International Journal of Gynecology and Obstetrics 127 (2014) 201–205

Contents lists available at ScienceDirect

International Journal of Gynecology and Obstetrics journal homepage: www.elsevier.com/locate/ijgo

CLINICAL ARTICLE

Incidence and risk factors for early neonatal mortality in newborns with severe perinatal morbidity in Uganda Moses Musooko a,⁎, Othman Kakaire a, Annettee Nakimuli a, Sarah Nakubulwa a, Jolly Nankunda b, Michael O. Osinde c, Scovia N. Mbalinda d, Nelson Kakande e, Dan K. Kaye a a

Department of Obstetrics and Gynecology, School of Medicine, College of Health Sciences, Makerere University, Kampala, Uganda Department of Pediatrics and Child Health Obstetrics and Gynecology, School of Medicine, College of Health Sciences, Makerere University, Kampala, Uganda Department of Obstetrics and Gynecology, Jinja Regional Hospital, Jinja, Uganda d Department of Nursing, School of Health Sciences, College of Health Sciences, Makerere University, Kampala, Uganda e Clinical, Operations and Health Services Research Program, Joint Clinical Research Centre, Kampala, Uganda b c

a r t i c l e

i n f o

Article history: Received 8 January 2014 Received in revised form 13 May 2014 Accepted 1 July 2014 Keywords: Incidence Neonatal mortality Neonatal near-miss morbidity Perinatal morbidity Quality of newborn care Risk factors

a b s t r a c t Objective: To determine the incidence and risk factors for early neonatal death among newborns with severe perinatal morbidity. Methods: A prospective cohort study was performed of 341 newborns with severe perinatal morbidity admitted to the neonatal intensive care unit of Mulago Hospital, Uganda. All newborns were followed up for 7 days or until time of death. Information surrounding the mother’s obstetric history and pregnancy, the birth, and the neonatal history was collected using an interviewer-administered questionnaire and by review of relevant records. Multivariate logistic regression analysis was performed to assess factors independently associated with early neonatal death. Results: A total of 37 (10.9%) neonates died within 7 days, giving an incidence of early neonatal death of 109 deaths per 1000 live births (3 per 100 person-days). In multivariate analysis, respiratory distress (adjusted risk ratio [aRR] 31.29; 95% CI, 4.17–234.20; P = 0.001) and inadequate fetal heart monitoring during labor (aRR 6.0; 95% CI 1.40–25.67; P = 0.016) were significantly associated with an increased risk of early neonatal death. Conclusion: Approximately one in 10 neonates with severe perinatal morbidity died within 7 days of birth. Respiratory distress and poor monitoring of labor were risk factors for early neonatal death. © 2014 Published by Elsevier Ireland Ltd. on behalf of International Federation of Gynecology and Obstetrics.

1. Introduction Early neonatal death is the death of a neonate within 7 days of birth, and accounts for three-quarters of all neonatal deaths [1]. Its major causes include birth asphyxia, neonatal infections, birth trauma, and complications of prematurity. Assessment of maternal near-miss morbidity is recommended as a measure of the quality of maternal health care [2–4]. As a related concept, a neonatal near-miss case refers to a neonate who presents with severe life-threatening complications but survives the neonatal period [5,6], while a perinatal near miss may refer to a neonate who survives severe life-threatening perinatal complications. Neonatal near-miss cases can be identified in several ways [7,8]: by clinical features (such as lethargy, failure to suckle, prematurity, low birth weight, respiratory complications, or hypothermia at birth); using organ-system dysfunction (metabolic, respiratory, neurological, or cardiovascular dysfunction such as hypoglycemia, jaundice, ⁎ Corresponding author at: Department of Obstetrics and Gynecology, School of Medicine, College of Health Sciences, Makerere University, P.O. Box 7072, Kampala, Uganda. Tel.: +256 414 534361; fax: +256 414 533451. E-mail address: [email protected] (M. Musooko).

