Household factors associated with infant
and under‑fve mortality in sub‑Saharan Africa
countries
Michael Ekholuenetale1*, Anthony Ike Wegbom2, Godson Tudeme3 and Adeyinka Onikan4
Abstract
Background: Child mortality has become a prominent public health issue in subSaharan Africa (SSA). The mortality rates can in part be translated to how communities
meet the health needs of children and address key household and environmental risk
factors. Though discussions on the trends and magnitude of child mortality continue as
to strategize for a lasting solution, large gap exists specifcally in family characteristics
associated with child death. Moreover, household dynamics of child mortality in SSA is
under researched despite the fact that mortality rates remain high. This study aimed to
examine the infuence of household structure on child mortality in SSA.
Methods: Secondary data from birth histories in recent Demographic and Health
Survey (DHS) in 35 SSA countries were used in this study. The total sample data of
children born in the 5 years prior to the surveys were 384,747 births between 2008 and
2017. Unadjusted and adjusted Cox proportional hazard regression model was ftted
to model infant and under-fve mortality. The measure of association was hazard ratio
(HR) with 95% confdence interval (CI). Statistical test was conducted at p < 0.05 level of
signifcance.
Results: Total infant mortality rates were highest in Sierra Leone (92 deaths per 1000
live births), Chad (72 deaths per 1000 live births) and Nigeria (69 deaths per 1000 live
births), respectively. Furthermore, total rates of under-fve mortality across 35 SSA
countries were highest in Cameroon (184 deaths per 1000 live births), Sierra Leone
(156 deaths per 1000 live births) and Chad (133 deaths per 1000 live births). The risk of
infant mortality was higher in households of polygyny, compared with households of
monogyny (HR=1.23; CI 1.16, 1.29). Households with large number of children (3–5
and ≥6) had higher risk of infant mortality, compared with those with 1–2 number
of children. Infants from mothers with history of multiple union had 16% increase in
the risk of infant mortality, compared with those from mothers from only one union
(HR=1.16; CI 1.09, 1.24). Furthermore, under-fve from female household headship
had 10% signifcant reduction in the risk of mortality, compared with those from male
household headship (HR=0.90; CI 0.84, 0.96). The risk of under-fve mortality was
higher in households of polygyny, compared with monogyny (HR=1.33; CI 1.28, 1.38).
Households with large number of children (3–5 and≥6) had higher risk of under-fve
mortality, compared with those with 1–2 number of children ever born. Under-fve
Open Access
© The Author(s) 2020. This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing,
adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and
the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material
in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material
is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the
permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativeco
mmons.org/licenses/by/4.0/.
RESEARCH
Ekholuenetale et al. ICEP (2020) 14:10
https://doi.org/10.1186/s40723‑020‑00075‑1
*Correspondence:
[email protected]
1 Department
of Epidemiology and Medical
Statistics, Faculty of Public
Health, College of Medicine,
University of Ibadan, Ibadan,
Nigeria
Full list of author information
is available at the end of the
article
Ekholuenetale et al. ICEP (2020) 14:10 Page 2 of 15
from mothers with history of multiple union had 30% increase in the risk of mortality,
compared with those from mothers from only one union (HR=1.30; CI 1.24, 1.36).
Conclusion: Household structure signifcantly infuences child mortality in SSA.
Knowledge of drivers of infant and child death is crucial in health policy, programmes
designs and implementation. Therefore, we suggest that policies to support strong
healthy families are urgently needed to improve children’s survival.
Keywords: Family structure, Infant death, Under-5 death, DHS
Background
Child mortality remains prevalent in SSA region (Burke et al. 2016; Ester et al. 2011;
Kazembe et al. 2012). Population surveillance of child mortality is of a great interest
to stakeholders in health system, and a key indicator of health and socioeconomic
development. Globally, SSA region still accounts for the highest absolute number
of under-5 deaths and the largest mortality rates (Guillot et al. 2012). Child mortality is reported at national level, particularly in resource-constrained settings, where
estimates from Global Burden of Disease studies depend on national-based surveys.
Nonetheless, more is yet to be known about urban vs rural spread of under-5 mortality across SSA region, pointing to key areas of evidence-based gaps in the knowledge
about death occurrences (Burke et al. 2016).
