Women in Somalia face one of the greatest lifetime risks of maternal death in the world, with a one in 22 lifetime risk of maternal death. These lifetime risks are the very main causes of maternal mortality in Somalia, which is among the highest in the world (Geleto et al., 2020). In order to tackle maternal deaths, clear understanding of the magnitude and associated factors of MNM in a given setting is crucial (Hogan et al., 2010). The study has identified that more than 80% of maternal near misses were severe anemia, severe pre-eclampsia, eclampsia, severe APH and abortion complications. This was almost in line with several studies conducted in Harare (Chikadaya et al, 2018; Kumela et al, 2020).
Most surprisingly, this study revealed that severe anemia was the leading cause of maternal near-misses. Although this result is quite contrary to most of the studies, it is consistent with a study conducted in Ethiopia which found almost similar results (Liyew et al, 2018). This may be explained by a number of factors, the first of which is that due to the country's poor health system, most pregnant women do not receive antenatal care. Secondly, many pregnant & lactating women do not have adequate nutrition due low economic profile. Thirdly, there is a common belief in the Somali community that if a pregnant mother eats a well-balanced diet, her baby will grow up and therefore be unable to give birth normally, necessitating surgery. This lead that the mothers deliberately avoid nutritious foods, resulting in severe anemia.
This study identified numerous sociodemographic, obstetric and healthcare factors associated with maternal near misses. Regarding socio-demographic characteristics, it has been found that residence, age, marital status, education, employment and family size were significantly associated with maternal near miss events. It has been revealed that mothers residing in the rural were almost three times more likely to have a maternal near miss event compared to mothers who resided in the urban [OR=2.685, 95%CI: (1.604-4.5640), p=<0.0001]. This can be explained by the fact that majority of the pregnant women in rural areas do not have access to basic maternity services. As a result, they are compelled to seek healthcare to the hospitals & health centers in large cities like, Mogadishu. The findings of this study are consistent with the results of other similar study conducted in Ethiopia's Oromia region (Danel, Graham and Boerma, 2011; Liyew et al, 2018). It is not surprising that the two studies are identical, as the two societies share many commonalities, including socio-economic status. However, on the other hand, the result is quite lower than the result of the same study conducted in Ethiopia, in which rural dwellers were six times more likely to have near misses compared to urban dwellers (Kumela, Tilahun, and Kifle, 2020).
Regarding the maternal age, the study found that odd of maternal near miss was almost three times higher among those younger than 20 years compared to those between 20-35 years of age [OR=2.728, 95%CI:(1.604-4.564). The findings of this study are consistent with other studies conducted in Brazil, Ethiopia and Lao which all found that young age was an important predictor of maternal near miss (CSA, 2012, De Moraes, 2011, Luexay, 2014; FMoH, 2015). This can be explained that young women are more susceptible to childbearing complications such as anemia in pregnancy, pregnancy-induced hypertensions, preterm labour, abortions, septicemia and so on. In addition, many young Somalis marry secretly without the knowledge of the families & communities or the pregnancy may occur outside the context of proper marriage, exposing the young women to adverse social consequences. Furthermore, many young Somalis marry surreptitiously without their families' or communities' knowledge, or pregnancies may occur outside of the framework of a traditional marriage, exposing young women to negative societal implications.
Early marriage and pregnancy are well-known social issues that have been widely discussed in a variety of social and health forums, but they continue to exist in the community, particularly in rural regions and among low-income families.
Similarly, the odd of MNM was somewhat rising with increasing age, however this was not statistically significant [OR=1.270, 95%CI: (0.699-2.306), p=0.433]. This is in line with the results of other studies conducted in Africa and Latin America (Tenaw et al, 2021; Oliveira etal, 2014).
According to study findings, women who were not in marriage contract at the time of the study were two times more likely to have maternal near miss event compared to those in marriage contract [OR=2.18, 95%CI, (1.247-3.81, p=0.006)]. The finding of this study regarding marriage contract is consistent with another study conducted in Addis Ababa, Ethiopia which found almost similar result (Liyew etal, 2018).
Moreover, the odds of maternal near miss were 3 times higher among women without formal education. It is not surprising that low-educated women face the greatest risk of maternal near miss as they are always the vulnerable to health risks. The result of this study coincided with that of Liyew et al (2018) which discovered that illiteracy increases maternal near miss by more than three folds.
