, 173 (65.5%), took place during daytime hours. Episiotomy was performed in 26 (9.8%) of the deliveries, and perineal tears were recorded in 13 (4.9%) mothers (Table 1). Table 1Maternal and obstetric-related characteristics of the study participants in selected South Wollo public hospitals, Ethiopia, 2025VariableCategoryFrequencyPercentageAge of the mother15–247428.025–3417867.4≥ 35124.5Maternal residenceUrban18268.9Rural8231.1Maternal Rh statusNegative3212.1Positive23287.9History of preeclampsiaYes2810.6No23689.4Premature rupture of MembraneYes114.2No25395.8Postpartum hemorrhageYes51.9No25898.1Mode of deliverySVD21581.4Instrumental delivery93.4Cesarean Delivery4015.2Time of deliveryDaytime17365.5Nighttime9134.5Episiotomy was doneYes269.8No23890.2Perineal tear occurredYes134.9No25195.1 Table 1Maternal and obstetric-related characteristics of the study participants in selected South Wollo public hospitals, Ethiopia, 2025VariableCategoryFrequencyPercentageAge of the mother15–247428.025–3417867.4≥ 35124.5Maternal residenceUrban18268.9Rural8231.1Maternal Rh statusNegative3212.1Positive23287.9History of preeclampsiaYes2810.6No23689.4Premature rupture of MembraneYes114.2No25395.8Postpartum hemorrhageYes51.9No25898.1Mode of deliverySVD21581.4Instrumental delivery93.4Cesarean Delivery4015.2Time of deliveryDaytime17365.5Nighttime9134.5Episiotomy was doneYes269.8No23890.2Perineal tear occurredYes134.9No25195.1 Maternal and obstetric-related characteristics of the study participants in selected South Wollo public hospitals, Ethiopia, 2025 Neonatal and birth attendant related characteristics Among the 264 deliveries, the nuchal cord was present in 15 (5.7%) cases. With respect to fetal birth weight, most newborns (248 (93.9%)) weighed between 2500 and 3999 g. The first‑minute Apgar score was less than 7 for only 9.1% of the neonates. Resuscitation at birth was required for 15 (5.7%) infants. The condition of the amniotic fluid was clear in 231 (87.5%) deliveries. The majority of the newborns, 252 (95.5%), were delivered at term. With respect to birth attendant characteristics, the majority of deliveries were attended by midwives (217, 82.2%), followed by integrated emergency surgical officers (IESOs) (24, 9.1%) and medical doctors (MD) in training (20, 7.6%). Most of the birth attendants held a Bachelor of Science (BSc) degree (181, 68.6%), followed by a diploma (12.5%). The vast majority of the birth attendants had five or more years of working experience (250, 94.7%) (Table 2). Table 2Neonatal and birth-attendant characteristics of the study participants in selected South Wollo public hospitals, Ethiopia, 2025VariableCategoryFrequencyPercentageNuchal CordYes155.7No24994.3Fetal weight< 2500gm134.92500-3999gm24893.9≥ 4000gm31.1First minute Apgar score< 7249.1≥ 724090.9Resuscitation was requiredYes155.7No24994.3Condition of amniotic fluidClear23187.5Meconium stained3312.5Gestational age< 37 weeks124.5≥ 37 weeks25295.5Profession of the birth attendantMidwifery21782.2Nurse31.1IESO249.1Resident207.6Qualification of the birth attendantDiploma3312.5BSc18168.6MSc3011.4MD in training207.6Working experience of the birth attendant< 5 years145.3≥ 5 years25094.7 Table 2Neonatal and birth-attendant characteristics of the study participants in selected South Wollo public hospitals, Ethiopia, 2025VariableCategoryFrequencyPercentageNuchal CordYes155.7No24994.3Fetal weight< 2500gm134.92500-3999gm24893.9≥ 4000gm31.1First minute Apgar score< 7249.1≥ 724090.9Resuscitation was requiredYes155.7No24994.3Condition of amniotic fluidClear23187.5Meconium stained3312.5Gestational age< 37 weeks124.5≥ 37 weeks25295.5Profession of the birth attendantMidwifery21782.2Nurse31.1IESO249.1Resident207.6Qualification of the birth attendantDiploma3312.5BSc18168.6MSc3011.4MD in training207.6Working experience of the birth attendant< 5 years145.3≥ 5 years25094.7 Neonatal and birth-attendant characteristics of the study participants in selected South Wollo public hospitals, Ethiopia, 2025 