A 10-year retrospective assessment of prevalence and factors associated with jaundice among neonates in selected facilities in Ondo State

Abstract

Background

Globally, jaundice affects about 6 in 10 term babies and 8 in 10 preterm newborns in their first week of life; West Africa, has the highest cases of neonatal jaundice, and it remains the leading cause of severe illnesses such as mental handicap, brain damage, physical disabilities and even early deaths among newborns in the region.

Methods

A hospital-based retrospective study design was conducted from perinatal data collected over 10 years in tertiary hospitals in Ondo State, Nigeria. A structured data extraction form was used to collect retrospective data from records of neonates and their respective mothers from the selected health facilities from 2015 to 2024. Statistical analysis was performed using SPSS version 23. The prevalence of jaundice was presented using a line graph. Participants’ socio-demographic and obstetric characteristics were assessed using frequency and percentage. Chi-square analysis was used to ascertain the relationship between the incidence of jaundice and socio-demographic and obstetric characteristics. P-value was set at 5% for significance.

Results

Among the 10,182 mother-neonates pairs analyzed, 8% were preterm and 92% were full-term neonates. The prevalence of jaundice among preterm and high-risk full-term between the year 2015–2024 were 37.4% and 5.7% respectively. Among pre-term neonates, jaundice was significantly more prevalent in babies aged 0–4 days, likewise among high-risk full-term neonates with 90.1% of jaundiced cases falling into this group.

Conclusion

Neonatal jaundice exhibits a pronounced disparity in Ondo State, with preterms dramatically more affected than high-risk full-terms. Low birth weight, prematurity, and structural disadvantages like low maternal education, unemployment, and rural residence were consistent risk amplifiers.

Introduction

Neonatal jaundice is a globally recognized, common, and worrisome condition among newborns, which causes the skin and the white part of the newborn’s eyes to turn yellow. This occurs when there is too much bilirubin in the baby’s body, and if not treated early, it can lead to more serious problems like preventable brain damage, hospital admission, mental handicap, and, in the worst scenario, even death [1, 2]. In Africa, especially West Africa, neonatal jaundice is one of the main causes of sickness and death among newborns [1–4]. Globally, jaundice affects about 6 in 10 term babies and 8 in 10 preterm newborns in their first week of life. (REF) Annually, about 1.1 million babies usually develop severe forms of jaundice, also known as severe hyperbilirubinemia with majority of these babies found in Sub-Saharan Africa and South Asia [2, 5]. In 2018, neonatal jaundice was found to be the seventh leading cause of mortality among babies during their first six days after birth and it also accounts for about 70% of morbidity and 10% of mortality among newborns [1, 2, 5, 6]. Also, over 20 million babies are at risk of hyperbilirubinemia-related complications, which is a condition that causes jaundice and, in some severe cases, leads to the death of about 114,000 newborns annually.

Studies have revealed that Sub-Saharan Africa (SSA), specifically West Africa, has the highest cases of neonatal jaundice, and it has remain the leading cause of complications such as mental handicap, brain damage, physical disabilities and even early deaths among newborns in the region, with about 667 per 10,000 live births being affected by severe neonatal jaundice [1, 4, 7]. Furthermore, it has been revealed that neonatal jaundice is influenced by several factors, especially in low and middle-income countries like Nigeria. Factors such as prolonged labour, male neonates, and maternal blood group “O” have been linked to a higher risk of jaundice [6]. Additionally, other factors such as the incompatibility between mother and baby’s blood group, inherited blood condition such as glucose-6-phosphate dehydrogenase (G6PD) deficiency, delayed cord clamping, exclusive breast feeding without early monitoring, premature birth, babies with low birth weight, primiparity, deliveries outside public hospitals, neonatal infection, and poor maternal knowledge about recognizing jaundice symptoms are additional risk factors that has been associated with jaundice occurrence [2, 8, 9].

Over the past 16 years, studies have been conducted in Nigeria’s tertiary facilities on jaundice among babies, some of which revealed that its prevalence is approximately 60% in the country and around 15% in the south-west [9].

However, it is important to note that all these studies tend to focus on either the country as a whole, the South-West region, the Northern regions, or are narrowed down to the community levels, with little or no specific attention given to neonates (the first 28 days after birth) in Ondo State.

Nevertheless, the Nigerian Government, in partnership with various organizations, has engaged in several initiatives to help reduce the prevalence of neonatal jaundice in the country. One of such initiatives is the Project Oscar (Light For Life), which took place on December 3rd, 2024, and was aimed at revolutionizing neonatal jaundice care by educating mothers, screening infants, training healthcare workers, and equipping facilities with necessary tools [10].

Therefore, this study aims to enhance the existing body of knowledge by thoroughly examining the prevalence of neonatal jaundice and its associated risk factors among selected facilities in Ondo State, Nigeria.

Methods and materials

Study design and setting

A hospital-based retrospective study design was conducted from perinatal data collected over 10 years in tertiary hospitals in Ondo State, Nigeria. Ondo State is one of the states in the southwestern region of Nigeria, and it has 3 senatorial districts (Ondo North, Ondo Central, and Ondo South senatorial districts) with 18 local government areas. Its state capital is Akure, which is 300 km southwest of Abuja (Nigeria’s capital city), 200 km north of Lagos, and 100 km east of Ibadan (Nigeria’s largest city). Its population is approximately 4.5 million, with a population density of approximately 230 people per square kilometer. Based on the availability of adequate equipment for the treatment of jaundice, the study was conducted in three purposively selected tertiary hospitals from the state, namely the Federal Medical Centre, Owo, University of Medical Sciences Teaching Hospital Complex, Ondo State, and University of Medical Sciences Teaching Hospital Complex, Akure.

