Burden and Determinants of Child Mortality in Somalia analysis of SDHS 2020

doi:10.21203/rs.3.rs-4958429/v1

Download PDF

Article

Burden and Determinants of Child Mortality in Somalia analysis of SDHS 2020

https://doi.org/10.21203/rs.3.rs-4958429/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Child mortality remains a significant public health burden in Somalia. There are many contributing factors for child mortality which varies across contexts. Limited UNICEF studies done in Somalia addressing level of child mortality in some part of country. Thus, this study investigates country wide prevalence of child mortality and important determinants of maternal and child characteristics using data from 2020 Somali Demographic and Health Survey (SDHS).

Stata version 16 with command of Prop49, 037 survivors (size) 3,161(outcome), exact (95%CI) was used to estimate the prevalence of child mortality rate using data extracted from SDHS, 2020. Binary logistic regression model is used to analyze the determinants of child mortality in Somalia after cleaning. Variables with P-values less 0.05 with 95% confidence intervals (C.I.) and Adjusted Odds Ratio (AOR) vary from 1 were reported as determinants.

Study shows prevalence of child mortality in Somalia of 65 per 1000 live births. Whereas no maternal visit to a health facilities in the previous 12 months, living in rural settings, short preceding and succeeding birth intervals, and multiple births were significantly increasing the odds of child mortality in Somalia respectively: (AOR = 1.57(0.97, 2.53), (AOR = 2.034, 95% CI: 1.15–3.61), 0.49(0.26, 0.94) and (AOR = 0.44, 95% CI: 0.29–0.66), and (AOR = 4.35, 95% CI: 1.919–23.73).

Prevalence of child mortality is critical burden in Somalia and this highlights poor situation of health care delivery. All identified risk factors are preventable by just improving maternal and child health access and quality of care. Improving key maternal and child health interventions including family planning service with due emphasis given on rural/nomad areas are critical measures for improving child survival in the country.

Health sciences/Risk factors

Health sciences/Health care/Paediatrics

Health sciences/Health care/Public health

child mortality

Somalia

SDHS

determinants

burden

The under-five mortality rate is the probability that an infant will die before reaching the age of five, expressed per 1,000 live births. It remains public health problem particularly low-income countries despite high impact intervention made ^{1, 29}. The burden of child mortality is problematic is Somalia where 106.1 deaths happen annually² which is off track to achieve the global target of 25 per 1,000 live births by 2030(SDG). Globally determinant factors of child mortality are multilevel and varies across contexts: maternal health service utilization in China and low- and middle-income countries ^{(3, 4)}. Residing in rural areas and distance from health care facilities ^{5, 6} in Nigeria and Burkina Faso, season, short birth interval in Malawi and Kenya, ^{7, 8} maternal undernutrition and young age in Latin America. ^{9, 10}

In Somalia, health conditions are exacerbated by instability, poor healthcare system, and pervasive poverty as well as weak governance. ^{11, 12} Recent finding provides generalized solutions that do not fully address the unique challenges faced by child health within Somalia. ^{13, 14, 15} To reduce child mortality, it is essential to bridge this gap by conducting more detailed analysis of local factors. Identifying these specific determinants will enable the development of targeted, contextually relevant interventions to improve child survival outcomes.¹⁶ Child mortality is poorly understood in Somalia and evidence regarding determinants remains unknown. Thus, aim of study is to identify the prevalence and determinants of child mortality in Somalia.

Data

The 2020 surveys provided the data sets used in this study. 3,161 deaths children under the age of five died out of the total 52,198 children registered in the SDHS data collection. Only women of reproductive age (15–49 years old) who had previous history of birth prior to the time of the survey and Important determinant variables with complete data were taken into account in the study's analysis.

Study design and period

A secondary data analysis was done based on first and recent Demographic and Health Survey (DHS) data of Somalia 2020 to investigate prevalence of child mortality and its determinants. SDHS 2020 is a community-based cross-sectional and nationally representative survey conducted by the Somalia National Bureau of Statistics between January 2018 and February 2019.

Sample size

A total of 52,198 children (3,161 deaths and 49,037 survivors) that is extracted from the 2020 SDHS

Included in the study.

