Risk Factors for Neonatal Deaths: An Audit of Neonatal Care: A Retrospective Cross-Sectional Study

Antenatal care for women and care during labour are of utmost importance in identifying the risks for mothers and neonates for prompt interventions to improve health outcomes. The objectives were to characterize the obstetric, antenatal and neonatal care indicators of pregnant women at birth and to calculate and estimate the risk factors for neonatal mortality. A cross-sectional retrospective study was undertaken at a health facility targeting all deliveries and neonates for two years. The study variables were summarized using descriptive summary measures. Cross-table analysis of independent and outcome variables was undertaken using chi-square (X²) and p-values. Significant exposure variables were entered into logistic regression to identify predictors. The results were expressed with adjusted odds ratios (OR) with corresponding two-sided 95% confidence intervals and associated p-values. A total of 1561 women gave birth to 1541 live-born neonates. Most of the neonatal care indicators were good (>90%) except for confirming HIV status for HIV exposed neonates (88.7%) and tuberculosis vaccination (69%). The birth outcomes of all 699 HIV infected pregnant women were recorded. Of them, 3 had neonatal deaths and 4 had stillbirths in this group. Among those (692) HIV exposed live-born neonates, 92.4% had PCR tests undertaken, and 95.9% of them had their results available in the birth register, of which 1.8% were PCR positive at birth. Considering all HIV exposed 692 (excluding stillbirths and NNDs) livebirths, the PCR results of the HIV exposed neonates were known to 88.8%. This means 11.2% HIV-exposed live-born neonates HIV status remained unknown at birth. Among those 681 received ART at birth HIV exposed live-born neonates, 91.7% had a record of receiving ART prophylaxis. This means nearly 1 in 10 HIV-exposed children who qualify for HIV prophylaxis at birth did not receive it. The neonatal death rate was 1.3% (7.5 per 1000 live births). Neonates born before 32 weeks of gestation had an OR of 15.9 (95% CI: 2.3-106.2, p = 0.004) and women who did not initiate antenatal care had an OR of 7.0 (95% CI: 1.1-42.3, p = 0.033). The compliance of neonatal care indicators from this facility was commendable. We emphasise the need to strengthen primary health care systems for tuberculosis and HIV-related care at the time of delivery to avoid missed opportunities for maternal and neonatal care.

Introduction

According to the Sustainable Development Goals (SDG) of the United Nations (UN), South Africa (SA) has identified improving the standard of maternal and newborn or neonatal health as a critical strategy and priority [1], [2]. Care given to women during pregnancy, labour or delivery (intrapartum), and the post-partum phase is thought to be of utmost importance in identifying the risks to both the mother and her unborn child or neonate and ensuring prompt interventions for improved health outcomes [3]. Studies from different countries show that maternal and neonatal health (MNH) does not adhere to the optimal standards [1], [4]. Thus, it is becoming more widely acknowledged that enhancing the standard of care provided in healthcare institutions is a key priority in the fight to eradicate preventable maternal and neonatal mortality and morbidity [4]. Significantly, more neonates are admitted to hospital neonatal intensive care units (NICU), and these admissions are associated with higher rates of morbidity and mortality because of intrapartum complications, prematurity, and low birth weight (LBW) births. According to estimates, 15 million newborns worldwide are delivered prematurely with LBWs [5]. Previous reports have identified the quality of care during pregnancy and care provided to neonates during and immediately after delivery is largely affected by the availability and implementation of standard management guidelines at primary level of care facilities [6], [7]. About 10% of new-borns in general are thought to need some help for breathing when they are born, while less than 1% needs intensive care [8].

According to the most recent SA Demographic and Health Survey (SADHS 2019), 94% of pregnant women receive antenatal care (ANC), 96% of women receive intrapartum care from experienced birth attendants, and 93% of infants receive all of their scheduled vaccinations. This indicates that access to skilled healthcare is improving in SA [8]. The perinatal mortality rate (PMR) remains high (28/1000 live births), even though these measures of healthcare facility utilization rates are at a satisfactory level and excellent [9]. These higher rates of PMN may be a result of the subpar antenatal and delivery care that pregnant women and newborns are receiving. Birth asphyxia, preterm births, and congenital abnormalities are the main recognized causes of neonatal and infant mortality in SA. About 31% of neonates in the LBW group (under 2.5 kg) are found to be born with intra-partum asphyxia alone [10], [11]. The SA National Department of Health (NDOH) has created and implemented the MPN Health Policy to reduce the neonatal death (NND) rate of 50% by 2030 in accordance with the UN’s SDGs. This policy aims to prevent avoidable NND in SA [12], [13].

