Effect of Selected Variables on Child Mortality in Ethiopia using 2019 EMNDHS: Application of Zero Inflated Negative Binomial Mixed Effect Regression Model

doi:10.21203/rs.3.rs-4807993/v1

Download PDF

Research Article

Effect of Selected Variables on Child Mortality in Ethiopia using 2019 EMNDHS: Application of Zero Inflated Negative Binomial Mixed Effect Regression Model

https://doi.org/10.21203/rs.3.rs-4807993/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

. Approximately 6.3 million under 5 age of children were died in 2013. An estimate of 2.4 million child die before their first month of birth. Struggling for the lives of new-born children must be the global concern. Under five child mortality rates was decreased from 32% in 1990 to 18% in developed country but it was decreased with lower rates in developing countries especially in Sub Sharan African and southern Asian countries. maternal and household characteristics such as place of child residence in rural areas, poor wealth index of household and uneducated mothers had the higher risk deaths of under-five child mortality rate. The data source for this analysis was 2019 EMDHS data. To analyse the effects of selected variables on under five child mortality, zero inflated negative binomial mixed effect regression model were appropriate. Goodness of fit model were based on AIC value and smaller AIC values were good fitted model. In This investigation the variables age of mother, age of mother at 1st birth, mother’s education level, types of cooking fuel, number of living children and types of birth were statistically significant factor for non-zero death of under-five child mortality in Ethiopia.

Zero inflated negative binomial

under-five child mortality

Ethiopia

Mixed Model

Reducing child mortality is a tool to compare the socio-economic improvement between countries and it is an indicator for improved quality of livelihood life [1]. An estimate of 2.4 million child die before their first month of birth [2]. Struggling for the lives of new-born children must be the global concern. A study done 2013 give an evidence that approximately 6.3 million under 5 age of children were died [3]. According to [3] under five child mortality rates was decreased from 32% in 1990 to 18% in developed country but it was decreased with lower rates in developing countries especially in Sub Sharan African and southern Asian countries. maternal and household characteristics such as place of child residence in rural areas, poor wealth index of household and uneducated mothers had the higher risk deaths of under-five child mortality rate [3].

Maternal biography, household’s exposure for media and child characteristics were an indicator of urban-rural under five child mortality gaps[4]. This difference of gaps was largest in Ethiopia and Kenya [4]. Chad, Sierra Leone and Cameroon had 133, 156 and 184 rates of child mortality per 1000 live births respectively among 35 Sub Saharan African countries studied by[5] from. Residency, maternal education status, age at first birth of mother [4], household wealth index sources of drinking water, delivery place, sex of children, preceding birth intervals, types of multiple birth and birth order were the major determinants of under-five child mortality rates [1], [6], [7], [8], [9], [10], [11]. In Kenya, maternal reproductive health and child and maternal nutritional status were a significant factor for child mortality [10]. Child mortality was related with the prevalence of malaria, baby postnatal care, health scheme[12] and breastfeeding [13]. Using Demographic Health Survey (DHS) data in Ghana, the odds of child mortality are higher for non-contraceptive use/intention as compare contraceptive [8], [14].

Factors such as place of residence province-wise, household head’s education level and source of drinking water have negative effect (high risk) on child mortality in Sri Lanka using logit regression model [15]. Ethiopia is one of the six leading countries that accounts half the global under five deaths [1]. The trends of child mortality rate in Ethiopia were 216, 163, 123,88 and 59 per 1000 live birth by year of 1990–9995, 2000, 2005, 2011 and 2016 based on EDHS data sources[16]. Many researchers done researches a bout mortality. A research done by[14], [17] using 2016 EDHS, the predictors of child mortality in rural Ethiopia were regions, household education level, multiple birth delivery place, occupation of parents, age of mother at first birth, month of breastfeeding, use of a contraceptive method, child vaccination status, family size, maternal antenatal care, and preceding birth interval.[15] also found that religion was another significant factor to determine child mortality in Ethiopia using bivariate and multilevel analysis of 2016 DHS data.

The livelihood of children must be improved; revise the health policy system and strategies to addressing health problems during conflict. During armed conflict crises, not only the conflict cause of child death but disease due to poor sanitation and insufficient nutrition like diarrheal diseases were occur and control protect and formulation of treating strategies can improve them maintain the life of children [18]. In many researches finding there is evidence that armed conflict and child mortality have direct relationships [19], [20]. The consequences of armed conflict or war were internal/external displacement, inadequate humanitarian demands extreme violence against children and women and lack of hyenine in the refugee camp and these were causes for child mortality [21].