Variable

The outcome variable of interest is child mortality happen before the age of 5 years which is categorized into dead with code 1 and alive with code 0. The selection of potential risk factors for childhood mortality was based on the framework developed by Mosley and Chen¹⁷ to describe factors impacting child survival in developing nations. The maternal level variables considered in this study are place of residence (urban vs rural including nomad), level of education (no formal education, primary, secondary and higher), and wealth index score (Lowest, Second, Middle, fourth, highest), maternal visit at health facility in the past year before the survey (yes, no). The child characteristics considered in the study are birth type (singleton or multiple), sex of the child (male or female), preceding birth interval (< 19 months label short vs 19–36 and more), Succeeding birth interval (< 20 months label short vs 20 and above) and birth order (2 vs 3–4)

Household wealth index is measured by collecting information on dwelling and household characteristics, consumer goods, and assets. Each household asset for which information is collected is assigned a ‘weight’ or ‘factor score’ generated through principal components analysis. The household wealth index variable was categorized into five quintiles (Lowest, Second, Middle, fourth, highest).

Data analysis

Data was extracted from SDHS, 2020 then cleaned and analyzed in SPSS version 20. A logistic regression model is used to analyze the determinants of child mortality in Somalia based on the data provided. To determine the significance of each variable, the analysis includes both crude odds ratios (COR) and adjusted odds ratios (AOR), with 95% confidence intervals (C.I.) and P-values. The variables found to be significant (P-value < 0.05) were reported as determinants of child mortality. Additionally, Stata version 16 with command of Prop49, 037 survivors (size) 3,161(outcome), exact (95%CI) was used to estimate the prevalence of child mortality rate.

Ethical consideration

The 2020 DHS of Somalia's data, which we got access upon request from Somalia National Bureau of Statistics, was used for the study. Since secondary sources provided the data for the analysis, no ethical considerations were made.

Maternal and child characteristics determinants of child mortality in binary logistic regression

Important maternal and child characteristics were included to assess determinants of child mortality. Households with lowest wealth index score experienced higher proportion (41.4%) of mortality rate compared to counterparts: second (18.9%), middle (17%), fourth (12.9%) and highest (9.8%). Out of 3,161 children who died, less proportion (21.3%) of child death is observed for mothers visited health facility in the past 12 months prior survey compared to those didn’t visit (78.7%). It is found that higher proportion (61.4%) of dead children lived in rural areas, compared to those lived at urban (38.6%). Higher mortality (54.0%) is observed for children had short (less than 19 months) preceding birth interval, compared to children with longer birth intervals, 19–23 months: 14.1%, 24–35 months: 21.6%, and ≥ 36 months:10.3% respectively.

Children with short succeeding birth interval (less than 20 months) has higher proportion (55.6%) of mortality compared to mortality of children with long interval (44.4%). Moreover, being multiple children at birth and being baby girl are linked lower survival rate; 1.9% and 47.3% compared to counterparts respectively; 98.1% and 52.7%. Additionally, being multiple birth (twin, triple), baby girl and birth order of 3 and above are increasing the risk of child mortality.

Determinants of Child Mortality

Variables considered into multivariate logistic regression model are all those considered in Bi-variate model regardless of its P- value due to its relevance in the literature. Forward logistic Regression method was also applied and below determinants remain in the last model.

Children whose mothers not visited health facility in past 12 months prior survey are 1.57 times more likely to die then counterparts: AOR = 1.57, however this association is marginal insignificant(P = 0.066). Children in rural locations are more than twice the risk of child death compared to those in urban residence, and this finding is statistically significant (AOR is 2.034, P-value = 0.015).

A child with longer preceding interval of 19–23 months is associated with a lower likelihood of child death than child with short interval of (less than 19 months). The AOR for this interval is 0.49, showing a 51% reduction in the probability of mortality, and the P-value is 0.031, making the finding significant. A longer succeeding birth interval of 20 months or more is significantly decreasing risk of child mortality compared to an interval of less than 20 months. The AOR is 0.44, showing a 56% reduction in the risk of mortality (P-value < 0.001). Multiple births (twins, triplets, etc.) have a much higher risk of child mortality than single births. Multiple births are associated with a significantly higher risk of mortality (AOR = 4.35, P < 0.001). (Table 2).

