Prevalence of non-reassuring fetal heart rate patterns and associated factors among labouring mothers at public hospitals in Wolaita Zone, Southern Ethiopia, 2024: A cross-sectional study

Abstract

Background

Non-reassuring fetal heart rate patterns are critical indicators of potential adverse outcomes for the fetus during pregnancy. Early detection and identification of predictors are essential for reducing preventable neonatal mortality. Despite being one of the leading causes of neonatal death, there is limited evidence regarding the prevalence and predictors of non-reassuring fetal heart rate patterns in Ethiopia.

Objective

This study aims to assess the prevalence of non-reassuring fetal heart rate patterns and the associated factors among labouring mothers in public hospitals within the Wolaita Zone of Southern Ethiopia.

Methods

A facility-based cross-sectional study was conducted among 334 labouring mothers in public hospitals within the Wolaita Zone from 1 to 30 July 2024. Study participants were selected using a systematic random sampling technique. Data were collected through interviewer-administered questionnaires and chart reviews. The collected data were entered into EpiData version 4.6 and subsequently exported to SPSS V.26 for analysis. Bivariate and multivariable logistic regression analyses were conducted to identify factors associated with the outcome variable. In the multivariable logistic regression analysis, adjusted ORs (AORs) with 95% CIs were reported, and a p-value of <0.05 was considered statistically significant.

Results

The prevalence of non-reassuring fetal heart rate patterns was 19.2% (95% CI 15.0 to 23.9). Labour augmentation (AOR 3.28, 95% CI 1.72 to 6.27), referred from other healthcare facilities (AOR 3.39, 95% CI 1.70 to 6.76), primigravida (AOR 2.00, 95% CI 1.07 to 3.74) and meconium-stained amniotic fluid (AOR 4.52, 95% CI 2.10 to 9.75) were significantly associated with non-reassuring fetal heart rate patterns.

Conclusion

The prevalence of non-reassuring fetal heart rate patterns was found to be high. Consequently, it is essential to enhance intrapartum monitoring and prompt management of these risk factors, particularly among primigravida women and mothers transferred from other healthcare institutions to reduce the risk of adverse fetal outcomes.

STRENGTH AND LIMITATIONS.

• This study provides updated data on fetal heart rate patterns in Wolaita Zone’s public hospitals.

• Recall bias may occur if participants misremember past pregnancy and labour events.

• The study was conducted only in public hospitals and may not be fully generalisable to private health facilities.

• The cross-sectional design cannot establish causality between fetal heart rate patterns and related factors.

• Medical record data may be incomplete or contain errors, affecting data quality.

Introduction

A non-reassuring fetal heart rate (FHR) pattern (NRFHRP) is characterised by abnormal FHRPs, including irregular FHR tones (such as fetal tachycardia or bradycardia), abnormal variability (including no, minimal and/or marked variability) and repeated FHR decelerations (such as late decelerations, variable decelerations and prolonged decelerations). This condition is considered when the FHR is <110 or >160 beats per minute.^{1 2}

FHRPs are classified into two categories: baseline FHR variability, accelerations and episodic or periodic decelerations.³ The third category, known as abnormal, is characterised by sinusoidal patterns that lack variability and do not exhibit recurrent late decelerations, varied decelerations or bradycardia.⁴

To indicate probable fetal hypoxia, the term non-reassuring fetal status is preferred over the more commonly used term fetal distress, also known as prenatal hypoxia; this condition results from insufficient oxygenation of the fetus. It can only be assessed indirectly, typically through electronic fetal monitoring,⁵ which is subject to significant interobserver variability in interpretation.²

The FHR is considered non-reassuring if it falls <110 beats bpm, a condition known as bradycardia, or exceeds 160 bpm, referred to as tachycardia. This range corresponds to the typical baseline FHR of 110–160 bpm. Fetal condition should be assessed through electronic fetal monitoring on the mother’s admission to identify and classify the risk of intrapartum fetal distress.¹ Intrapartum monitoring of FHR is one of the most critical metrics for evaluating fetal health and can help reduce the risk of intrapartum fetal death or irreversible fetal injury.⁶

Practical labour ward carelessness may result from failing to recognise NRFHRPs as labour progresses and from misinterpreting cardiotocography that shows persistent and unexplained tachycardia.² When performing intrapartum fetal monitoring, the WHO advises periodically auscultating the FHR.⁴

