The effect of nighttime or weekend births on perinatal outcomes in Africa, 2025: Systematic review and meta-analysis

Abstract

Background

Differences in staffing, clinical supervision, emergency preparedness, and the organization of services outside normal working hours may be related to variations in perinatal outcomes for off-hours and weekend deliveries. Evidence from Africa is sparse and conflicting, with some studies reporting increased neonatal morbidity or mortality for births occurring at nighttime/off-hours or weekends while others have reported no significant differences. Hence, this systematic review and meta-analysis aimed to estimate the association between nighttime/off-hours or weekend births and adverse perinatal outcomes in African health facilities.

Methods

This systematic review and meta-analysis followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses 2020 guideline. An internet-based search of noninterventional studies in Africa was performed in CINAHL, EMBASE, PubMed/MEDLINE, Cochrane Library, Scopus, Web of Science, African Index Medicus, African Journals Online, Africa-Wide Information (EBSCOhost) and Google Scholar. A total of 3430 studies were screened, and only 8 studies satisfied the inclusion criteria. The Newcastle–Ottawa Scale and Joanna Briggs Institute critical appraisal checklist were used. The random- and fixed-effects meta-analyses were done in Stata version 17, and forest plots, tests of heterogeneity, and funnel plots were all generated.

Results

Eight studies were included. The majority of single-centric studies report increasing risk of early neonatal death (OR ≈ 1.70), fresh stillbirths (OR ≈ 1.95), low Apgar scores, and neonatal intensive care unit admissions with off-hour deliveries. Meta-analyses showed the borderline relationship of fresh stillbirth (OR 1.33; 95% CI 0.99–1.67) with I² = 88.7%, early neonatal death (OR 1.62; 95% CI 1.20–2.05; I² = 0%), but not with a 5-minute Apgar score of < 7 (OR 1.13; 95% CI 0.86–1.40; I² = 0%) with night/weekend birth. No evidence for publication bias was demonstrated.

Conclusion

Off-hour delivery within African hospitals is associated with increased risk of early neonatal death and possibly fresh stillbirth, but the impact on Apgar score and other outcomes is heterogeneous. Staffing, case mix, and service organization were the explanations for differences.

Background

Evening and nighttime deliveries that required obstetric interventions were linked to a higher risk of adverse perinatal outcomes compared to similar deliveries during the daytime [8]. Research also found that births occurring in the evening, at night, on weekends, or on public holidays were more likely to result in unfavourable Appearance, Pulse, Grimace, Activity, and Respiration (Apgar) scores or abnormal umbilical pH levels [4]. Findings indicate that weekend deliveries consistently pose a risk for various perinatal complications, and there may be differences in how hospitals manage increased patient volumes [22]. However, some studies did not find significant differences in perinatal or maternal outcomes based on the time of day, except for a higher rate of third- and fourth-degree perineal tears during evening deliveries. Additionally, neonatal outcomes following obstetric emergencies such as uterine rupture, partial placental abruption, and cord prolapse did not show variations between day and night [23].

Different times of delivering babies were linked to more negative outcomes for newborns. Researchers suggested that maintaining the correct mix of health care workers during the day and night shifts could be more important than primarily relying on having more senior obstetricians available at night [18]. There were no differences in how often mothers experienced health issues based on when they gave birth. However, babies born during the evening or night had a higher chance of getting infections and needing care in the neonatal intensive care unit. Other health problems in newborns were not affected [25]. Similarly, deliveries occurring during weekends, nighttime shifts, or at regional perinatal care centers have been associated with an increased risk of cerebral palsy following emergency cesarean section [24].

A study in Africa showed that perineal injury, obstetric complications, and neonatal outcomes showed no significant difference by day of delivery, and these results contradict the idea that deliveries on weekends and holidays are more risky for patients and their children [28]. The “night & weekend effect” did not exist among patients who were in spontaneous labour and whose delivery date was not planned. Weekday adverse obstetric outcomes may be less likely when “low-risk” inductions and cesarean sections are scheduled [5]. These findings suggest that variation in perinatal outcomes by timing of birth may be dependent on clinical setting, staffing patterns and organization of maternity and neonatal services. Evidence from African settings is limited and inconsistent and it is not clear whether nighttime/off-hours or weekend births are consistently associated with adverse perinatal outcomes across different hospital contexts. Thus, it is important to synthesize the available African evidence to clarify the magnitude and consistency of these associations and to identify potential context-specific areas for quality improvement.

Objectives

To estimate the association between nighttime or weekend births at a health facility and adverse perinatal outcomes in African countries.