Population

The study population were neonates 0-28days (preterm and full-term) and their respective mothers in the selected health facilities. The inclusion criteria for the selection were: Neonates who were born or admitted in the selected health facilities between January 2015 and December 2024, and Neonates whose mothers’ records were available, and Neonates who had information on neonatal jaundice in the hospital records. Neonates who had other serious health conditions and were critically ill were excluded from the study.

Sample size determination and sampling procedures

We purposively selected Ondo state in southwest Nigeria because neonatal jaundice affects neonates in the state, and it often leads to complications, thereby leading to the death of the neonates [11]. Three hospitals which served as refferal centres for all cases of jaundice and swith adequate equipment for the treatment of jaundice were purposively selected from the state, which included 3 tertiary health facilities: Federal Medical Centre, Owo, University of Medical Sciences Teaching Hospital Complex, Ondo state, and University of Medical Sciences Teaching Hospital Complex, Akure. Medical records of neonates with jaundice admitted to the selected hospitals from January 2015 to December 2024 were retrieved and examined from the health facilities’ records (Tables 1 and 2).

Table 1.

Neonates with completed data born and admitted in the hospital between 2015–2024

Year	Number of neonates		Total
	Male	Female
2015	421	353	774
2016	404	444	848
2017	386	319	705
2018	383	288	671
2019	782	581	1363
2020	728	587	1315
2021	599	471	1070
2022	523	409	932
2023	612	497	1109
2024	760	635	1395
Total	5598	4584	10,182

Table 2.

Distribution of neonates per facility

Year	Federal medical centre, owo		Unimedth, akure		Unimedth, ondo		Total
	Male	Female	Male	Female	Male	Female	Male	Female
2015	220	164	116	102	85	87	421	353
2016	208	221	112	156	84	67	404	444
2017	222	164	95	77	69	78	386	319
2018	225	164	90	89	68	35	383	288
2019	287	246	290	171	205	164	782	581
2020	253	198	276	291	199	98	728	587
2021	260	166	201	180	138	125	599	471
2022	237	179	178	120	108	110	523	409
2023	267	186	149	159	196	152	612	497
2024	294	234	291	209	175	192	760	635
Total	2473	1922	1798	1554	1327	1108	5598	4584

The data from the three selected hospitals were merged and used for the study. Data quality was ensured through systematic cleaning and validation. Records with incomplete or inconsistent data on neonatal jaundice status and key predictors—gestational age, sex, maternal education, parity, residence, and ABO/Rh incompatibility—were excluded. The final analysis included 10,182 neonates with complete data. The dataset was checked for accuracy, consistency, and outliers prior to analysis.

Data collection procedures

A structured data extraction form was used to collect retrospective data from records of neonates and their respective mothers from the selected health facilities for 10 years, from 2015 to 2024. The data extraction form was developed by reviewing previous similar studies that consist of all the variables that can be used to achieve the objective of the study. During the study period, neonatal admittance cards from the selected hospitals were separated, tallied, and their numbers were sorted by admission order. Under the guidance of the primary investigator, two nurses and one resident carefully gathered the required data. Information was collected from the records of the mother-infant dyad, which includes their sociodemographic data and obstetric characteristics, including access to health care facility, parity, and gravidity, rh incompatibility, and ABO incompatibility.

Operational definition

Neonatal jaundice

Infants who are diagnosed as jaundiced by doctors (general practitioners, pediatricians, and neonatologists) based on their medical history, clinical signs and symptoms, and/or laboratory tests [12].

ABO incompatibility

An antibody reaction that happens when a mother and child have different blood groups usually the mother’s blood group is O and the baby’s blood group is A, B or AB [12].

Rh incompatibility

A condition in which a pregnant Rh-negative mother has a Rh-positive fetus, which triggers the production of antibodies by her immune system that target the baby’s red blood cells and may result in Rh illness [13].

High risk full-term

High-risk full-term neonate for jaundice’ refers as any infant born at ≥ 37 weeks gestation presenting with at least one major or moderate risk factor for severe hyperbilirubinemia, including early-onset jaundice (< 24 h), hemolytic disease (e.g., ABO/Rh incompatibility), G6PD deficiency, significant bruising, suboptimal breastfeeding, or presence of neurotoxicity risk factors such as sepsis or hypoxia.

Outcome measure

The diagnosis and classification of neonatal jaundice were based on clinician assessment (general practitioners, pediatricians, and neonatologists) supported by laboratory bilirubin measurements recorded in hospital records [14].

Neonates were categorized as pre-term if they were less than 37 weeks at birth and full term if they were 37 weeks and above [15].

Data management and analysis

Data were exported to SPSS version 23 for analysis after being entered into an Excel worksheet. The prevalence of jaundice was presented using a line graph. Participants’ socio-demographic and obstetric characteristics were assessed using frequency and percentage. Chi-square analysis was used to ascertain the relationship between the incidence of jaundice and socio-demographic and obstetric characteristics. P-value was set at 5% for significance.

Figure 1 shows the prevalence of jaundice among preterm and high-risk full-term between the year 2015–2024. About 37.4% of the participants had jaundice in preterm while only 5.7% had only jaundice among high-risk full term.

Fig. 1.

a Prevalence of Jaundice among Pre-term and High-risk Full term between 2015–2024. b Prevalence of Jaundice among neonates (10182) between 2015–2024

Figure 1b shows the prevalence of jaundice among neonates between 2015 and 2024. Less than 10% (8.3%) of the respondents had jaundice while 91.7% of them are not jaundiced.

Figure 2 shows the prevalence of jaundice among pre-term and high-risk full-term between the year 2015–2024.

Fig. 2.

Trend of jaundice between 2015–2024

Among preterm neonates, the prevalence of jaundice demonstrated marked fluctuations over time. It increased sharply from 16.1% in 2015 to 57.9% in 2016, remained relatively high between 2016 and 2018 (54.9%–57.4%), and then declined to 34.2% in 2019 and 22.1% in 2020. A notable peak was observed in 2021 (62.0%), followed by a decline to 18.6% in 2022, a slight increase to 28.6% in 2023, and a decrease again to 18.9% in 2024. Overall, the pattern suggests substantial year-to-year variability without a consistent upward or downward trend.