According to the National Institute for Health and Care Excellence guidelines, intrapartum fetal monitoring facilitates the transfer of the mother from midwifery to obstetrician-led care.⁷ Furthermore, the clinical management guidelines from the American College of Obstetricians and Gynaecologists state that the effectiveness of electronic fetal monitoring during labour should be assessed based on its ability to reduce complications such as infant convulsions, cerebral palsy or intrapartum fetal death while minimising the need for unnecessary obstetric interventions, such as operative vaginal delivery or caesarean section.⁸

Intermittent auscultation techniques involve assessing fetal cardiac performance by counting the number of beats over a specific duration, typically using a Pinard fetoscope or a handheld Doppler device.⁹

Globally, the prevalence of NRFHRPs ranges from 9.9% to 30.7%.10,12 The following risk factors have been associated with NRFHRPs: antepartum haemorrhage, intrauterine growth restriction, amniotic fluid abnormalities, maternal medical conditions, induction or augmentation of labour, use of anaesthesia and the presence of meconium-stained amniotic fluid.12,14

Approximately 5 million perinatal deaths caused by avoidable stillbirths and neonatal deaths occur annually.¹⁵ The most common causes of perinatal mortality include preterm birth-related intrapartum events (24%), complications (35%) and sepsis (15%). Perinatal mortality rates are higher in cases with NRFHRPs than in those with normal FHRPs.¹⁵ In 2021, the global neonatal death rate decreased to 11 per 1000 live births. Approximately 6400 babies died each day, and 2.3 million children died within their first month of life.¹⁶

In the UK, antepartum deaths account for 5%, with a prevalence of intrapartum deaths also at 5%. In comparison, South Africa reports a 50% prevalence of antepartum fetal deaths and an 11% of intrapartum deaths. ¹⁷Intrapartum FHR monitoring is a crucial tool for assessing fetal health during childbirth.⁶

Although its false-positive rate is approximately 99%, failing to identify or address an abnormal FHR can lead to a range of serious complications and catastrophic outcomes, such as cerebral palsy, hypoxia/anoxia, partial or total brain damage, paralysis, nerve damage and stillbirth. Ultimately, the primary objective of intrapartum FHR monitoring is to detect metabolic acidosis and fetal hypoxia, thereby reducing the need for surgical intervention in cases of uncomplicated hypoxia. However, if the condition deteriorates before causing tissue damage or fetal mortality, surgical intervention may still be necessary.¹⁸

Despite initiatives aimed at enhancing the accessibility and quality of healthcare services, NRFHRPs remain the most common indicators for surgical deliveries and signs of complications requiring hospitalisation in the neonatal intensive care unit, thereby increasing costs to society.¹⁹

NRFHRPs are one of the leading causes of neonatal death and continue to pose a significant public health challenge in Ethiopia, despite the implementation of development programmes aimed at improving access to and the quality of healthcare services.²⁰ The majority of these losses are attributed to hypoxia, which can be prevented through effective foetal monitoring. Consequently, identifying NRFHRPs can alert healthcare professionals to the fetal condition, enabling timely interventions to reduce the risk of morbidity and mortality in newborns.

Despite the critical role of NRFHRPs in influencing birth outcomes, research on the prevalence of these patterns and their associated factors in Ethiopia is limited, and no studies have been specifically conducted in Southern Ethiopia. Understanding the extent of NRFHRPs and identifying maternal, fetal and institutional factors contributing to these patterns within the context of local public hospitals are essential for improving labour management and reducing neonatal morbidity and mortality. Therefore, this study aimed to assess the prevalence of NRFHRPs and their associated factors among labouring mothers in public hospitals in the Wolaita Zone of Southern Ethiopia.

Methods and materials

Study setting and period

This study was conducted in public hospitals within the Wolaita Zone of Southern Ethiopia from 1 July to 30 July 2024. The Wolaita Zone is located 394 km south of Addis Ababa, the capital city of Ethiopia, and 134 km from Hawassa, the capital of the Southern Nations, Nationalities and Peoples’ Region. The zone is bordered by Gamo Gofa to the south, Omo River to the west and Dawro to the northwest. It has a total population of 5,385,282, comprising 2,687,021 males and 2,698,261 females.

The study area has a population of 419 198. According to the zonal health department report, Wolaita Zone’s healthcare infrastructure includes 13 hospitals (9 governmental and 4 private), 69 health centres, 24 private medium clinics, 342 health posts and 1419 health workers. These facilities provide a range of services, such as normal delivery, basic obstetrical care, management of complicated delivery, caesarean section and maternal and child healthcare.