Methods

Search strategy and sources

This Systematic Review and Meta-Analysis (SRMA) was conducted via the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 guidelines [16] to investigate the effect of nighttime or weekend births at health facilities on perinatal outcomes in Africa. We searched all noninterventional studies electronically in CINAHL, EMBASE, PubMed/MEDLINE, the Cochrane Library, Scopus, Web of Science, African Index Medicus (AIM), African Journals Online (AJOL), Africa-Wide Information (EBSCOhost), and Google Scholar. The references of the retrieved studies were also searched for further articles. Additionally, a manual search for a thesis/dissertation was carried out on university websites. The search was performed under the population, exposure, comparison and outcome (PECO) framework between August 20 and September 10, 2025. Population: Pregnant women and their newborns delivered in African countries; Exposure: Nighttime birth and/or weekend birth; Comparator: Daytime birth or weekday birth; Outcome: Perinatal outcomes. All studies reporting the effect of nighttime and /or weekend births in Africa were included. The principal aim of this review was to determine the effect of nighttime or weekend births on perinatal outcomes in Africa. In this SRMA, a systematic evaluation of the included studies concerning the effect of nighttime or weekend births was conducted.

We checked the PROSPERO database (https://www.crd.york.ac.uk/prospero/) to determine whether there are any recently published, completed or ongoing projects about the subject in a bid to prevent any further unnecessary duplication. The findings established that there were no ongoing or published articles on this topic. Consequently, we tried to register, but the system was restricted for noninterventional protocols and we did not register in the PROSPERO database.

All relevant terms were searched interchangeably with either the Boolean operator “AND” or “OR” as indicated, and two independent reviewers (YDM and AAM) who reached a consensus on every occasion without disagreements were involved.

The search terms used in PubMed was ((((“Pregnant People“[Text Word] OR “Pregnant Person“[Text Word] OR “Pregnant Woman“[Text Word] OR “Women, Pregnant“[Text Word] OR “Pregnant Women“[Text Word] OR “Woman, Pregnant“[Text Word] OR “People, Pregnant“[Text Word]) OR “Pregnant People“[Mesh] OR (“Infant, Newborn“[Text Word] OR “Infants, Newborn“[Text Word] OR “Newborn“[Text Word] OR “Newborn Infant“[Text Word] OR “Newborn Infants“[Text Word] OR “Neonate“[Text Word] OR “Neonates“[Text Word] OR “Newborns“[Text Word]) OR “Infant, Newborn“[Mesh]) AND (“Africa“[Text Word] OR “Africa“[Mesh])) AND ((“night“[Text Word] OR “nighttime“[Text Word] OR “night-time“[Text Word] OR “evening“[Text Word] OR “after-hours“[Text Word] OR “out-of-hours“[Text Word] OR “off-hours“[Text Word] OR “duty hours“[Text Word] OR “weekend“[Text Word] OR “Saturday“[Text Word] OR “Sunday“[Text Word] OR “public holiday“[Text Word] OR “Time Factors“[Mesh] OR “Parturition“[Mesh]))) AND ((“perinatal mortality“[Text Word] OR “perinatal death“[Text Word] OR “stillbirth“[Text Word] OR “fresh stillbirth“[Text Word] OR “macerated stillbirth“[Text Word] OR “early neonatal death“[Text Word] OR “neonatal mortality“[Text Word] OR “Apgar score“[Text Word] OR “NICU admission“[Text Word] OR “resuscitation“[Text Word] OR “birth asphyxia“[Text Word] OR “hypoxia“[Text Word] OR “neonatal sepsis“[Text Word]) OR (“Infant Mortality“[Mesh] OR “Perinatal Mortality“[Mesh] OR “Perinatal Death“[Mesh] OR “Infant Death“[Mesh] OR “Fetal Death“[Mesh] OR “Stillbirth“[Mesh] OR “Apgar Score“[Mesh] OR “Intensive Care Units, Neonatal“[Mesh] OR “Resuscitation“[Mesh] OR “Asphyxia Neonatorum“[Mesh] OR “Hypoxia“[Mesh] OR “Neonatal Sepsis“[Mesh]))

Inclusion and exclusion criteria

Inclusion

This SRMA incorporates (1) all observational studies conducted on the effect of nighttime or weekend births on perinatal outcomes in Africa; (2) all noninterventional studies (cross-sectional, mixed methods, qualitative, and observational analytical descriptive studies); and (3) both published and unpublished (preprints, theses/dissertations).

Exclusion

Interventional studies (RCTs, quasi-experimental), case series/reports, books, reviews, editorials, conference abstracts, correspondence, and letters were excluded. Studies not reporting time/day exposure, non-African settings, multi-country studies without extractable Africa-specific data, or studies where exposure was unrelated to time/day were excluded.

Interventional studies including randomized controlled trials and quasi-experimental studies were excluded as this review aimed to estimate the association between naturally occurring night-time/off-hours or weekend delivery, and perinatal outcomes rather than the effect of an intervention to modify off-hours care. Case reports and case series were excluded as they generally do not have an appropriate comparison group and cannot reliably estimate the association between time/day of birth and perinatal outcomes. These criteria were used to ensure methodological consistency with the review question, although this resulted in a relatively small number of studies being eligible. The detailed process of identifying eligible studies and the reasons for exclusion are presented in Fig. 1.

