Egypt’s commitment to the successful achievement of the Millennium Development Goals has been clearly observed in the past few years, since the endorsement of the Millennium Declaration in 2000. Egypt has already achieved the goal of reducing under-five mortality rate (U5MR) by two-thirds between 1990 and 2015. Thanks to public health interventions, under-five and infant mortality rates are decreasing at a faster pace than neonatal mortality (NM); consequently, neonatal deaths will represent an increasing proportion of child deaths. One of the challenges that Egypt currently faces is how to reduce perinatal mortality (PM) and NM 1,2. In developing countries, the average neonatal mortality rate in 2005 was 31/1000 live births (LB), with a range of 2–70 3. Comparison of estimates from surveys conducted at different time points indicated a decrease in neonatal mortality rate from 20/1000 LB in 2005 to 16/1000 LB in 2008. The corresponding figures for the perinatal mortality rate were 23/1000 pregnancies and 19/1000 pregnancies, respectively 1,2.
The perinatal period commences at 22 completed weeks (154 days) of gestation and ends 7 completed days after birth 4. The distinction between a stillbirth and an early neonatal death is often a fine one, depending on observing and then recalling sometimes-faint signs of life following delivery. The causes of stillbirths and early neonatal deaths are closely linked, and just examining one or the other can understate the actual level of mortality around delivery. For every baby who dies in the first week after birth, another is born dead (fetal deaths or stillbirths) 1.
Stillbirth is a professional and lay term that refers to a stillborn fetus. Intrauterine death occurs either before the onset of labor (antepartum death) or during labor (intrapartum death). Fetuses may die intrauterine, before the onset of labor, and because of pregnancy complications or maternal diseases; however, no special reason can be found for many antepartum intrauterine deaths. Complications arising during birth are the main cause of death among almost all infants who were alive when labor started, but were born dead 5.
The neonatal period begins with birth and ends 28 complete days after birth. Neonatal deaths may be subdivided into early neonatal deaths, occurring during the first 7 days of life (0–6 days), and late neonatal deaths, occurring after the seventh day but before the 28th day of life (7–27 days). Babies die after birth because they are severely malformed, are born very prematurely, have obstetric complications before or during birth, have difficulty adapting to extrauterine life, or because of harmful practices after birth that lead to infections. Early neonatal deaths occur during the perinatal period, and have obstetric origins, similar to those leading to stillbirths 5.
Globally, three causes of death infections (pneumonia, diarrhea, and tetanus), preterm birth, and asphyxia account for the majority of neonatal deaths. The causes of death vary between early and late neonatal periods, with deaths caused by preterm birth, asphyxia, and congenital defects occurring predominantly during the first week of life and infection being the major cause of neonatal death thereafter 3.
Neonatal deaths and stillbirths stem from poor maternal health, inadequate care during pregnancy, inappropriate management of complications during pregnancy and delivery, poor hygiene during delivery and the first critical hours after birth, and lack of newborn care. There are many other risk factors that have been considered to play a role in PM and NM such as socioeconomic factors, age of the mother and the father, age at marriage, sex of the fetus or the newborn, antenatal care, and other many factors 5. Variations in the causes and risk factors of perinatal and neonatal deaths have been found between and within countries. Information about the local epidemiology is important in prioritizing interventions. The present study aims to explore the risk factors associated with PM and NM in Alexandria, Egypt.
Participants and methods
Study setting and design
This case–control study was carried out at Elshatby University Hospital of Obstetrics and Gynecology and the Elshatby University Hospital of Pediatrics. These are university hospitals with ∼300 beds in Alexandria, with a catchment area including the three governorates: Alexandria, Matrouh, and Beheira. Most of the attendants are from low or middle socioeconomic levels.
The study populations were:
- Cases: mothers with a history of perinatal (from the 28th week of pregnancy to less than 7 days after birth) or neonatal deaths (from the seventh day to less than the 28th day after birth) 4,6 attending the University Hospital of Obstetrics and Gynecology.