encephalopathy, sepsis, electrolyte imbalance, or thrombocytopenia); and from interventions used in the management of complications (such as tracheal intubation or blood transfusion). Assessment of stillbirths, neonatal morbidity, and neonatal mortality can generate useful data for establishing targeted interventions to address these problems [9]. The aim of the present study was to determine the incidence and risk factors for early neonatal death among neonates with severe perinatal morbidity. 2. Materials and methods The present study was conducted in the labor ward and the neonatal intensive care unit (NICU) of Mulago Hospital, Kampala, Uganda. On average, 2500 mothers are admitted to the labor ward and 300 newborns to the NICU every month, and the neonatal mortality rate is 26%–29% (Mulago hospital records 2012). A prospective cohort study of newborns admitted to the NICU with severe perinatal morbidity was undertaken from February 1 to March 31, 2013. The inclusion criteria were delivery in Mulago Hospital, a diagnosis of severe perinatal morbidity, a gestational age of 28 weeks or more, birth weight of at least 800 g, admission to the NICU within 24 hours of birth, and mothers’

http://dx.doi.org/10.1016/j.ijgo.2014.05.017 0020-7292/© 2014 Published by Elsevier Ireland Ltd. on behalf of International Federation of Gynecology and Obstetrics.

202

M. Musooko et al. / International Journal of Gynecology and Obstetrics 127 (2014) 201–205

informed consent for newborns to participate in the study. Severe perinatal morbidity was defined as birth asphyxia (5-minute Apgar score b 7) with or without fetal distress (defined as intrapartum bradycardia, tachycardia, or irregular fetal heartbeat, fetal skull molding, or meconium-stained liquor), respiratory distress, or meconium aspiration. Ethical approval for the study was granted by the Research and Ethics Committee of Mulago Hospital, the Ethics and Research committee of the Makerere University School of Medicine, and the Uganda National Council for Science and Technology. Consecutive sampling was used, whereby all neonates meeting the inclusion criteria were recruited until the sample size of 341 was achieved. This sample size was calculated from an estimated incidence of early neonatal mortality of 33.3% in the NICU for neonates with low Apgar scores (unpublished data from Mulago NICU records 2012), with an acceptable error of 5%, a power of 80%, and a 95% significance level. All newborns were followed up for 7 days or until time of death. Data were obtained from review of maternal and neonatal records (antenatal, laboratory, and delivery records) for newborn or obstetric complications and subsequent care. For neonates, data collected included breastfeeding, appearance, temperature, pallor, cyanosis, jaundice, results of investigations, and management provided. An interviewer-administered questionnaire was used to assess mothers’ sociodemographic characteristics (maternal age, level of education, occupation, marital status, and urban/rural residence), prenatal and prior obstetric history (parity, pregnancy, or childbirth complications such hypertensive disease, hemorrhage, cesarean delivery, and postpartum complications), childbirth complications, newborn care (neonatal resuscitation, cord care, immunization, eye care, and antibiotic treatment), and neonatal complications (convulsions, hyperthermia, hypothermia, hypoglycemia, septicemia, hyperbilirubinemia, and anemia) and their management. Neonates who died in the 7 days constituted the early neonatal deaths, and survivors constituted the neonatal near-miss cases. Data were analyzed using Stata version 10 (StataCorp, College Station, TX, USA). At analysis, characteristics of the neonates who died and survivors (near-miss cases) were compared. For continuous variables, means and standard deviations are reported. For categorical variables, frequencies and percentages are presented. The incidence of early neonatal death was estimated as the proportion of newborns who died within 7 days of birth. To assess risk factors for early neonatal death, characteristics of the neonates and their mothers were compared with those of survivors and their mothers. For continuous/numerical variables and categorical variables, t tests and χ2 tests, respectively, were used at the 5% confidence level. To assess independent contribution of these risk factors for mortality, hierarchical multivariate analysis was performed, in which all independent variables with a P value of less than 0.2 were included. 3. Results During the study period, 635 neonates were admitted to the NICU. A total of 341 met the inclusion criteria. Reasons for non-inclusion were very low birth weight (67 newborns weighed b 800 g, 49 of whom died within the first 72 hours after birth), gestational age of less than 28 weeks (53 newborns, 28 of whom died within the first 72 hours after birth), admission to the NICU later than 24 hours after birth (78 newborns), admission for observation with no overt morbidity (22 newborns), mother or caretaker unavailable to give informed consent (45 newborns, five of whom died within the first 72 hours after birth), and severe congenital anomalies (three with hydrocephalus, two with anencephaly, and two with omphalocele). A total of 79 (12.4%) of the 635 neonates admitted to the NICU died within 7 days (early neonatal deaths). Of the 341 neonates included in the present study, 37 (10.9%) died within 7 days, giving an early neonatal mortality incidence in the study population of 109 deaths per 1000