Study design

A facility-based cross-sectional study was conducted.

Source and study population

All mothers who gave birth at public hospitals in the Wolaita Zone of the Southern Ethiopia region constituted the source population. Of these, mothers who gave birth at public hospitals in the Wolaita Zone of the Southern Ethiopia region during the data collection period constituted the study population.

Inclusion criteria

Mothers in labour admitted to the designated public hospitals within the Wolaita Zone of Southern Ethiopia during the data collection period were included. These mothers had comprehensive chart documentation regarding the essential intrapartum variables of interest: FHR, cervical dilation/effacement, membrane status, amniotic fluid colour, uterine contraction frequency, oxytocin administration, analgesia/anaesthesia, maternal temperature/infection indicators, mode of delivery, newborn weight and Apgar scores.

Exclusion criteria

A mother in labour who was diagnosed with intrauterine fetal demise on admission and cases transferred from other institutions lacking complete referral documentation regarding intrapartum monitoring were excluded.

Explicit elimination criteria

These included charts exhibiting >20% missing data for any essential variables required for the classification of NRFHRPs, uninterpretable or technically invalid FHR recordings, incomplete referral cases identified post-inclusion and significant discrepancies among data sources.

Sample size determination

The sample size for the first specific objective was calculated using a single population proportion formula, based on the following assumptions: a margin of error (d) of 4%, confidence level of 95% (Zα/2 = ±1.96) and proportion of NRFHRPs (p) of 15.1%, as determined by a previous study conducted in Finote Selam, Ethiopia.¹ Using these parameters, an initial sample size of 308 participants was calculated, which was considered sufficient to reliably estimate the proportion of NRFHRPs in the population. To accommodate a potential 10% non-response rate, an additional number of participants was added to the initial estimate. This adjustment resulted in a final sample size of 339 participants, ensuring that the study maintained adequate statistical power.

Sampling procedure and technique

The sample was drawn from randomly selected public hospitals in the Wolaita Zone of Southern Ethiopia, with proportional allocation based on the average number of deliveries over the past 3 months. A systematic random sampling technique was employed. Of the nine public hospitals providing delivery services, five were selected by simple random sampling: Wolaita Sodo University Comprehensive Specialised Hospital, Tabela Primary Hospital, Badesa Primary Hospital, Bale Primary Hospital and Bitena Primary Hospital. The sample size was proportionally allocated to each hospital based on the average monthly number of deliveries: Wolaita Sodo University Comprehensive Specialised Hospital with 499; Tabela Primary Hospital, 126; Badesa Primary Hospital, 72; Bale Primary Hospital, 58; and Bitena Primary Hospital, 55. The number of study units sampled from each hospital was determined using proportional allocation, employing the formula: (nf×ni) / N, where nf is the final sample size, ni is the number of labouring mothers at each selected hospital and N is the total number of labouring mothers across the selected hospitals. Study participants were selected by systematic random sampling, with the sampling interval (K) calculated as K=N/n=810/339 ≈ 2. The lottery method was used to select the first mother, and the selection proceeded with every two labouring mothers’ intervals until the required sample size was obtained (figure 1).

Figure 1. Schematic presentation of the sampling procedure to assess the prevalence of non-reassuring fetal heart rate patterns and its associated factors in public hospitals in Wolaita Zone, Southern Ethiopia 2024. PH, primary hospital; WSUCSH, Wolaita Sodo University Comprehensive Specialised Hospital.

Data collection instruments

A structured questionnaire was adapted from relevant literature.^{1 2 12} Data were collected through chart reviews and face-to-face interviews. Chart reviews were conducted to gather information on FHR when abnormalities were detected, cervical dilation, amniotic fluid status, amniotic fluid colour, total duration of labour, fetal presentation, cord prolapse, nuchal cord, newborn weight, APGAR scores at the first and fifth minutes and adverse outcomes. Face-to-face interviews were used to collect data on the socio-demographic characteristics of participants and assess various antenatal factors.

The questionnaire was initially prepared in English, translated into Amharic and subsequently back-translated into English to ensure consistency by different individuals. Five midwives with a degree in Bachelor of Science (BSc) in Midwifery and three maternity and reproductive health nursing supervisors with a Master of Science (MSc) collected the data. Data collection was conducted over 1 month, from 1 July to 30 July 2024. The overall data collection process was supervised by the nursing supervisors and the principal investigator.

Data collectors

Data collectors were selected based on their experience in data collection. The process was carried out by five trained BSc midwives and supervised by three MSc nurses specialising in maternity and reproductive health.