Incidence and time to early cord clamping Among 264 neonates included in the analysis, 113 experienced early cord clamping, representing a cumulative incidence of 42.8% (95%CI: 36.8%-49.0%). The total follow‑up time was 13,666 person‑seconds, yielding an overall incidence rate of 8.27 per 1000 person‑seconds (95% CI: 6.88–9.94). The observed clamping times ranged from 5 to 60 s across the full sample, with an overall mean clamping time of 51.76 ± 13.4 s. The Kaplan‑Meier survival function estimate for early umbilical cord clamping is shown in Fig. 1. The survival curve demonstrates a gradual decline over the 60‑second observation period, reflecting the cumulative occurrence of cord clamping. At 60 s, the survival probability was 0.572 (95% CI: 0.511–0.633), indicating that 57.2% of neonates remained unclamped by the end of the observation period. The curve shows a relatively steady decline during the first 40 s, followed by a more gradual decrease thereafter, suggesting that the highest hazard of clamping occurred in the early post‑birth period. The median time to clamping could not be estimated because the survival curve did not cross the 0.50 probability threshold during the observation window. The 25th percentile of clamping time was 47 s, indicating that one‑quarter of the neonates had their cords clamped by this time(Fig. 1). Fig. 1The overall Kaplan‒Meier survival function estimate of early umbilical cord clamping among neonates born in the southern Wollo zone in northeastern Ethiopia, 2025 The overall Kaplan‒Meier survival function estimate of early umbilical cord clamping among neonates born in the southern Wollo zone in northeastern Ethiopia, 2025 Comparisons of survival status across different categorical variables Kaplan‒Meier survival curves and log-rank tests were computed to estimate and compare the timing of early cord clamping across the categorical variables. In these analysis, the event of interest was early clamping; thus, a curve situated below another signifies a shorter duration until clamping occurs (infants in the lower-curve cohort experience a lower probability of remaining unclamped over time). The log-rank test was used to assess statistical significance between survival functions. Neonates born to mothers with a history of preeclampsia experienced a markedly shorter time to early cord clamping compared to those born to normotensive mothers. Within the first 20 s post‑delivery, the probability of remaining unclamped dropped sharply to approximately 40% in the preeclampsia group, whereas the non‑preeclamptic group maintained a probability of nearly 80% at the same time point. By 60 s, the survival probability was approximately 10% among neonates of preeclamptic mothers compared to nearly 63% among their counterparts. The log‑rank test confirmed that this difference was statistically significant (p < 0.001) (Fig. 2). Fig. 2Kaplan‒Meier estimates of time to early cord clamping among neonates born to mothers with and without a history of preeclampsia (log-rank test, p < 0.001) Kaplan‒Meier estimates of time to early cord clamping among neonates born to mothers with and without a history of preeclampsia (log-rank test, p < 0.001) The timing of cord clamping varied substantially by mode of delivery. Neonates delivered via cesarean section experienced the fastest time to clamping, with the survival curve dropping precipitously within the first 20 s; by approximately 30 s, the probability of remaining unclamped had fallen below 20%. In contrast, neonates delivered by spontaneous vaginal delivery maintained a high probability of remaining unclamped throughout the observation period, with approximately 65% remaining unclamped at 60 s. Neonates delivered by instrumental delivery showed an intermediate pattern. The log‑rank test confirmed that these differences were statistically significant (p < 0.001) (Fig. 3). Fig. 3Kaplan‒Meier estimates of time to early cord clamping among neonates delivered by spontaneous vertex delivery (SVD), instrumental delivery, or Cesarean section (C/S) (log-rank test, p < 0.001) Kaplan‒Meier estimates of time to early cord clamping among neonates delivered by spontaneous vertex delivery (SVD), instrumental delivery, or Cesarean section (C/S) (log-rank test, p < 0.001) Neonates born to mothers who underwent episiotomy experienced a considerably shorter time to early cord clamping compared to those without episiotomy. Within the first 20 s post‑delivery, the probability of remaining unclamped dropped to approximately 40% in the episiotomy group, whereas the non‑episiotomy group maintained a probability of nearly 75% at the same time point. By 60 s, the survival probability was approximately 27% among neonates in the episiotomy group compared to nearly 60% among those without episiotomy. The log‑rank test confirmed that this difference was statistically significant (p < 0.001) (Fig. 4). Fig. 4Kaplan‒Meier estimates of the time to early cord clamping among neonates with and without episiotomy (log-rank test, p < 0.001) Kaplan‒Meier estimates of the time to early cord clamping among neonates with and without episiotomy (log-rank test, p < 0.001) Neonates presenting with a nuchal cord experienced a markedly faster time to early cord clamping compared to those without. Within the first 10 s post‑delivery, the probability of remaining unclamped dropped sharply to approximately 60% in the nuchal cord group, declining further to approximately 25% by 30 s. In contrast, neonates without a nuchal cord maintained a high probability of remaining unclamped throughout the observation window, with nearly 60% remaining unclamped at 60 s. The log‑rank test confirmed that this difference was statistically significant (p < 0.001) (Fig. 5). Fig. 5Kaplan‒Meier estimates of time to early cord clamping among neonates with and without a nuchal cord (log-rank test, p < 0.001) Kaplan‒Meier estimates of time to early cord clamping among neonates with and without a nuchal cord (log-rank test, p < 0.001) Neonates requiring resuscitation at birth experienced a substantially shorter time to early cord clamping compared to those who did not require resuscitation. Within the first 20 s post‑delivery, the probability of remaining unclamped dropped to approximately 40% in the resuscitation group, whereas the non‑resuscitation group maintained a probability of nearly 80% at the same time point. By 60 s, the survival probability was approximately 33% among neonates requiring resuscitation compared to nearly 58% among those who did not. The log‑rank test confirmed that this difference was statistically significant (p < 0.001) (Fig. 6). Fig. 6Kaplan‒Meier estimates of time to early cord clamping among neonates with and without need for resuscitation (log-rank test, p < 0.001) Kaplan‒Meier estimates of time to early cord clamping among neonates with and without need for resuscitation (log-rank test, p < 0.001) Neonates with a first‑minute Apgar score below 7 experienced a significantly shorter time to early cord clamping compared to those with a score of 7 or higher. Within the first 20 s post‑delivery, the probability of remaining unclamped dropped to approximately 40% in the low Apgar group, whereas the normal Apgar group maintained a probability of nearly 80% at the same time point. By 60 s, the survival probability was approximately 30% among neonates with a low Apgar score compared to nearly 60% among those with a normal score. The log‑rank test confirmed that this difference was statistically significant (p < 0.001) (Fig. 7). Fig. 7Kaplan‒Meier estimates of time to early cord clamping among neonates with first-minute APGAR scores < 7 and ≥ 7 (log-rank test, p < 0.001) Kaplan‒Meier estimates of time to early cord clamping among neonates with first-minute APGAR scores < 7 and ≥ 7 (log-rank test, p < 0.001) Predictors of early cord clamping In the bivariable Cox regression analysis, predictors with p < 0.25 (maternal Rh status, preeclampsia, premature rupture of