In Ethiopia the issue of child mortality was still continue in public health practices and unable to provide equitable access within different cluster levels of household. To reduce the gaps of health access variation, the intervention practices focused on households who lives on under poverty, offer education opportunity, and access to sanitation and hygiene could assist in lowering child mortality rates in Ethiopia [11]. In the last two or more decay Ethiopia passed through horrid crises by conflict, natural disasters like El Nino-driven drought and desert locust infestation and recently the pandemic of COVID-19 and in the Northern part Conflict [22]. These crises all in together affect more than 30 million and causes to occur food insecurity problem, internal and external displacement. As studied by [22], an estimate of more than 2.5 million people displaced from their home and living areas due to Northern conflict-driven humanitarian crisis in the from the regions of Tigray, Amhara and Afar.

In the last four years the Northern Ethiopian armed war conflict results 51% of destructions of health facilities; damaged and interrupt health services of patients such as chronic disease follows up, non-communicable and chronic infectious disease mental health problem services and maternal and child health services in the affected region and hindering improvements of children to survive from their death [23], [24]. In Ethiopia there is a significant intervention activity through governments and non-governmental agencies to lower the rate of child mortality and to achieve the sustainable development goals of WHO organization. However, the root is not easy to implement these policies due to the political issues especially from 2016 to 2018 in Amhara and Oromia Region. These issues still continue for internal problems of the country for high internal immigrations primarily in Tigray, Amhara, Somali and Oromia regions. These conditions prohibited the government as well as the humanitarian agencies to apply health policies. The government of Ethiopia gives high attention for child and maternal health improvements up to the amendment of policies related to maternal benefits or maternal leave from four months to six months for all public and private workers. This policy helps to reduce infant mortality and morbidity, and maternal mortality related to death due to birth.

In Bangladesh [25] found that male neonates and child has a lower significant risk of dying during the neonatal and child’s period of 2011 and mother’s education has a significant of higher risk for neonatal mortality rate; but lower risk of child mortality in 2011. They also found that Father’s education had a great influence on the survival of young children. Their findings suggest that in 2014 gender of children and mother’s education has been improved.

In this study the data was recent 2019 mini EDHS data to estimate infant mortality rate, infant mortality. Even if Ethiopia as a long history greater, there is an internal conflict for supremacy of power between different ethnic groups and due to wrong narration or tale. As a result, there is a high internal migration and women delivery of birth in hospitals or health sectors was impossible. It is known that during conflicts and wars women and children primarily face for prejudice.

From the mid 2014 in Ethiopia, there is high conflict in the people against the government. Internal displace persons (IDP) have experienced vulnerability and navigate their daily lives [26]. He indicates that access to healthcare services was impeded by lack of finances, unavailability of drugs, and poor-quality service at healthcare facilities. Insecurity has negatively aggravated access to basic needs such as food, shelter, clothing, and healthcare, causing uncertainty in IDP livelihoods.

Internally displaced women in Ethiopia due to Tigrian defence force (TDF) and FDRE are a group of targeted and victimized civilians suffering from a wide range of atrocities that are rooted in their gender identity, marginalized status, and extreme vulnerability in Dabat and Debark District [27].

Sources of data: - the data source secondary data from 2019 EMDHS.

Variables of the study: - the response variables were under 5 child mortality (the death of live birth within 5 year of birth). There are two independent variables considered in this study. These are fixed effect and random effect variables. Fixed effect variables were region (coded as Tigray, Afar, Amhara, Oromia, Somali, Benishangul, SNNPR, Gambella, Harari, Addis Ababa, and Dire Dawa), residence (coded as rural, and urban), sex of child and household head (coded as male and female), maternal age, women education (recoded as no, primary, and secondary and above), wealth index (coded as poor, middle and rich), type of birth (coded as single and multiple), place of delivery (coded as home and Institution), and sex of child (coded as male and female). Source of drinking water (coded as improve or unimproved), types of toilet facility (coded as modern, traditional or no toilet facility), types of cooking fuel (recoded as improved or unimproved), ages of mother at first birth (recoded as 20 and less than or more than 20), and season of child birth (recoded as autumn, winter spring or summer). The only random effect variable was cluster number or (enumeration area) used as sampling technique.