Data collection procedure

Two days of intensive training were provided to both data collectors and supervisors to ensure they were well-versed in data collection methods. This training covered appropriate techniques for gathering accurate and reliable information. Data were collected during the postpartum period, particularly within 4 h after delivery. NRFHRPs were evaluated throughout labour using intermittent auscultation with a fetoscope. The diagnosis was based on identifying abnormal FHRPs during either the first or second stage of labour. On assessment, the stage of labour, cervical dilation and effacement were recorded from maternal charts. Data were collected from study participants using a pre-tested questionnaire, supplemented by information obtained from medical records. To ensure accuracy, participants’ diagnoses were confirmed by experienced physicians. The entire data collection process spanned 1 month, during which careful attention was paid to completeness and precision. This comprehensive approach helped maintain the integrity of the data collected for the study.

Dependent variable

NRFHRPs (yes/no).

Independent variables

Maternal socio-demographic factors: mothers’ age and place of residence.

Intrapartum-related factors

Gestational age, meconium-stained amniotic fluid, induction and augmentation of labour, mothers with referral paper and low birth weight.

Antepartum related factors

Antenatal care (ANC) follow-up, parity, gravidity, medical illness and history of previous adverse outcome.

Operational definitions

NRFHRPs occur when one or more of the following abnormal FHRPs are recorded more than twice during the intrapartum period: tachycardia, bradycardia, reduced fetal FHR variability or decelerations.¹

Baseline FHR status: normal between 110 and 160 bpm.²¹

Tachycardia: FHR of >160 bpm.¹⁸

Bradycardia: FHR of <110 bpm.²²

Decelerations occur when the FHR drops below the baseline, with amplitude of >15 bpm and duration of >15 s. These can be divided into variable decelerations, characterized by a rapid drop (onset to nadir in <30 s), late decelerations, characterised by delayed beginning and gradual return to baseline, and reduced variability and prolonged decelerations that continue >3 min.²³

Variability

This refers to the oscillations in the FHR signal, which are measured at 1 min intervals as the average bandwidth amplitude of the signal includes normal variability (5–25 bpm), reduced variability (< 5 bpm) and increased variability (> 25 bpm).²⁴

Gestational age

This refers to age of pregnancy in weeks, which is calculated from the last normal menstrual period, early ultrasound estimation or assessing Ballard score complemented with the medical chart of the mother.²⁰

Data quality control

To maintain data quality, the questionnaire was evaluated by researchers familiar with the study topic, and a pre-test was conducted on 5% of the sample size at Arba Minch General Hospital before the actual data collection period. This pre-test aimed to ensure that the questionnaire was clear, error-free and comprehensive. Based on the pre-test results, necessary adjustments were made, and the revised tool was used for the actual data collection. During this period, the supervisor and principal investigator closely monitored and supervised the process to ensure data quality. Daily evaluations assessed the completeness of data, and any challenges encountered were addressed promptly. Finally, all collected data were reviewed by the supervisor and principal investigator for completeness, consistency and clarity during data management.

Data processing and analysis

The collected data were cleaned, coded and entered into EpiData version 4.6. Subsequently, data were exported to SPSS V.26 for analysis. Descriptive analysis was used to characterise the participants, and the results were presented in the form of text, tables and figures. A binary logistic regression model was employed to conduct bivariate analysis, examining the relationship between each independent variable and the outcome variable. Independent variables with a p-value of <0.25 were included in the multivariable analysis to control for potential confounders and identify true associations. Multicollinearity was assessed using the variance inflation factor (VIF) to evaluate linear correlation among independent variables. None of the variables had a VIF >5. Model fit was evaluated with the Hosmer–Lemeshow goodness-of-fit test, which yielded an insignificant result (p value=0.417), indicating that the model was appropriately fitted. In the multivariable logistic regression analysis, adjusted ORs (AORs) with 95% CIs were reported, and variables with a p-value of <0.05 were considered significantly associated with NRFHRPs.

Ethical considerations

Ethical clearance was obtained from the Institutional Health Research Review Board of Arba Minch University, College of Medicine and Health Sciences (Reference number IRB/23172/2024), after confirming that there would be no harm to study participants. A cooperation letter was issued to public hospitals in the study area by the School of Nursing. Before data collection commenced, the Medical Directorates of each public hospital in the study setting provided written and signed permission. Each participant was informed about the study’s objectives and potential benefits before agreeing to participate.