In contrast, among full-term neonates, the prevalence remained consistently low throughout the study period, with relatively minor fluctuations. It decreased from 19.6% in 2015 to 9.1% in 2016, reached lower levels between 2017 and 2020 (4.2%–9.8%), and dropped to 0.5% in 2021. A slight increase was observed thereafter, with values ranging from 3.0% to 5.7% between 2022 and 2024. Overall, the trend among full-term neonates indicates a generally low prevalence with modest fluctuations over time.

Association between socio-demographic characteristics and prevalence of neonatal jaundice

Between 2015 and 2024, among pre-term neonates (n = 819), the prevalence of jaundice was 306 (37.4%). Age of the neonate was significantly associated with jaundice (χ²=4.954, p < 0.001), with a higher proportion of jaundiced neonates observed among those aged 5–12 days (46.4%) compared to non-jaundiced neonates, who were predominantly aged 0–4 days (99.2%). Neonates sex was also significantly associated with jaundice prevalence (χ²=34.370, p < 0.001), with male neonates (53.6%) constituting a greater proportion of jaundiced cases compared to females.

Birth weight showed a strong association with jaundice (χ²=195.510, p < 0.001). Jaundice was more common among neonates with birth weight < 2500 g (67.9% combined for < 1500 g and 1500–2499 g), whereas non-jaundiced neonates were predominantly within the 2500–3999 g category (63.4%). Maternal age was significantly associated with neonatal jaundice (χ²=18.168, p < 0.001), with most cases occurring among mothers aged 27–36 years (57.5%). Ethnicity was significantly associated (χ²=14.198, p = 0.003), with a higher proportion of Hausa ethnicity among jaundiced neonates relative to non-jaundiced neonates. Marital status showed a significant association (χ²=10.040, p = 0.002), with a higher proportion of jaundice among neonates born to married mothers (92.5%). Maternal education was significantly associated with jaundice (χ²=220.277, p < 0.001). Jaundiced neonates were more frequently born to mothers with primary or secondary education, whereas non-jaundiced neonates were more likely to have mothers with tertiary education (54.0%). Religion was also associated (χ²=15.617, p < 0.001), with a relatively higher proportion of jaundice among neonates born to Muslim mothers (22.5%) compared to non-jaundiced neonates. Maternal employment status was significantly associated (χ²=8.573, p = 0.003), with a higher proportion of unemployed mothers among jaundiced neonates. Similarly, paternal unemployment was markedly higher among jaundiced neonates (27.1%) compared to non-jaundiced neonates (2.3%) (χ²=114.826, p < 0.001). Place of residence also showed a strong association (χ²=163.811, p < 0.001), with jaundice more prevalent among neonates from rural areas (62.1%), whereas non-jaundiced neonates were predominantly from urban areas (81.9%). Lastly, family type was significantly associated (χ²=14.208, p < 0.001), with a higher proportion of jaundice among neonates from polygamous families (27.1%) compared to non-jaundiced neonates.

Among high-risk full-term neonates, neonatal age was significantly associated with jaundice (χ²=249.795, p < 0.001). Jaundiced neonates were more frequently aged 0–4 days (68.2%) and 5–12 days (29.7%), while non-jaundiced neonates were predominantly aged 0–4 days (90.1%).

Birth weight showed a strong association (χ²=293.516, p < 0.001). Jaundice was more common among neonates with birth weight < 2500 g, particularly 1500–2499 g (39.2%), while non-jaundiced neonates were predominantly within the 2500–3999 g category (71.1%). Maternal age was significantly associated (χ²=216.777, p < 0.001), with a higher proportion of jaundiced neonates born to mothers aged 37–47 years (11.7%) compared to non-jaundiced neonates (0.7%). Ethnicity was also significantly associated (χ²=16.675, p < 0.001), with relatively higher proportions of Igbo and Hausa ethnicities among jaundiced neonates. Marital status showed a significant association (χ²=20.446, p < 0.001), with a slightly higher proportion of jaundice among neonates born to single mothers (7.2%) compared to non-jaundiced neonates.

Maternal education demonstrated a strong association (χ²≈713.725, p < 0.001). Jaundiced neonates were more commonly associated with lower educational levels, while non-jaundiced neonates were predominantly born to mothers with tertiary education (54.1%). Religion was significantly associated (χ²=29.214, p < 0.001), with a higher proportion of jaundice among neonates born to Muslim mothers (20.6%). Maternal and paternal employment statuses were both significantly associated (χ²=208.258 and χ²=142.742 respectively; p < 0.001), with higher proportions of unemployment among parents of jaundiced neonates. Place of residence showed a strong association (χ²=302.569, p < 0.001), with jaundice more common among neonates from rural areas (48.7%), whereas non-jaundiced neonates were predominantly urban. Household characteristics were significantly associated (χ²=914.940, p < 0.001), with a notably higher proportion of jaundice among neonates not living with both parents (38.1%). Family type was also significantly associated (χ²=307.237, p < 0.001), with jaundice more common among neonates from polygamous families (29.7%) Table 3.

Table 3.