Methods of Data Analysis

The data were extracted using SPSS software and the analysis were done using R software version 4.1.2 Zero inflated models allow for both over dispersion and zero inflated count data. The frequent model for count data poison regression and negative binomial regression model were usually used. Poison regression model applied if the assumption of poison distribution; that is if the mean and variance approximately equal. It is recognized that counts often display extra Poisson variation or over dispersion relative to Poisson model. To accommodate this over dispersion or variation negative binomial regression model is used instead.

Thus, the mean and variance of Poisson regression model were 0.37 and 0.624 which is not equal and the frequencies show that 4374 zero out of 5753 cases. It gives evidence to use Zero inflated negative binomial regression model and the random variation was cluster number (EA).

An arithmetic mean for the number of children who died before their 5th birth day is 0.366 and having variances of 0.624. This shows that it is not equally-dispersed; rather it is over dispersed. Which mean it is better to use negative binomial regression family model instead of using Poisson regression family model.

Table 1

summary of descriptive statistics for variables included in our study
Variable	Category	Child death		Total	X² value (p-value)
Variable	Category	yes	percent	Total	X² value (p-value)
Region	Tigray	14	4.1	450	0.000
	Afar	35	10.5	651
	Amhara	23	6.8	501
	Oromia	41	12.1	715
	Somali	57	16.9	637
	Benishangul	45	13.3	522
	SNNP	26	7.7	657
	Gambela	35	10.4	444
	Harari	29	8.6	437
	Addis Ababa	6	1.8	287
	Dire Dawa	27	8.0	394
Place of residence	Urban	72	21.3	1306	0.462
Place of residence	Rural	266	78.7	4389	0.462
Mothers age at 1st birth	Less 20 More than 20	243 95	71.9 28.1	4103 1592	0.949
Level of education	No education	190	56.2	3135	0.004
	Primary	122	36.1	1802
	Secondary and above	26	7.7	758
Source of drinking water	Improved/ protected	213	63.0	3632	0.403
Source of drinking water	Unimproved/unprotected	125	37	2063	0.403
Types of cooking fuel	Improved	21	6.2	464	0.180
Types of cooking fuel	Unimproved	317	93.8	5231	0.180
Types of toilet facility	Modern facility	143	42.3	2553	0.616
	No facility	147	43.5	2388
	traditional	48	14.2	754
Sex of hh head	Male	279	82.5	4561	0.244
Sex of hh head	Female	59	17.5	1134	0.244
Wealth index	Poor	193	57.1	2940	0.100
	Middle	45	13.3	797
	Rich	100	29.6	1958
Sex of child	Male	195	57.7	2942	0.022
Sex of child	Female	143	42.3	2753	0.022
Types of birth (twin)	Single birth Multiple birth	294 44	87 13	5528 167	0.000
Place of delivery	Home	196	58.0	2864	0.004
Place of delivery	Institution	142	42.0	2831	0.004
Season of birth	Autumn Winter Spring Summer	111 88 84 55	32.8 26 24.9 16.3	1632 1600 1454 1009	0.358
Model selection criteria

An appropriate model that has used to fit the given data was selected by comparing the Akaike Information (AIC) criteria and Bayesian Information (BIC) values. The smallest IC value is the better fit model [28]. We have compared the total AIC and BIC values. The smaller IC value is used as model selection criteria. Therefore, AIC was used to select appropriate model and hence zero inflated negative binomial mixed (ZINBM) model was good-fitted model.

Table 2

output of model selection criteria
Model type	AIC	BIC	Deviance
NBM	8054.3	8227.1	8002.3
ZINB	8068.0	8393.8	7970.0
ZINBM	7999.6	8332.0	7899.6

Estimation of risk of child deaths

From Table 3 below the variable age of mother, age of mother at 1st birth, mother’s education level, types of cooking fuel, number of living children, types of birth and from region Addis Ababa were a significant factor on non-zero death of under-five child mortality in ZINBM model.