Informed, voluntary, written consent was obtained from each participant before the start of data collection. Participants were also made aware of their rights to refuse participation, ask questions or withdraw at any time during the process without any repercussions. The contact information for the principal investigator and the Institutional Health Research Review Board of Arba Minch University was provided to participants in case they had any questions or concerns.

Results

Socio-demographic characteristics of mothers

A total of 334 mothers participated in the study, resulting in a response rate of 98.5%. Their mean age was 32.04 (SD ±6.3) years. The majority of respondents (63.2%) were between 26 and 35 years old. Most participants resided in urban areas (58.7%). Regarding educational attainment, 39.5% of mothers had completed primary education, whereas 35.3% had completed secondary education. A significant majority of mothers were housewives (75.1%), and 11.1% were government employees. Most of their husbands completed secondary education (71.5%), and some had a college degree or higher (20.1%) (table 1).

Table 1. Socio-demographic characteristics of mothers at public hospitals of Wolaita Zone in Southern Ethiopia, 2024 (n=334) .

Variables	Categories	Frequency (n)	Percent (%)
Age (years)	≤ 25	76	22.8
	26–35	221	63.2
	≥ 36	47	14.0
Place of residence	Urban	196	58.7
	Rural	138	41.3
Religion	Protestant	173	51.8
	Muslim	14	4.2
	Orthodox	144	43.1
	Catholic	3	0.9
Mothers’ education status	No formal education	44	13.2
	Primary	132	39.5
	Secondary	118	35.3
	College and above	40	12
Mothers’ occupation	Housewife	251	75.1
	Government employee	37	11.1
	Private work	38	11.4
	Daily labourer	8	2.4
Fathers’ education	No formal education	7	2.1
	Primary school	21	6.3
	Secondary school	239	71.5
	College or above	67	20.1

Antepartum profile

A majority (62.3%) of the mothers were multigravida. Most mothers (92.2%) delivered at term (≥37 weeks). Nearly all participants (96.7%) received ANC follow-up; however, 65.9% had between one and three visits, whereas only 34.1% attended four or more visits. More than half (52.6%) of the ANC visits took place in government hospitals. Only 6.6% of the mothers had a history of obstetric complications, including conditions such as pregnancy-induced hypertension, hyperemesis gravidarum and antepartum haemorrhage. Additionally, 11.4% reported experiencing medical illnesses during pregnancy, with urinary tract infections being the most common, followed by malaria, cardiac disease and dyspepsia (table 2

Table 2. Antepartum profiles of mothers at public hospitals of Wolaita Zone in Southern Ethiopia, 2024 (n=334).

Variables	Categories	Frequency (n)	Percent (%)
Gravidity	Primigravida	126	37.7
	Multigravida	208	62.3
GA during labour and delivery	Preterm (<37 weeks)	26	7.8
	Term (≥37 weeks)	308	92.2
ANC follow-up	Yes	323	96.7
	No	11	3.3
Number of ANC visits (n=323)	1–3	213	65.9
	≥4	110	34.1
Place of ANC follow-up (n=323)	Health centre	105	32.5
	Government hospitals	170	52.6
	Private institutions	48	14.9
History of obstetric complications	Yes	22	6.6
	No	312	93.4
Type of adverse obstetric history (n=22)	Pregnancy-induced hypertension	6	27.3
	Hyperemesis gravidarum	4	18.2
	Antepartum haemorrhage	7	31.8
	Others*	5	22.7
Medical illness during pregnancy	Yes	38	11.4
	No	296	88.6
Types of medical illness (n=38)	Urinary tract infection	16	42.1
	Malaria	8	21.0
	Cardiac disease	6	15.8
	Dyspepsia	5	13.2
	Others†	3	7.9

^*Others1: preterm labour, stllbirth and abortion

^†Others2: diabetes mellitus, hypertension and anemia

ANC, antenatal care; GA, gestational age.

Intrapartum profile

Labour augmentation was performed in 30.5% of mothers, and labour began spontaneously in most of them. On admission, 93.1% of mothers had ruptured membranes, and 16.5% exhibited meconium-stained amniotic fluid. Nearly half of the mothers (49.4%) were referred from other institutions, primarily due to NRFHRPs, prolonged labour, antepartum haemorrhage, cephalopelvic disproportion, malpresentation and premature rupture of membranes. Regarding the mode of delivery, 48.5% of mothers experienced spontaneous vaginal delivery, 39.8% underwent caesarean sections and 11.7% had instrumental deliveries (table 3).

