Risk Factors Related to the Neonatal Mortality in Kurdistan Province, Iran: A Population-Based Case-Control Study

Neonatal mortality rate (NMR) as an indicator of public health reflects the demographic, biological, cultural, and economic status of the population (1). The NMR also represents the growth trend and public health of the population (2). Reducing the NMR is still an important health topic in the developing countries and it is also the 4th objective of the Millennium Development Goals, which drags huge attentions (3, 4). Neonatal mortality is defined as the death of newborns in their first 28 days of life or the neonatal period (5, 6).

Nearly, 40% of the under-5 mortality occurs within the 28 days of life (7). The rates and causes of the neonatal mortality vary in different regions and states (8). Many previous studies investigated the causes of the neonatal mortality to improve health care and reduce neonate deaths. According to the world health organization (WHO) reports, the leading causes of 80% of all neonatal mortality include prematurity, low birth weight, infections, birth asphyxia, and birth trauma (9, 10). Due to the use of innovative solutions/interventions to prevent and care in many parts of the world, the NMR has decreased (11). Based on the latest reports, the NMR in the world decreased from 33 deaths per 1000 live births in 1990, to 21 deaths per 1000 live births in 2012 (12). In Iran, after the implementation of the integrated children health care programs, the NMR decreased from 26 deaths per 1000 live births in 1990, to 16 per 1000 live births in 2012. This value in Kurdistan province also decreased from 13.5 deaths per 1000 live births in 1990 to 12.2 per 1000 live births in 2012 (13, 14).

Despite efforts to reduce the NMR, its trend is slow and this subject led to an increase in the deaths of children under 5 years old (8); 1 study assessed under-5 mortality rate in Kurdistan province during 2007 to 2011 and showed that the NMR did not have a significant decreasing pattern (14). The current study aimed at investigating the influencing factors on the NMR in Kurdistan province, Iran, in 2013.

The current population-based case-control study was conducted from October to December 2013. Cases were neonates who died within the first 28 days of their lives in Kurdistan province, Iran. Control group was randomly selected from neonates born in the closest time and place to dead neonates. The inclusion criterion was neonatal death within the study period in Kurdistan province. Exclusion criterion for case and control groups was death occurred in the children who came from other parts of Iran to Kurdistan province and died there. Data were collected from the deputy of health and civil registration. In the year of conducting the study, 198 neonates died that based on the inclusion and exclusion criteria, 181 cases were selected as the case group, while 17 neonates were excluded from the study. The control group had twice the size of the case group. After reviewing the literature and national guidelines for maternal death and the neonatal care, the causes of the neonatal mortality were identified by referring to the references and preparing the initial checklist. Data were collected using a checklist contained various questions, including demographic variables, maternal history of diseases, problems during pregnancy and childbirth, and questions related to the neonates. The checklist was checked, revised, and approved by a panel of 3 experts. To determine the reliability of the checklist, it was completed by 2 interviewers with 10 cases in 1 of the health centers. Data were collected by trained interviewers from urban health centers, hospitals (including maternity hospitals), medical records, and interviewing with the parents of neonates in both case and control groups.

Analysis of the data was made by SPSS version 19. First, the univariate analysis was conducted using the Chi-square and Fisher exact test, then, multivariate analysis was performed using logistic regression.

Among all mothers, 2 cases (0.04%) gave birth at home. Of all pregnant mothers, 415 (76.4%) were covered by health centers, 124 (22.8%) were visited by private gynecologist and 4 cases (0.7%) were not covered by any health services during pregnancy. The average distance from households to the closest health center in the case and control groups was 15.8 ± 16.9 and 18.9 ± 19.3 minutes, respectively. Most of the study participants were Kurdistan province residents. There was a significant difference between the place of residence and the NMR (P = 0.027). A significant association was also observed in the maternal education between the case and control groups (P = 0.04). Most of the neonate deaths were observed among illiterate mothers. Moreover, there were no significant relationships between the case and control groups regarding mother’s job (P = 0.7), father’s education (P = 0.1), and father’s job (P = 0.7) (Table 1).

There were no significant differences between the studied groups regarding unwanted pregnancy, number of pregnancies, pre-pregnancy care, high-risk pregnancy, complete vaccination during pregnancy, exposure to the X-rays during pregnancy, and history of medication during pregnancy. However, there were statistically significant differences in maternal passive smoking (OR = 2.1, P = 0.002), history of neonatal mortality (OR = 3.4, P = 0.005), having prior stillbirths (OR = 5.8, P = 0.002), having prior abortion (OR = 1.6, P = 0.05), and history of hospitalization (OR = 1.7, P = 0.02) between the studied groups. In the case group, mothers were more exposed to passive smoking during pregnancy than the control group. Moreover, the history of stillbirth and hospitalization was higher in the case group (Table 2).