Fig. 1.

PRISMA 2020 flow chart for the included studies [16]

Data extraction

Data were independently extracted by two reviewers (YDM and AAM). All studies that matched the selection criteria were imported into the EndNote 20 database. After the removal of duplicates, the titles and abstracts were evaluated for eligibility before full-text assessment. Full-text reviews were checked for quality thereafter via the Newcastle‒Ottawa Quality Assessment Scale (NOS), which was adapted for cross-sectional studies, and the JBI critical appraisal checklist, which was used for retrospective descriptive (cohort-type) research. Data extraction was performed via a table and an Excel spreadsheet. The following information was extracted from each study: (1) identification details (surname of first author with year of publication), (2) study country, (3) study design, (4) study setting, (5) characteristics of population, (6) sample size, (7) perinatal outcomes, (8) data collection method, (9) comparison and (10) quality score.

Quality assessment

Two authors assessed the quality of the methodology, study population, sample size, comparability, and statistical analysis of each study. In cases of disagreement (10%), a third author (HA) was involved, and the discussion led to complete resolution. The quality of the selected quantitative articles was assessed via the Newcastle‒Ottawa Scale (NOS), which was adapted for cross-sectional studies [15]. The score ranges from 0 to 10 points. Articles with a score of 7 and above were considered high quality, those with scores between 5 and 6 were deemed average quality, and those with a score of 4 and below were rated poor quality. No studies were removed because of poor quality in this review. Overlapping studies were excluded by retaining the most comprehensive, recent studies and were methodologically fit with this SRMA. After the quality analysis of the included studies was completed, a data display matrix about quality was created and summarized.

The quality of one retrospective descriptive study was assessed independently by using the Joanna Briggs Institute (JBI) critical appraisal checklist adapted for cohort studies [2], which contains 11 critical questions [12]. Disagreements were also resolved by the third author (HA). The JBI checklist promises rapid evaluation because it comprises items under a comprehensible checklist and applies to different studies. The quality of the study was, thus, categorized using this tool as YES to mean that this item was appropriate for the study, NO to mean this item was not appropriate for the study, and UNCLEAR to indicate that it was not clear whether this item could be applied to the study, and NOT APPLICABLE when the item was not applicable to the method. A score was given for each item that fell into the YES category, and the included study was judged as having good-quality results by scoring 6 out of 11.

Definitions/outcome measurement

The concept of day and night in Africa is almost the same, where day refers to working hours and night to the after-hours. However, the specific hours differ based on the hospital shift hours and the study framework. Some studies divide the 24 h into 3 shifts, namely morning, evening, and night [14]. In South Africa and Nigeria, the daytime aligns with working hours (8:00 am to 5:00 pm) and night (5:00 pm to 2:00 am). Due to heterogeneity of shift classifications among the included studies, terms nighttime, off-hours, and evening/night were used as defined in the original studies. Evening was not a standalone primary exposure category in this review. However, studies were eligible if evening was included within an off-hours/nighttime exposure, or if evening and nighttime were reported within the study’s assessment of time-of-delivery effects. Differences in definition of exposure were considered in the interpretation of heterogeneity across studies.

Weekend Versus Weekday

12:00 AM Saturday – 11:59 PM Sunday versus Monday–Friday.

Perinatal Period

This includes ≥ 28 weeks of gestation and the first seven days of post-delivery.

Perinatal Outcomes

Stillbirth, either fresh or macerated; early neonatal death (ENND); perinatal mortality which includes stillbirth and ENND; neonatal death before discharge; low Apgar score (1-min or 5-min < 7); neonatal intensive care unit (NICU) admission/ referral; birth asphyxia/need for resuscitation; and birth trauma/injuries.

Statistical analysis

Statistical pooling (meta-analysis) was performed to quantify the effect of nighttime or weekend births on perinatal outcomes. A check for any possible publication bias was performed via a funnel plot and Egger’s regression test. The significance level for the heterogeneity test and Egger’s test was considered statistically significant for a p-value < 0.05. The choice of meta-analysis model was made depending on the level of heterogeneity between studies calculated with the help of Cochran’s Q and I² tests. The I² index of less than 50%, 50 to 75%, and more than 75% were assumed as the signs of low, moderate, and high heterogeneity, respectively [3]. The fixed-effect model was employed in the case of low heterogeneity, while the random-effects model was used in the case of moderate heterogeneity. Thus, the fixed-effect model was used for pooled results of 5-minute Apgar score < 7 (I²=0%) and early neonatal death (I²=0%), whereas a random-effects model was used for fresh stillbirth due to the presence of significant heterogeneity (I²=88.65%). Stata version 17.0 was used for the meta-analysis. Analyses were performed by the type of outcomes identified in at least 2 studies.