live births (or 3 deaths per 100 person-days). Seventeen (45.9%) of the 37 early neonatal deaths in the study population occurred on the first day of life, 7 (18.9%) on the second day, 4 (10.8%) on the third day, and 9 (24.3%) between the fourth and seventh days. After discharge, no neonatal deaths occurred, although four newborns who were not discharged and remained in the NICU died between 8 and 28 days. During the study period, there were 4551 live births at Mulago Hospital. The 79 early neonatal deaths in the NICU constitute an early neonatal mortality rate of 17.4 per 1000 live births, while the 37 early neonatal deaths in the study population constitute an early neonatal mortality rate of 8.1 per 1000 live births. Table 1 shows the sociodemographic characteristics of the mothers of the 341 neonates in the study population. Most mothers were aged 20–29 years (68.7%), had given birth once or twice previously (58.9%), and had attained at least a secondary level of education (97.9%). A total of 310 (90.9%) had attended prenatal care and booked for delivery in a health facility. A total of 174 (51.0%) were emergency referrals to Mulago Hospital from peripheral health units. Of the 341 newborns, the average gestational age at delivery was 36.0 ± 1.3 weeks. A total of 209 (61.3%) were male, 334 (97.9%) required neonatal resuscitation, and 162 (47.5%) showed clinical evidence of respiratory distress at admission to the NICU. Fifteen (4.4%) had birth injuries: 5 (1.5%) had fractures, 2 (0.6%) cephalohematomas, and 8 (2.3%) large caput succedaneums. The average time from delivery to initial assessment in the NICU was 4.6 ± 0.2 hours, and all initial assessments occurred within 8 hours of birth. Table 2 shows a stratified analysis of the risk of early neonatal mortality by specific clinical characteristics and mothers’ demographic characteristics. A blue skin color (cyanosis of extremities and mucus membranes) was associated with approximately a five-fold increase in risk of death (P b 0.001), while respiratory distress increased the risk of death by more than 3.5 times (P = 0.003) (Table 2). Mortality of newborns of mothers who had formal or informal employment was significantly lower than when mothers were unemployed or housewives (P = 0.042) (Table 2). Neonates whose mothers had attended prenatal care at least four times and were booked to deliver at a health facility actually had a higher incidence risk of mortality compared with those whose mothers had attended prenatal care three times or fewer,

Table 1 Baseline sociodemographic characteristics of mothers of neonates included in the study population (n = 341). Characteristic Marital status Married Unmarried Level of education Primary Secondary Tertiary Mothers’ occupation Not employed Salary earnera Businesswomanb Housewife Mother’s age, y 15–19 20–24 25–29 30–34 N35 Parity 1–2 3–4 N5 a b

Women in formal employment. Women in informal employment.

No. (%) 269 (78.9) 72 (21.1) 7 (2.1) 101 (29.6) 233 (68.3) 18 (5.3) 54 (15.8) 91 (26.7) 178 (52.2) 48 (14.0) 134 (39.3) 100 (29.3) 39 (11.4) 20 (5.9) 201 (58.9) 92 (27.0) 48 (14.1)

M. Musooko et al. / International Journal of Gynecology and Obstetrics 127 (2014) 201–205

203

Table 2 Stratified analysis of the risk of early neonatal death by selected demographic and clinical characteristics among newborns in the study population.a Characteristic Sex of neonate Male Female Skin color of neonate Pink Blue or pallor Respiratory distress No Yes Able to breastfeed No Yes Marital status Unmarried Married Employment status Not employed Salary earnerb Businesswomanc Housewife Formal or informal employmentd Education level Primary/secondary Tertiary Mode of delivery Vaginal delivery Caesarean delivery Prenatal care attendance: Attended ≥4 times and booked to deliver at a health facility Attended ≤3 times