There were statistically significant differences in the history of placental abruption (OR = 13.9, P < 0.001), placenta previa (OR = 8.2, P = 0.045), premature rupture of membranes (PROM) (OR = 2.9, P = 0.02) and difficult childbirth (OR = 3.3, P = 0.03) between the 2 studied groups. The placental abruption, placenta previa, PROM, and difficult childbirth were higher in the case group. A significant difference was also observed in umbilical cord problems between the 2 groups (OR = 6.08, P = 0.1) (Table 3).

The association between problems related to high-risk pregnancy and neonatal mortality is detailed in Table 4. There were no statistically significant associations between the 2 studied groups regarding abnormal body mass index (BMI), pregnancy under 18 years old, first pregnancy (first gravidity), mother’s height less than 150 cm, and breastfeeding during pregnancy. In the current study, there were significant associations between the 2 groups regarding pregnancy after 35 years old (OR = 1.79, P = 0.04), pregnancy intervals less than 3 years (OR = 2.5, P = 0.004), 5th and more pregnancies, (OR = 4.6, P = 0.005) and multiple pregnancies (OR = 8.1, P = 0.000). Meanwhile, pregnancy after 35 years, pregnancy intervals of less than 3 years, 5th pregnancy, and multiple pregnancies were higher in the case group (Table 4). No significant association was observed between the mean age of the mothers in the case (232.3 ± 6.0 years) and control groups (696.3 ± 936.5 years), (P = 0.8). Gestational age in the case and control groups was 31.8 ± 1.7 and 38.7 ± 5.8 weeks, respectively (P < 0.001). There was a significant difference in the gestational age and birth weight between the groups. In the case group, the mean gestational age, and birth weight were higher than those of the control group.

One hundred and seventy-six (48.6%) neonates in the control group and 181 (44.8%) neonates in the case group were female and the rest were male. There was no significant difference in the neonatal gender between the studied groups (OR = 1.2, P = 0.4). There were statistically significant differences in the birth weight (less and more than 25000 g), type of childbirth (vaginal or cesarean) (P = 0.004, OR = 1.709), type of feeding (breast or formula milk), (OR = 7.9, P < 0.001), sepsis (OR = 0.300, P < 0.001), asphyxia (OR = 174.6, P < 0.001), breathing problems (such as hyaline membrane disease, respiratory distress syndrome, and pneumonia) (OR = 113.9, P < 0.001,), history of hospitalization (OR = 41.8, P < 0.001), medication (OR = 74.4, P < 0.001), birth injuries (OR = 12.4, P = 0.007), and malformations (OR = 19.2, P < 0.000) between the 2 groups. Therefore, the number of caesarean sections, formula milk feeding, sepsis, asphyxia, breathing problems, need for the neonatal resuscitation, hospitalization history, receiving medication, birth injuries, and malformations were higher in the case group than the control group (Table 5). It is noteworthy that after transferring the variables into the logistic regression model, factors associated with the neonatal death were classified into (i) maternal factors and (ii) the neonatal factors (Table 5).

In 2001, the NMR was 183.3 deaths per 1000 live births in Iran and this rate decreased to 15.3 deaths per 1000 live births, according to the Iran’s Multiple-Indicator Demographic and Health Survey (IrMIDHS) report (15). Achieving the Millennium Development Goals to reduce NMRs requires the identification of risk factors associated with the neonatal mortality. The current study attempted to determine factors influencing the NMR in Kurdistan province, Iran.

According to the IrMIDHS report, as well as other related studies, the NMR was higher in rural areas in 2010 (15). In various studies, lack of access to health centers and long distance between health facilities in rural areas were listed as factors influencing the NMR (16, 17). Moreover, previous studies in rural areas of Iran indicated that the NMR decreased due to the coverage of health services (13, 15). In the present study, the NMR in rural areas was higher than urban areas. In the recent years, the following issues came to light more: Considering the coverage and convenient access to health services in rural areas, improving the quality of maternal and neonatal health care services, application of various approaches focusing on the health services instead of treatment in family physician program, and using the current midwifery personnel after considering in service training.

Several studies investigated the effect of various factors influencing the NMR such as illiteracy, lack of knowledge, and parental employment with the focus on mothers (18, 19). In the current study, the NMR was higher in uneducated mothers. This result was consistent with those of other findings. Results of a national study in 2015 showed that the level of females` health literacy in Kurdistan province was lower than that of the national health literacy (15). Due to the protective effects of mother’s education on the NMR (20, 21), more training programs should be provided to improve the neonatal health care in mothers. Also, employed mothers should be encouraged to spend more time before and after childbirth.

Low birth weight is one of the most important factors associated with the NMR in different parts of the world (4, 22, 23), as well as different areas of Iran (24-26). The current study results also showed that low birth weight (less than 2500 g) increases the risk of death in the neonates by 32.6-fold. The prematurity and low birth weight are other important factors influencing the NMR, which in some cases are associated with economic conditions of the family. To prevent the birth of neonates with less than 2500 g weight, increasing the quantity and quality of health care services in the prenatal period and strengthening of nutritional programs for low income pregnant mothers should be considered as a priority (27, 28).