Result

Literature search findings

The first search gave results in 3,422 citations, along with eight citations obtained from other sources such as Google Scholar and reference list screening. After removing 42 duplicate records, 3,388 citations were left for the next step of screening through title and abstract. In the step of title and abstract screening, 589 citations were selected for full-text screening. Of these 589 citations, 581 were removed because of non-compliance with eligibility criteria, which led to eight records being included in the systematic review and meta-analysis [16] in Fig. 1.

Characteristics of included studies

We used the Newcastle‒Ottawa Scale (NOS) adapted for cross-sectional studies [15] and the JBI critical appraisal checklist for retrospective descriptive studies for quality assessment, as described in the methodology section (Table 1).

Table 1.

Description of the studies included on the effect of nighttime or weekend births on perinatal outcomes, 2025

Ref no, Authors (Years)	Study Country	Study Design	Study setting	Sample Size	Study population	Outcomes Identified
[14]	Tanzania	Hospital- based cross-sectional	Tertiary referral hospital	2636	Low-risk pregnant women	10% of Apgar scores < 7 at 5th min); 4% of ENND and 6% of fresh stillbirths. Apgar at 5 min Neonatal (OR, 1.48, p = 0.02), ENND (OR, 1.70; p = 0. 03), and fresh stillbirths (OR, 1.95; p = 0.001) were more significantly associated with deliveries occurring during night shifts compared to evening and morning shifts.
[6]	Sudan	Hospital-based cross-sectional	Maternity hospital	388	Low-risk pregnant women	The nighttime total neonatal complication was 40 (48.8%) and significant at p < 0. 005. Newborn birth injuries 10 (2.6%), birth asphyxia 62 (16%), and NICU admission 103 (26.5%). The evening time was 62 (46.3%), which was significant at p < 0.005
[7]	Nigeria	Hospital-based cross-sectional	Tertiary hospital	4556	All term deliveries were included.	NICU admission between morning and evening hours (p = 0.009) and between morning and night hours (p = 0.795). ENND during the evening was twice as high (1.2%) as in the morning (0.5%); p = 0.047 and night hours (0.6%); p = 0.623. There was no difference in the risk of fresh stillbirths between morning and evening (p = 0.560), as well as morning and night hours (p = 0.75); there was also no difference in fresh stillbirths between weekdays and weekends (p = 0.895). There was no difference in low Apgar scores at the 1st minute between morning and evening (p = 0.053) and night (p = 0.221), and between weekdays and weekends (p = 0.524). Similarly, there was no difference in low 5th -minute Apgar scores between morning and evening (p = 0.165) and night (p = 0.944), as well as between weekdays and weekends (p = 0.529). However, ENND was twice (p = 0.085) and 1.3 times higher (p = 0.526) for evening and night hours, respectively, while there was no difference between weekends and weekdays (p = 0.652).
[26]	Kenya	Secondary analysis of prospective cohort data	National referral hospital	25,911	neonates delivered more than 24 weeks included	Nighttime/weekend delivery was associated with a reduced risk of adverse composite birth outcomes (a RR 0.92, 95% CI: 0.88, 0.96). No association was found between nighttime/weekend delivery and neonatal death. Nighttime/weekend deliveries were associated with better maternal and birth outcomes, perhaps due to decreased emergency Caesarean deliveries for high-risk pregnancies or other factors.
[13]	Kenya	Prospective cohort study	Tertiary referral hospital	240	Postnatal mothers who have undergone unscheduled C/S	Poor neonatal outcomes (neonatal death, respiratory difficulties, trauma, and neonatal sepsis) did not show a significant difference depending on the time of surgery within 72 h.
[27]	Senegal	Cross-sectional	Maternity health center	48,270	All mothers delivered and their newborns after 22 weeks of gestation.	Evening shift deliveries had 1.1 times the odds of poor perinatal outcome (Apgar score < 7 at 1 min) compared to morning shift deliveries (OR 1.1, 95% CI; 1.02–1.18, p = 0.012). No significant difference was found in the odds of neonate referrals and deaths across the three shifts. Night shift deliveries had 1.1 times the odds of perineal injuries compared to morning shift deliveries (OR 1.11, 95% CI 1.04–1.18, p = 0.001 for episiotomy and OR, 1.14, 95% CI 1.04–1.26, p = 0.008 for perineal tears).
[9]	South Africa	retrospective descriptive	Hospital-based delivery registry	9397	Infant births weighing more than 1,000 g	The perinatal mortality rate was similar across day, early night, and late night (no statistically significant difference).
[28]	Senegal	Retrospective descriptive & analytical (10-year)	Urban maternity health center	42,870	a pregnancy of more than 22 weeks that gave birth	The rate of Apgar score < 7 at 5 min was 10.8% on weekdays compared to 10.7% on weekends and holidays (p = 0.437). Deliveries on weekends and holidays had a higher risk of neonatal death compared to those on weekdays; the risk was low for transfers/referrals. Perineal injury, World Health Organization obstetric complications, and neonatal outcome showed no significant difference by day of delivery.