Disparities exist in child mortality rates amongst family structure types (Omariba
and Boyle 2007). A study reported on the role that marital status plays in predicting child survival outcome, and obtained a consistent pattern that polygyny signifcantly decreased the survival of infants from polygynous families (Smith-Greenaway
and Trinitapoli 2014). Furthermore, infants born to unmarried mothers had higher
mortality rates than those born to married mothers (Freeman and Brewer 2013).
Understanding the diferences in infant and child mortality rates between married
and cohabiting mothers, or cohabiting and single mothers is paramount. Based on
the pattern of birth outcomes and prenatal health behaviours, one might entirely conclude that infants of married or cohabiting mothers could have less of a mortality risk
than infants of single mothers. An evidence-based study reported single motherhood
is a risk factor of children’s nutritional status and chances of survival before 5 years of
age in SSA (Ntoimo and Odimegwu 2014).
Historically, large family sizes were regarded as a source of sustenance and respect
in many SSA communities. Unfortunately, this exposed children to high risk of death
due to economic constraints of large households, in terms of accessibility and availability of resources (Kabagenyi and Rutaremwa 2013). Children are the most vulnerable group subjected to the risk of death as a result of unfavourable household factors
(Ajao et al. 2010). Tis can be explained by the capacity of a household to adequately
meet the needs of all members been afected by household structure comprising
household size, household type, place of residence amongst others. In a previous
report, the infuence of household structure and household economic status on the
well-being of a child in SSA was established (Akinyemi et al. 2016; Omariba and Boyle
2007). Te issue of household structure in relation to child survival requires adequate
explanations to enhance sustainable programmes on child survival.
Ekholuenetale et al. ICEP (2020) 14:10 Page 3 of 15
Te implications of changing the structure of family and what it means for the nation’s
children have been the subject of extensive research in recent decades (Freeman and
Brewer 2013). Te need to understand this relationship is underscored by the welldocumented impact of prenatal health behaviours on birth outcomes (Kabagenyi and
Rutaremwa 2013). Tus, gaining an understanding of the ways in which family structure afects the healthy development of children is vital in this time of rapid changes to
family life. Certainly, diferent family structures provide very diferent environments for
children, but the impact that family structure has on early child health outcomes is most
salient. Since family is usually the basic unit of interaction for everyone, family structure
and dynamics impact child survival outcomes immensely. Te family process literature
describes the family as a unit that makes decisions and allocates resources to achieve
goals. Family processes such as fexibility, caring communication and supervision are
important factors infuencing the functioning of families and the well-being of individual
members (Freeman and Brewer 2013; Omariba and Boyle 2007). Household dynamics
of child mortality in SSA is under researched despite the fact that mortality rates remain
high. Tis study aimed to examine the infuence of household structure on child mortality in SSA.
Methods
Data source
Te pooled multi-country nationally representative DHS data from birth histories in 35
SSA countries were analysed in this study. Te total sample data of children born in the
5 years prior to the surveys were 384,747 births between 2008 and 2017. DHS data for all
SSA countries were retrieved from the Measure DHS online data archive after necessary
approval for data use. DHS are routinely conducted every 5 years using similar methodologies and instruments across several countries. Te data are in the public domain and
were accessed at http://dhsprogram.com/data/available-datasets.cfm. Details of DHS
data have been reported elsewhere (Corsi et al. 2012).
Sampling procedure
DHS were based on a stratifed multi-stage cluster sampling technique. Te stratifcation strategy divides the population into groups. For instance, all DHS employ a region
crossed by urban–rural stratifcation. A multilevel stratifcation approach is used to
divide the population into frst-level strata and to subdivide the frst-level strata into
second-level strata, and so on. A two-level stratifcation in DHS is region and urban/
rural stratifcation. Globally, DHS are comparable household surveys that have been
conducted in more than 85 countries since 1984. Tough it was designed to expand on
demographic, family planning, and fertility data collected in the World Fertility Surveys
and Contraceptive Prevalence Surveys, DHS continue to provide an important resource
for the monitoring of vital statistics and population health indicators in resource-constrained settings. It collects a wide range of objective and self-reported data with a strong
focus on indicators of fertility, reproductive health, maternal and child health, mortality, nutrition and self-reported health behaviours amongst adults. Key advantages of
the DHS include high-quality interviewer training, national coverage, standardised data
collection procedures across countries and consistent content, allowing comparability
Ekholuenetale et al. ICEP (2020) 14:10 Page 4 of 15
across populations cross-sectionally and over time. Data from DHS facilitate epidemiological research focused on monitoring of prevalence, trends and inequalities. A variety
of robust observational data analysis methods have been used, including cross-sectional
designs, repeated cross-sectional designs, spatial and multilevel analyses, intra-household designs and cross-comparative analyses.