The high incidence of HIV infections in mothers and children in SA is one of the major issues facing pregnant women and young children. Numerous studies have revealed a higher incidence of perinatal mortality in HIV-infected neonates who were delivered prematurely, with low birth weights, and with intrauterine growth retardation [14], [15]. In SA, newborns are now screened for HIV infection using the polymerase chain reaction (PCR), and all newborns (of HIV exposed mothers) are given preventative care as well as immediate treatment with anti-retroviral treatment (ART) if their PCR result is positive [16]. The maternal and neonatal care at primary health care (PHC) facilities in SA start with a midwife as the initial caregiver who should be able to manage ANC and childbirth of low-risk pregnancies, and to be able to identify complications and emergencies those require resuscitation and referrals (due to maternal and or neonatal causes) to hospital [10]–[18].

In order to speed up immunization, the World Health Organization (WHO) advises neonatal vaccination against tuberculosis, hepatitis B, and polio [19]. BCG and the oral polio vaccine (OPV) are administered in SA at birth. A single dose of the live tuberculosis vaccine (BCG) given at birth has been shown to offer lifetime protection [20]. Additionally, each neonate receives a single intramuscular dose of vitamin K (0.5 mg for birthweights under 1500 g or 1.0 mg for birthweights over 1500 g) after birth in order to avoid haemorrhagic disease [21]. The degrees of compliance with neonatal care and vaccination in SA, however, have not been well-documented. The majority of studies from SA focus on the antenatal and delivery care coverage at healthcare facilities, and there are few reports on the specific regular neonatal care activities. Hence, this study aimed to quantify the extent of the compliance with the neonatal care at the study facility and thus would expect to contribute to the understanding of the problem of non-compliance in a midwife obstetric unit (MOU) delivering neonates in SA. The objectives are to characterize the obstetric, ANC, and neonatal care indicators of pregnant women at birth, as well as to calculate and estimate the risk factors for neonatal deaths.

Methods and Materials

Study Design

The “birth register” from the labour ward was used to gather data from January 2020 to December 2021 for a facility-based cross-sectional retrospective study.

Study Setting, Population and Sampling

The research was carried out at Kwadabeka Community Health Centre (KCHC), a public health facility in the SA province of KwaZulu-Natal (KZN) that offers maternity and neonatal care. The facility is located in black peri-urban communities in the eThekwini district’s northern and western borders. More than 150 000 predominantly black people who live primarily in informal housing make up the facility’s catchment area. They are largely regarded as being in poverty and sharing strong traditional ties with the rural populations of the provinces of KZN and Eastern Cape. The community’s residents rely heavily on the state healthcare system. For low-risk pregnancies, ANC is provided at 21 community sites and 3 additional PHC fixed clinics including KCHC under the same administration. Pregnant women from those centres are then directed to KCHC MOU for delivery care. According to KCHC’s unpublished data, there are over 210 000 headcounts per year and, on average, 5000 maternity-related instances. About 1500 pregnant women initiate antenatal care each year and 850 vaginal deliveries at KCHC.

According to SA’s national health policy and priority, the KCHC offers the complete range of PHC services [22]. A minimum of two trained advanced midwives are always on duty to provide maternity services. These services are also supported by a medical officer and other administrative personnel. According to the national guideline and protocol, ANC and delivery services are provided at KCHC [23].

Inclusion and Exclusion Criteria

All women who gave births and all neonates who were born alive irrespective of birth weight during the study period were included in the study.

Care During ANC and Delivery for Mothers and Neonates

In SA, it is the normal practice to perform standard tests for anaemia, syphilis, HIV, and Rhesus (Rh) blood grouping are undertaken at different ANC visits. Voluntary counselling and testing for HIV are offered to all pregnant mothers for inclusion in the universal ART programme (lifelong ART regardless of gestation, CD4 count, or clinical stage) on the same day of the test if there are no contraindications (such as tuberculosis, hepatitis etc.) on acceptance. For women starting ART for the first time, the recommended regimen is a fixed-dose combination (FDC) of Tenofevoir (TDF), Lamivudine (3TC), and Dolutegravir (DTG). Women with newly confirmed or known HIV positivity who are not on ART are offered one FDC with one dose of Nevirapine (NVP) right away. After ruling out any ART contraindication, lifelong ART is commenced.