Among the seven quantitative studies, four articles were identified with a score of high quality (7–10), and the remaining 3 articles were found to have 6 and were deemed average quality by using the NOS (Table 2).

Table 2.

Modified Newcastle‒Ottawa Scale adapted for cross-sectional studies to assess quality of studies, 2025.

Reference No, Author (Years)	Criteria
	Selection (5*)				Comparability (2*)		Outcome (3*)		10*
	Represent-ativeness of the sample	Sample size	Non-res-pones rate	Ascertainment of exposure	Study controls for the most important factor	Control for any additional factor	Assessment of the outcome	Statistical test	Quality score
[14]	*			*	*		**	*	6*
[6]	*			*	*		**	*	6*
[7]	*			*	*		**	*	6*
[26]	*	*		**	*	*	**	*	9*
[13]	*	*		**	*		**	*	8*
[27]	*	*		**	*		**	*	8*
[28]	*	*		**	*		**	*	8*

Keywords: *=1 Point and **=2 Points

The remaining one retrospective descriptive article was independently evaluated by the JBI appraisal checklist adapted for cohort studies and judged as moderate quality with sore of 6 out of 11 items (Table 3).

Table 3.

JBI Critical Appraisal Checklist for retrospective descriptive study.

(adapted from cohort studies),2025

JBI Critical Appraisal Questions (11 items)		21 [9],
		Yes	No	Unclear	Not Applicable
1	Were the two groups similar and recruited from the same population?				☑
2	Were the exposures measured similarly to assign people to both exposed and unexposed groups?	☑
3	Was the exposure measured in a valid and reliable way?	☑
4	Were confounding factors identified?	☑
5	Were strategies to deal with confounding factors stated?	☑
6	Were the groups/participants free of the outcome at the start of the study (or at the moment of exposure)?			☑
7	Were the outcomes measured in a valid and reliable way?	☑
8	Was the follow up time reported and sufficient to be long enough for outcomes to occur?			☑
9	Was follow up complete, and if not, were the reasons to loss to follow up described and explored?			☑
10	Were strategies to address incomplete follow up utilized?		☑
11	Was appropriate statistical analysis used?	☑
	Total Score	Moderate quality (6/11)

Key: ☑ align for the top of the column (Yes, No, Unclear or Not Applicable). ☑ under Yes=1 point, but in others= 0 point

Effect of nighttime or weekend births on perinatal outcomes

Eight hospital-based observational studies were included from Africa; below are the per-study reported perinatal findings.

In Tanzania [14], off-hours deliveries have been associated with greater chances of neonatal distress (OR 1.48), ENND (OR 1.70), and fresh stillbirth (OR 1.95). The study further reported that the proportions of low 5-minute Apgar and physical birth trauma were higher during the night shift. The NICU admission differed by shift (higher in the evening than in the morning; p = 0.009). The ENND was higher in the evening (1.2%) than the morning (0.5%) (p = 0.047); however, no significant difference was recorded in fresh stillbirths by shift or weekday/weekend. Low 1- and 5-minute Apgar scores did not record significant differences by shift overall [7].

According to the large-scale multivariable models, nighttime/weekend deliveries did not carry a risk for neonatal death. In contrast, night- and weekend-time deliveries appeared to be slightly protective for the composite outcome of maternal and composite birth outcome (adjusted RR ~ 0.92; 95% CI 0.88 to 0.96) and had no association with neonatal death [26]. The large-scale facility series found slightly higher odds for low Apgar at 1 min during the evening shift (evening vs. morning: OR ~ 1.10, 95% CI 1.02 to 1.18) and an increased neonatal referral during the evening (OR ~ 1.46, 95% CI 1.03 to 2.08). No statistically significant differences were reported in the neonatal death rates across different working shifts, and overall neonatal outcomes on weekdays versus the weekend were not grossly discrepant [27].

Unscheduled/emergency Caesarean delivery was more common at night. Still, the study gave an overall conclusion of there being no consistent difference in immediate neonatal outcomes (neonatal morbidity/mortality) between night and day Cesarean section. The authors argue about longer operative times at night and system factors [13]. In a retrospective registry analysis of spontaneous births, there was no difference in perinatal mortality (and distribution of early/late neonatal deaths) between day and night, and the timing of spontaneous birth did not affect adverse perinatal outcomes in this cohort [9]. In a low-risk population in Sudan, the authors found no significant differences in neonatal outcomes (stillbirth, neonatal referral, or low Apgar) by day of delivery (weekday/weekend) or time of day. This study reported descriptive comparisons only; there were no elevated ORs for perinatal outcomes [6]. In a large Dakar facility dataset, authors reported that though no difference was established for neonatal death across different shifts, a slightly increased odds of Apgar < 7 was found for deliveries in the evening (OR 1.10, 95% CI 1.02–1.18), with neonates having higher odds of referral; there were no increased rates of birth trauma or neonatal death through the night [28].