If a mother age increases by one year, the incidence rate ratio of non-zero death increases by 1.125 assuming other variables unchanged. The incidence rate ratio of non-zero death for age of mother at first birth less than 20 years were 1.6456 times that of non-zero death of age of mother greater than 20 years old assuming other variables constant. The incidence rate of non-zero under five child death of no education was 1.9785 and primary education was 2.002 times that of non-zero death education level secondary and above. Incidence rate of non-zero under five child mortality for unimproved cooking fuel were 2.2496 times that for non-zero death of improved cooking fuel. Incidence rate of non-zero death of under-five child mortality in Addis Ababa was 0.382 times that of non-zero death of Tigray region. For a household the number of living children increase by one child the non-zero incidence rate of under-five child death decreases by 0.821. Incidence rate of non-zero death of under five child mortality for a multiple type of birth was 1.6635 times that incidence rate of non-zero death of under-five child mortality.

Table 3

Conditional effect estimate of child mortality
variable	Estimate	St. Error	Z-value	Exp(\(\:\beta\:\))	Pr(>\|z\|)
Constant	-5.378	0.468	-11.493	0.0046	< 2e-16 ***
Current mother age	0.1178	0.0079	15.01	1.125	< 2e-16 ***
Residence Rural	-0.2762	0.1496	-1.846	0.7587	0.064898 .
Age of mother at 1st birth Less than 20 years	0.4981	0.0988	5.04	1.6456	4.67e-07 ***
Education level No education primary	0.6823 0.6942	0.1957 0.1913	3.487 3.63	1.9785 2.0020	0.000488 * 0.000284 *
Toilet facility No facility Traditional toilet facility	-0.1159 -0.0753	0.0915 0.1689	-1.266 -0.446	0.8905 0.9274	0.205494 0.655550
Source of drinking water Unprotected /unimproved	0.0863	0.086	1.004	1.0901	0.315429
Wealth index middle Poor	0.2213 0.2403	0.1369 0.1237	1.617 1.942	1.2478 1.2716	0.105873 0.052130 .
sex of child female	0.0614	0.0650	0.944	1.0633	0.345084
Types of birth is twin Multiple birth	0.508915	0.104357	4.877	1.6635	1.08e-06 ***
Number of living children	-0.1972	0.0193	-10.210	0.8210	< 2e-16 ***
Types of cooking fuel unimproved	0.8108	0.3100	2.618	2.2496	0.008834 **
Sex of house hold head Female	0.0336	0.0987	0.340	1.0341	0.733893
Region Afar Amhara Oromia Somalia Benishangul SNNP Gambela Harari Addis Ababa Diredawa	0.0714 -0.0802 -0.1742 0.3703 0.1282 -0.0769 -0.0100 0.1469 -0.962 0.0819	0.282 0.2693 0.2491 0.2655 0.2699 0.2618 0.2794 02823 0.4888 0.2892	0.253 -0.298 -0.699 1.394 0.475 -0.294 0.036 0.520 -1.968 0.283	1.074 0.9229 0.8401 1.4482 1.1368 .92594 0.9901 1.1582 0.3821 1.085	0.800091 0.765778 0.484389 0.163180 0.634851 0.768832 0.971650 0.603120 0.049046 * 0.777060
Season of birth Winter Spring summer	-0.0637 0.0203 0.0728	0.0860 0.0844 0.0965	-0.741 0.241 0.754	0.9383 1.0205 1.07551	0.458855 0.809712 0.450682
Place of delivery Home	0.1008	0.0782	1.288	1.1060	0.197610
---

Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

The variable sex of child and household head, all regions except Addis Ababa, delivery place, types of toilet facility, source of drinking water season of child birth, wealth index and place of residence were not statistically significant variable for under five child mortality.