membranes, postpartum hemorrhage, mode of delivery, time of delivery, episiotomy, nuchal cord, Apgar score, need for resuscitation, amniotic fluid status, and gestational age) were considered for multivariable analysis. After adjustment, six factors (preeclampsia, mode of delivery, episiotomy, nuchal cord, resuscitation, and first minute APGAR score) remained significant predictors of time to cord clamping. Neonates born to mothers with preeclampsia were approximately three times more likely to experience early cord clamping than those without preeclampsia (AHR = 3.06, 95% CI: 1.81–5.19). Delivery via cesarean section was associated with a threefold increased hazard relative to spontaneous vaginal delivery (AHR = 3.08, 95% CI: 1.90–5.01). Compared with no episiotomy, episiotomy was associated with a 3.48-fold greater hazard of early clamping (AHR = 3.48, 95% CI: 2.01–6.07). The presence of a nuchal cord was associated with a fourfold greater hazard of the event (AHR = 4.08, 95% CI: 2.09–7.96). Neonates with a first‑minute Apgar score < 7 were more likely to experience early clamping than those with a score ≥ 7 (AHR = 2.22, 95% CI: 1.05–4.67). Those requiring resuscitation were approximately three times more likely to experience the event (AHR = 2.98, 95% CI: 1.20–7.37) (Table 3). Table 3Bivariate and multivariate Cox regression analyses of factors associated with time to early cord clamping among neonates born in southern Wollo, Northeast Ethiopia, 2025VariableSurvival StatusCHR (95%CI)AHR (95%CI)P valueEvent (%)Censored (%) Maternal Rh status Negative19(59.4)13(40.6)1.74(1.06, 2.85)1.26(0.73, 2.17)0.404Positive94(40.5)138(59.5) 1 1 Preeclampsia Yes25(89.3)3(10.7)3.21(2.04, 5.03)3.06(1.81, 5.19) <0.001 No88(37.3)148(62.7) 1 1 Premature rapture of Membrane Yes7(63.6)4(36.4)1.85(0.86, 3.97)1.59(0.72, 3.52)0.249No106(41.9)147(58.1) 1 1 Postpartum Hemorrhage Yes4(80.0)1(20.0)2.36(0.87, 6.40)0.99(0.32, 3.01)0.986No109(42.1)150(57.9) 1 1 Mode of Delivery SVD75(34.9)140(65.1) 1 1 Instrumental5(55.6)4(44.4)1.87(0.75, 4.64)2.02(0.78, 5.22)0.147C/S33(82.5)7(17.5)3.28(2.17, 4.97)3.08(1.90, 5.01) <0.001 Time of delivery Daytime69(39.9)104(60.1)0.76(0.52, 1.11)0.70(0.47, 1.03)0.076Nighttime44(48.4)47(51.6) 1 1 Episiotomy Yes19(73.1)7(26.9)2.61(1.59, 4.28)3.48(2.01, 6.07) <0.001 No94(39.5)144(60.5) 1 1 Nuchal Cord Yes11(73.3)4(26.7)2.89(1.55, 5.38)4.08(2.09, 7.96) <0.001 No102(41.0)147(59.0) 1 1 Apgar score <717(70.8)7(29.2)2.31(1.37, 3.87)2.22(1.05, 4.67) 0.035 ≥796(40.0)144(60.0) 1 1 Need for resuscitation Yes10(66.7)5(33.3)3.27(1.71, 6.29)2.98(1.20, 7.37) 0.018 No103(41.4)146(58.6) 1 1 Amniotic fluid status Clear96(41.6)135(58.4) 1 1 Meconium Stained17(51.5)16(48.5)1.47(0.88, 2.47)1.51(0.85, 2.68)0.153 Gestational Age <37 weeks7(58.3)5(41.7)1.72(0.80, 3.71)1.52(0.64, 3.61)0.337≥37 weeks106(42.1)146(57.9) 1 1 Table 3Bivariate and multivariate Cox regression analyses of factors associated with time to early cord clamping among neonates born in southern Wollo, Northeast Ethiopia, 2025VariableSurvival StatusCHR (95%CI)AHR (95%CI)P valueEvent (%)Censored (%) Maternal Rh status Negative19(59.4)13(40.6)1.74(1.06, 2.85)1.26(0.73, 2.17)0.404Positive94(40.5)138(59.5) 1 1 Preeclampsia Yes25(89.3)3(10.7)3.21(2.04, 5.03)3.06(1.81, 5.19) <0.001 No88(37.3)148(62.7) 1 1 Premature rapture of Membrane Yes7(63.6)4(36.4)1.85(0.86, 3.97)1.59(0.72, 3.52)0.249No106(41.9)147(58.1) 1 1 Postpartum Hemorrhage Yes4(80.0)1(20.0)2.36(0.87, 6.40)0.99(0.32, 3.01)0.986No109(42.1)150(57.9) 1 1 Mode of Delivery SVD75(34.9)140(65.1) 1 1 Instrumental5(55.6)4(44.4)1.87(0.75, 4.64)2.02(0.78, 5.22)0.147C/S33(82.5)7(17.5)3.28(2.17, 4.97)3.08(1.90, 5.01) <0.001 Time of delivery Daytime69(39.9)104(60.1)0.76(0.52, 1.11)0.70(0.47, 1.03)0.076Nighttime44(48.4)47(51.6) 1 1 Episiotomy Yes19(73.1)7(26.9)2.61(1.59, 4.28)3.48(2.01, 6.07) <0.001 No94(39.5)144(60.5) 1 1 Nuchal Cord Yes11(73.3)4(26.7)2.89(1.55, 5.38)4.08(2.09, 7.96) <0.001 No102(41.0)147(59.0) 1 1 Apgar score <717(70.8)7(29.2)2.31(1.37, 