When a pregnant woman attends with labour pain at KCHC, an examination and assessment are undertaken to diagnose or classify any obstetric and foetal risks. Women without any apparent risk or imminent complications of pregnancy or delivery are allowed to continue to deliver at KCHC. A partogram (a chart with all maternal and foetal observations, fluid intake and output and medications) is used to monitor the progress of labour. Alert and action lines on the partogram, together with other observations (e.g., temperature of mother, blood pressure (BP), foetal heart rate, etc.), are used to identify labour complications of mothers and the foetuses during labour. Maternal complications or risk factors (e.g., raised BP of mothers, eclampsia, foetal distress, etc.) and for the foetus that can’t be managed at the MOU are referred to the hospital using an ambulance of emergency medical rescue services. For those women who deliver at the MOU without any problem are observed (mothers and neonates) for 6 to 8 hours after delivery. All observation and examination findings during and after delivery, with the demographic and obstetric information, are recorded in the birth register. Any complications identified in mothers and or neonates are also referred to hospitals. Mothers and neonates without any complications are observed for 6 hours in KCHC and then discharged home after proper counselling for maternal post-natal and neonatal care, vaccinations, breastfeeding, family planning, etc. At KCHC, only vaginal deliveries are performed without the use of any tools like vacuums or forceps. This MOU does not make use of labour augmentation.

To detect HIV infection in utero, a birth PCR test on all HIV-exposed neonates is undertaken. All HIV-exposed neonates also receive a minimum of six weeks post-exposure prophylaxis with NVP. All high-risk neonates who are opted for breastfeeding should receive additional AZT for the first six weeks of life and should receive NVP for a minimum of 12 weeks. NVP should only be stopped when the breastfeeding mother has a VL of less than 1000 c/ml, or until four weeks after the mother stops breastfeeding. All high-risk infants who are exclusively formula fed should receive AZT and NVP for 6 weeks. The other neonatal interventions are oral polio vaccine, BCG and intramuscular dose of vitamin K as per the Maternity and Neonatal Care Guidelines. BCG vaccination is avoided if a baby is TB-exposed and will be receiving TB prophylaxis [16]. All HIV positive neonates should be initiated on ART irrespective of birth weight [23].

Data Collection and Data Source

From January to May 2022, information was gathered from the delivery unit’s “birth register” and entered into Microsoft 365 (Microsoft, USA). The register included a summary of the demographic, ANC, labour and delivery, and postpartum treatment for mothers and neonates that were recorded immediately after childbirth. The demographic and obstetric variables of the study samples, such as age, gravidity, initiation and number of ANC visits, gestational age (GA), HIV and syphilis status at birth, Rhesus factor (Rh) for blood grouping, were collected from the birth register. Selected indicators measured for neonatal care were: recorded time of delivery, gender, APGAR scores assigned in 1 and 5 minutes, birth weight, complications during delivery, resuscitation, referrals to hospital, record of polio and BCG vaccinations, injection of vitamin K, polymerase chain reaction (PCR) test for live birth neonates to HIV infected mothers, HIV prophylaxis to PCR tested negative neonates and ART for PCR positive neonates. Pregnancy outcomes were measured with neonatal death (NND) and stillbirths.

Data Analysis

Data were coded and analysed using the Statistical Package for Social Science (SPSS 22.0) for Windows version. Age was categorized into <20 (teenage), 20–24, 25–29, 30–34 and >35 years; gravidity into 1, 2–3, 4–5 and >6 (multi-gravida), HIV and syphilis status of mothers were expressed as positive or negative (reactive or non-reactive). Gestational age (GA) was determined by a woman’s reported last menstrual period (LMP) or by measuring the fundal height in centimetres (cm) in cases where the LMP was unclear or uncertain. GA at delivery was categorised into: <32 weeks (extreme preterm), 32–36 weeks (preterm) and >37 weeks (term) pregnancy. The APGAR scores for neonates were categorised as <7 and >8 at 1 and 5 minutes. Recording of neonatal complications, resuscitation and hospital referral rates were also calculated for live birth neonates. Neonatal death was assigned as yes (1) and no (0) as a dichotomous variable. The study variables were summarized using descriptive summary measures: expressed as mean and standard deviation (SD) for continuous variables and percent for categorical variables. Cross-table analysis of independent (exposure) and outcome variables was undertaken using chi-square (X²) and p-values. Exposure variables that had significant differences in the outcome variable (using p-values < 0.05) were entered into a step-by-step (backward) logistic regression analysis to identify predictors for the outcome variable. For regression models, the results were expressed as an effect with an adjusted odds ratio (OR) for binary outcomes with corresponding two-sided 95% confidence intervals (95% CI), and associated p-values. Variables had p-values < 0.05 were considered significant predictors.