Number of deaths (n = 37)

Incidence relative risk (95% CI)

P value

24 (64.9) 13 (35.1)

1 0.81 (0.38–1.64)

24 (64.9) 13 (35.1)

1 5.26 (2.46–10.75)

9 (24.3) 28 (75.7)

1 3.63 (1.67–8.75)

0.003

30 (81.1) 7 (18.9)

1 0.49 (0.18–1.15)

0.083

10 (27.0) 27 (73.0)

1 0.73 (0.34–1.70)

0.404

18 (5.3) 54 (15.8) 91 (26.7) 178 (52.2 145 (78.9)

1 0.07 (0.01-0.41) 0.22 (0.06-0.80) 0.23 (0.07-0.75) 0.05 (0.02–0.89)

0.003 0.021 0.015 0.042

11 (29.7) 26 (70.3)

1 1.18 (0.56–2.64)

0.664

33 (89.2) 4 (10.8)

0.42 (0.11–1.17)

0.077

35 (94.6) 2 (5.4)

1 0.44 (0.05–1.71)

0.250

0.537

b0.001

Abbreviation: CI, confidence interval. a Values are given as number (percentage) unless otherwise stated. b Women in formal employment. c Women in informal employment. d Businesswomen and salary earners.

although this was not statistically significant (Table 2). The possible explanation for this finding is that women with prepartum obstetric complications attended prenatal care more than the routine four times, because they had to be reviewed more frequently, while women without obstetric complications were seen fewer times. Table 3 shows univariate and multivariate analyses of complications of pregnancy or labor as risk factors for early neonatal death. In univariate analysis, higher gestational age was associated with a significantly lower risk for early neonatal death (P = 0.006) (Table 3). Increasing parity seemed to be associated with a lower risk for mortality although this was not statistically significant (Table 3). Inadequate fetal monitoring during labor was significantly associated with an increased risk of early neonatal death, although this was not significant on univariate analysis (P = 0.070) (Table 3). Neonates whose mothers were emergency referrals in labor were more likely to die in the early neonatal period (P = 0.003) (Table 3). Blue skin color or pallor, respiratory distress, and inability to breastfeed were statistically significant risk factors for early neonatal death (P b 0.05) (Table 3). The number of prenatal care visits did not have a significant effect on mortality in univariate analysis (P = 0.097) (Table 3). Distance to the neonatal care unit was not a predictor for early neonatal death (Table 3). On multivariate analysis, inadequate fetal heart monitoring during labor, respiratory distress, inability to breastfeed, and mother’s referral status were independent predictors of early neonatal death (P b 0.05) (Table 3). Cyanosis in the newborn was not statistically significant as a risk factor for early neonatal death (P = 0.081) (Table 3). Likewise, gestational age at delivery was not an independent risk factor for early neonatal death (P = 0.336) (Table 3). A higher maternal age was significantly associated with lower risk of early neonatal death (P = 0.022) (Table 3).

4. Discussion In the present study, 10.9% of neonates with severe perinatal morbidity died within 7 days, giving an incidence of early neonatal mortality of 109 per 1000 newborns, or 3 deaths per 100 person-days. Nearly 50% of the deaths occurred within the first 24 hours of birth. Inadequate fetal heart monitoring during labor was associated with a six-fold increase in the risk of early neonatal death in these newborns, while respiratory distress was associated with a 30-fold increase in risk of early neonatal death. The results are similar to those of a study on newborns admitted to the Neonatal Intensive Care Unit at the Medical University of Lodz, Poland, within the first 12 hours after birth [10], in which the frequency of early neonatal death was 11.5%. However, the incidence in the present study was much lower than at Moi Teaching and Referral Hospital, Eldoret, Kenya [11], where the early neonatal death rate was shown to be 19.7%. The present study’s finding that nearly 50% of deaths occurred on the first day of life is similar to that of Rajaratnam et al. [1], who showed that 30%–40% of early neonatal deaths occurred on the first day of life. Most early neonatal deaths are attributable to severe respiratory complications, hypothermia, and hypoglycemia, all of which combine to lead to electrolyte imbalance, neonatal infections, coagulation defects, acidosis, delayed fetal-to-newborn circulatory adjustment, hyaline membrane disease, necrotizing enterocolitis, and intracerebral hemorrhage [12–14]. Hypothermia can be prevented through simple measures such as ensuring the environment is warm during delivery, early breastfeeding and skin-to-skin contact with the mother, delayed and proper bathing, keeping the newborn dry, and prompt identification and rewarming of neonates who develop hypothermia. Inadequate monitoring during labor was a significant risk factor for early neonatal death in the present study. This finding is similar to