Table 3. Intrapartum profile of mothers at public hospitals of Wolaita Zone in Southern Ethiopia, 2024 (n=334).

Variables	Categories	Frequency (n)	Percent (%)
Augmentation of labour	Yes	102	30.5
	No	232	69.5
Membrane status on admission	Intact	23	6.9
	Ruptured	311	93.1
Meconium-stained amniotic fluid	Yes	55	16.5
	No	279	83.5
Total duration of labour	≤ 12 h	288	86.3
	> 12 h	46	13.7
Onset of the labour	Spontaneous	298	89.2
	Induced	36	10.8
Referred from other institutions	Yes	165	49.4
	No	169	50.6
Mode of delivery	Spontaneous vaginal delivery	162	48.5
	Instrumental delivery	39	11.7
	Caesarean section	133	39.8

Newborn-related Factors

A total of 334 newborns were delivered, all of whom were singleton births. Among these, 59.4% were female. The majority (79.6%) presented in a cephalic position, whereas 12.6% were in the breech position. Cord prolapse occurred in 8.1% of the deliveries. 89.2% of the newborns fell within the normal range of 2.5 to 4.0 kg.

Birth injuries were reported in 11.7% of cases. The APGAR score at the first minute was <7 in 32.3% of the newborns, whereas at the fifth minute, only 2.1% had an APGAR score <7 (table 4).

Table 4. Newborn-related factors at public hospitals of Wolaita Zone in Southern Ethiopia, 2024 (n=334).

Variables	Categories	Frequency (n)	Percent (%)
Sex of the newborn	Male	139	41.6
	Female	195	59.4
Cord prolapse	Yes	27	8.1
	No	207	91.9
Nuchal cord	Yes	12	3.6
	No	322	96.4
Fetal presentation	Cephalic presentation	266	79.6
	Breech presentation	42	12.6
	Compound and transverse	26	7.8
Neonatal weight (kg)	< 2.5	31	9.3
	2.5–4.0	298	89.2
	˃ 4	5	1.5
Birth injury	Yes	39	11.7
	No	295	88.3
APGAR at the first minute	<7	108	32.3
	≥ 7	226	67.7
APGAR at fifth minute	<7	7	2.1
	≥ 7	327	97.9

APGAR, appearance, pulse, grimace, activity and respiration score.

Prevalence of non-reassuring fetal heart rate pattern

In this study, NRFHRPs were observed in 64 cases (19.2%), with a 95% CI of 15.0% to 23.9% (figure 2).

Figure 2. Prevalence of non-reassuring fetal heart rate patterns among labouring mothers at public hospitals of Wolaita Zone in Southern Ethiopia, 2024 (n=334).

Among these 64 cases, tachycardia was the most common abnormality, accounting for 56.2%. Various interventions were implemented in response to these FHR abnormalities, with the majority of cases (64.1%) resulting in caesarean section deliveries. Resuscitation efforts were successful in 31.3% of cases, whereas instrumental delivery was utilised in 4.7% (table 5).

Table 5. Abnormal fetal heart rate patterns among labouring mothers at public hospitals of Wolaita Zone in Southern Ethiopia, 2024 (n=334).

FHR abnormality	Frequency (n)	Percent (%)
Tachycardia	36	56.2
Bradycardia	10	15.7
Late decelerations	8	12.5
Variable decelerations	5	7.8
Prolonged decelerations	3	4.7
Reduced variability	2	3.1
Total	64	100

.FHR, fetal heart rate.

Factors associated with non-reassuring fetal heart rate patterns

In the bivariable logistic regression analysis, factors such as place of residence, gravidity, referral from other health facilities, medical illnesses during the current pregnancy, labour augmentation, meconium-stained amniotic fluid and fetal presentation were identified as associated with NRFHRPs (p<0.25). These factors were subsequently considered candidates for multivariable analysis. In the multivariable logistic regression, gravidity, referral from other health facilities, labour augmentation and meconium-stained amniotic fluid were found to be significantly associated with NRFHRPs at a 95% CI (p<0.05).