Variables’ selection and measurement
Outcome variables
We recorded two response variables in this study to represent child mortality: infant and
under-fve mortality. Infant mortality was defned as the probability of death before the
frst birthday. In addition, under-fve mortality was defned as the probability of death
before the ffth birthday.
Household characteristics
Several household variables were included in this study.
• Sex of household headship; male/female);
• Family type: monogyny/polygyny;
• Family size based on the number of children ever born: 1–2/3–5/6+;
• Marital status: never married/in union or living with a man/formerly in union or living with a man;
• Number of union for mothers: once/more than once;
• Current residence of mothers: living with husband/partner/staying elsewhere;
• Duration of couples in union: 1 month–7 years/> 7 years–14 years/> 14 years–
21 years/> 21 years/never in union.
• Maternal educational level: None/primary/secondary/tertiary.
• Place of residence: urban/rural.
• Maternal age: 15–19/20–24/25–29/30–34/35–39/40–44/45–49.
• Household wealth quintile: For the computation of wealth index, principal components analysis (PCA) was used to assign the wealth indicator weights. Tis procedure
assigned scores and standardised the wealth indicator variables such as foor type,
wall, roof, water source, sanitation facilities, radio, electricity, television, refrigerator,
cooking fuel, furniture, and number of persons per room. Tereafter, the factor coeffcient scores (factor loadings) and z-scores were calculated. Finally, for each household, the indicator values were multiplied by the loadings and summed to produce
the household’s wealth index value. Te standardised z-score was used to disentangle the overall assigned scores to poorest/poorer/middle/richer/richest (Rutstein and
Staveteig 2012).
Ethical consideration
We used publicly available data in this study. Te ethical procedures for data collection
were the responsibility of the institutions that commissioned, funded, or managed the
surveys. All DHS are approved by ICF international and Institutional Review Board
(IRB) to ensure that the protocols are in compliance with the U.S. Department of Health
Ekholuenetale et al. ICEP (2020) 14:10 Page 5 of 15
and Human Services regulations for the protection of human subjects. Terefore, this
study did not require further ethical approval.
Data analysis
Te collinearity testing method utilised the correlation analysis to detect interdependence between variables to reduce multicollinearity. A cut-of of 0.7 was used to examine the multicollinearity known to cause major concern in multicollinearity (Midi et al.
2010). Maternal education was retained in this study as it was found to have strong association with paternal education. Other variables were retained in the model due to lack
of multicollinearity. We used complex survey module (‘svy’) command to adjust for clustering, stratifcation and sampling weights. Te rates of infant and under-fve mortality
were computed for the 5 years preceding the surveys and disaggregated by sex of children and place of residence. To explore the relationship child mortality and household
factors, the unadjusted and adjusted Cox proportional hazard regression model was ftted to model infant and under-fve mortality. Te survival time for dead children was
taken as their age at death. For children who were alive, their survival time was right
censored at their current age at the time of survey. Te measures of association were
hazard ratio (HR) with 95% confdence interval (CI). All statistical test was conducted
at p < 0.05 level of signifcance. Data were analysed using Stata version 14.0 (StataCorp,
College Station, TX) (StataCorp; http://www.stata.com).