204

M. Musooko et al. / International Journal of Gynecology and Obstetrics 127 (2014) 201–205

Table 3 Univariate and multivariate analysis of the risk of early neonatal death by complications of pregnancy or labor among newborns in the study population. Characteristic

Birth weight, kg Skin color of neonate Pink Blue color or pallor Respiratory distress No Yes Newborn able to breastfeed Yes No Mothers’ age, y Parity Estimated gestational age at delivery Fetal heart monitored regularly during labor Yes No Number of prenatal care visits Mother referred No Yes Mode of delivery Vaginal Assisted vaginal Caesarean Distance to nearest health unit, km HIV status Negative Positive

Univariate analysis

Multivariate analysis

Unadjusted RR (95% CI)

P value

Adjusted RR (95% CI)

P value

0.69 (0.47–1.02)

0.060

1.29 (0.39–4.27)

0.673

1 6.04 (2.75–13.31)

0.001

5.04 (0.82–31.08)

0.081

1 3.95 (1.80–8.60)

0.001

1 31.29 (4.17–234.20)

0.001

1 1.36 (1.15–1.84) 0.93 (0.87–1.00) 0.92 (0.76–1.12) 0.89 (0.82–0.97)

0.018 0.051 0.406 0.006

1 3.63 (1.67–8.75) 0.83 (0.71–0.97) – 1.30 (0.76–2.23)

0.003 0.022 – 0.336

1 1.91 (0.95–3.84) 0.80 (0.62–1.04)

0.070 0.097

1 6.00 (1.40–25.67) –

0.016 –

1 3.21 (1.50–6.85)

0.003

1 3.24 (1.09–9.57)

0.034

1 0.82 (0.10–6.85) 0.53 (0.25–1.15) 0.72 (0.33–1.61)

0.856 0.109 0.431

– – –

– – –

2.11 (0.43–10.35)

0.356

–

–

Abbreviations: RR, risk ratio; CI, confidence interval.

that of a study from Tanzania [15], in which inadequate fetal heart monitoring during labor was associated with over 40% of perinatal deaths. Respiratory distress in the newborn was also significantly associated with risk for early neonatal death in the present study. Asphyxiated newborns may have suffered significant organ damage due to poor intrapartum fetal oxygenation (especially to the brain) during labor or after birth. They may also have concurrent morbidities such as neonatal infection acquired in utero, particularly after obstructed labor and chorioamnionitis. Therefore, birth asphyxia increases their risk for early neonatal death. Studies from Tanzania [15,16] showed that birth asphyxia was significantly associated with neonatal mortality, accounting for 51.3% of deaths [15]. Poor-quality obstetric care is a risk factor for perinatal mortality. A community-based cohort study in Mbale, Uganda [17], found that complicated deliveries, cord prolapse, obstructed labor, antepartum hemorrhage, and malpresentations were associated with perinatal death, and that early neonatal death was three times more likely if mothers delivered at home rather than in a health facility. Where there is inadequate newborn care and resuscitation, newborns with birth asphyxia may be regarded as stillborn, or may be transferred late to a neonatal care facility [7]. This might be due to poor healthseeking practices associated with cultural, logistical, and financial barriers [18,19]. Acute intrapartum emergencies associated with insufficient fetal oxygenation contribute to stillbirth, severe perinatal morbidity, and early neonatal death [20–23]. Maternal conditions such as anemia, pre-eclampsia/eclampsia, and malaria are known to cause hypoxic ischemic encephalopathy and eventual early neonatal death [21,22]. Prepartum identification of mothers with high-risk conditions (such as pre-eclampsia, sickle cell disease, and diabetes) or situations in which a fetus is compromised (such as intrauterine growth restriction) necessitates screening for evidence of poor oxygenation (such as chronic fetal distress or poor biophysical profile) [20–23]. Screening in higher-risk pregnancies may involve monitoring of amniotic fluid levels, serial biophysical profile assessment, continuous electronic fetal