Association between maternal and neonatal socio-demographic characteristics and the prevalence of jaundice among pre-term neonates between 2015 to 2024

Variable	Pre-term		Statistics	Full term		Statistics
	Prevalence of Jaundice			Prevalence of Jaundice
	Present (n = 306)	Not present (n = 513)
Foetal risk
Age of the baby
0–4 days	160(52.3)	509(99.2)	x2=4.954	367(68.2)	7952(90.1)	x2=249.795
5–12 days	142(46.4)	1(0.2)	P < 0.001	160(29.7)	563(6.4)	P < 0.001
13–26 days	4(1.3)	3(0.6)		11(2.0)	310(3.5)
Gender
Male	164(53.6)	167(32.6)	x2=34.370	301(55.9)	4966(56.3)	x2=0.022
Female	142(46.4)	346(67.4)	P < 0.001	237(44.1)	3859(43.7)	P = 0.883
Birth weight
< 1500	102(33.3)	57(11.1)	x2=195.510	40(7.4)	167(1.9)	x2=293.516
1500–2499	106(34.6)	128(25.0)	P < 0.001	211(39.2)	2197(24.9)	P < 0.001
2500–3999	62(20.3)	325(63.4)		246(45.7)	6276(71.1)
> 4000	16(5.2)	3(0.6)		13(2.4)	185(2.1)
Not recorded	20(6.5)	0(0.0)		28(5.2)	0(0.0)
Mother’s characteristics
Mother’s Age
17–26 years	114(37.3)	199(38.8)	x2=18.168	138(25.7)	3195(36.2)	x2=216.777
27–36 years	176(57.5)	311(60.6)	P < 0.001	337(62.6)	5572(63.1)	P < 0.001
37–46 years	16(5.2)	3(0.6)		63(11.7)	58(0.7)
Ethnicity
Yoruba	256(83.7)	434(84.6)	x2=14.198	433(80.5)	8038(91.1)	x2=16.675
Igbo	36(11.8)	74(14.4)	P = 0.003	73(13.6)	650(7.4)	P < 0.001
Hausa	10(3.3)	5(1.0)		20(3.7)	137(1.6)
Others	4(1.3)	0(0.0)		12(2.2)	0(0.0)
Marital status
Single	23(7.5)	77(15.0)	x2=10.040	39(7.2)	300(3.4)	x2=20.446
Married	283(92.5)	436(85.0)	P = 0.002	499(92.8)	8525(96.6)	P < 0.001
Level of education
No formal education	7(2.3)	3(0.6)	x2=220.277	20(3.7)	105(1.2)	x2713.725
Primary	118.6)	11(2.1)	P < 0.001	185(34.4)	316(3.6)	P < 0.001
Secondary	109(35.6)	222(43.3)		270(50.2)	3628(41.1)
Tertiary	72(23.5)	277(54.0)		63(11.7)	4776(54.1)
Religion
Christianity	237(77.5)	451(87.9)	x2=15.617	427(79.4)	7724(87.5)	x2=29.214
Islam	69(22.5)	62(12.1)	P < 0.001	111(20.6)	1101(12.5)	P < 0.001
Mother’s Employment status
Unemployed	82(26.80	93(18.1)	x2=8.573	146(27.1)	735(8.3)	x2=208.258
Employed	224(73.2)	420(81.9)	P = 0.003	392(72.9)	8090(91.7)	P < 0.001
Father’s employment status
Unemployed	83(27.1)	12(2.3)	x2=114.826	128(23.8)	741(8.4)	x2=142.742
Employed	223(72.9)	501(97.7)	P < 0.001	410(76.2)	8084(91.6)	P < 0.001
Place of residence
Urban	116(37.9)	420(81.9)	x2=163.811	276(51.3)	7240(82.0)	x2=302.569
Rural	190(62.1)	93(18.1)	P < 0.001	262(48.7)	1585(18.0)	P < 0.001
Household characteristics
Living with spouse	224(73.2)	378(73.7)	x2=0.023	333(61.9)	841(95.3)	x2=914.940
Not living with spouse	82(26.8)	135(26.3)	P = 0.880	205(38.1)	41(4.7)	P < 0.001
Type of family
Monogamous	223(72.9)	430(83.8)	x2=14.208	378(70.3)	8154(92.4)	x2=307.237
Polygamous	83(27.1)	83(16.2)	P < 0.001	160(29.7)	671(7.6)	P < 0.001

Association between obstetric characteristics and prevalence of jaundice

Table 4 shows association between obstetric characteristics and prevalence of jaundice. For pre-term neonates, access to a healthcare facility was significantly associated with jaundice (χ²=172.490, p < 0.001). A lower proportion of jaundiced neonates had access to healthcare (55.9%) compared to non-jaundiced neonates (94.2%), while lack of access was more common among jaundiced cases (44.1% vs. 5.8%). Distance to healthcare facility was also significantly associated (χ²=49.863, p < 0.001), with a higher proportion of jaundiced neonates residing far from healthcare facilities (38.9%) compared to non-jaundiced neonates (16.6%). Parity was strongly associated with jaundice (χ²=131.569, p < 0.001). A higher proportion of jaundiced neonates were born to multiparous mothers (62.7%), whereas non-jaundiced neonates were predominantly born to primiparous mothers (77.4%).

Table 4.

Association between Obstetric factors and the prevalence of jaundice among pre-terms

Variable	Preterm		Statistics	Full term		Statistics
	Prevalence of Jaundice			Prevalence of Jaundice
	Present	Not Present
Access to healthcare facility
Yes	171(55.9)	483(94.2)	x2=172.490	370(68.8)	8374(94.9)	x2=560.188
No	135(44.1)	30(5.8)	P < 0.001	168(31.2)	451(5.1)	P < 0.001
Distance to healthcare facility
Not far	187(61.1)	428(83.4)	x2=49.863	332(61.7)	7415(84.0)	x2=176.782
Far	119(38.9)	85(16.6)	P < 0.001	206(38.3)	1410(16.0)	P < 0.001
Parity
Primiparous	114(37.3)	397(77.4)	x2=131.569	79(14.7)	3540(40.1)	x2=138.286
Multiparous	192(62.7)	116(22.6)	P < 0.001	459(85.3)	5285(59.9)	P < 0.001
Place of delivery
Home	18(5.9)	0(0.0)	x2=	24(4.5)	749(8.5)	x2=10.853
Health facility	288(94.1)	513(100.0)	P < 0.001	514(95.5)	8076(91.5)	P < 0.001
Type of delivery
Spontaneous vaginal delivery	239(78.1)	398(77.6)	x2=0.030	354(65.8)	7427(84.2)	x2=120.423
Caesarean section	67(21.9)	115(22.4)	P = 0.862	184(34.2)	1398(15.8)	P < 0.001
ABO Blood incompatibility
Yes	80(26.1)	0(0.0)	x2=148.636	106(19.7)	0(0.0)	x2=1758.665
No	226(73.9)	513(100.0)	P < 0.001	432(80.3)	8825(100.0)	P < 0.001
Rh incompatibility
Yes	63(20.6)	0(0.0)	x2=151.179	64(11.9)	0(0.0)	x2=1057.039
No	243(79.4)	513(100.0)	P < 0.001	474(88.1)	8825(100.0)	P < 0.001