Results
In Table 1, we presented the total rates of infant and under-fve mortality across 35 SSA
countries. In addition, these rates were reported by the place of residence and sex of
children. Total infant mortality rates were highest in Sierra Leone (92 deaths per 1000
live births), Chad (72 deaths per 1000 live births) and Nigeria (69 deaths per 1000 live
births), respectively. In rural settlements with higher mortality rates; Sierra Leone (92
death per 1000 live birth), Nigeria (75 deaths per 1000 live births), Guinea (73 death per
1000 live birth), Chad and Cote d’Ivoire (72 deaths per 1000 live births) had the leading rates of infant mortality in SSA countries. Majority of male children had higher
rate; Sierra Leone (98 deaths per 1000 live births), Cote d’Ivoire (88 deaths per 1000 live
births), Chad (80 deaths per 1000 live births), Nigeria (74 deaths per 1000 live births)
and Guinea (71 deaths per 1000 live births) had the leading infant mortality rates. Furthermore, the total rates of under-fve children mortalities across 35 SSA countries were
highest in Cameroon (184 deaths per 1000 live births), Sierra Leone (156 deaths per
1000 live births), Chad (133 deaths per 1000 live births), Burkina-Faso (129 deaths per
1000 live births), Nigeria (128 deaths per 1000 live births), Niger (127 deaths per 1000
live births), Guinea (123 deaths per 1000 live births), Cote d’Ivoire (108 deaths per 1000
live births) and Democratic Republic of Congo (104 deaths per 1000 live births), respectively. More so, for rural settlements and male children with higher mortality rates; these
countries accounted for some of the highest under-fve mortality rates. See Table 1 for
details.
The summary statistics of household/family structure is presented in Table 2; here,
approximately one-fifth (20.6%) of the households had female headship. In addition,
about one-quarters (24.4% and 24.8%) of the households practiced polygyny (men
Ekholuenetale et al. ICEP (2020) 14:10 Page 6 of 15
Table 1 Distribution of infant and under-fve mortality rates in SSA countries; DHS 2008–2017
Country Survey year N Infant mortality/1000 live births (95%CI) Under-5 mortality/1000 live births (95%CI)
Place of residence Sex of a child Total Place of residence Sex of a child Total
Urban Rural Male Female Urban Rural Male Female
Angola 2015/2016 14,322 36 (29, 42) 57 (48, 66) 51 (44, 59) 37 (30, 44) 44 (39, 50) 57 (47, 67) 85 (74, 96) 74 (65, 83) 62 (52, 73) 68 (61, 76)
Benin 2012 13,407 40 (33, 47) 43 (38, 48) 46 (41, 52) 37 (32, 42) 42 (38, 46) 59 (50, 68) 77 (69, 84) 75 (67, 82) 65 (58, 72) 70 (64, 76)
Burkina-Faso 2010 15,044 46 (37, 55) 69 (63, 75) 69 (62, 76) 61 (54, 68) 65 (60, 70) 82 (69, 94) 138 (128, 147) 133 (122, 144) 124 (114, 134) 129 (120, 137)
Burundi 2016/2017 13,192 56 (25, 87) 46 (41, 51) 49 (41, 57) 45 (38, 52) 47 (42, 52) 64 (34, 94) 79 (72, 87) 80 (71, 90) 76 (67, 85) 78 (71, 85)
Cameroon 2011 11,732 55 (46, 64) 68 (60, 76) 68 (60, 76) 57 (49, 65) 62 (56, 68) 90 (77, 102) 145 (134, 156) 131 (119, 142) 113 (102, 124) 184 (168, 201)
Chad 2014/2015 18,623 75 (63, 87) 72 (65, 79) 80 (73, 88) 64 (57, 71) 72 (66, 78) 132 (115, 149) 133 (124, 143) 143 (132, 154) 122 (113, 131) 133 (125, 141)
Comoros 2012 3,149 23 (12, 34) 41 (29, 52) 38 (26, 51) 33 (21, 45) 36 (27, 45) 32 (18, 47) 56 (41, 71) 50 (40, 59) 50 (36, 64) 50 (38, 61)