Definition of Terms

Low birth weight was defined as the birth of a neonate with a weight of less than 2500 g, irrespective of gestational age.

Neonatal death (NND) was defined (for this study) as the death of a live-born neonate within 6 hours of delivery at the facility, irrespective of duration of pregnancy.

The neonatal and stillbirth rates were calculated using the total number of NNDs and stillbirths per 1000 live births, respectively.

The study further calculated other outcome variables e.g., mother to child transmission of HIV (MTCT), neonatal complications and referral rates to hospital.

Ethical Considerations

The UMgungundlovu Health Ethics Review Board has given ethical permission (Reference no. UHERB 015/2020). Permission was sought from the KZN provincial health Research Committee and the management of KCHC.

Results

A total of 1561 women delivered 1541 live-born singleton neonates and 20 stillbirths during the study period. The mean age with standard deviation (SD) of the mothers was 25.8 (5.6) years, with the minimum and maximum ages of 14 and 43 years, respectively. The descriptive statistics (frequency and percent) of the baseline variables are shown in Table I. There were missing values for variables measured; thus, actual numbers (denominators) of each variable is shown as n values. Teenage (age < 20 years) pregnancy constituted 14.9%. Ages between 20 to 34 years were the most (72.2%). Nearly a third (31.1%) of them were having their first baby (primi-gravida). Only a few (0.5%) were grand multi-gravidity. The rate of not having ANC was 5.1%. Over half (56.3%) of them had first ANC initiation before 20 weeks of gestation. However, only a quarter (25.0%) of pregnant women received 8 or more ANC visits. Only a few (1.5%) women had Rh blood group negative. The estimates of HIV and syphilis infections among these women were 43.8% (699/1561) and 2.1% (32/1557), respectively. The preterm and LBW delivery rates were 14.8% and 9.3%, respectively.

Table I. Frequency Distribution of Demographic, Obstetric and Antenatal Care of the Sample and the Results of Cross Table Analysis with X² and p Values for Exposure and Outcome Variables

Variables	Actual sample frequency	Actual sample percentages	No. of neonatal deaths	% Neonatal death rates	X2-values	P-values
Age distribution (n = 1560)
<19 years	233	14.9	2	18.2	2.57	0.632
20–24 years	452	29.0	5	45.4
25–29 years	452	29.0	2	18.2
30–34 years	300	19.2	2	18.2
⮚35 years	123	7.9	0	0,0
Gravidity (n = 1554)
Primigravida	483	31.1	5	45.5	1.112	0.774
Gravida 2–3	870	56.0	5	45.5
Gravida 4–5	193	12.4	1	9.0
Gravida > 6	8	0.5	0	0
Initiated ANC (n = 1553)
Yes	1456	94.9	5	45.5	48.211	0.000
No	77	5.1	6	54.5
Number of antenatal visit categories (n = 1511)
Un-booked (o visit)	77	5.1	4	44.5	24.534	0.000
1–3 visits	448	29.6	3	33.3
4–7 visits	607	40.3	1	11.1
=>8 visits	379	25.0	1	11.1
Initiated ANC before 20 weeks GA (n = 1456)
Yes	820	56.3	3	27.3	2.949	0.078
No	636	43.8	8	72.7
GA at delivery recorded (n = 1499)
<32 weeks	34	2.3	6	54.5	68.401	0.000
32–36 weeks	187	12.5	2	18.2
=> 37 weeks	1278	85.3	3	27.3
Rh blood group (n = 1557)
Positive	1533	98.5	11	100	0.173	0.842
Negative	24	1.5	0	0
Birth weight category
<2.5 kg	145	9.3	9	81.8	68.908	0.000
>2.5 kg	1411	90.7	2	18.2
HIV status at birth (n = 1561)
Positive	699	43.8	3	27.3	1.367	0.195
Negative	862	55.2	8	72.8
Syphilis status (n = 1557) at birth
Non-reactive	1525	97.9	10	90.9	2.719	0.205
Reactive	32	2.1	1	9.1