- Controls: mothers with LB 28 days of age and older (up to 2 months to avoid recall bias) attending the University Hospital of Pediatrics.4,6
Using the maternal age as a risk factor and to detect an odds ratio (OR) of at least 2.5 with a power of 80% using a 5% level of significance when the prevalence of young or old maternal age among controls is 26%, the sample size was calculated to be 150 cases and 150 controls 7.
Sample type and selection
The days of the week were allocated randomly to visit each hospital 3 days a week. Consecutive cases were selected over 3 days/week from the antenatal outpatient clinic for those mothers with fetal deaths who were attending the clinic from 9 to 12 a.m. and from the Neonatal Care Unit for those with neonatal deaths from 12 a.m. till the required sample size was reached.
Controls were selected over 3 days/week from outpatient clinics in the ElShatby University Hospital of Pediatrics until the required sample size was reached.
The interview questionnaire method was used for data collection. It included sociodemographic data for parents, the medical history of the mothers, and smoking status of the parents during index pregnancy. Other items included past and present obstetrical history, antenatal care visits, vaccination with the tetanus toxoid, and the causes of PM/NM.
Data were analyzed using SPSS, version 16.0 (SPSS Inc., Chicago, Illinois, USA) and Epi Info, version 2003 (Atlanta, Georgia, USA). Socioeconomic level was classified into low, middle, and high using the modified Fahmy and El-Sherbini score on the basis of the crowding index, parental education, working status of the mother, and income 8. Descriptive statistics such as frequency and percents were used. OR and its corresponding 95% confidence intervals (CI) were computed to estimate the risk of PM/NM in relation to the independent variables studied. An OR greater than 1 indicates a higher risk of PM/NM and a CI that does not include 1 indicates a statistically significant difference in risk. Multivariate stepwise logistic regression analysis was used to model the risk of PM/NM as a function of the sociodemographic data, medical history, past and current obstetric history as well utilization of antenatal services. To control for confounders and to minimize the effect of multicollinearity among the risk factors, only significant (P<0.05) variables in bivariate analysis were included in the stepwise logistic regression analysis with the forward Wald technique. If two variables are significantly highly related, only one of them is introduced into the model. All tests were two tailed and the 0.05 level of significance was used 9.
Consent was obtained from all mothers after an explanation of the study objectives was provided and anonymity of the results was assured. Approval for the study was obtained from the Faculty of Medicine and Vice President for Graduate Studies and Research of Alexandria University.
The study included 300 mothers, of whom 150 were cases (mothers with perinatal or neonatal deaths) and 150 were controls (mothers with live babies 28 days of age or older). Among the cases, 62% were stillbirths whereas the remaining percent (38%) included neonatal deaths; 58.8% of these neonatal deaths occurred in the first week of life (early neonatal deaths) whereas 41.2% died after the first week up to the age of 28 days (late neonatal deaths).
Considering the causes of PM (Fig. 1), the most frequently reported causes were congenital anomalies (30.1%), preterm baby (24.7%), and asphyxia (15.1%), with a huge percentage (21.5%) with an unknown cause. NMs were mainly because of respiratory distress, preterm baby (19.3%), diarrhea (15.8%), jaundice (12.3%), and low birth weight (10.5%).
Table 1 shows that very young (≤20 years) and very old mothers (>35 years) had a higher risk of PM/NM compared with middle-aged mothers (OR=4.1; 95% CI: 1.7–10.1 and OR=6.7; 95% CI: 2.9–14.9, respectively). Those of a low social status had a higher risk of PM/NM by about 1.5 times than those of a middle social status (OR=1.5; 95% CI: 0.95–2.37), which was statistically insignificant (although it was of borderline significance). A higher percentage of cases were living in rural areas (42%) in comparison with the controls (32%), with an insignificantly higher risk of having PM/NM by about 1.5 times (95% CI: 0.96–2.46).
A history of maternal chronic health problem was associated with a significantly higher risk of PM/NM (OR=2.28; 95% CI: 1.38–3.71). Grand multiparity (>5 previous deliveries) was also significantly associated with a higher risk of mortality in the perinatal or the neonatal period (OR=3.40; 95% CI: 1.7–7.1). Mothers with a history of PM/NM had almost 10 times risk of having PM/NM compared with those without a history, a difference that was statistically significant. About 30% of cases (29.3%) had a history of previous pregnancy complications, whereas those with the same history among the controls were significantly fewer (12.5%), a pattern similar to that observed in women with previous delivery complications (OR=2.6; 95% CI: 1.3–5.2).