heart rate monitoring, and Doppler blood flow measurements [22,23]. During labor, skilled care with regular fetal heart monitoring is crucial for identifying compromised fetuses. Quick intervention by assisted or cesarean delivery markedly reduces intrapartum death and early neonatal death. Neonates admitted to the NICU were excluded from the present study when the gestational age was less than 28 weeks or the birth weight was less than 800 g, because these factors are associated with higher neonatal mortality. When these neonates were included in analyses, the neonatal mortality rate more than doubled, yet their deaths might not have been preventable. In a multicountry review of neonatal mortality in east Africa [24], in which data were stratified by weight-forgestational age, neonates with low birth weight contributed over 50% of neonatal deaths, preterm newborns had the greatest risk of death, and size-for-gestational age significantly modified risk of death. Surviving the first week does not mean surviving the neonatal period, and four newborns included in the present study died between 8 and 28 days. However, detailed data are available for only the 341 newborns in the cohort. Therefore, the present study is not representative of the community and is not generalizable. Additionally, evaluation of organ-system dysfunction was limited by availability of imaging or laboratory investigations. The present study did not include an in-depth analysis of variables that potentially affect birth outcomes, such as duration of labor, duration of premature rupture of membranes, repeated vaginal examinations, and presence of chorioamnionitis. Lastly, there was no assessment of interaction between risk factors, which is known to modify risk. However, the present study is a step toward development and validation of clinical markers for severe neonatal morbidity, interventions that might be useful in reducing the burden of early neonatal deaths, and utility of the neonatal near-miss concept in newborn care. In conclusion, the frequency of early neonatal death in neonates with severe perinatal morbidity in Mulago Hospital was 10.9% (or 3 deaths per 100 person-days). Respiratory distress and inadequate intrapartum monitoring were significant risk factors for early neonatal

M. Musooko et al. / International Journal of Gynecology and Obstetrics 127 (2014) 201–205

death. Strategies to ensure timely availability of obstetric and newborn care, timely referrals of women with childbirth complications, adequate intrapartum monitoring, training of caregivers in neonatal resuscitation, and improved care of asphyxiated newborns would improve neonatal survival. Acknowledgments This study was part of a postdoctoral research project for D.K.K funded by the Swedish International Development Cooperation Agency Department of Research Cooperation through the Makerere University– Karolinska Institutet postdoctoral research grants program. The findings and conclusions of this article are those of the authors and do not necessarily represent the views of the funders. Conflict of interest The authors have no conflicts of interest. References [1] Rajaratnam JK, Marcus JR, Flaxman AD, Wang H, Levin-Rector A, Dwyer L, et al. Neonatal, postneonatal, childhood, and under-5 mortality for 187 countries, 1970–2010: a systematic analysis of progress towards Millennium Development Goal 4. Lancet 2010;375(9730):1988–2008. [2] Say L, Souza JP, Pattinson RC. WHO working group on Maternal Mortality and Morbidity classifications. Maternal near miss–towards a standard tool for monitoring quality of maternal health care. Best Pract Res Clin Obstet Gynaecol 2009;23(3):287–96. [3] Mantel GD, Buchmann E, Rees H, Pattinson RC. Severe acute maternal morbidity: a pilot study of a definition for a near-miss. Br J Obstet Gynaecol 1998;105(9):985–90. [4] Waterstone M, Wolfe C, Hooper R, Bewley S. Postnatal morbidity after childbirth and severe obstetric morbidity. BJOG 2003;110(2):128–33. [5] Pileggi C, Souza JP, Cecatti JG, Faúndes A. Neonatal near miss approach in the 2005 WHO Global Survey Brazil. J Pediatr (Rio J) 2010;86(1):21–6. [6] Avenant T. Neonatal near miss: a measure of the quality of obstetric care. Best Pract Res Clin Obstet Gynaecol 2009;23(3):369–74. [7] Marsh DR, Darmstadt GL, Moore J, Daly P, Oot D, Tinker A. Advancing newborn health and survival in developing countries: a conceptual framework. J Perinatol 2002;22(7):572–6.