Table II describes the neonatal care indicators compliance and shows that the time of delivery was recorded for all (99.6%). However, the recording of the gender of neonates was 92%, and among them, female neonates (51%) outnumbered males. GA at delivery was recorded for 96% of women, with the majority of them (85%) delivered at term gestation. Virtually all neonates had birth weight recorded, and the low birth weight rate was 9.3%. Almost all livebirth neonates had APGAR scores in 1 and 5 minutes recorded (97.7% and 98%, respectively), with poor APGAR scores (scores < 7) in 1 and 5 minutes being 5.3% and 2.3%, respectively.

Table II. Recorded Neonatal Care Indicators at the Time of Delivery for the Sample

Variables	Frequency	Percentage
Time of delivery recorded (n = 1561)	1555	99.6
Gender of the neonates recorded (n = 1541)	1417	91.9
Male	695	49.0
Female	722	51.0
GA recorded (1561)	1499	96.0
<32 weeks	34	2.3
32–36 weeks	187	12.5
=> 37 weeks	1278	85.3
Birth weight (n = 1561)	1556	99.6
<2.5 kg	145	9.3
>2.5 kg	1411	90.7
Birth outcomes of HIV positive women (n = 699)	699	100.0
PCR test undertaken to HIV exposed live birth neonates (n = 692)	640	92.4
Targeted birth PCR test results available (n = 640)	614	95.9
PCR test positive (n = 614)	11	1.8
Missing PCR results (n = 640)	26	4.0
HIV status not known to HIV exposed neonates (PCR not tested and no test results) n = 692	78	11.2
HIV exposed neonates received prophylactic ART (n = 681) (3 NND, 4 stillbirths & 11 on ART excluded)	625	91.7
PCR positive baby received ART (n = 11)	11	100.0
APGAR score in 1 minute recorded (n = 1540)	1505	97.7
APGAR score < 7	81	5.3
APGAR scores > 8	1455	94.7
APGAR score in 5 minutes recorded (n = 1540)	1510	98.0
APGAR score < 7	36	2.3
APGAR scores > 8	1507	99.8
Neonatal complication recorded for live birth (n = 1541) (stillbirths excluded)	1540	99.9
Neonatal complication required resuscitation (n =1541)	78	5.1
Neonates required referral to hospital (n = 1521) (20 stillbirth excluded)	36	2.3
Injection vitamin K given (n = 1530) (11 NND & 20 stillbirth excluded)	1529	99.9
Polio vaccine given (n = 1530)	1450	94.7
BCG vaccine given (n = 1530)	1050	68.6
Birth outcome of neonates recorded (n = 1561)	1561	100.0
Live birth rate (n = 1561) (of all births)	1541	98.7
Still birth (n = 1541)	20	1.3
Neonatal deaths (n = 1541)	11	0.7

The birth outcomes of all 699 (100%) HIV infected pregnant women were recorded. Of them, 3 (0.5%) had neonatal deaths and 4 (0.6%) had stillbirths in this group. Among those (692) HIV exposed live-born neonates, 92.4% (640/692) had PCR tests undertaken and 614/640 (95.9%) of them had their results available in the birth register, of which 11 (1.8%) were PCR positive (HIV) at birth. Considering all HIV exposed 692 (excluding stillbirths and NNDs) livebirth neonates, the PCR results of the HIV exposed neonates were known to be 88.8% (614/692). This means 11.2% HIV HIV-exposed live-born neonates HIV status remained unknown at birth. Among those 681 (692-11) who received ART at birth) HIV exposed live-born neonates, 625 (91.7%), had a record of receiving ART prophylaxis. This means nearly 1 in 10 HIV exposed children who qualify for HIV prophylaxis at birth did not receive it. Among those live-born neonates that had PCR results available (614), 11 were found with PCR test positive, resulting in a neonatal HIV transmission rate at birth of 1.8% (mother-to-child transmission rate of HIV, MTCT rate). However, all HIV positive neonates received ART.