3.87)2.22(1.05, 4.67) 0.035 ≥796(40.0)144(60.0) 1 1 Need for resuscitation Yes10(66.7)5(33.3)3.27(1.71, 6.29)2.98(1.20, 7.37) 0.018 No103(41.4)146(58.6) 1 1 Amniotic fluid status Clear96(41.6)135(58.4) 1 1 Meconium Stained17(51.5)16(48.5)1.47(0.88, 2.47)1.51(0.85, 2.68)0.153 Gestational Age <37 weeks7(58.3)5(41.7)1.72(0.80, 3.71)1.52(0.64, 3.61)0.337≥37 weeks106(42.1)146(57.9) 1 1 Bivariate and multivariate Cox regression analyses of factors associated with time to early cord clamping among neonates born in southern Wollo, Northeast Ethiopia, 2025 Discussion This prospective observational study, utilizing a time-to-event analytic approach, revealed that early umbilical cord clamping remains highly prevalent in public hospitals of Northeast Ethiopia, with 42.8% (95% CI: 36.8%–49.0%) of neonates clamped within 60 s of birth. The overall mean clamping time was 51.76 ± 13.4 s, and the incidence rate of early clamping was 8.27 per 1000 person-seconds (95% CI: 6.88–9.94).These findings indicates that a substantial proportion of newborns in this setting are being deprived of the physiological benefits of placental transfusion, including up to 60% more red blood cells and 30% greater blood volume, which are critical for preventing iron deficiency anemia and supporting neurodevelopmental outcomes [27, 28]. The cumulative incidence of early clamping observed in our study (42.8%) aligns closely with findings from other resource-limited and high-income settings. A cross-sectional study from Brazil reported that 46.3% of newborns underwent early cord clamping [6], while a national survey in the United States revealed that 46.6% of neonates received early clamping [29]. The consistency of these estimates across diverse geographic and economic contexts suggests that the barriers to implementing delayed umbilical cord clamping are deeply rooted in entrenched clinical traditions and real-time risk benefit assessments by birth attendants, rather than being solely determined by resource availability. This pattern is concerning given that the World Health Organization and the Ethiopian Federal Ministry of Health have long recommended delayed clamping of between one and three minutes [3, 25]. However, in Nepal, the prevalence was higher at 52% [8]. The possible explanation for the observed discrepancy may be attributable to differences in study design and population. The Nepalese research was conducted in a tertiary hospital where higher-risk deliveries are concentrated, potentially increasing the perceived urgency for immediate clamping [8]. In contrast, our multicenter design encompassing diverse facility types may have captured a broader spectrum of clinical practices and risk profiles, potentially yielding a more representative estimate for the region. The presence of a nuchal cord emerged as the strongest predictor of early clamping, with a fourfold increased hazard (AHR = 4.08, 95% CI: 2.09–7.96). This finding reflects a deeply ingrained clinical response: when the umbilical cord is found wrapped around the neonate’s neck, birth attendants often perceive an urgent need to clamp and cut the cord to facilitate delivery and prevent fetal compromise. However, evidence suggests that the nuchal cord, in the absence of fetal distress, does not necessarily require immediate clamping, as most cases can be easily reduced over the baby’s head, and even when tight, delayed clamping may still be possible with careful management [30, 31]. Neonates delivered by cesarean section were three times more likely to experience early clamping compared to those born via spontaneous vaginal delivery (AHR = 3.08, 95% CI: 1.90–5.01). This finding is consistent with a previous study conducted in Brazil [6]. The observed association is likely driven by multiple factors: the perceived urgency of completing the operative procedure, concerns about uterine incision closure and maternal blood loss, and entrenched surgical traditions of immediate cord clamping. Importantly, accumulating