The working-hour phenomenon, a description of the adverse outcomes in facility-based births that take place outside the routine work hours, suggests that quality of care of obstetric patients may not be uniform round the clock.1 , 2 Since the timing of labour falls outside the control of the doctors (except for induced cases), medical and nursing staff in the labour rooms are generally busy throughout the day.3 Studies have linked outcomes of care with institutional organizational characteristics and staff competence,4 5 6 7 8 9 10 and these factors appear to be modulated by volume of clients/deliveries11 time/day of birth,12 , 13 availability of advanced surveillance techniques, and population/gestational ages of infants studied.13 , 14 Furthermore, occupational stress has been shown to negatively impact on the sleep quality of nurses. This chronic sleep deprivation, which can occur with nurses on prolonged night shifts, can lead to diminished ability to function at work, and possibly poor birth outcomes. This may be particularly so in hospitals with high patient-to-nurse ratios.15 , 16 However, other studies have found no association between time of conduct of obstetric procedures/deliveries and selected adverse perinatal outcomes.17 18 19 20 21 22

Most health facilities in the same tier or cadre of service provision in Nigeria have similar organizational framework, with advanced fetal surveillance services being virtually non-existent, even in the tertiary facilities. Thirty-nine percent of the deliveries in Nigeria take place outside a health facility, with many of the parturients presenting for care after the occurrence of a complication (‘unbooked’ patients). Of the total deliveries in Nigerian public and private facilities, women with induced labour accounted for less than 10%, and only 3% occurred via caesarean section, with scheduled operations making up 1%.23 Thus, a large number of births in Nigeria are unscheduled and occur around the clock.

Although there are data from audits of hospital performance using various indices in Nigeria, well-designed studies assessing the effect of time of delivery on birth outcomes are sparse. What are the predictors of/factors associated with out-of-hours’ deliveries in Nigeria? Does the general lack of advanced fetal surveillance technology have any impact on the pregnancy outcomes in out-of-hours’ births? This study is aimed at providing answers to these questions, and suggesting appropriate recommendations for clinical practice.

This study was conducted in the Maternity wing of Ekiti State University Teaching Hospital (EKSUTH), Ado – Ekiti, south-western Nigeria. The 46-bedded public facility with a delivery rate of about 1,500/annum, provides specialist obstetric services to clients within Ekiti and its neighbouring states.

The study population included obstetric patients who either had antenatal care in EKSUTH or were referred from other maternity care providers following the onset of a complication during pregnancy/delivery. The Department of Obstetrics and Gynaecology, EKSUTH, daily reviews all obstetric cases attended to during the previous 24 hours. Thereafter, with the aid of a proforma, data including the socio-demographic characteristics, obstetric events, past medical and obstetric history, details of parturition and the puerperium, complications, and perinatal outcomes were extracted from records and nurses’ sheets into a comprehensive database, by a full-time research assistant.

All deliveries occurring between October, 2013 and March, 2015 were purposively included in the study. However, women with incomplete details, who had severe medical conditions or died during pregnancy or childbirth were excluded from this study. At the time of this study, the hospital did not have facilities for advanced fetal surveillance, and almost all the patients had to pay for prenatal care and delivery services out-of-pocket. Of the 2,077 births that occurred during the period, 1,807 deliveries met the inclusion criteria, and were therefore included in the study and analyses. The institution’s Ethics and Research Committee gave approval for the study.