Regarding husband’s employment, it has been identified that the odds of maternal near-miss events were three times higher among women whose husbands were unemployed compared to those whose husbands were employed [OR=2.992, 95%CI:(1.886-4.745, p<0.0001]. Although no studies investigating the association between husband's employment and maternal near miss events were conducted, this result could be explained by the fact that when husbands are unemployed, the pregnant woman becomes the family's breadwinner, and as a result of this double burden, the pregnant mother experiences a variety of problems, including a higher incidence of maternal near miss events.
Moreover, the study has revealed that the odds of developing maternal near-miss events among women with family income of less than 100 USD were 3.33 times more compared to those with more than 500 USD [OR=3.333, 95%CI= (1.055-10.530)]. The result of this study was consistent with the work of Asaye (2020) divulged that women with the lowest monthly income (1000 ETB) had 3.99 times the chance of having maternal near-miss than those with a monthly income more than or equal to 3001 ETB. This is not surprising since over 99% of maternal deaths occur in low- and middle-income countries due to extreme poverty resulting in lack of access to quality healthcare and education of women (WHO, 2015). The magnitude of maternal near miss varies between and within countries; however, the highest rates are found in low- and middle-income countries (Abdel-Raheem etal, 2016; Yasmin etal, 2016, Jyoti and Garima, 2016 and Assarag etal, 2015). Education increases women’s access to relevant information and may facilitate the financial means required to pay for transportation to care (Mekango et al, 2017).
Regarding obstetric factors, the study revealed that maternal near miss events were associated with the maternal age at marriage, maternal age at first pregnancy, mother’s birth interval, previous history of obstetric complication and last birth outcome.
This has found that age at marriage and age at first pregnancy were significant predictor of maternal near-miss occurrence. The odd of maternal near-miss occurrence was 3.4 times higher among women who had married at the age of 15 years or earlier than those aged twenty-one years (OR=3.4, 95%CI, 1.912-6.091). On the other hand, when one's age approached 35, there was a growing odd (Oliveira et al, 2006). The observed association between maternal age at marriage and first pregnancy and maternal near miss is consistent with another study in Ethiopia (Mekango et al, 2017).
The study has also shown that odds of maternal near-miss were 3.1 times higher among women who had their first pregnancy before 18 years of age compared to those who had their first pregnancy after 18 years of age (OR=3.091, 95% CI, 2.044-4.674). Almost similar study results were presented by several studies (Habte and Wondimu, 2021; Adedokun, Adeyemi, and Dauda, 2016). This could be explained that younger women are often not physically capable of childbearing. Furthermore, girls married early are more likely to experience violence, abuse and forced sexual relations due to unequal power relations, exposing women to adverse social consequences. They are also vulnerable to sexual transmitted infections and severe pregnancy symptoms.
The study has also shown that the odds of a maternal near-miss were six times higher among women with a birth interval of less than two years compared to those with a birth interval of more than two years [OR=5.922, 95%CI, (3.891-9.014)]. This is in line with several studies conducted in several counties in East Africa (Habte and Wondimu, 2021; Mekango et al. 2017; Bauserman et al, 2020).
Another important predictor those with previous history of obstetric complications. This showed that the women with previous history obstetric complications were 6.5 times more likely to have maternal near miss than those without history of obstetric complications [OR=6.568, 95%CI, (4.286-10.066]. Several studies found the almost same results (Sikder, et al, 2014; Mekango et al. 2017). On the other hand, contrary to study conducted in Bangladesh, the study has found that maternal near miss events were not significantly associated with maternal gravidity and parity (Sikder et al, 2014). This can be attributable to many factors, including differences in the socio-economic situations and healthcare systems of the two countries.
With respect to the association between respondent’s healthcare service characteristics and maternal near miss events, the study has identified that odd of maternal near miss events were associated with ANC attendance, ANC booking, means of transportation, referral point, autonomy in seeking medical help, delays and birth outcomes.