These findings revealed that the odds of developing NRFHRPs were three times higher among newborns from mothers who underwent labour augmentation than among newborns from mothers who did not (AOR 3.28, 95% CI 1.72 to 6.27). Additionally, the odds of developing NRFHRPs were 3.4 times higher among newborns born from mothers who were referred from other health facilities than among newborns born from mothers who came directly from home (AOR 3.39, 95% CI 1.70 to 6.76). Primigravida mothers were twice as likely to develop NRFHRPs than multigravida mothers (AOR 2.00, 95% CI 1.07 to 3.74). The odds of developing NRFHRPs were 4.5 times higher among newborns born from mothers with meconium-stained amniotic fluid than among newborns born from mothers with clear amniotic fluid (AOR 4.52, 95% CI 2.10 to 9.75) (table 6

Table 6. Factors associated with non-reassuring fetal heart rate patterns among labouring mothers at public hospitals of Wolaita Zone in Southern Ethiopia, 2024 (n=334).

Variables	NRFHRP		COR (95% CI)	AOR (95% CI)	P-value
	Yes (n)	No (n)
Gravidity
Primigravida	33	93	2.02 (1.16 to 3.51)	2.00 (1.07 to 3.74)	0.029*
Multigravida	31	177	1.00	1.00
Place of residence
Urban	30	166	1.00	1.00
Rural	34	104	1.80 (1.04 to 3.13)	1.02 (0.52 to 2.02)	0.935
Referred from other health facilities
Yes	43	122	2.48 (1.39 to 4.41)	3.39 (1.70 to 6.76)	0.001*
No	21	148	1.00	1.00
Medical illness during the current pregnancy
Yes	11	27	1.86 (0.87 to 4.00)	2.41 (0.99 to 5.85)	0.056
No	53	243	1.00	1.00
Augmentation of labour
Yes	30	72	2.42 (1.38 to 4.24)	3.28 (1.72 to 6.27)	0.001*
No	34	198	1.00	1.00
Fetal presentation
Vertex presentation	44	222	1.00	1.00
Breech presentation	10	32	1.57 (0.72 to 3.44)	1.15 (0.46 to 2.88)	0.754
Compound and transverse	10	16	3.15 (1.34 to 7.40)	1.59 (0.52 to 4.79)	0.408
Meconium-stained amniotic fluid
Yes	21	34	3.38 (1.79 to 6.38)	4.52 (2.10 to 9.75)	<0.001*
No	43	236	1.00	1.00

p-value less than 0.001 indicates a highly statistically significant result

AOR, adjusted odds ratio; COR, crude odds ratio; NRFHRP, non-reassuring fetal heart pattern.

Discussion

This study aimed to assess the prevalence of NRFHRPs and associated factors among labouring mothers in public hospitals in the Wolaita Zone of Southern Ethiopia. The findings revealed NRFHRP prevalence of 19.2% (95% CI 15.0 to 23.9). The study identified gravidity, referral from other healthcare facilities, labour augmentation and meconium-stained amniotic fluid as factors significantly associated with NRFHRPs.

The prevalence of NRFHRPs in this study was consistent with findings from studies conducted at Finote Selam Primary Hospital in Northwest Ethiopia (15.1%)¹ and the Jigjiga, Ethiopia (22.4%).¹² This consistency may be attributed to similarities in clinical practices, healthcare infrastructure and maternal risk factors across these locations. Furthermore, the use of comparable methodologies and criteria for assessing FHRPs likely contributed to these similar findings.

However, this finding was higher than those reported in studies conducted at hospitals in South India (11.5%),²⁵ Tanzania (9.9%)²⁶ and Zimbabwe Harare (11.2%),¹⁴ whereas it was lower than the result of a study conducted in Thailand (30.7%).⁵ The discrepancy may be attributed to variations in healthcare infrastructure, quality of prenatal care and criteria used to assess FHR, such as cardiotocography monitoring in developed countries. Additionally, the broader classification of NRFHRPs in other studies compared with the narrower focus on bradycardia and tachycardia in this study may also account for the differences.

The findings of this study revealed that the odds of developing NRFHRPs were three times higher among newborns from mothers who underwent labour augmentation than among those from mothers experienced spontaneous labour. This result aligns with studies conducted in hospitals of Denmark.²⁷

This increased risk may be attributed to the physiological stress induced by labour augmentation interventions, such as oxytocin administration or artificial rupture of membranes. These interventions can result in stronger and more frequent uterine contractions, which may reduce uteroplacental blood flow and oxygen delivery to the fetus, thereby increasing the likelihood of fetal distress.²¹

The odds of developing NRFHRPs were 3.4 times higher among mothers referred from other health facilities than among those who came directly from home. This finding is consistent with results from studies conducted in Finote Selam, Northwest Ethiopia,¹ and in Nigerian hospitals.²⁸ This discrepancy may be attributed to the fact that referred mothers are often more likely to have complicated or high-risk pregnancies, necessitating transfer to higher-level facilities. Such complications can include prolonged labour, pre-existing maternal conditions or fetal distress, all of which increase the likelihood of NRFHRPs. Additionally, delays in referral or suboptimal management at the initial health facility may contribute to deterioration in fetal condition, thereby increasing the risk of fetal distress on arrival.