In the current study, various risk factors including history of abortion (with probability of 1.6), stillbirth (with odds ratio of 5.8), and history of hospitalization (with odds ratio of 1.665) increased the risk of neonatal mortality. The occurrence of placental abruption, placenta previa, PROM, and difficult childbirth was higher in the case group. These factors are associated with the quality of prenatal care and emergency midwifery with particular attention to the high-risk pregnancy. Therefore, it is necessary to pay more attention to the organization of structures, standardization of services in high-risk pregnancy centers, and increasing the knowledge and skills in midwifery personnel in the province.

Several studies on the effect of smoking on newborns showed that cigarette smoke (active or passive) competes with oxygen in reaching the fetus (29, 30). This phenomenon causes adverse effect on pregnancy, increases the costs of health care, sudden death syndrome, and especially low birth weight (31). There was a statistically significant relationship between the passive smoking and NMR. The results of multivariate analysis also showed that passive smoking increased the NMR by 2.121-fold. The effect of smoking on newborn mortality requires further and more complete studies; after inclusion of confounding factors. Pre conception cares for mothers should include dangers of smoking and its high risk in neonatal mortality, the benefits of quitting smoking, and avoidance of smoking. Previous studies reported that maternal age and interval between consecutive births were the effective factors in the NMR (32, 33). Mortality rate in the neonates whose mothers were aged 20-30 years was low, and the interval between the births more than 3 years could reduce the neonatal mortality by 40% (34).

The results of the current study showed that the NMR increased by 1.25-fold in mothers over 35 years. Also, pregnancy interval more than 3 years decreased the mortality rate by 1.25-fold. Regarding the health of mother and child, high rate of fertility is one of the current challenges in Iran. Therefore, the current status of various factors such as increased age of marriage, increased age of the first pregnancy after the marriage, and unwanted pregnancy are considered. To improve maternal and neonatal health, the following points should be emphasized in the educational programs: Marriage at an appropriate age, decreased maternal age of the 1st pregnancy, and increased interval between pregnancies.

Due to its anti-hypothermia and anti-hypoglycemia effects, breastfeeding is one of the most important factors influencing neonatal mortality (35, 36). Breastfeeding is also introduced by WHO as an effective strategy to reduce the mortality rate (37, 38). Moreover, feeding by cow milk is introduced as a risk factor for neonatal mortality in various studies in the world (18, 39). According to the IrMIDHS study, the breastfeeding rate was 53.13% in Iran, and 74.82% in Kurdistan province, in 2010 (15). In this study, feeding method was associated with neonatal mortality. Therefore, the following points play a key role in the improvement of breastfeeding program: Mother’s education on the benefits of exclusive breastfeeding in the first 6 months of life, skin to skin contact in the first hours after birth, and allocation of specialized counseling centers.

Normal vaginal delivery in health centers and under the supervision of a trained person may help to prevent birth injuries and complications of the delivery (40). In previous studies conducted in different parts of the world, it was reported that the cesarean was a risk factor for the neonatal mortality (41-43). In Iran, 45.55% and in the Kurdistan province 38.89% of females give birth by caesarean section, while the normal rate of cesarean is 20% to 30% according to the National Health Improving Program (15). In the present study, cesarean section was a risk factor for the neonatal mortality. Therefore, encouraging mothers to natural childbirth in the health center, developing natural childbirth, and eliminating unnecessary cesarean are the priority programs in the healthcare system.

Neonatal mortality rate is higher in males than females. This may be due to immune deficiency, higher prevalence of respiratory problems, infectious diseases, congenital disorders, and urinary system malformations (24, 44). While, there was no difference in healthcare services between both genders. However, various studies reported that gender was another factor affecting the neonatal mortality and the mortality rate was different between the two genders in different regions (7, 8, 45, 46). It was noted that although greater rate of neonatal mortality in males than females was reported in some studies (7, 13), the current study results were consistent with those of the study conducted in Pakistan (10), the neighboring country of Iran.

Although many different factors were implicated based on the geographic region of neonatal death including pneumonia, diarrhea, infection, asphyxia, trauma, sepsis, congenital abnormalities, metabolic disorders, pulmonary hemorrhage, necrotizing enter-colitis, and meningitis (10, 13, 47-53), the main causes of neonatal mortality in Iran were respiratory distress syndrome, congenital anomalies, sepsis, infection, asphyxia, and pulmonary hemorrhage (47).

One of the limitations of the current study was incomplete data on neonatal mortality. However, through contact with the parents and review of other records, the researchers tried to fill the questionnaires and complete the data as much as possible.

In summary, the current study findings could be utilized to determine priorities, planning, evaluating of services, and improving the health care for mothers and the neonates. In the current study, 3 factors including anomalies, asphyxia, and respiratory problems were the main causes of neonatal mortality. Therefore, educational and therapeutic interventions, as well as improvement of structural defects in the body of healthcare system should be a priority for health policy-makers and health providers.