Considering the index pregnancy, when the father was a smoker, the risk of PM/NM increased by 20% and when the mother was a smoker, the risk was almost three times (OR=2.8), although both effects were not statistically significant. Complications during index pregnancy were reported significantly more among the cases (40%) compared with the controls (20%). Antenatal visits of less than four during index pregnancy were associated with four times the risk of PM/NM compared with those with four visits or more antenatal visits. A very small percent of controls did not receive any vaccination against tetanus during the index pregnancy (8.0%); in contrast, the corresponding percentage among the cases was 26.0% (OR=4.1; 95% CI: 2.0–8.1).
Considering the index delivery, 35.3% of the cases had delivered by cesarean section or assisted delivery compared with 32.7% of controls, a difference that was not statistically significant. However, delivery at home was significantly associated with a higher risk of PM/NM (OR=2.6; 95% CI: 1.5–4.4). Also, the skill level of the birth attendant significantly affected the risk of PM/NM. Mothers attended by a nurse or a Daya had about two times the risk (OR=1.9) of PM/NM compared with deliveries attended by a physician. This risk increased to eight times when the mother was alone at birth. About 41% of the cases experienced problems during the delivery stage compared with 13.3% of the controls. This huge difference was statistically significant (OR=4.6; 95% CI: 2.6–8.1).
Table 2 shows the results of a stepwise logistic regression analysis of the significant risk factors related to PM/NM. After adjusting for medical and obstetric history, the model extracted five factors as the most significant predictors of PM/NM: maternal age, past history of PM/NM, parity, frequency of antenatal visits, and index pregnancy complications. A higher risk of PM/NM was significantly associated with very young or very old mothers (OR=4.1 and 6.46, respectively), a positive past history of PM/NM (OR=12.7; 95% CI: 5.04–29.54), multiparity (OR=2.23; 95% CI: 1.50–5.42), infrequent antinatal visits (OR=3.88; 95% CI: 2.26–6.65), and the occurrence of complications during the index pregnancy (OR=9.21; 95% CI: 3.40–24.95).
On the basis of The Egypt National Perinatal/Neonatal Mortality Study 2000 10, PM was mainly attributed to congenital anomalies, prematurity, asphyxia, and a considerable percentage of unknown causes. However, the main causes of neonatal death were prematurity, respiratory distress, infections, and unknown causes.
Many studies 11–14 have reported an increased risk of perinatal/neonatal deaths among infants born to very young or very old mothers. Chen et al.14 have argued that teenage mothers are more likely to have perinatal/neonatal deaths, preterm babies, babies with congenital anomalies, and babies with low birth weight. All these factors finally decrease the probability of survival of the babies. Rahman and Abidin 15 have reported that mothers younger than 25 years of age at pregnancy had an increased risk for neonatal deaths. Del Rossario et al.16 studied the relation between maternal age at pregnancy and the pregnancy outcome in terms of stillbirths and found that those mothers who became pregnant at the age of 25 years or younger had a higher risk of having a stillborn baby. The results of our study are in agreement with all these studies as 13.3% of the cases were pregnant at 20 years or younger compared with 4.7% of controls, with a four-fold higher risk of having PM/NM.
In contrast to many previous studies, 17–19 the results of the present study indicated a nonsignificant effect of socioeconomic or urban–rural classification on PM/NM. However, a prospective cohort study carried out in Ludhiana, Punjab 11, reported a nonsignificant effect of urban–rural categorization on PM. Considering the socioeconomic level, most of the cases and controls in the present study were from a low to moderate social class, with mild variation among them leading to the dilution of the variability accuring from socioeconomic factors. Moreover, the Egypt Demographic and Health Survey, 2008, confirmed the absence of urban–rural differences in PM/NM 1.