Summary of descriptive findings

Quantitatively, some of the single-center studies [7, 14] found increased odds/percentages of ENND, fresh stillbirths, low Apgar, and NICU admission with evening/night or off-hours deliveries (ORs for ENND ≈ 1.70 and for fresh stillbirth ≈ 1.95; ENND to be higher in the evening with p ≈ 0.047, and NICU admissions to be higher in the evening with p ≈ 0.009). Larger datasets and several facility-series studies (secondary analysis in Kenya, Senegal, Sudan, and South African registry) generally had no findings of increased neonatal death with nighttime/weekend delivery; some showed no difference or very small effect sizes for Apgar or referrals, e.g., evening Apgar OR ≈ 1.10 in Senegal, while the secondary analysis study in Kenya adjusted for factors and did not find the association to neonatal death [6, 9, 26–28].

Generally, the evidence is mixed: the majority of single-center studies suggest that evening (and especially nighttime) deliveries carry greater neonatal compromise; however, larger/adjusted analyses show heterogeneity in confirming a consistent increase in neonatal death. The heterogeneity across studies perhaps stems from the case mix (e.g., referrals vs. antenatal admissions), differences in staffing and organization, and outcome definitions (e.g., fresh vs. macerated stillbirth and early vs. all neonatal death).

Meta-analysis/statistical pooling

Initially, a forest plot was generated for the outcomes via the fixed effect model, and heterogeneity was identified for some outcomes. We assessed the differences in the reported effect between the studies via Cochran’s Q test and I² statistics. A p-value of less than 0.05 from the Cochran’s Q test indicated statistical significance, with variation levels indicated by the I² statistic ranging from 0% to 100%: 0%—no variation, 25%—minor variation, 50%—moderate variation, and ≥ 75%—greater variation. In case of heterogeneity observed in some studies, we applied a random effects model otherwise we used fixed model. This provided a complete picture and understanding of the data being aggregated.

We had made pooled synthesis of the effect of nighttime or weekend birth on perinatal outcomes reported in at least two studies in same measurement. These outcomes are Apgar score < 7 at 5-minute, ENND and fresh stillbirth. We also carried out Egger’s regression tests, publication bias and sensitivity analysis. Unfortunately, there were no identified small study effects for the three outcomes in the analysis of publication bias. We also conducted Heterogeneity (Galbraith plot) and funnel plot for the three perinatal outcomes.

Forest plot of APGAR score less than 7 at five minutes

As shown in the figure below, the pooled outcome of APGAR score < 7 at 5 min is 1.13 with 95% CI: [0.86–1.40], I² = 0.00, and P = 0.76 (Fig. 2). This indicated that no significant heterogeneity was identified between studies and the effect is not significant. For publication bias, the regression-based Egger test for small-study effects with the random-effects model identified no small-study effect; hence, we used the fixed-effect model for this case. The pooled estimate is dominated by one large study [27].

Fig. 2.

Forest plot for the pooled prevalence of Apgar score < 7 at 5 min in Africa, 2025

Funnel plot of Apgar score less than 7 at 5 min

The funnel plot of Apgar score less than 7 at 5 min of delivery is presented below (Fig. 3).

Fig. 3.

Funnel plot of Apgar score < 7 at 5-minutes,2025

Forest plot of early neonatal death (ENND)

The pooled effect of the nighttime or weekend birth effect on ENND showed that the pooled OR was 1.62 with 95% CI: [1.20–2.05], I² =0.00 and p < 0.0001. This indicates there is no significant heterogeneity identified, but the effect is significant. Here is also one large study dominance. For publication bias, the regression-based Egger test for small-study effects with the fixed-effects model identified no small-study effect (Fig. 4).

Fig. 4.

Forest plot of the early neonatal Death in Africa, 2025

Forest plot of fresh stillbirth

The pooled analysis of fresh stillbirth yield OR:1.33 with 95% CI: [0.99–1.67], I² = 88.65%, and p = 0.00 (Fig. 5). This indicates the effect is significant, and heterogeneity was identified with the random effect model. For publication bias, the regression-based Egger test for small-study effects with a random-effects model was used; no small study effect was identified.

Fig. 5.

Forest plot of the fresh stillbirth analysis in Africa, 2025

Discussion

This SRMA has analyzed evidence from eight hospital-based observational studies in African settings to ascertain the influence of nighttime or weekend deliveries on perinatal outcomes. Our findings provide important insights into whether temporal patterns in childbirth time translate to neonatal compromise.