Table 4

Zero-inflation model:
variable	Estimate	St. Error	Z-value	Exp(\(\:\beta\:\))	Pr(>\|z\|)
Constant	2.4382	1.6213	1.504	11.4521	0.132605
Current mother age	-0.0826	0.0281	-2.943	0.92072	0.003255 **
Residence Rural	-0.6841	0.3385	-2.021	0.50455	0.043264 *
Age of mother at 1st birth Less than 20 years	-1.5062	0.2803	-5.374	0.2217	7.71e-08 ***
Education level No education primary	0.3469 0.6935	0.6300 0.5937	0.551 1.168	1.4147 2.001	0.581835 0.242761
Toilet facility No facility Traditional toilet facility	-0.7245 0.1975	0.2959 0.4538	-2.449 0.435	0.4845 1.218	0.014318 * 663379
Source of drinking water Unprotected /unimproved	-0.1643	02457	-0.669	0.8485	0.503621
Wealth index middle Poor	0.4226 0.4557	0.4203 0.3813	1.005 1.195	1.5259 1.577	0.314699 0.232052
sex of child female	0.1368	0.2176	0.629	1.0633	0.529585
Child is twin Multiple birth	-19.16039	2394.34080	-0.008	4.772542e-09	0.993615
Number of living children	0.0626	0.0679	0.921	1.0645	0.356869
Sex of house hold head Female	0.5473	0.2895	1.891	1.7286	0.058676 .
Region Afar Amhara Oromia Somalia Benishangul SNNP Gambela Harari Addis Ababa Diredawa	-0.9310 -0.7452 -2.5980 -2.2273 -2.1384 -1.6488 -2.5352 -1.7395 -0.9464 -1.9062	0.5197 0.4398 0.6376 0.5612 0.5933 0.5078 0.8725 0.5310 0.9746 0.6916	-1.791 -1.694 -4.075 -3.968 -3.604 -3.247 -2.906 -3.276 -0.971 -2.756	0.3942 0.4746 0.0744 0.1078 0.1178 0.1922 0.0793 0.1756 0.3881 0.1486	0.073217 0.090197 4.61e-05 * 7.24e-05 * 0.000313 * 0.001166 0.003665 0.001054 0.331492 0.005846 **
Types of cooking fuel	1.8191	1.2627	0.629	6.1666	0.1497
Season of birth Winter Spring summer	-0.0703 -0.0799 -0.3337	0.2880 0.2904 0.2880	-0.244 -0.275 1.159	0.93208 0.9231 1.3962	0.8070 0.7831 0.2466
Place of delivery Home	-0.1294	0.2423	-0.534	-1.379	0.5933
---

Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

From Table 4, current mother age, age of mother at 1st birth, no toilet facility, place of delivery and all regions except Addis Ababa and the reference category Tigray were significant variables of excess zero. For a one-year increase of mother age, the estimated odds of the number of child death becomes zero increases by 0.9207. The odds of number of child deaths become zero with ages of mother at first birth less than 20 years old were 0.247 that odds of number of child deaths become zero for ages of mother at first birth greater than 20 years. The odds of number of child death become zero for no toilet facility were 0.4845 times that of odds of number of child death become zero for modern toilet facility user household.

The variables education level, traditional toilet facility, types of birth, sex of child and household head, types of cooking fuel season of child birth, place of delivery and source of drinking water were not statistically for excess zero deaths.

The rate of child mortality in Ethiopia was highest in the world and it was slightly decreasing; however internal conflicts became problems to meet the sustainable and development goals set by countries. Most studies on multilevel logistic models and the concept of random effect was miss used. Only categorical variables were used as a random factor and it is considered as if it was not covered all categories but generalise about all categories. The present study considered only cluster number (enumeration area) as random factors. This finding includes mother of age, place of residences ages of mother at first birth, mother’s education level, source of drinking water, source of toilet facility, types of cooking fuel, sex of child and household head, types of birth, place of delivery, number of living children, wealth index, season of birth and region.

Ages of mother at first birth in this study were a significant factor for child mortality. There is a similar finding in in 35 Sub Sharan African countries done by[4], in Nigeria by[10] and in Ethiopia by [14]. Mothers who give their first birth at earlier have no adequate awareness about all antenatal postnatal treatment follow up and had no experience to get healthy baby. Mothers education level were highly related with the risk of child mortality in this study using 2019 EMDHS data. It is also coincided with the investigation by [4], [6], [8], [14], [15], [17], [29], [30]. This implies that mother’s education level was still a problem for child mortality. Governments formulate health infrastructures to afford the problems and to reach sustainable and development goals. There is health facility structure till kebele level to educate maternal about pregnancy, child nutrition, contraceptive methods and soon. However, the problem of mother education was a problem for child mortality.

Multiple types of birth also had high risk for child mortality in this investigation this study was supported by[1], [14] and [30]. For multiple birth, mothers got problems in breast feeding and highly challenged to give proper care. Even starting from pregnancy, if they are multiple they compute the nutrient and unable to get enough nutrition and they will have low birth weights. These made to death for children. Types of cooking fuels was another risk factor for child mortality in this investigation. The majority of Ethiopian peoples were getting their sources of energy from unimproved sources of energy. This was respiratory and related health problem and causes for child mortality. This case was highly explained on experts that thy had investigate about child mortality and planetary health problems. Mothers age a negative effect and numbers of living children had a positive effect on child mortality. But these two findings are contradicted with other studies done 2000–2016 EDHS data sets. when a house hold had a greater number of living children, they share burdens to care child and prevent from contaminated water food and other communicable diseases.