Congo 2011/2012 9,329 40 (30, 50) 38 (33, 43) 48 (38, 57) 31 (24, 39) 39 (33, 46) 68 (54, 81) 68 (61, 75) 78 (65, 91) 58 (47, 69) 68 (59, 76)
Cote d’Ivoire 2011/2012 7776 61 (48, 74) 72 (59, 85) 88 (74, 103) 47 (39, 55) 68 (59, 77) 95 (82, 82, 109) 116 (100, 131) 132 (116, 126) 83 (71, 96) 108 (97, 119)
Democratic Republic of
Congo
2013/2014 18,716 53 (45, 62) 61 (54, 67) 60 (53, 67) 57 (50, 63) 58 (53, 63) 91 (79, 103) 110 (101, 119) 108 (98, 118) 100 (91, 110) 104 (97, 112)
Ethiopia 2016 10,641 43 (21, 64) 49 (42, 56) 60 (50, 70) 35 (27, 43) 48 (41, 55) 49 (28, 70) 69 (59, 79) 81 (69, 93) 52 (42, 62) 67 (58, 76)
Gambia 2013 8088 37 (28, 47) 32 (24, 39) 37 (30, 44) 32 (23, 40) 34 (28, 40) 54 (39, 70) 53 (43, 63) 54 (45, 64) 53 (42, 64) 54 (45, 62)
Gabon 2012 6067 42 (33, 51) 45 (34, 56) 47 (35, 60) 37 (28, 47) 43 (35, 50) 62 (52, 72) 78 (65, 91) 77 (60, 94) 53 (41, 64) 65 (56, 74)
Ghana 2014 5884 42 (31, 54) 40 (32, 49) 44 (34, 54) 38 (29, 48) 41 (34, 48) 56 (44, 68) 63 (52, 73) 67 (55, 80) 52 (42, 63) 60 (52, 68)
Guinea 2012 7039 49 (37, 60) 73 (63, 84) 71 (60, 81) 63 (52, 74) 67 (58, 75) 78 (63, 92) 139 (124, 153) 128 (113, 143) 117 (102, 133) 123 (111, 135)
Kenya 2014 20,964 43 (35, 52) 36 (32, 40) 42 (36, 47) 36 (30, 41) 39 (35, 43) 57 (48, 66) 50 (45, 55) 54 (49, 60) 50 (44, 56) 52 (48, 57)
Lesotho 2014 3138 55 (33, 77) 61 (50, 72) 65 (51, 79) 54 (39, 69) 59 (49, 70) 88 (60, 116) 84 (72, 97) 89 (72, 106) 81 (64, 99) 85 (73, 97)
Liberia 2013 7606 52 (40, 65) 55 (47, 63) 53 (43, 64) 54 (45, 64) 54 (46, 61) 89 (73, 105) 98 (87, 110) 94 (81, 107) 94 (81, 106) 94 (84, 104)
Madagascar 2008/2009 12,448 42 (31, 52) 49 (43, 54) 51 (43, 58) 45 (38, 52) 48 (43, 53) 57 (44, 69) 74 (67, 81) 72 (64, 81) 71 (62, 80) 72 (65, 78)
Malawi 2015/2016 17,286 42 (28, 56) 42 (37, 46) 50 (43, 56) 34 (29, 38) 42 (38, 46) 55 (39, 72) 65 (60, 71) 73 (65, 81) 55 (49, 61) 64 (59, 69)
Mali 2012/2013 10,326 39 (30, 47) 60 (52, 68) 63 (55, 72) 48 (41, 56) 56 (49, 63) 59 (49, 69) 103 (92, 114) 109 (97, 120) 81 (71, 91) 95 (86, 104)
Mozambique 2011 11,102 61 (51, 71) 65 (57, 73) 69 (59, 78) 59 (52, 67) 64 (58, 71) 91 (78, 103) 99 (90, 109) 104 (94, 115) 89 (80, 99) 97 (89, 105)
Namibia 2013 5046 34 (24, 44) 44 (34, 54) 42 (33, 51) 36 (27, 45) 39 (32, 46) 50 (39, 62) 58 (47, 70) 56 (45, 67) 53 (43, 64) 54 (46, 63)
Niger 2012 12,558 33 (24, 42) 53 (47, 60) 55 (48, 63) 46 (39, 53) 51 (45, 56) 66 (52, 80) 136 (127, 146) 134 (124, 145) 120 (109, 132) 127 (119, 136)
Ekholuenetale et al. ICEP (2020) 14:10 Page 7 of 15
Table 1 (continued)
Country Survey year N Infant mortality/1000 live births (95%CI) Under-5 mortality/1000 live births (95%CI)
Place of residence Sex of a child Total Place of residence Sex of a child Total
Urban Rural Male Female Urban Rural Male Female
Nigeria 2013 31,482 55 (49, 62) 75 (70, 81) 74 (69, 80) 63 (58, 68) 69 (64, 73) 89 (80, 97) 149 (139, 158) 135 (126, 144) 121 (113, 130) 128 (121, 135)
Rwanda 2014/2015 7856 28 (20, 36) 33 (28, 38) 37 (30, 43) 28 (22, 34) 32 (28, 37) 40 (29, 51) 52 (46, 59) 54 (47, 62) 64 (57, 71) 50 (45, 56)
Sao Tome & Principe 2008/2009 1931 39 (26, 51) 38 (22, 53) 58 (40, 75) 18 (8, 27) 38 (28, 48) 76 (44, 109) 49 (33, 66) 93 (63, 123) 30 (15, 45) 63 (44, 81)