Impact on perinatal outcomes

Our meta-analysis of hospital-based observational studies in Africa showed no pooled association with low 5-minute Apgar score, a significant association with ENND, and an equivocal heterogeneous signal with regard to fresh stillbirth. The ENND finding—an increase in risk seen consistently across the studies partially accords with global syntheses in that modest off-hour associations have been reported for several perinatal adverse outcomes; nevertheless, the existence and magnitude of an “off-hours” or “weekend” effect differ across settings and outcomes. This is supported by evidence of a higher chance of stillbirth or hospital death within the first seven days for babies born on the weekend, the rising incidence of newborn injury and three-day neonatal emergency readmissions [10, 17], and an increased risk of infection and the requirement for admission to an intensive care unit linked to evening or nighttime births [25].

Other global studies showed weekend deliveries were still substantially linked to higher risks of Apgar score < 7, newborn intensive care unit admission, longer maternal duration of stay, and infant convulsions on high-volume weekdays, even after controlling for between-hospital heterogeneity [22]. Neonates with a 5-minute Apgar score < 7 and umbilical artery pH < 7.0 were more likely to be born to women who gave birth at night, but none of the other outcomes that were looked at showed any differences [11]. Riskier pregnancies delivered by senior staff in a tertiary center favor a better outcome [18]. Hence, achieving a better balance in the total number of labor ward workers during the day and the night appears to be a greater priority than boosting the continuous presence of senior obstetrical staff on the labor ward during the out-of-hours period.

Evidence identified weekend births were associated not only with higher rates of neonatal death but also with lower birth weight and ethnicity of the mother [19]. A markedly increased risk of cerebral palsy in emergency cesarean sections was linked to labor on the weekends, during the night shift, and at regional perinatal medical care centers [24]. Newborns delivered during the evening and night shifts were more likely to stay longer than five days than those delivered during the morning shift. Despite physician seniority, the nighttime shift is linked to the highest likelihood of unfavorable newborn outcomes in term pregnancies [21]. These findings are relatively in line with our findings to some extent but not fully supported.

On the contrary, weekends, holidays, or at nighttime births did not detect any higher rates of morbidity or fatality [20]. With the exception of a higher frequency of third- and fourth-degree tears in the evening, there were no variations in maternal or perinatal outcomes throughout the day. Additionally, there were no day-night variations in the neonatal outcome following obstetric crises (uterine rupture, partial placental abruption, and cord prolapse). The availability of skilled medical personnel may help prevent unfavorable evening results [23]. This implies that the effect of weekend, night, or off-hours birth on perinatal outcomes is determined by many factors, including hospital setting, senior staff, caseload, skill of providers, and other system-related variables.

Systematic reviews and multi-center studies have yielded conflicting findings, with some suggesting there may be an increased risk for a composite of adverse outcomes or specific complications at weekends/off-hours, while others find nil or negligible effects on neonatal mortality once case-mix and confounding have been adjusted [10]. The single-center studies in our review, for example [14], reported higher risks for ENND, low Apgar, and NICU admission during night or evening shifts, in keeping with literature describing greater perinatal compromise during off-hours in some hospitals. However, at times such associations are attenuated or nullified after multivariable adjustment in large population or facility-level analyses. Large recent analyses in low- and middle-income countries (LMICs) context have shown no increase in the risk of neonatal death with night or weekend births, with some adjusted models showing slight protective effects for composite maternal-neonatal outcomes. This implies that strong signals in smaller single-center studies and weaker to null effects in larger adjusted analyses align with findings from global systematic work showing heterogeneity across hospitals and studies. Conflicting results reinforce an idea: that the “off-hours effect” depends upon context and is not uniform [10].

The high differences in fresh stillbirth rates, other perinatal outcomes, and the variation across studies likely reflect several interacting sources: (1) case mix and referral patterns: Tertiary or referral centers receive sicker women and may have different case mixes at night; (2) staffing models and supervision: Reduced availability of senior clinicians, midwives, or immediate access to operating rooms and neonatal resuscitation during nighttime or weekend periods may plausibly increase adverse events in some settings. However, the included studies did not provide standardized staffing data by shift or weekday/weekend period, limiting our ability to quantify this contribution; (3) measurement and outcome definitions: Differences in classification (fresh vs. macerated stillbirth), timing, and identification of early neonatal deaths along with varying reporting practices; and (4) decisions and system factors: Longer operation times at night, delayed responses, and differences in transfer and referral thresholds all contribute. These factors often appear in global reviews and research as possible reasons for the observed associations during off-hours [22].

Our findings are in part consistent with smaller single-center reports of increased risks of neonatal compromise in night or off-shifts.