Neonatal birth complications were recorded for all and only 5.1% of them had any complications that required resuscitation, and 2.3% of neonates required referral for higher-level of care at the hospital. Virtually all (99.9%) neonates received vitamin K after birth. However, 94.7% and 68.6% of neonates received polio and BCG vaccines, respectively at birth. Birth outcomes (live births, neonatal deaths and stillbirths) of all deliveries were recorded. Among all childbirths, a total of 1541 (98.7%) had livebirths and 20 (1.3%) had stillbirths. The outcomes of all live-born neonates (1541) were recorded, with 11 (0.7%) that demised after delivery and resulting in a neonatal death rate of 7.5 per 1000 live births.

Cross table analysis (Table I) indicates that women who had no ANC, the number of ANC visits, and GA at delivery had significant differences in neonatal death rates. Therefore, these variables were entered into step-by-step binary logistic regression analysis to identify predictors for neonatal deaths. The results in Table III show that women delivered at lower GA (<32 weeks) and had no ANC are significant predictors for neonatal deaths. Neonates those were born before 32 weeks of GA had a higher OR of 15.9 (95% CI: 2.3–106.2, p = 0.004). Similarly, women who did not attend for ANC had an OR of 7.0 (95% CI: 1.1–42.3, p = 0.033).

Table III. Logistic Regression Output for ENND

Exposure variables	p-values	OR	95% CI for OR
			Lower	Upper
GA coded	0.016
GA < 32 weeks	0.004	15.954	2.395	106.287
GA 32–36 weeks	0.505	2.168	0.222	21.139
Had no ANC	0.033	7.033	1.168	42.341
Constant	0.000	0.002

Discussion

We undertook this study at an MOU run by midwives. The results thus reflected the performance of midwives undertaking delivery and neonatal care in SA. The results must be interpreted with the context of the demographic profile of the population that is keeping in with a poor socio-economic status, unemployment and informal types of residence. Furthermore, the results of this study sample pertaining to ANC, delivery and neonatal care provided to women and neonates were categorised as having low-risk pregnancies. Therefore, the findings were limited to those women who had childbirths at KCHC. The antenatal and neonatal services provided but not recorded in the birth register were considered as not being provided, but the clients (mothers and neonates) might have benefited from those unrecorded services that led to information bias regarding our results. Similarly, services not provided and recorded as services provided also led to false measurements. The characteristics of this study sample are also similar to the previous studies of the same health facility [24], [25]. For example, the sample mean age, teenage pregnancy rate, ANC booking patterns and HIV positivity rate of women who had childbirths at KCHC. The mean age of the sample of mothers was 25.6 years, with a total of 14.9% being teenage pregnancies. This figure of teenage pregnancy is encouraging compared to the overall global prevalence of 19%, with a higher rate in East Africa (21%) [25]. However, strategies are required to reduce teenage pregnancy in this community as it is a risk factor for negative pregnancy outcomes and hinders personal academic development [26].

Most (92.4%) of the HIV exposed neonates were tested for PCR and 96% of them had their results available, of which 1.8% neonates were HIV positive. Among those HIV-exposed neonates, 91.76% received ART prophylaxis as per national protocol. The undertaking of PCR test in our study was good, but 4% of them did not have PCR results. The blood sample collection and sending of PCR tests to the laboratory rate were higher than the rate found from another study [14]. However, the missing of PCR results is considered a missed opportunity for neonates’ identification of HIV status and to initiate ART earlier (at birth). Nearly 11% of the HIV-exposed neonates’ HIV status was unknown at birth. These included either the PCR test was not done (blood sample not collected) and or the blood samples sent to the laboratory for the test, but the results were unavailable. Further inquiry is required by the unit manager to deal with the matter to solve the problem of PCR test in the unit. PCR positivity rate (HIV transmission rate) at birth in our study was lower than the rate found from other study in SA where it was found higher of 2.5% [14]. However, the MTCT rate at birth (1.8%) in our study is higher than reported from an earlier study (1.5% in 2017 from SA) [27]. Studies from SA demonstrate that MOUs can attain the provision of cost-effective and high-quality maternal care [28]. However, in settings such as KZN where the maternal HIV prevalence is high, it can be difficult for MOUs to attain all the targets required to eliminate HIV. Current PMTCT services at MOUs involve the confirmation of HIV diagnosis in the antenatal period, initiation/continuation of antiretroviral therapy (ART) for HIV infected mothers, early infant HIV diagnosis and treatment and initiating HIV prophylaxis to HIV exposed neonates [16], [29].