Place of delivery was significantly associated (p < 0.001), with all non-jaundiced neonates delivered in health facilities, while a small proportion of jaundiced neonates were delivered at home (5.9%). Type of delivery was not significantly associated with jaundice (χ²=0.030, p = 0.862), with similar distributions between spontaneous vaginal delivery and caesarean section among both groups. ABO blood incompatibility was significantly associated with jaundice (χ²=148.636, p < 0.001), observed in 26.1% of jaundiced neonates and none of the non-jaundiced neonates. Similarly, Rh incompatibility showed a significant association (χ²=151.179, p < 0.001), present in 20.6% of jaundiced neonates and absent among non-jaundiced neonates.

Among high-risk full-term neonates, access to healthcare was significantly associated (χ²=560.188, p < 0.001). A smaller proportion of jaundiced neonates had access to healthcare (68.8%) compared to non-jaundiced neonates (94.9%), while lack of access was more common among jaundiced neonates (31.2% vs. 5.1%). Distance to healthcare facility was also significant (χ²=176.782, p < 0.001), with a higher proportion of jaundiced neonates living far from healthcare facilities (38.3%) compared to non-jaundiced neonates (16.0%). Parity showed a strong association (χ²=138.286, p < 0.001), with a higher proportion of jaundiced neonates born to multiparous mothers (85.3%), compared to non-jaundiced neonates (59.9%). Place of delivery was significantly associated (χ²=10.853, p < 0.001), with a lower proportion of jaundiced neonates delivered in health facilities (95.5%) compared to non-jaundiced neonates (91.5%), and a slightly higher proportion delivered at home. Type of delivery was significantly associated (χ²=120.423, p < 0.001). Caesarean section was more common among jaundiced neonates (34.2%) compared to non-jaundiced neonates (15.8%), while spontaneous vaginal delivery was less common.

ABO incompatibility was significantly associated with jaundice (χ²=1758.665, p < 0.001), present in 19.7% of jaundiced neonates and absent among non-jaundiced neonates. Rh incompatibility also showed a significant association (χ²=1057.039, p < 0.001), observed in 11.9% of jaundiced neonates and none of the non-jaundiced group.

Predictors of Jaundice among preterm and high-risk full-term neonates

Table 5 shows the predictors of jaundice among preterm and high-risk full- term neonates between 2015 and 2024. For preterm neonates, male sex was a strong independent predictor of jaundice (OR = 7.97, p < 0.001). Birth weight of 2500–3999 g was protective (OR = 0.09, p = 0.040), while lower weight categories were not significant. Younger maternal age (17–36 years) markedly reduced odds (OR ≈ 0.002, p < 0.001). Lower maternal education levels were associated with increased risk, particularly primary education (OR = 109.76, p < 0.001). Urban residence (OR = 0.055, p < 0.001), access to healthcare (OR = 0.198, p = 0.002), and shorter distance to facilities (OR = 0.062, p < 0.001) were protective. Several years (2016–2019, 2023) showed significantly higher odds compared to 2024, while employment status was not significant.

Table 5.

Predictors of jaundice among the preterm neonates between 2015–2024

Variable	Preterm				Variable	Full term
	Odds ratio	p-value	95% confidence interval			Odds ratio	p-value	95% confidence interval
			Lower	Upper				Lower	Upper
Sex					Birth weight
Male	7.970	< 0.001	3.941	16.119	< 1500	< 0.001	47.100	18.249	121.562
Female					1500–2499	< 0.001	8.688	3.898	19.366
Birth weight					2500–3999	0.001	3.549	1.640	7.681
< 1500	0.758	0.813	0.077	7.491	> 4000
1500–2499	0.276	0.265	0.029	2.656	Mother’s age
2500–3999	0.091	0.040	0.009	0.895	17–26 years	< 0.001	0.022	0.012	0.039
> 4000					27–36 years	< 0.001	0.036	0.021	0.063
Mother’s age					37–46 years
17–26 years	0.002	< 0.001	0.000	0.031	Level of education
27–36 years	0.002	< 0.001	0.000	0.029	No formal education	< 0.001	35.758	17.115	74.708
37–46 years					Primary	< 0.001	49.742	33.566	73.715
Level of education					Secondary	< 0.001	8.596	6.168	11.979
No formal education	12.725	0.019	1.511	107.168	Tertiary
Primary	109.757	< 0.001	29.486	408.559	Employment status
Secondary	3.653	< 0.001	1.692	7.888	Unemployed	< 0.001	1.956	1.444	2.649
Tertiary					Employed
Employment status					Place of residence
Unemployed	1.015	0.969	0.452	2.283	Urban	0.378	< 0.001	0.293	0.486
Employed					Rural
Place of residence					Access to healthcare facility
Urban	0.055	< 0.001	0.028	0.108	Yes	< 0.001	0.064	0.045	0.091
Rural					No
Access to healthcare facility					Distance to health facility
Yes	0.198	0.002	0.071	0.553	Far	0.936	0.653	0.700	1.251
No					Not far
Distance to health facility					Year
Far	0.062	< 0.001	0.017	0.223	2015	< 0.001	15.404	9.852	24.085
Not far					2016	< 0.001	3.356	2.107	5.346
Year					2017	0.059	1.714	0.980	2.998
2015	0.075	0.020	0.008	0.668	2018	< 0.001	4.795	2.955	7.779
2016	23.976	0.004	2.817	204.080	2019	0.385	1.231	0.770	1.969
2017	17.234	0.011	1.946	152.673	2020	0.522	0.845	0.505	1.414
2018	22.483	0.005	2.625	192.609	2021	< 0.001	0.096	0.028	0.325
2019	8.882	0.046	1.040	75.857	2022	0.146	1.455	0.878	2.410
2020	4.390	0.195	0.469	41.050	2023	0.644	0.882	0.516	1.505
2021	7.408	0.065	0.881	62.307	2024
2022	4.170	0.230	0.406	42.806
2023	9.566	0.042	1.088	84.091
2024