In this study, the following definitions were used: booked clients had prenatal care at EKSUTH, while ‘unbooked’ patients had no antenatal care, or were referred to EKSUTH from other maternity facilities when pregnancy or labour became complicated. ‘Out-of-hours’ deliveries were those that occurred between 00:00 hours and 07:59 hours the following day, a period that forms part of the doctors’ call duties and nurses’ night shifts in the Nigerian health care system. Hypertensive disorders included the spectrum of chronic hypertension, pregnancy-induced hypertension and pre-eclampsia. Rupture of the fetal membranes was regarded as premature when it occurred before the onset of labour.24 Babies with weights less than the 10^th centile for the gestational age were regarded as having intrauterine growth restriction.25 The Composite Adverse Obstetric Outcome referred to the occurrence of at least one of the following: hypertensive disorder in the index pregnancy, human immune deficiency virus (HIV) infection, premature rupture of membranes and intrauterine growth restriction. Still born babies did not have any signs of life at delivery. Early neonatal death referred to demise that occurred within the first week of life, and perinatal mortality was the sum of stillbirths and early neonatal deaths. The attending midwife or paediatrician assigns the APGAR Scores at the first and fifth minutes of baby’s life. The Composite Adverse Perinatal Outcome involved the occurrence of at least one of the following: 5-minute APGAR Score below 7, stillbirth or admission into the neonatal intensive care unit within 72 hours of birth.

The extracted data were entered into, and analyses done using, the Statistical Software for the Social Sciences (SPSS) package version 20. Pearson’s Chi-square (and whenever appropriate, Fisher’s exact) test was used to explore the association between maternal sociodemographic and baseline pregnancy characteristics with out-of-hours’ delivery. Multivariate regression model was employed to identify the obstetric and perinatal outcomes independently associated with births occurring out-of-hours in our institution. These were adjudged significant when p < 0.05.

Of the 1,807 patients eligible for inclusion in the study, 524 (29%) had out-of-hours’ deliveries. The mean age of the women was 30.0 ± 4.7 years, with a range of 14 – 49 years. Eight hundred and twenty-five (45.7%) women were below 30 years, while the majority, 1666 (92.2%) were of Yoruba extraction. Most of the participants were employed 1414 (78.3%), married 1757 (97.2%) and multiparous 1058 (58.6%) women, but without health insurance package 1775 (98.2%). At least three-quarters (76.4%) were educated to the tertiary level, and more than two-thirds 1244 (68.8%) were booked for prenatal care at EKSUTH (Table 1).

The association between patients’ characteristics and out-of-hours’ delivery was highlighted in Table 2. All the sociodemographic characteristics of women who had out-of-hours’ births were comparable with those of the women who did not. Significantly more women who had reduced perception of fetal movements were delivered out-of-hours (6.5% versus 3.6%; odds ratio: 1.866, p = 0.008). Women with hypertensive disease in the pregnancies studied (3.2% versus 6.7%; odds ratio: 0.467, p = 0.004) and those with at least one adverse obstetric outcome (9.5% versus 14.6%; odds ratio: 0.618, p = 0.004) had lower odds of delivering out-of-hours.

Table 3 displayed the results of the multivariate logistic regression analysis of the pregnancy outcomes following out-of-hours’ delivery. Women who delivered out-of-hours were more likely to have augmentation of labour (crude odds ratio [COR]: 0.73; 95% confidence interval [CI]: 0.58 – 0.93, p = 0.009), vaginal birth (COR: 1.70; 95% CI: 1.36 – 2.12, p < 0.001) of term babies (COR: 1.38; 95% CI: 1.01 – 1.87, p = 0.041), with fifth-minute APGAR scores at least 7 (COR: 1.65; 95% CI: 1.07 – 2.52, p = 0.023), and less likely to have at least one adverse perinatal outcome on the composite scale (COR: 1.38; 95% CI: 1.01 – 1.89, p = 0.046). The mean duration of labour augmentation (minutes) was significantly longer in women who had out-of-hours’ delivery (275 ± 190 versus 324 ± 205, p = 0.032). However, when the confounders were kept constant, vaginal birth (adjusted odds ratio [AOR]: 1.65; 95% CI: 1.32 – 2.06, p < 0.001), fifth-minute APGAR scores at least 7 (AOR: 1.94; 95% CI: 1.04 – 3.62, p = 0.038), and augmentation of labour (AOR: 0.78; 95% CI: 0.61 – 0.99, p = 0.041) were found to be independently associated with out-of-hours’ birth.

Table 1: Distribution of the patients