Table 1

Bi-variable logistic regression analysis of maternal and child characteristics by child mortality
Characteristics		Child status at the time of interview		Child status at the time of interview	P-value
Characteristics		Dead (%), n = 3161	survived (%), n = 45876
Wealth Index score	Lowest	1308(41.4%)	17958(39.1%)	1
	Second	597(18.9%)	8969(19.6%)	0.88(0.77, 0.99)	0.045
	Middle	538(17.0%)	7674(16.7%)	0.96(0.83, 1.11)	0.572
	Fourth	409(12.9%)	6439(14.0%)	0.91(0.79, 1.05)	0.208
	Highest	309(9.8%)	4836(10.5%)	1.00(0.86, 1.17)	0.939
Mothers education	No education	2751(87.0%)	39518(86.1%)	1
	Primary	322(10.2%)	4796(10.5%)	0.66(0.40, 1.09)	0.108
	Secondary	72(2.3%)	1215(2.6%)	0.69(0.41, 1.15)	0.152
	Higher	16(0.5%)	347(0.8%)	0.78(0.45,1.35)	0.375
Maternal visit at health facility in the past 12 months	Yes	64(21.3%)	1264(29.2%)	1
Maternal visit at health facility in the past 12 months	No	237(78.7%)	3064(70.8%)	1.53(1.15, 2.03)	0.003
Place of residence	Rural	1941(61.4%)	27455(59.8%)	1.07(0.99, 1.15)	0.084
Place of residence	Urban	1220(38.6%)	18421(40.2%)	1
Preceding child interval	< 19	791(54.0%)	11856(33.7%)	1
	19–23	207(14.1%)	5293(15.0%)	0.32(0.66, 0.37)	< 0.001
	24–35	317(21.6%)	10922(31.0%)	0.54(0.44, 0.67)	< 0.001
	≥ 36	151(10.3%)	7139(20.3%)	0.73(0.59, 0.88)	< 0.001
Birth Order	2	1605(50.8%)	18515(40.4%)	1
	3	438(13.9%)	7578(16.5%)	1.50(1.35,1.67)	< 0.001
	4	1118(35.4%)	19783(43.1%)	1.53(1.42,1.66)	< 0.001:
Succeeding birth interval	< 20	886(55.6%)	12558(35.8%)	1
Succeeding birth interval	≥ 20	707(44.4%)	22525(64.2%)	0.44(0.40, 0.49)	< 0.001
Birth type	Singleton	28851(98.1%)	19271(98.1%)	1
Birth type	Multiple	545(1.9%)	370(1.9%)	3.682(3.11, 4.35)	< 0.001
Child Sex	Male	1818(57.5%)	24156(52.7%)	1
Child Sex	Female	3161 (42.5%)	45876 (47.3%)	1.22(1.13, 1.31)	< 0.001

Table 2

Multivariable logistic regression analysis of maternal and child characteristics by child mortality.
Mother and child Characteristics		Child status at the time of interview		COR 95% C.I	AOR 95% C.I	P-value
Mother and child Characteristics		Dead (%), n = 3161	Survived (%), n = 45876
Maternal Visit at health facility past 12 months	Yes	64(21.3%)	1264(29.2%)	1
Maternal Visit at health facility past 12 months	No	237(78.7%)	3064(70.8%)	1.53(1.15, 2.03)	1.57(0.97, 2.53)	0.066
Place of residence	Rural including nomad	1941(61.4%)	27455(59.8%)	1.07(0.99, 1.15)	2.034(1.15, 3.61)*	0.015
	Urban	1220(38.6%)	18421(40.2%)	1	1
Child characteristics
Preceding child interval	< 19	791(54.0%)	11856(33.7%)	1
	19–23	207(14.1%)	5293(15.0%)	0.32(0.66, 0.37)	0.49(0.26, 0.94)*	0.031
	24–35	317(21.6%)	10922(31.0%)	0.54(0.44, 0.67)	0.96(0.41, 2.26)	0.928
	≥ 36	151(10.3%)	7139(20.3%)	0.73(0.59, 0.88)	0.97(0.47,2.02)	0.947
Succeeding birth interval	< 20	886(55.6%)	12558(35.8%)	1
Succeeding birth interval	≥ 20	707(44.4%)	22525(64.2%)	0.44(0.40, 0.49)	0.44(0.29, 0.66)**	< 0.001
Birth type	Singleton	28851(98.1%)	19271(98.1%)	1
Birth type	Multiple	545(1.9%)	370(1.9%)	3.682(3.11, 4.35)	4.35(1.94, 9.73)**	< 0.001
** means P-value < 0.001, * means P-value < 0.05