The chronic health problems of the mother were found to be a predictive factor for perinatal/neonatal deaths. Conde-Agudelo et al.20 have reported that mothers with chronic hypertension were at a higher risk of having stillbirths (OR=2.25) and mothers with anemia had double the risk of having a stillborn baby. Weiner et al.21 found that poor maternal nutrition and maternal anemia were associated with increased PM in the developing countries. These is in agreement with the finding in the present study that mothers with preexisting chronic health problems had an increased risk for perinatal/neonatal deaths (OR=2.3; CI: 1.4–3.7).
Amoa et al.22 have reported that mothers with a history of neonatal deaths were at a higher risk of having neonatal deaths in the subsequent pregnancies (OR=2.59). The same finding was obtained by Conde-Agudelo et al.20. This may be associated with the presence of previous complications in pregnancy, which may reoccur during the index pregnancy or the presence of preexisting unavoidable factors (e.g. chronic health problem, lack of medical care, and poverty).
Index pregnancy complications were considered as the most important predictor for adverse pregnancy outcomes 11,23. This relation has been confirmed in the present study as several complications were found during the index pregnancy that were significantly associated with a higher risk of PM/NM.
In agreement with other studies, the number of previous deliveries was associated significantly with PM/NM. Joshi 23 found that the neonates with birth order greater than 3 had reduced probability for longer life (OR=4.4) compared with the first-order newborns. Also, a previous study carried out in Alexandria (2001) indicated that mothers with more than five previous deliveries were more likely to have PM/NM 13. This association may be because of the older maternal age for those with high multiparity. Older maternal age is usually associated with chronic health problems, pregnancy complications, and delivery complications. In addition, the first and second births are usually the most desired by families and such children are treated with more care, whereas children born later receive less care.
Schoeps et al.12 reported that prenatal care was considered adequate when the first appointment took place in the first trimester of pregnancy and at least four appointments were reported, and in at least one of these appointments maternal blood pressure and fetal heart rate were assessed. They also reported the negative effect of inadequate or absent prenatal care on early NM. The present study also came to this conclusion: mothers with poor prenatal care reported a higher rate for PM/NM (OR=4.3).
Multiple gestations are a major contributor to PM. The high rates of death for twin and multiple pregnancies are mainly related to the complications specific to multiple pregnancy (such as twin-to-twin transfusion syndrome) and an increased risk of complications common to singletons and multiples, especially fetal abnormalities and growth restriction. This risk factor has been researched consistently in studies on stillbirths and perinatal deaths in developed countries because of the high rate of twin and multiple pregnancy following the widespread use of assisted reproductive technology. Assisted reproduction techniques are becoming increasingly available in less developed countries, and there is reason to believe that the proportion of multiple pregnancies is increasing 5. The present study did not include this risk factor as other risk factors are considered more important.
The causes of PM and NM were presented in some cases on the basis of the mothers’ reports. Recall bias may have contributed to the order of these causes. For example, according to mothers’ reports, respiratory distress was the most commonly mentioned cause of neonatal deaths whereas it is well known that infection (e.g. pneumonia) is the leading cause of NM in developing countries including Egypt. The dyspnea found in the case of pneumonia was most probably perceived by the mother as respiratory distress and the cause was recorded as per the reason cited by the mother.
Study of the risk factors of PM and NM should include information on many other variables such as mothers’ nutritional status before and during index pregnancy, birth weight, and gestational age. This information could not be obtained from this case–control study because of recall bias of the mothers.
PM and NM should be analyzed separately as they may have different risk factors. However, this was inconvenient in the current study as a larger sample size and a longer duration are required.
Conclusion and recommendations
The causes behind PM/NM included various social, obstetrical, and medical factors. Very young or very old mothers, those with grand multiparity, with inadequate medical care during pregnancy, a history of PM/NM, and pregnancy complications were the most significant risk factors.
This study highlights the importance of improving access to quality health services, with special care given to young mothers and high multiparous women. Information on mothers with previous PM or NM should be obtained to offer more suitable conditions to the follow-up of these pregnant women. Efforts in the coming years must thus focus on implementing preconception education and counseling programs to ensure more widespread pregnancy planning with appropriate interpregnancy intervals. Registration of pregnant mothers, detection of high-risk cases, and a timely intervention, which can be ensured by a minimum of four antenatal visits, delivery by trained personnel, and limiting of family size, will help in achieving the goal of improving maternal health and reducing PM/NM.