Senegal 2017 12,185 36 (28, 44) 45 (39, 51) 45 (38, 52) 38 (32, 45) 42 (37, 46) 43 (34, 51) 63 (56, 70) 61 (55, 69) 50 (44, 57) 56 (50, 61)
Sierra Leone 2013 11,938 95 (81, 109) 92 (83, 100) 98 (88, 108) 87 (78, 96) 92 (85, 100) 153 (134, 171) 157 (146, 168) 168 (155, 180) 144 (132, 156) 156 (146, 166)
South Africa 2016 3548 34 (21, 47) 39 (28, 49) 41 (29, 53) 29 (18, 41) 35 (26, 44) 38 (25, 51) 49 (37, 61) 49 (36, 61) 35 (23, 47) 42 (33, 51)
Tanzania 2015/2016 10,233 55 (44, 66) 39 (33, 45) 48 (41, 55) 38 (32, 45) 43 (38, 48) 81 (66, 95) 62 (55, 70) 70 (61, 79) 64 (55, 73) 67 (60, 74)
Togo 2013/2014 6979 41 (31, 51) 53 (45, 61) 54 (44, 64) 43 (36, 51) 49 (42, 55) 60 (46, 73) 103 (92, 115) 93 (80, 107) 83 (73, 94) 88 (79, 98)
Uganda 2016 15,522 39 (30, 49) 44 (39, 48) 49 (43, 54) 37 (32, 42) 43 (39, 47) 52 (41, 63) 68 (62, 73) 72 (65, 79) 56 (50, 63) 64 (59, 69)
Zambia 2013/2014 13,457 47 (39, 56) 43 (37, 49) 49 (42, 57) 40 (34, 45) 45 (40, 49) 71 (61, 81) 77 (69, 85) 80 (71, 90) 69 (61, 76) 75 (68, 81)
Zimbabwe 2015 6132 39 (30, 49) 55 (47, 63) 56 (46, 66) 44 (35, 53) 50 (44, 56) 52 (41, 63) 77 (66, 88) 76 (64, 87) 62 (52, 73) 69 (60, 78)
Ekholuenetale et al. ICEP (2020) 14:10 Page 8 of 15
Table 2 Summary statistics for pooled data; DHS 2008–2017
Variable N %
Household headship
Male 305,652 79.4
Female 79,094 20.6
Family type
Monogyny 242,784 75.6
Polygyny 78,491 24.4
Family size
1–2 127,684 33.2
3–5 161,578 42.0
≥6 95,484 24.8
Marital status for mothers
Never married 22,860 5.9
In union/living with a man 335,672 87.3
Formerly in union/living with a man 26,212 6.8
Number of unions for mothers
Once 309,179 85.8
More than once 51,188 14.2
Current residence of mother
Living with husband/partner 280,723 86.3
Staying elsewhere 44,747 13.7
Duration of couple in union
1 month–7 years 120,187 33.2
> 7 years–14 years 126,387 34.9
> 14 years–21 years 78,382 21.7
> 21 years 36,758 10.2
Never in union 170 0.1
Household wealth quintiles
Poorest 98,297 25.6
Poorer 84,607 22.0
Middle 75,708 19.7
Richer 67,919 17.6
Richest 58,215 15.1
Mother’s educational level
No formal education 159,327 41.4
Primary 134,113 34.9
Secondary 81,467 21.2
Higher 9788 2.5
Place of residence
Urban 114,475 29.8
Rural 270,271 70.2
Maternal age
15–19 24,514 6.4
20–24 87,151 22.7
25–29 103,929 27.0
30–34 80,195 20.8
35–39 54,946 14.3
40–44 26,038 6.8
45–49 7973 2.1
Ekholuenetale et al. ICEP (2020) 14:10 Page 9 of 15
having more than one wife at a time) and have at least 6 children ever born. However, approximately one-third (33.2%) of households have at most 2 children ever
born. Whilst 5.9% of women were never married, about 6.8% reported formerly in
union, others were currently in union/living with a man (87.3%). Results showed that
about 14.2% of women had been in more than one union and approximately 13.7%
currently living alone (staying elsewhere). There were disparities in the duration of
couples in union and this varied between 1 month–7 years and above 21 years in
union. About 41.4% of women had no formal education, whilst only 2.5% had higher
education. Approximately, 70.2% dwell in rural residence and disparities existed
across maternal age categories. Furthermore, about 93.5% of under-five children
were alive at the time of the survey. See Table 2 for the details.