Primigravida mothers are twice as likely to develop NRFHRPs compared with multigravida mothers. This finding is consistent with studies conducted in Finote Selam, Northwest Ethiopia,¹ Jigjiga Public Hospitals in Eastern Ethiopia,¹² Thailand⁵ and Israel.¹¹ This increased risk may be attributed to primigravida mothers, who are experiencing their first pregnancy and often have less uterine tone and limited labour experience compared with multigravida mothers, who have previously given birth. This lack of experience can lead to less effective uterine contractions and a higher incidence of labour complications. Additionally, the fetus in a primigravida pregnancy may be more likely to present in a sub-optimal position or encounter more challenges during labour. These factors can contribute to an elevated risk of NRFHRPs .²⁴

The likelihood of developing NRFHRPs was 4.5 times higher among mothers with meconium-stained amniotic fluid than among those with clear amniotic fluid. This finding aligns with results from studies conducted in Jigjiga Public Hospitals in Eastern Ethiopia.¹² This increased risk may be attributed to the association of meconium-stained amniotic fluid with fetal hypoxia, which can lead to abnormal FHRPs resulting from decreased oxygen levels. Additionally, meconium can cause umbilical cord compression, disrupting blood flow and oxygen delivery to the fetus. Furthermore, meconium may provoke an inflammatory response, increasing the risk of infection and further jeopardising fetal well-being. These combined mechanisms contribute to the high likelihood of NRFHRPs in such cases.²⁹

This study emphasises the clinical implications and scientific contributions by identifying significant predictors of NRFHRPs. These predictors include labour augmentation, referrals from other healthcare facilities, primigravidity and meconium-stained amniotic fluid. The findings highlight the importance of early identification and monitoring of at-risk pregnancies to prevent adverse fetal outcomes. Additionally, the results provide valuable evidence regarding the prevalence and risk factors associated with NRFHRPs in Ethiopia, a context where data are limited. This research can inform clinical practices and guide interventions aimed at reducing neonatal mortality in similar settings.

Limitations

This study has several limitations that should be considered when interpreting the findings. Self-reported information may be affected by recall bias. The study was conducted in only public hospitals, so the results may not be fully generalisable to private health facilities. The cross-sectional design precludes drawing causality between FHRPs and related factors. Medical record data may be incomplete or contain errors, affecting data quality.

Conclusion

The prevalence of NRFHRPs in the study area was notably high and was significantly associated with factors such as gravidity, referrals from other health facilities, labour augmentation and meconium-stained amniotic fluid. To address these issues and reduce adverse fetal outcomes, it is essential to enhance intrapartum monitoring and prompt management of these risk factors, particularly among primigravida women and mothers transferred from other healthcare institutions to reduce the risk of adverse fetal outcomes.

Recommendations

Based on the findings of this study, the following recommendations were given:

To the regional health bureau

• Improving referral systems between health facilities to ensure timely and effective care for labouring mothers.

• Providing training for healthcare providers on the management of labour augmentation.

• Enhancing the detection and management of meconium-stained amniotic fluid.

• Efforts should focus on targeted interventions for primigravida mothers to minimise risks and improve maternal and fetal outcomes.

To public hospitals in Wolaita Zone

• Implement continuous and thorough FHR monitoring, especially for primigravida mothers and those referred from other facilities, to promptly identify and manage non-reassuring patterns.

• Ensure that labour augmentation is carefully monitored and used only when clinically necessary, with clear guidelines in place to minimise potential risks to the foetus.

• Provide training for healthcare providers to effectively detect and manage cases of meconium-stained amniotic fluid, reducing the likelihood of fetal complications.

• Strengthen communication and coordination with referring health facilities to ensure timely and comprehensive care for labouring mothers, reducing delays that may affect fetal outcomes.

To future researchers

• Expand the study to other regions or include private healthcare settings to assess whether the same risk factors are prevalent, enhancing the generalisability of the findings.

• Conduct longitudinal studies to better understand the causal relationships between the identified factors and NRFHRPs .

Not applicable

Data availability statement

Data are available upon reasonable request.