For instance, Mgaya et al., in Tanzania [14], reported increased risk of ENND and unmacerated stillbirth among night births, while Eze et al., [7] reported increased neonatal admissions and low Apgar scores during night shifts. Similarly, Wade et al., in Senegal, also reported slightly increased risk of low Apgar scores and neonatal referrals among evening deliveries [28]. However, bigger and more controlled/adjusted analyses did not always demonstrate an increased risk. Virk et al., in their analysis of over 25,000 deliveries, demonstrated no association between off-hour births and neonatal mortality, actually having slightly protective effects for composite maternal and neonatal outcomes [26]. Wade et al., also found no difference in neonatal death between shifts, although they did find greater odds of low 1-minute Apgar scores in evening births [27]. Makori et al., found higher nighttime emergency cesarean sections but concluded that neonatal morbidity and mortality did not have a day-night difference in cases [13]. Similarly, Hoque & Hoque [9] and Eltayeb [6] were unable to establish differences in perinatal outcomes between daytime delivery and nighttime delivery in their respective samples.

Thus, even though the evidence for an association between ENND and night/weekend deliveries is provided by our pooled analysis, individual large-scale studies show that the association is not in all settings.

Strengths and limitations

A strength of this analysis is the utilization of multi-country African data to provide a broader regional view. As an additional measure, our publication bias test using Egger’s test revealed no suggestion of small-study biases. One of the main limitations of this systematic review is the small number of eligible studies and small numbers of studies included in each of the meta-analyses performed. Even though the search process yielded a considerable amount of literature, only eight studies fulfilled the predefined eligibility criteria. As a result, it was not possible for us to perform proper subgroup analyses or meta-regression analyses based on the country, hospital level, staffing system, exposure measure, or other contextual variables. Furthermore, there were several instances where the pooled estimates had been significantly influenced by large individual datasets. In addition, the observational study designs could be subject to some uncontrolled confounding.

Conclusion

Inconsistent evidence on the effects of nighttime/weekend delivery in Africa was found from this meta-analysis and systematic review. Single-center studies show raised risks of adverse outcomes, but larger analyses have less consistent results. Pooled estimates were as follows: no significant effect on Apgar < 7 at 5 min, raised risk of ENND, borderline raised risk of fresh stillbirth with extreme heterogeneity. Overall, there appears to be a relation between nighttime/weekend delivery and higher early neonatal mortality, with some heterogeneity across settings.

Implications for practice and recommendations

Even when pooled estimates are small, the consistent direction of increasing ENND risk in multiple African studies suggests system potential for intervention, strengthening night-shift staffing and ensuring timely availability of experienced senior clinical staff, including senior midwives and obstetric clinicians, to provide supervision, clinical decision-making, and rapid emergency response. Quality and preparedness audits that are critical time-critical activities have the potential to reduce inequalities observed in some hospitals. International evidence suggests that reducing organizational variability (stable senior cover, intensive night staffing, normalized protocols) will probably decrease off-hours outcome inequalities when they exist [1, 10].

The hospitals better to perform a thorough analysis to identify any hospital-specific gaps for each of the selected perinatal outcomes by timing and delivery day, such as early neonatal death, fresh stillbirth, low Apgar score, neonatal referral/NICU admission, and others that could be used at particular hospitals. The gap analyses must identify possible factors that affect the selected perinatal outcomes, such as staffing level and senior medical supervision, emergency response time, neonatal resuscitation and operating room access, referral and transfer process, handover, and workload during nights and weekends. In accordance with identified gaps, the hospital needs to design quality improvement programs that include measurable indicators of change and further assessment of their efficiency in decreasing the disparity in off-hours outcomes.

Generally, the following recommendations were forwarded:

Significantly enhance health systems: augment staffing, supervision, and resources during off-peak times.

Establish context-specific interventions: modulate service delivery to workload, staff competence, and emergency capacity.

Promote more research: conduct large multi-center studies with confounder adjustment.

Enable policy action: maintain equity in quality of care at all times of delivery.

Abbreviations

CI

Confidence interval

JBI

Joanna briggs institution

NOT

Newcastle‒ottawa scale

PRISM

Preferred reporting items for systematic review and meta-analysis

NICU

Neonatal intensive care unit

LMICs

Low- and middle-income countries

SRMA

Systematic review and meta-analysis

Author contributions

YDM, AAM, GTM, MGF, and HA contributed to the conceptualization and methodology of the study, performed formal analysis, interpreted the findings, and participated in writing, reviewing, and editing the manuscript. YDM additionally contributed to data curation, project administration, visualization, validation, and preparation of the original draft manuscript. GTM and MGF also participated in the investigation and drafting of the original manuscript. AAM, HA, AG, MG, AAE, and TDK supervised the study, contributed to validation and formal analysis, and critically reviewed and edited the manuscript. YDM, GG, AAA, AL, AAA, MS, and MDB contributed to data curation, investigation, validation, and manuscript review and editing. All authors read and approved the final manuscript.

Funding

None.

Data availability

Data supporting the findings of the analysis presented in this review can be submitted upon request to the corresponding author.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.