Using ZINBMM model, mother’s education level, mother’s age at first birth types of cooking fuel, types of birth and number of living children were had higher risk of death for child mortality in Ethiopia. ZINBMM were best fitted model to analyse count data from 2019 EMDH.S data

AIC

akaki information criteria

BIC

Bayesian information criteria

DHS

demographic and health survey

enumeration area

EDHS

Ethiopian demographic and health survey

EMDHS

Ethiopian mini demographic and health survey

HIV

human immunodeficiency virus

information criteria

IYCF

infant and young child feeding

NBM

negative binomial mixed effect

ZINBM

zero inflated negative binomial mixed effect

ZINBMM

zero inflated negative binomial mixed model

Competing interest

The authors declare that they have no competing interest

Ethics Approval and Consent to Participate

the study was from secondary sources of data of 2019 EMDHS which is public available data for researchers and ethical approvals were by Ethiopian Public Health Institutes Review Board, Institutional Review Board IYCF international, National Research Ethics Review Committee. Written consents were obtained from mothers/primary care givers and data collected from 2019 EMDHS.

Consents for Publication

• I confirm that I have read, understand, and agreed to the submission guidelines, policies, and submission declaration of the journal.

Competing Interest

The authors declare that they have no competing interest.

Funding Sources

Since the sources data for this research was secondary sources of data, there is no any fund during the entire research works.

Author Contribution

Author 1. write the background of the study, the methodology and interpretation the result section.Author 2. extracting data using SPSS software, organize and prepare R codes, analyse extracted data using R software and prepare output result in tabular format

Acknowledgement

The authors would like to forward acknowledgement for Ethiopian Central Statistical Agency and Ethiopian Public Health Institution for their freely available dataset for researchers.