Conflicts of interest
There are no conflicts of interest.
1. El-Zanaty F, Way A Egypt
Demographic and Health Survey 2008. 2009 Cairo, Egypt
Ministry of Health, El-Zanaty and Associates, and Macro Internationa
2. Tabet M Egypts progress towards achieving the millennium development goal. 2010 Egypt
The Minister for Economic Development
3. Lawn JE, Zupan J, Begkoyian G, Knippenberg KJamison DT, Breman JG, Measham AR, Alleyne G, Claeson M, Evans DB, et al. Newborn Survival. Disease Control Priorities in Developing Countries. 20072nd ed. New York Oxford University Press:531–549
4. Make every mother and child count. 2005 Geneva WHO
5. Park K Park’s textbook of preventive and social medicine. 200719th ed. Jabalpur India Bhanot Publishers
6. Reducing perinatal
and neonatal mortality
. Report of a meeting Baltimore, May 10–12. 1999;Vol. 3 Maryland WHO:1–50
7. Lwanga SK, Lemeshow S Sample size determination in health studies. A practical manual. 1991 Geneva WHO
8. Fahmy SI, El Sherbinie AF. Determining simple parameters for social classifications for health research. Bulletin of HIPH. 1983;13:95–107
9. Streiner DL, Norman G Biostatistics: the bare essentials. 20083rd ed. USA PMPH
10. Campbell O, Gipson R, el-Mohandes A, Issa AH, Matta N, Mansour E, Mohsen L. The Egypt
Study 2000. J Perinatol. 2004;24:284–289
11. Benjamin AI, Sengupta P, Singh S. Perinatal mortality
and its risk factors
in Ludhiana: a population-based prospective cohort study. Health Popul Perspect Issues. 2009;32:12–20
12. Schoeps D, Furquim de Almeida M, Alencar GP, França I Jr, Novaes HM, Franco de Siqueira AA, et al. Risk factors
for early neonatal mortality
. Rev Saúde Pública. 2007;41:1–9
13. Mokhtar MM, Abdel-Fattah MM. Consanguinity and advanced maternal age as risk factors
for reproductive losses in Alexandria
. Eur J Epidemiol. 2001;17:559–565
14. Chen XK, Wen SW, Fleming N, Yang Q, Walker MC. Increased risks of neonatal
and postneonatal mortality
associated with teenage pregnancy had different explanations. J Clin Epidemiol. 2008;61:688–694
15. Rahman MM, Abidin S. Factors affecting neonatal mortality
in Bangladesh. J Health Manag. 2010;12:137–152
16. Del Rossario GR, Lewis T, Irons B, Campbell-Forrester S, Weiss HL, Jolly PE. Assessment of risk factors
for stillbirth among pregnant women in Jamaica. J Obstet Gynaecol. 2004;24:750–755
17. Peña R, Wall S, Persson LA. The effect of poverty, social inequity, and maternal education on infant mortality
in Nicaragua, 1988–1993. Am J Public Health. 2000;90:64–69
18. Perinatal mortality
2005: England, Wales and Northern Ireland. 2007 London CEMACH
19. Godson MC, Nnamdi J. Environmental determinants of child mortality
in Nigeria. J Sus Dev. 2012;5:65–75
20. Conde-Agudelo A, Belizán JM, Díaz-Rossello JL. Epidemiology of fetal death in Latin America. Acta Obstet et Gynecol Scand. 2000;79:371–378
21. Weiner R, Ronsmans C, Dorman E, Jilo H, Muhoro A, Shulman C. Labour complications remain the most important risk factors
for perinatal mortality
in rural Kenya. Bull World Health Organ. 2003;81:561–566
22. Amoa AB, Klufio CA, Amos L. A case–control study of early neonatal
deaths at the Port Moresby General Hospital to determine associated risk factors
. PNG Med J. 2002;45:185–196
23. Joshi R Perinatal
and neonatal mortality
in rural Punjab. A community based case–control Study. Working Papers Series No. 3 from eSocialSciences, 2003, pp. 1–29