However, for full term neonates, low birth weight was the strongest predictor, with a dose–response relationship (< 1500 g: OR = 47.10; 1500–2499 g: OR = 8.69; 2500–3999 g: OR = 3.55; all p ≤ 0.001). Younger maternal age remained protective (OR = 0.022–0.036, p < 0.001). Lower education levels were strongly associated with increased odds, especially primary education (OR = 49.74, p < 0.001). Unemployment modestly increased risk (OR = 1.96, p < 0.001). Access to healthcare was highly protective (OR = 0.064, p < 0.001), whereas distance to facility and residence were not significant. Temporal variation was observed, with increased odds in 2015, 2016, and 2018, and reduced odds in 2021.

Discussion of findings

This study assessed the prevalence and risk factors associated with the incidence of jaundice among neonates in selected facilities in Ondo State, Nigeria. A hospital-based retrospective research design was employed to gather information and opinions, aiming to achieve a deeper understanding and insights into the incidence of jaundice, its prevalence, and associated risk factors among neonates in Ondo State, Nigeria.

The sociodemographic and obstetric profile of the study population revealed notable disparities between preterm and high-risk full-term neonates, aligning with established regional patterns. The study observed that among the neonates observed 8% of them were preterm which falls within the 4.7–16.9% range reported in Nigerian hospital-based studies and is comparable to the sub-Saharan African average of approximately 9.9% [16, 17]. The markedly higher prevalence of low birth weight among preterm neonates, particularly those under 1500 g, shows the combined influence of reduced gestational age and possible intrauterine growth. Although prematurity alone substantially contributes to lower birth weight, some cases may also involve imbalanced fetal growth, as reported in Nigerian neonatal studies [18, 19]. An unexpected observation was the female predominance among preterm births (59.6% vs. 43.7% in full-term). While some literature suggests a male disadvantage in neonatal survival, sex distribution in preterm births has been inconsistent across studies, and this divergence may be due to sampling variability or context-specific survival patterns [20, 21]. Preterm births were more common among rural residents, unemployed mothers, those with lower education, single marital status, and polygamous households. These findings align with those of Yahaya et al., who identified low education, rural residence, and inadequate antenatal access as modifiable risk factors for adverse neonatal outcomes in Nigeria [22]. Notably, a higher proportion of preterm births was observed among rural residents compared to their urban counterparts, a finding not similar to findings from other studies where higher proportion was observed in urban areas [23, 24], possibly due to differences in transport infrastructure and healthcare distribution in Ondo State compared to more urbanized regions. Finally, the higher rates of ABO (9.8%) and Rh (7.7%) incompatibility among preterm neonates compared to high-risk full-term births (1.1% and 0.7%, respectively) highlight a dual cause: these incompatibilities can lead to fetal distress resulting in earlier delivery and also predispose newborns to hemolysis and subsequent hyperbilirubinemia [25]. The numbers here are higher than the reports of Kolawole et al. [26] where they reported a rate of 5.3%, but lower than the reports of Bujandric & Grujic [27], Osuorah et al. [28] which recorded a rate of 38% and 17.6% ABO incompatibility, respectively, which could be due to better detection from routine antenatal blood typing and early newborn screening in the participating facilities.

The study assessed the prevalence of jaundice among the neonates between the year 2015 and 2024. The study showed that the prevalence of jaundice was substantially higher (37.4%) among preterm neonates compared to high-risk full-term neonates (5.7%), reaffirming the strong epidemiological association between prematurity and neonatal hyperbilirubinemia. This magnitude is consistent with reports from Nigerian tertiary hospitals, where jaundice prevalence among preterm neonates ranges from 35% to over 50% [29, 30], and with broader sub-Saharan African data showing that preterm neonates have a 3–5 times higher risk of developing clinically significant jaundice [31]. The lower rate among high-risk full-term neonates aligns with international norms, where physiological jaundice remains common, but severe forms are less frequent in healthy term neonates [32]. The temporal trends between 2015 and 2024 provide further insight. The sustained predominance of jaundice among preterms mirrors findings by Olusanya et al., who reported a persistent high burden despite modest declines in term cases, attributing this to the disproportionate survival of vulnerable preterm neonates in Nigerian facilities without corresponding improvements in jaundice prevention and management [33]. The surges in prevalence observed in 2016, 2021, and 2024 in this study may reflect episodic stressors on the health system, such as resource shortages, industrial actions, or infectious disease outbreaks, that disrupt continuity of neonatal care. Notably, the marked dip in 2020 could plausibly be linked to reduced hospital admissions during the COVID-19 lockdowns, as seen in other Nigerian facilities where maternal presentation rates declined sharply [34, 35]. By contrast, the steady decline in jaundice prevalence among high-risk full-term neonates over the study period is encouraging and parallels patterns observed in parts of Asia and Latin America following improvements in maternal education, increased facility delivery rates, and early newborn follow-up [36]. The difference in trends between preterm and term neonates may indicate that while preventive measures, such as early feeding initiation and risk-based screening, are reaching healthier term newborns, similar interventions are not yet sufficiently adapted for preterm-specific vulnerabilities, such as immature liver conjugation pathways and higher rates of hemolysis [37].