Data Availability

the data had obtained from online data from Ethiopian 2019 mini DHS data sources

Fenta SM, Fenta HM. Risk factors of child mortality in Ethiopia: application of multilevel two-part model. PLoS ONE. 2020;15(8):e0237640.
Asghar N, Amjad MA, Rehman HU. Analyzing the impact of access to electricity and biomass energy consumption on infant mortality rate: a global perspective. Environ Sci Pollut Res. 2023;30(11):29550–65.
Vakili R, Khademi G, Vakili S, SAEIDI M. Child mortality at different world regions: a comparison review, 2015.
Yaya S, Uthman OA, Okonofua F, Bishwajit G. Decomposing the rural-urban gap in the factors of under-five mortality in sub-Saharan Africa? Evidence from 35 countries. BMC Public Health. 2019;19:1–10.
Ekholuenetale M, Wegbom AI, Tudeme G, Onikan A. Household factors associated with infant and under-five mortality in sub-Saharan Africa countries. Int J Child Care Educ Policy. 2020;14:1–15.
Van Malderen C, Amouzou A, Barros AJD, Masquelier B, Van Oyen H, Speybroeck N. Socioeconomic factors contributing to under-five mortality in sub-Saharan Africa: a decomposition analysis. BMC Public Health. 2019;19:1–19.
Kravdal Ø. New evidence about effects of reproductive variables on child mortality in sub-Saharan Africa. Popul Stud (NY). 2018;72(2):139–56.
Aheto JMK. Predictive model and determinants of under-five child mortality: evidence from the 2014 Ghana demographic and health survey. BMC Public Health. 2019;19:1–10.
Ezeh OK, Agho KE, Dibley MJ, Hall JJ, Page AN. Risk factors for postneonatal, infant, child and under-5 mortality in Nigeria: a pooled cross-sectional analysis. BMJ Open. 2015;5(3):e006779.
Adetoro GW, Amoo EO. A statistical analysis of child mortality: Evidence from Nigeria. J Demography Social Stat. 2014;1(1):110–20.
Kuse KA, Chikako TU, Hagan Jr JE, Seidu A-A, Ahinkorah BO. Assessing Individual-and Community-Level Variability in Predictors of Neonatal, Infant, and Under-Five Child Mortality in Ethiopia Using a Multilevel Modeling Approach. Children. 2022;9(7):1071.
Bello R, Joseph A. Determinants of Child Mortality in Oyo State, Nigeria. Afr Res Rev. Feb. 2014;8(1):252. 10.4314/afrrev.v8i1.17.
Bello RA, Joseph AI. Determinants of child mortality in Oyo State, Nigeria. Afr Res Rev. 2014;8(1):252–72.
Bedada D. Determinant of under-five child mortality in Ethiopia. Am J Theoretical Appl Stat. 2017;6(4):198–204.
Zewudie AT, Gelagay AA, Enyew EF. Determinants of Under-Five Child Mortality in Ethiopia: Analysis Using Ethiopian Demographic Health Survey, 2016, Int J Pediatr, vol. 2020, no. 1, p. 7471545, 2020.
Demographic E. Health Survey: 2005: Central Statistical Agency Addis Ababa, Ethiopia. Maryland: ORC Macro Calverton, 2006.
Gebremichael SG, Fenta SM. Under-Five Mortality and Associated Risk Factors in Rural Settings of Ethiopia: Evidences from 2016 Ethiopian Demographic and Health Survey, Adv Public Health, vol. 2020, no. 1, p. 8430246, 2020.
Schedwin M, et al. Under-five mortality in the Democratic Republic of the Congo: secondary analyses of survey and conflict data by province. Bull World Health Organ. 2022;100(7):422.
Ali HE, Adan UA. Conflict types and child and infant mortality rates: Evidence from panel data. J Political Military Sociol, pp. 137–61, 2013.
Jawad M, Hone T, Vamos EP, Roderick P, Sullivan R, Millett C. Estimating indirect mortality impacts of armed conflict in civilian populations: panel regression analyses of 193 countries, 1990–2017. BMC Med. 2020;18:1–11.
Biset G, Goshiye D, Gedamu S, Tsehay M. The effect of conflict on child and adolescent health in Amhara region, Ethiopia: Cross-Sectional Study. BMC Pediatr. 2023;23(1):463.
Abebe H, Belaineh G. Key considerations: social science perspectives for emergency response to the conflict in Northern Ethiopia. Social Sci Humanitarian Action Platf, 2022.
Abebaw N, Mekonen H, Animut Y, Tefera Z, Walle F, Zeleke A. Narrative descriptive reports on public health impact of conflict among war-affected zones at Amhara Region, 2022, 2022.
Joshi M. Comprehensive peace agreement implementation and reduction in neonatal, infant and under-5 mortality rates in post-armed conflict states, 1989–2012. BMC Int Health Hum Rights. 2015;15:1–10.
Maniruzzaman M et al. Risk factors of neonatal mortality and child mortality in Bangladesh. J Glob Health, 8, 1, 2018.
Kemei J, Salami B, Soboka M, Gommaa HIM, Okeke-Ihejirika P, Lavin T. The forms and adverse effects of insecurities among internally displaced children in Ethiopia. BMC Public Health. 2023;23(1):200.
Desalegn S, Kasseye E, Gebeyaw G, Meshelemiah JCA. The Challenges of Women Housed in Internally Displaced Persons (IDP) Camps During an Armed Conflict in Ethiopia. Affilia. 2023;38(1):55–74.
Akaike H. A new look at the statistical model identification. IEEE Trans Automat Contr. 1974;19(6):716–23.
Jayathilaka R, Adikari H, Liyanage R, Udalagama R, Wanigarathna N. Cherish your children: socio-economic and demographic characteristics associated with child mortality. BMC Public Health. 2021;21(1):1217.
Karaye IM, Stone KW, Horney JA. Determinants of under-five mortality in an armed Conflict Setting: Empirical findings from the demographic and health surveys. Int J Environ Res Public Health. 2022;19(21):14179.

No competing interests reported.

Download PDF

Version 1

posted

You are reading this latest preprint version

Effect of Selected Variables on Child Mortality in Ethiopia using 2019 EMNDHS: Application of Zero Inflated Negative Binomial Mixed Effect Regression Model

Status:

Version 1

Abstract

Figures

Background

Methods

Results

Estimation of risk of child deaths

Discussion

Conclusion

Abbreviations

Declarations

Competing interest

Competing Interest

Funding Sources

Author Contribution

Acknowledgement

Data Availability

References

Additional Declarations

Status:

Version 1