Multivariable Cox Proportional Hazard model was used to examine the variables
of household structure with infant mortality. The risk of infant mortality was higher
in households of polygyny, compared with households of monogyny (HR=1.23; CI
1.16, 1.29). Households with large number of children (3–5 and ≥6) had higher risk
of infant mortality, compared with those with 1–2 number of children ever born.
Furthermore, infants from mothers with history of multiple unions had 16% increase
in the risk of infant mortality, compared with those from mothers from only one
union (HR=1.16; CI 1.09, 1.24). Children from households with longer duration in
union, high household wealth and maternal education had reduction in the risk of
infant mortality, respectively. See Table 3 for the details.
In Table 4, we presented the results of multivariable Cox model used to examine variables of household structure for under-five mortality. The under-five from
female household headship had 10% significant reduction in the risk of mortality,
compared with male household headship (HR =0.90; CI 0.84, 0.96). On the other
hand, the risk of under-five mortality was higher in households of polygyny, compared with households of monogyny (HR =1.33; CI 1.28, 1.38). Households with
large number of children (3–5 and≥6) had higher risk of under-five mortality, compared with those with 1–2 number of children ever born. In addition, under-five
children from mothers with history of multiple union had 30% increase in the risk
mortality, compared with those from mothers from only one union (HR =1.30; CI
1.24, 1.36). Children from households with longer duration in union, high household
wealth, maternal education and advanced maternal age had reduction in the risk of
under-five mortality, respectively. There was increased risk of under-five mortality
in rural residence, when compared with the urban (HR =1.07; CI 1.01, 1.12). See
Table 4 for details.
Table 2 (continued)
Variable N %
Child status
Dead 25,174 6.5
Alive 359,572 93.5
Ekholuenetale et al. ICEP (2020) 14:10 Page 10 of 15
Table 3 Household factors associated with infant mortality in SSA; DHS 2008–2017
Variable Model I Model II
HR (95%CI) P HR (95%CI) P
Household headship
Male 1.00
Female 0.98 (0.93, 1.03) 0.435 – –
Family type
Monogyny 1.00 1.00
Polygyny 1.45 (1.37, 1.52) < 0.001* 1.23 (1.16, 1.29) < 0.001*
Family size
1–2 1.00 1.00
3–5 1.59 (1.51, 1.68) < 0.001* 1.74 (1.61, 1,90) < 0.001*
≥6 2.27 (2.15, 2.40) < 0.001* 2.54 (2.27, 2.83) < 0.001*
Marital status for mothers
Never married 1.00 1.00
In union/living with a man 1.49 (1.35, 1.64) < 0.001* n/a
Formerly in union/living with a man 2.17 (1.92, 2.45) < 0.001* n/a
Number of unions for mothers
Once 1.00 1.00
More than once 1.33 (1.25, 1.41) < 0.001* 1.16 (1.09, 1.24) < 0.001*
Current residence of mother
Living with husband/partner 1.00 1.00
Staying elsewhere 0.90 (0.84, 0.96) 0.002* 0.97 (0.90, 1.04) 0.416
Duration of couple in union
1 month–7 years 1.00 1.00
> 7 years–14 years 1.34 (1.27, 1.41) < 0.001* 0.89 (0.82, 0.97) 0.006*
> 14 years–21 years 1.49 (1.40, 1.59) < 0.001* 0.82 (0.74, 0.92) 0.001*
> 21 years 2.13 (1.98, 2.29) < 0.001* 0.90 (0.78, 1.04) 0.147
Never in union 0.26 (0.04, 1.85) 0.178 0.31 (0.04, 2.17) 0.236
Household wealth quintiles
Poorest 1.00 1.00
Poorer 1.02 (0.96, 1.08) 0.565 1.06 (0.99, 1.14) 0.061
Middle 0.93 (0.87, 0.99) 0.027* 1.00 (0.93, 1.07) 0.970
Richer 0.82 (0.77, 0.88) < 0.001* 0.92 (0.85, 0.99) 0.028*
Richest 0.66 (0.61, 0.71) < 0.001* 0.82 (0.74, 0.90) < 0.001*
Mother’s educational level
No formal education 1.00 1.00
Primary 0.79 (0.75, 0.83) < 0.001* 0.88 (0.83, 0.93) < 0.001*