According to the neonatal assessment and immediate care, the neonates were assessed for APGAR scores (98%) in 1 and 5 minutes, and for those babies who required assistance, the resuscitation process was undertaken for (5%). Those required a higher level of care (2.3%) were transferred to the hospital. The rate of identification of neonatal complications in our study setting was 5%, a rate lower than global estimates (10%) [8]. This could be due to the setting of our study, where only low-risk pregnant women delivered compared to high-risk pregnancies that deliver in hospitals or in general. Furthermore, women during labour from this MOU were also referred to the hospital due labour complications (e.g., Cephalo-pelvic disproportion, foetal distress etc.) Therefore, this lower rate of identification of neonatal complications is likely. However, the proportion of neonates that required hospital neonatal ICU and had further neonatal deaths in the hospital is not known.

It is found from this study that 94.7% and 68.6% neonates at birth received polio and BCG vaccines, respectively. These rates are similar to global neonatal immunization with polio (94%) and BCG (68.7%) [30]. The low BCG vaccination rate is not uncommon in SA due to administrative reason, such as stock shortage reported from other provinces [31]. Successful widespread vaccination against polio in developed countries has virtually eradicated the disease [32]. However, there are sporadic cases of polio found in places where vaccine coverage is insufficient, like sub-Saharan Africa and Southern Asia [33]. Over 95% of those who receive the oral or injectable polio vaccine as children are protected against polio [33]. Therefore, we can conclude that the polio vaccination rate at birth in this MOU is commendable. The remaining neonates could be vaccinated during the routine EPI programme from the catchment population. Contrarily, despite its well-known robust protection that greatly lowers infection rates, the Hepatitis B vaccination (HPV) is not given in SA [34]. However, unlike BCG, which provides protection after just one administration, the HPV vaccine is only effective for protecting neonates and requires 3 doses. It would be acceptable or essential to regularly examine the efficacy of the vaccination program in SA in order to ensure continuing compliance with neonatal immunization for any preventable infection or condition [35].

Despite a recent trend of an increase in the number of mothers who refuse injectable vitamin K for their neonates, we discovered that almost all (99.4%) neonates in this MOU received the vitamin K injection after birth, which led to an increase in the cases of late-onset vitamin K deficiency bleeding disorder (VKDB) [36]. According to a recent practice review, consistent injectable vitamin K supplementation at birth significantly reduces VKDB [37]. The risk for harm to the infant must also be taken into account, even though clinical decisions should always be based on the best available data. Since there have been no serious complications associated with vitamin K injections in neonates, higher compliance is acceptable [38].

The early neonatal death rate (7/1000 live births) in our study is lower than the rate reported for SA during 2016 of 10.2% [39]. The ENND rate in our study is lower due to referrals of very sick neonates to the hospital and delivery of low-risk pregnancies. The unit was unaware of the outcomes of neonates born to referred women during labour. Prematurity (47.9%) and diseases related to intrapartum, such as intrapartum hypoxia (24.3%) and infection (12%), are associated with higher proportions of ENND in SA [39]. In our study, GA of 32 weeks (OR = 16) and non-attendance at ANC throughout pregnancy (OR = 7.0) are risk factors (predictors) for ENND. This is consistent with WHO research that found that access to ANC services around the world has a significant impact on the lives of pregnant women and their neonates [40]. The identification of pregnancy-related risk factors and the resulting actions relating to the start-up and maintenance of ANC, which result in positive pregnancy outcomes.

Limitations

The exposure variables for measuring the association between neonatal death and pre-existing conditions in women, such as hypertension, diabetes mellitus, TB, etc., were restricted in a retrospective analysis. Unexpectedly, the birth register did not contain CD4 and viral load values for HIV-positive mothers. The outcomes of the one-centre study couldn’t be generalized widely. Future multicentre studies with comparable study variables will be taken into consideration.

Conclusion

The study from a MOU that provided a priority health programme in SA shows good compliance on neonatal care and emphasise the need to strengthen primary health care systems for early HIV-related care at the time of delivery to avoid a missed opportunity. We demonstrated poor uptake of PCR test, initiation of ART and BCG vaccination of neonates delivered at this PHC facility.

We thank the midwives at KCHC maternity unit for participating in data capture.

Conflict of Interest

Conflict of Interest: The authors declare no competing interests.