The study also assessed the association between maternal and neonatal socio-demographic characteristics and the prevalence of jaundice among pre-term neonates between 2015 and 2024 and discovered variations in statistical significance across the neonates and maternal characteristics and prevalence of jaundice. For preterm neonates, the result from the study showed that neonates age is significantly associated with jaundice and predominant among neonates aged 5–12 days, which aligns with the clinical knowledge of neonatal jaundice where the levels of bilirubin typically rise between days 3–7 after birth. This pattern likely shows physiological jaundice, which is driven by immature liver function and increased breakdown of red blood cell [38, 39]. The sex of the neonates also significantly associated with jaundice prevalence with the males having a higher proportion. This susceptibility experienced by the male neonates have been observed in several studies and have been linked up to genetic and hormonal differences affecting the metabolism of bilirubin and some even attached this susceptibility to Glucose-6 phopshate dehydrogenase deficiency[26, 29, 40]. Birth weight was also found significant and it was found common among neonates with low birth weight. Low birth weight is often associated with prematurity, which is a major risk factor because of the immature functioning of the liver which bring about reduction in the conjugation of bilirubin [36, 37, 41]. Several maternal socio-demographic factors are significantly associated with neonatal jaundice. Notably, maternal age was significantly associated with cases occurring more among women aged 27–36 years, which showcases a higher birth rates in the age group rather than a direct biological effect, thus, maternal age may serve as a distributional factor, and not necessarily a causal risk factor. Similarly, ethnicity was significantly associated with jaundice prevalence with a higher proportion among Hausa neonates. This may be explained by genetic predispositions, for instance, higher prevalence of conditions like G6PD deficiency) or cultural/healthcare access differences [30, 42]. More so, the marital status of mothers showed an association with neonates born to married mothers having a higher proportion of jaundice, and this pattern can be associated with the fact that married women generally account for the majority of births than marital status serving as a biological determinant. Also, maternal education showed a strong association with jaundice prevalence, and the proportion higher among neonates born to mothers with primary or secondary education. This pattern shows that lower maternal education may increase risk indirectly through delayed care-seeking and poor neonatal care practices. Religion also indicated a significant association with the prevalence of jaundice, with a higher proportion observed among neonates of Muslim mothers, which may likely reflect underlying socio-cultural differences or differences in access to healthcare. Maternal employment status also significantly associated with neonatal jaundice indicating that lower socioeconomic status has the potential to limit the access to quality antenatal care, early detection, and timely management of neonatal conditions [11, 43]. Further, the place of residence is significantly associated with the prevalence of jaundice, with a higher proportion of prevalence in neonates in rural areas, reflecting the inadequacy of healthcare facilities and delayed care-seeking behavior in rural areas. Additionally, family type is also associated with the prevalence of jaundice, with a higher proportion from polygamous families, which may showcase an instance of the dilution of resources and reduction in maternal attention or utilization of healthcare [33, 43, 44].

Among high-risk full-term neonates, the study found that neonatal age showed a substantial correlation with jaundice in high-risk full-term newborns; greater proportions were seen in the early neonatal period (0–12 days), highlighting the crucial window of bilirubin buildup. Jaundice was more common in low birth weight neonates (< 2500 g), suggesting physiological immaturity as a major risk factor, and birth weight was substantially correlated. Neonates born to older mothers had higher rates of maternal age, indicating potential obstetric or metabolic impacts. Similar correlations were found between marital status and ethnicity, suggesting underlying socio-cultural and support-related disparities [36, 37, 41]. Maternal education showed a strong association, with higher jaundice prevalence associated with lower educational attainment, perhaps as a result of differences in health literacy and care use. Significant correlations were found between parental employment position and religion, with greater percentages among Muslim moms and unemployed parents, suggesting socioeconomic factors. Living in a rural area indicated a disadvantage, which is consistent with having little access to newborn care services. Neonates from polygamous households and those who did not live with both parents had a greater prevalence of jaundice, indicating the impact of the caregiving environment and resource distribution on newborn outcomes, and more so, household structure and family type were also significant [33, 43, 44].

The study also identified the association between obstetric factors and the prevalence of jaundice among preterms and high-risk full-terms. The study showed that among pre-term neonates, access to a healthcare facility was significantly associated with jaundice, with a lower proportion of jaundiced neonates having access and a higher proportion lacking access. This suggests that limited healthcare access may delay early detection and management of hyperbilirubinemia, thereby increasing risk. Distance to healthcare facility further reinforces this, as a higher proportion of jaundiced neonates resided far from facilities, indicating that geographic barriers may impede timely intervention. Parity was also strongly associated, with multiparity more common among jaundiced cases, possibly reflecting cumulative maternal physiological stress or reduced per-child care attention.

Place of delivery was significantly associated, as all non-jaundiced neonates were delivered in health facilities, while some jaundiced neonates were delivered at home, highlighting the protective role of skilled birth attendance and immediate neonatal assessment. However, type of delivery showed no significant association, indicating that mode of delivery may not independently influence jaundice occurrence in pre-term neonates. In contrast, ABO and Rh incompatibility were strongly associated with jaundice, consistent with established pathophysiology where hemolytic processes increase bilirubin production beyond neonatal clearance capacity [8, 28, 45, 46] More so, among high-risk full-term neonates, access to healthcare and distance to facilities remained significantly associated, mirroring the pattern observed in pre-term neonates and emphasizing systemic barriers to care. Parity also showed a strong association, with multiparity more prevalent among jaundiced neonates, suggesting similar maternal and caregiving dynamics [47, 48]. Place of delivery was significant, with slightly lower facility-based delivery among jaundiced neonates, again indicating the importance of institutional care. Unlike pre-term neonates, the type of delivery was significantly associated with higher caesarean section rates among jaundiced neonates, which may reflect underlying obstetric complications predisposing to neonatal morbidity [36, 49]. Finally, ABO and Rh incompatibility showed extremely strong associations with jaundice among high-risk full-term neonates, with notable presence among jaundiced cases and absence among non-jaundiced neonates. This finding aligns with the biological mechanism of immune-mediated hemolysis, reinforcing these factors as critical clinical predictors of neonatal jaundice [50, 51].