205

[8] Goldenberg RL, McClure EM. Reducing intrapartum stillbirths and intrapartumrelated neonatal deaths. Int J Gynecol Obstet 2009;107(Suppl. 1):S1–3. [9] Engmann C, Matendo R, Kinoshita R, Ditekemena J, Moore J, Goldenberg RL, et al. Stillbirth and early neonatal mortality in rural Central Africa. Int J Gynecol Obstet 2009;105(2):112–7. [10] Fendler W, Walenciak J, Mlynarski W, Piotrowski A. Higher glycemic variability in very low birth weight newborns is associated with greater early neonatal mortality. J Matern Fetal Neonatal Med 2012;25(7):1122–6. [11] Ayaya SO, Esamai FO, Rotich J, Liechty E. Perinatal mortality in the Special Care Nursery of Moi Teaching and Referral Hospital, Eldoret, Kenya. East Afr Med J 2004;81(11):555–61. [12] Lawn JE, Cousens S, Zupan J. Lancet Neonatal Survival Steering Team. 4 million neonatal deaths: when? Where? Why? Lancet 2005;365(9462):891–900. [13] Lawn J, Shibuya K, Stein C. No cry at birth: global estimates of intrapartum stillbirths and intrapartum-related neonatal deaths. Bull World Health Organ 2005;83(6):409–17. [14] Feresu SA, Harlow SD, Welch K, Gillespie BW. Incidence of stillbirth and perinatal mortality and their associated factors among women delivering at Harare Maternity Hospital, Zimchildwe: a cross-sectional retrospective analysis. BMC Pregnancy Childbirth 2005;5(1):9. [15] Kidanto HL, Mogren I, van Roosmalen J, Thomas AN, Massawe SN, Nystrom L, et al. Introduction of a qualitative perinatal audit at Muhimbili National Hospital, Dar es Salaam, Tanzania. BMC Pregnancy Childbirth 2009;9:45. [16] Mmbaga BT, Lie RT, Olomi R, Mahande MJ, Kvåle G, Daltveit AK. Cause-specific neonatal mortality in a neonatal care unit in Northern Tanzania: a registry based cohort study. BMC Pediatr 2012;12:116. [17] Nankabirwa V, Tumwine JK, Tylleskär T, Nankunda J, Sommerfelt H, PROMISE EBF Research Consortium. Perinatal mortality in eastern Uganda: a community based prospective cohort study. PLoS One 2011;6(5):e19674. [18] Jewkes R, Wood K. Competing discourses of vital registration and personhood: perspectives from rural South Africa. Soc Sci Med 1998;46(8):1043–56. [19] Ahmed S, Sobhan F, Islam A, Barkat-e-Khuda. Neonatal morbidity and care-seeking behaviour in rural Bangladesh. J Trop Pediatr 2001;47(2):98–105. [20] Halloran DR, McClure E, Chakraborty H, Chomba E, Wright LL, Carlo WA. Birth asphyxia survivors in a developing country. J Perinatol 2009;29(3):243–9. [21] Azra Haider B, Bhutta ZA. Birth asphyxia in developing countries: current status and public health implications. Curr Probl Pediatr Adolesc Health Care 2006;36(5):178–88. [22] Goldenberg RL, McClure EM, Bann CM. The relationship of intrapartum and antepartum stillbirth rates to measures of obstetric care in developed and developing countries. Acta Obstet Gynecol Scand 2007;86(11):1303–9. [23] Darmstadt GL, Yakoob MY, Haws RA, Menezes EV, Soomro T, Bhutta ZA. Reducing stillbirths: interventions during labour. BMC Pregnancy Childbirth 2009;9(Suppl. 1):S6. [24] Marchant T, Willey B, Katz J, Clarke S, Kariuki S, ter Kuile F, et al. Neonatal mortality risk associated with preterm birth in East Africa, adjusted by weight for gestational age: individual participant level meta-analysis. PLoS Med 2012;9(8):e1001292.