Prevalence of child mortality

The magnitude of child mortality rate of Somalia was estimated to be 65; 95% CI (62.32–66.67) per 1000 live births. That is 65 children couldn’t celebrate their 59 months in every 1000 live births. It was estimated from 49,037 live births with 3,161 children dead before reaching 59 months of age. The data is extracted from the 2020 Somali Demographic and Health Survey (DHS).

In our present study, the magnitude of child mortality rate is 65; 95% CI (62.32–66.67) per 1000 live births in Somalia. The study also identified maternal visit at health facilities, rural residence, birth interval (children born before and after inadequate birth interval) and multiple birth as an important determinant factors of child mortality in Somalia.

65 children are dying before celebrating their 5th year of live which reflects poor health care delivery and nutritional status of children in the context. This child mortality is much higher than 2022 global estimate of child mortality (37 every 1000 live birth)¹ and east African countries including Ethiopia (47 every 1000 live birth).^17,18 This must be due to improper management of preventable childhood diseases resulted from poor access of health care service.¹⁹

Our study shows maternal visit at health facilities is pretty much reducing child mortality. this is in line with studies done in China and low and middle-income countries.^3,4,20 This might be the fact that maternal visit at health facilities exposes mothers and their children to high impact maternal and child health interventions that reduces both maternal and child deaths.³

The current study identified living in rural settings increases the odds of child mortality in Somalia. The finding is consistent with meta-analysis study done in low and middle-income countries.²¹ However the finding is contradicting to studies done in India and Tanzania^19,22 this must be due to different health care access in different context. Access to health care in Somalia is problematic and this gets more difficult in rural setting.^14,23

In our study shows children born before and after inadequate birth interval have higher mortality rate compared to children with optimum birth interval. It is consistent with study done in Malawi and Kenya where both before and after short birth intervals reduce child survival.^7,8 This must be due to poor birth outcome like low birth and preterm as well as maternal undernutrition are linked to shorter birth intervals and ultimately higher under-five death rates.^24,25 Additionally, short birth interval less than WHO recommended 33-month birth-to-birth interval is associated with maternal undernutrition,²⁴ that eventually results adverse birth outcome according to large scale systematic review done in Iran.The finding is also consistent with studies done in Pakistan,²⁶ East Africa,²⁷ Kenya,⁷ Ethiopia,²⁸ Bangladesh,²⁹ and sub-Saharan Afica.³⁰ However, the finding is contradicting with study done in Bangladesh that shows child born after a short birth interval has higher chances to survive its infancy if the previous sibling died than if it survived due to extensive family health service presence.³¹

Moreover, the current study reveals that multiple births increase the risk of child death compared to children born with single birth. This finding is consistent with studies done in Rwanda, Togo and Ethiopia.^32–34 This is the fact that multiple pregnancy/birth compromises both maternal and child nutrition and health status.^34,24

The study finding is estimated from nationwide recent survey that has acceptable external validity(representative) to all children in Somalia. Despite this strength, the study finding is secondary analysis of first Somalia DHS which is subject to measurement and editorial errors happen during data collection. To overcome this limitation, appropriate measures were taken during variable selection and analysis.

The finding of the study reflects the burden of child mortality and existing health care access/delivery in Somalia. It also addresses behavior of women in reproductive age regarding child spacing.

The burden (prevalence) of child mortality in Somalia is very high and its determinant risk factors are preventable. The study finding highlights the importance of improving health care access and quality to address child mortality. The study also capitalizes the importance of birth spacing as child survival intervention strategy since context experiences high fertility rate with unmet need of contraceptive and the double risk of short birth intervals (preceding and succeeding birth intervals). Thus, ministries of health and its supporting partners must improve access (universal health coverage) and utilization of health care particularly rural setting. They must also improve the uptake of family planning service by working its acceptability to address short birth intervals.