The study also assessed the predictors of jaundice among preterm and full-term neonates between 2015 and 2024. Male sex was found to be a significant independent predictor of jaundice in preterm newborns, suggesting a potential biological vulnerability associated with sex-specific variations in hepatic enzyme activity and bilirubin metabolism. While lower weight categories were not statistically significant in this model, the protective impact of birth weight between 2500 and 3999 g indicates that relative physiological maturity improves bilirubin conjugation and elimination. In comparison to older mothers, younger mothers (17–36 years old) had significantly decreased odds of jaundice, which is probably due to their greater physiological resilience and lower obstetric risk [36, 37, 41]. The risks of newborn jaundice were significantly increased by lower levels of maternal education, especially primary school, according to a clear gradient effect. This result highlights the importance of early detection of neonatal danger indicators, prenatal care use, and health literacy. Environmental and structural factors were also important; being in an urban area, having access to healthcare, and being closer to facilities were all protective, highlighting the significance of early intervention and prompt access to newborn treatments. Changes in healthcare quality, policy implementation, or external systemic pressures may be the cause of temporal variations, with some years exhibiting higher odds than 2024. Parental job status, on the other hand, did not show a significant independent effect, indicating that other socioeconomic or access-related factors may mitigate its impact [33, 43, 44].

For full-term neonates, low birth weight was found to be the most potent predictor of jaundice in full-term neonates, with a clear dose-response relationship that biologically represents increasing hepatic immaturity and decreased bilirubin conjugation capacity as weight declines. Younger maternal age continued to be highly protective, perhaps because to improved mother physiological capacity and decreased obstetric risk. Lower levels of maternal education, especially primary school, significantly increased the likelihood, indicating the importance of health literacy, care-seeking behavior, and neonatal monitoring practices. Risk was slightly elevated by unemployment, suggesting underlying socioeconomic barriers that might limit access to prompt medical attention and wholesome food. The importance of early detection and management of hyperbilirubinemia was reinforced by the high protective effect of access to healthcare; however, the lack of significance for domicile and distance implies that service availability may be more important in this group than geographic location. The influence of evolving healthcare practices, regulatory contexts, or systemic issues across time is further implied by observed temporal fluctuations.

Strengths and limitations of the study

This study’s core strength lies in its exceptionally large sample (10,182 pairs) spanning a full decade, granting exceptional power and temporal granularity. The bifurcation between preterm and high-risk full-term neonates deepens interpretative clarity, enabling discerning analysis of physiological versus haemolytic contributions to jaundice. However, the retrospective design creates inherent limitations, including potential data gaps. Facility-based sampling may overrepresent severe cases and under-represent community births. The study was also limited to tertiary hospitals which limits generalizability Lastly, comparing findings across regions is constrained by a paucity of analogous data from Ondo State, though national studies offer partial benchmarks.

Recommendations

The findings of this study highlight an urgent need for a multi-pronged approach to addressing neonatal jaundice in Ondo State, encompassing health system strengthening, community engagement, capacity building, and improved surveillance systems. The high prevalence of neonatal jaundice observed calls for robust and standardized screening and management protocols to be implemented across healthcare facilities in the state. Universal bilirubin screening within the first 24–48 h of life, combined with prompt initiation of phototherapy or exchange transfusion when indicated, should become routine practice. To support this, the Ministry of Health should ensure the availability of functional phototherapy units and encourage the adoption of diagnostic tools such as transcutaneous bilirubin meters to enhance the accuracy of early detection.

Conclusion

Neonatal jaundice exhibits a pronounced disparity in Ondo State, with preterms dramatically more affected than high-risk full-terms. Low birth weight, prematurity, and structural disadvantages like low maternal education, unemployment, and rural residence were consistent risk amplifiers. Among high-risk full-terms, hemolytic etiologies (ABO and Rh incompatibility) emerged prominently, while preterm jaundice was largely driven by physiological immaturity. The temporal trends suggest evolving healthcare dynamics, including encouraging declines in term jaundice and concerning surges among preterms in recent years. These findings highlight the imperative for complex, dual-pathway interventions that strengthen both physiological care for preterms and hemolytic screening and resilience in healthcare delivery systems.

Abbreviations

SPSS

Statistical package for the social sciences

G6PD

Glucose-6-phosphate dehydrogenase

ABO

Presence of antibodies and antigens in red blood cells / Classification of human blood into four main types

Rh

Rhesus factor

Authors’ contributions

A.O and G.B.O conceptualize the study, wrote the main manuscript text . All authors reviewed the manuscript.”

Funding

The research was self-funded. Hence did not receive funding from any organization.

Grant Information: NIL.

Data availability

The data set for this research is available upon reasonable request from the authors.

Declarations

Ethics approval and consent to participate

This study was performed in accordance with the Declaration of Helsinki and was approved by the Health Research Ethical Committee of Federal Medical Centre, Owo, (FMCOWO/ HREC/2025/14), the Ondo State Health Research Ethics Committee (0SHREC/31/01/2025/799) and University of Medical Sciences Teaching Hospital (UNIMEDTH/ERC/025/016) because this study was a retrospective study with no direct contact with human beings or human tissue samples and the data used in the study were medical record data, the need for informed consent from individual patients was waived by the Ethics Committee of the of the three tertiary health facilities used. No confidential information was involved in this research.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.