evidence demonstrates that delayed clamping during cesarean delivery is both feasible and safe, with no increased risk of maternal hemorrhage or other adverse outcomes [32–34]. A novel finding of this study is the association between episiotomy and early cord clamping, with episiotomy increasing the hazard by 3.48-fold (AHR = 3.48, 95% CI: 2.01–6.07). While episiotomy has not been widely investigated as a predictor of clamping timing in the literature, our finding likely reflects the clinical context in which episiotomy is performed. Episiotomy often serves as a marker for complicated or urgent deliveries, such as those involving fetal distress or operative vaginal birth, where the clinical priority shifts toward immediate neonatal assessment and resuscitation [35, 36]. However, episiotomy itself does not physiologically necessitate early clamping. Our finding suggests that providers may conflate the urgency of the delivery with the necessity of immediate cord clamping, representing an opportunity for targeted quality improvement. Neonates born to mothers with preeclampsia are were three times more likely to experience early clamping (AHR = 3.06, 95% CI: 1.81–5.19). This association may be driven by concerns about maternal hemodynamic instability, placental insufficiency, or the perception that the neonate may be compromised and require immediate assessment [37, 38]. However, current evidence does not support routine early clamping in the context of preeclampsia, and delayed clamping may actually benefit neonates by providing additional blood. Neonates requiring resuscitation were nearly three times more likely to experience early clamping (AHR = 2.98, 95% CI: 1.20–7.37). This finding was supported by previous studies conducted in Brazil and Nepal [6, 8]. The observed association underscores that when a newborn requires immediate intervention, birth attendants may prioritize moving the baby to a resuscitation corner over awaiting placental transfusion. However, current evidence suggests that effective resuscitation can be initiated with an intact umbilical cord and that delayed clamping may benefit depressed newborns by providing an additional volume of oxygenated blood and that innovative approaches such as bedside resuscitation with an intact cord have been found to be feasible, safe, and effective [39–41]. Similarly, the study revealed that neonates with a first-minute Apgar score < 7 had more than twice to the hazards of receive early cord clamping. This finding was consistent with previous findings from Brazil [1]. However, this finding contradicts a study from Nepal [15]. The observed discrepancy may be attributed to differences in the study period and sample size, as the sample size of the current study is double that of the Nepal study. However, the limitations should be acknowledged: the currents study was conducted in hospital settings, and the findings may not be generalizable to home births or health center deliveries where different practices and challenges may exist. Conclusion and recommendation This study revealed that early cord clamping remains highly prevalent in South Wollo public hospitals, with more than two-fifths of neonates clamped within the first minute. Presence of Nuchal cord, the need for resuscitation, a low Apgar score, cesarean delivery, preeclampsia, and episiotomy were significant predictors of early cord clamping. These findings highlight the gap between policy recommendations and clinical practice and identify specific targets for intervention. Therefore, addressing this gap through targeted training, institutional protocol reinforcement, and quality improvement initiatives could improve neonatal iron stores and contribute to better health outcomes in this resource-limited setting. Healthcare facilities should consider implementing bedside resuscitation with an intact cord and developing specific protocols for high-risk situations to ensure that neonates receive the full benefits of placental transfusion.