Acknowledgments

We are grateful for the access to the Somali Health and Demographic Survey 2020 provided by the Somalia National Bureau of Statistics.

Author Contributions

Conceptualization: Hamze G. Dahir, Ahmed T. Ahmed, Mohamed A. Abdilahi.

Formal analysis: Hamze G. Dahir, Ahmed T. Ahmed.

Methodology: Hamze G. Dahir, Ahmed T. Ahmed, Mohamed A. Abdilahi

Writing, Review and Editing: Hamze G. Dahir, Ahmed T. Ahmed, Mohamed A. Abdilahi. All authors reviewed the manuscript.

Competing interest

The author(s) declare no competing interests.

Data availability

The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.

UNICEF. Key demographic indicators,(2024). https://data.unicef.org/country/som/
UNICEF. Monitoring the situation of children and women. (2024). https://data.unicef.org/country/som/
Karra, M., Fink, G. & Canning, D. Facility distance and child mortality: a multi-country study of health facility access, service utilization, and child health outcomes. Int. J. Epidemiol. 46 (3), 817–826 (2017).
Zhang, J. et al. Association between maternal health service utilization and under-five mortality rate in China and its provinces, 1990–2017. BMC Pregnancy Childbirth. 24 (1), 326 (2024).
Ezeh, O. K. et al. Risk factors for postneonatal, infant, child and under-5 mortality in Nigeria: a pooled cross-sectional analysis. BMJ Open. 5, e006779 (2015).
Heiko, B. et al. Risk factors of infant and child mortality in rural Burkina Faso. Bull. World Health Organ. 82, 265–273 (2004).
Fotso, J. C., Cleland, J., Mberu, B., Mutua, M. & Elungata, P. Birth spacing and child mortality: an analysis of prospective data from the Nairobi urban health and demographic surveillance system. J. Biosoc. Sci. 45 (6), 779–798 (2013).
Manda, S. O. Birth intervals, breastfeeding and determinants of childhood mortality in Malawi. Soc. Sci. Med. 48 (3), 301–312 (1999).
Felisbino-Mendes, M. S., Moreira, A. D. & Velasquez-Melendez, G. Association between maternal nutritional extremes and offspring mortality: a population-based cross-sectional study, Brazil, demographic health survey 2006. Midwifery. 31 (9), 897–903 (2015).
Myrskylä, M. & Fenelon, A. Maternal age and offspring adult health: evidence from the health and retirement study. Demography. 49, 1231 (2012).
Hassan-Kadle, A. A. et al. One Health in Somalia: Present status, opportunities, and challenges. One Health. 18, 100666. 10.1016/j.onehlt.2023.100666 (2023). PMID: 38226139; PMCID: PMC10788489.
World Bank, world Corruption index(. (2023). https://qz.com/somalia-is-the-most-corrupt-country-in-the-world-1850063626
WHO, Child mortality (under five) fact sheets. (2022). https://www.who.int/news-room/fact-sheets/detail/levels-and-trends-in-child-under-5-mortality-in-2020
Gele, A. Challenges facing the health system in Somalia and implications for achieving the SDGs. Eur. J. Pub. Health. 30 (Supplement_5), ckaa165–1147 (2020).
Morrison, J. & Malik, S. M. Health equity in Somalia? An evaluation of the progress made from 2006 to 2019 in reducing inequities in maternal and newborn health. Int. J. Equity Health. 23 (1), 46 (2024).
Mohamoud, S. A. et al. Determinants and prevalence of zero-dose children in Somalia: Analysis of the 2020 Health Demographic Survey data. PLOS Global Public. Health. 4 (7), e0002612 (2024).
Mosley, W. H. & Chen, L. C. An analytical framework for the study of child survival in developing countries. Popul. Dev. Rev. 10, 25–45 (1984).
Ethiopia, M. Demographic and Health Survey (2019).
Srivastava, S., Upadhyay, S. K., Chauhan, S. & Alagarajan, M. Preceding child survival status and its effect on infant and child mortality in India: An evidence from National Family Health Survey 2015–16. BMC Public. Health. 21 (1), 1–13. 10.1186/s12889-021-11569-z (2021).