In terms of ANC attendance, the odds of maternal near-miss occurrence were 2.7 times higher among women who never received ANC services [OR=2.687, 95%CI, (1.802-4.006)]. This finding is consistent with many other studies conducted in several places in the world (Tenaw, et al, 2021; Kumela, Tilahun,and Kifle, 2020; Dessalegn et al, 2020). Moreover, the study has also found out that odd of maternal near miss events were 70% lower among pregnant women who booked ANC services before 12 weeks of gestation compared to those who never booked antenatal services at all (OR=0.301, 95%CI, 0.197-0.459) and 11% lower among pregnant who booked ANC services late gestation compared to those who never booked antenatal services at all [OR=0.894, 95%CI, (0.514-1.554)]. These findings are consistent with several studies conducted in several countries (Habte and Wondimu, 2021; Kabakyenga et al, 2011; Mekango et al. 2017; Assarag et al, 2015). This can be explained that utilization of ANC reduces the maternal morbidity and mortality rates by screening high-risk mothers for complications and facilitating a rapid diagnosis and management of life-threatening obstetric conditions (Carroli, Rooney and Villar, 2001).
Similarly, the odds of maternal near events were 4.7 times higher among mothers giving birth in non-health facility setting compared to those who gave birth in health facilities [OR=4.672, 95%CI, (3.105-7.029)]. This is in line with a study conducted in Morocco (Assarag et al, 2015). The advantages of hospital birth over home delivery are widely established, including adequate pain management, access to a NICU in the event of an emergency, professional staff assistance, and the availability of advanced interventions.
Women who were non-autonomous in seeking medical help were 3.5 times more likely to have maternal near-miss events [OR=3.538, 95%CI, (1.468-8.524)]. This is consistent with the work of Habte and Wondimu (2021). This could be explained that respecting the autonomy of the women allows them to make decisions that are in their best interests, as they are usually the best judges of those interests.
The study discovered that women who took more than an hour to reach health facility were almost two times more likely to be maternal near miss than whose who arrived in less than one hour [OR=1.773, 95% CI, (1.212-2.595)]. Although statistically insignificant, the other two delays also exhibited a greater odd. The findings of this study correspond to the results of other studies conducted in some parts of Africa like Ethiopia and Morocco (Habte and Wondimu, 2021; Assarag etal, 2015). Delays in obtaining care were collected according to the 3-delay model (Thaddeus and Maine, 1994), which was adapted as follows. Firstly, delay in decision to seek care as a consequence of women’s low socio-economic status, lack of understanding of life-threatening complications and risk factors in pregnancy and when to seek medical help, previous poor experience of health care and financial implications. Secondly, delay in reaching care due to; distance to health centres and hospitals, unavailability of transportation, lack of cost of transportation, insecurity and poor roads and infrastructure. Thirdly, delay in accessing and receiving adequate health care mainly due to poor facilities and lack of medical supplies, inadequately trained and poorly motivated medical staff and inadequate referral systems (Maternity worldwide, 2014).
A lot of reasons contribute to this, including poor public infrastructure such as roads, low community awareness, and a health system that is too limited in terms of availability, accessibility, affordability and skilled birth attendants especially in remote areas.
In this study, still birth as the outcome of last pregnancy was positively related to maternal near-miss [OR=5.543, 95%CI: (2.880-10.668)] and is compatible with studies conducted in Nigeria and Ethiopia (Adeoye, Onayade and Fatusi, 2013; Habte and Wondimu, 2021).
This can be explained by the fact that women with a history of stillbirth may experience a variety of psychological and relational issues, which may raise the risk of maternal complications in subsequent pregnancies. The link between maternal chronic hypertension and stillbirth may also be an alternative explanation. Moreover, the argument of Todd et al was handy here, in which women who have experienced a stillbirth might have a history of chronic hypertension, and thereby increase the odds of maternal near-miss. Todd et al. in their study on correlates of severe acute maternal morbidity in Kabul also established that prior stillbirth or spontaneous abortion was an immense predictive factor for maternal near-miss (Todd et, 2015).
Strength and limitation of the study
The strength of the study is that it is the first of its kind in Somalia to document the determinants of maternal near-miss events among women admitted to tertiary hospitals using the WHO case identification criteria of near-miss mothers. The study looked into the women's experiences as well as the timeliness of their medical care using multiple sources of information like patient cards, hospital records and interviews with women represented another strength of our study. Furthermore, using a regression analysis model that incorporates interrelationships between variables reduces the likelihood of erroneous conclusions.
The selection of women was limited to those who reached the study hospitals. On the other hand, even though the study adapted WHO’s maternal near-miss criteria, due to limited resources, few advanced laboratory investigations and management-based criteria were not used, however, this had no impact on diagnosis and outcome since all other clinical-based criteria were applied.