Ahmed, T. et al. Healthcare utilization and maternal and child mortality during the COVID-19 pandemic in 18 low-and middle-income countries: An interrupted time-series analysis with mathematical modeling of administrative data. PLoS Med. 19 (8), e1004070 (2022).
Forde, I. & Tripathi, V. Association of place of residence and under-five mortality in middle-and low-income countries: A meta-analysis. Children. 5 (4). 10.3390/children5040051 (2018).
Ogbo, F. A. et al. Determinants of trends in neonatal, post-neonatal, infant, child and under-five mortalities in Tanzania from 2004 to 2016. BMC Public. Health. 19 (1), 1–12. 10.1186/s12889-019-7547-x (2019).
Bogren, M. et al. Health workforce perspectives of barriers inhibiting the provision of quality care in Nepal and Somalia–A qualitative study. Sex. Reproductive Healthc. 23, 100481 (2020).
Karami, B., Abbasi, M. & Tajvar, M. Determinants of Neonatal, Infant and Child Mortalities in Iran: A Systematic Review. Iran. J. Public. Health. 53 (1), 104–115. 10.18502/ijph.v53i1.14687 (2024).
Rutstein, S. O. Effects of preceding birth intervals on neonatal, infant and under-five years mortality and nutritional status in developing countries: evidence from the demographic and health surveys. Int. J. Gynecol. Obstet. 89, S7–24 (2005).
Asif, M. F. et al. The Interaction Effect of Birth Spacing and Maternal Healthcare Services on Child Mortality in Pakistan. Children. 10 (4), 1–11. 10.3390/children10040710 (2023).
Tesema, G. A., Worku, M. G. & Alamneh, T. S. Estimating the impact of birth interval on under-five mortality in east african countries: a propensity score matching analysis. Published online 2023:1–10 .
Gebremichael, S. G. & Fenta, S. M. Under-Five Mortality and Associated Risk Factors in Rural Settings of Ethiopia: Evidences from 2016 Ethiopian Demographic and Health Survey. Adv Public Heal. ;2020. doi: (2020). 10.1155/2020/8430246
slam Rahman, M. Z. & Khan, M. M. Effects of short birth interval on different forms of child mortality in Bangladesh: Application of propensity score matching technique with inverse probability of treatment weighting. PLoS One. 18 (4 April). 10.1371/journal.pone.0284776 (2023).
Yaya, S., Bishwajit, G., Okonofua, F. & Uthman, O. A. Under five mortality patterns and associated maternal risk factors in sub-Saharan Africa: A multi-country analysis. PLoS One. 13 (10), 1–14. 10.1371/journal.pone.0205977 (2018).
Van Soest, A. & Saha, U. R. Relationships between infant mortality, birth spacing and fertility in Matlab, Bangladesh. PLoS One. 13 (4), 1–21. 10.1371/journal.pone.0195940 (2018).
Nuwabaine, L. et al. Factors associated with anaemia among pregnant women in Rwanda: an analysis of the Rwanda demographic and health survey of 2020. BMC Pregnancy Childbirth. 24 (1), 1–20. 10.1186/s12884-024-06528-6 (2024).
Levano, S. R. et al. Identifying determinants of under-five child mortality in northern Togo. J. Glob Health. 14 (February). 10.7189/JOGH.14.04019 (2024).
Gebrerufael, G. G. & Hagos, B. T. Predictors of mortality among under-five children in rural Ethiopia: a cross sectional study. BMC Pediatr. 23 (1), 1–11. 10.1186/s12887-023-04440-0 (2023).

No competing interests reported.

Download PDF

Reviewers invited by journal
27 Sep, 2024
Editor assigned by journal
27 Sep, 2024
Editor invited by journal
02 Sep, 2024
Submission checks completed at journal
30 Aug, 2024
First submitted to journal
22 Aug, 2024

You are reading this latest preprint version

Burden and Determinants of Child Mortality in Somalia analysis of SDHS 2020

Status:

Version 1

Abstract

Introduction

Methods

Data

Study design and period

Sample size

Variable

Data analysis

Ethical consideration

Results

Maternal and child characteristics determinants of child mortality in binary logistic regression

Determinants of Child Mortality

Prevalence of child mortality

Discussion

Conclusions

Declarations

References

Additional Declarations

Status:

Version 1