Secondary 0.60 (0.56, 0.64) < 0.001* 0.82 (0.76, 0.89) < 0.001*
Higher 0.38 (0.32, 0.47) < 0.001* 0.64 (0.50, 0.81) < 0.001*
Place of residence
Urban 1.00 1.00
Rural 1.28 (1.22, 1.35) < 0.001* 1.03 (0.96, 1.10) 0.372
Maternal age
15–19 1.00 1.00
20–24 1.32 (1.21, 1.45) < 0.001* 1.04 (0.93, 1.16) 0.491
25–29 1.48 (1.36, 1.62) < 0.001* 0.90 (0.80, 1.02) 0.102
30–34 1.56 (1.42, 1.71) < 0.001* 0.80 (0.70, 0.92) 0.001*
35–39 1.76 (1.59, 1.93) < 0.001* 0.80 (0.69, 0.93) 0.004*
40-44 2.43 (2.17, 2.71) < 0.001* 0.94 (0.79, 1.12) 0.492
45-49 3.58 (3.08, 4.17) < 0.001* 1.32 (1.07, 1.63) 0.011*
Ekholuenetale et al. ICEP (2020) 14:10 Page 11 of 15
Discussion
In this study, we examined the diferences in infant and under-fve mortality across SSA
countries, by place of residence and sex of children. In addition, we investigated the
association between household structure and child mortality using multivariable Cox
models. Based on the results, SSA countries showed high rates of infant and under-fve
mortality. Comparing with the third Sustainable Development Goal (SDG-3) of ending
preventable deaths of newborns and children under 5 years of age, with all countries
aiming to reduce under-5 mortality to a minimum of 25 per 1000 live births by 2030
(Rosa, 2017); the fndings showed that most SSA countries still have high child mortality. Te rates appear to arise largely from household structure. Poor household structure could adversely increase the risk of child mortality. Te fndings are consistent with
reports from a previous study (Guillot et al. 2012).
Tere was reduction in the risk of under-fve mortality in female-headed households.
Similar fndings have been reported by Adhikari and Podhisita in which female-headed
household had reduction in the risk of under-fve mortality (Adhikari and Podhisita
2010). Tis study found that amongst many other factors, household headship was
a strong determinant of under-fve mortality. Tis shows that women’s autonomy and
empowerment through improved maternal literacy, ability to decide independently on
the use of maternal healthcare services including paediatric care, could help to reduce
under-fve mortality. Furthermore, we found reduction in the risk of infant and underfve mortality with longer marriage duration. Tis is contrary to a previous report, which
found the proportions of child death increased with marital duration (Islam, Rahman,
Rahman 2013). Tis could be due to the fact that the average parities reported from the
study increased monotonically with the duration of marriage. Moreover, women with
longer duration in marriage may have more knowledge to prevent preterm birth complications, pneumonia, interpartum-related events, neonatal sepsis, diarrhoea, malaria,
severe undernutrition amongst other factors which could contribute to high risk of child
mortality.
High household wealth status, maternal education and advanced maternal age were
found to signifcantly reduce the risk of child mortality. Tis is consistent with reports
from previous studies which found maternal education and improved household wealth
index to be associated with reduction in the risk of child death (Adebowale et al. 2012;
Yaya et al. 2017). Notably, mothers’ socioeconomic status, specifcally education and
wealth status could be linked with quality health practices and proper health behaviour
including childcare and optimal feeding habits. Te socioeconomic status of a mother
could modify her role in the family and equip her in taking measures to improve child’s
health by adequately using modern and innovative health services (Buor, 2003).
In addition, children within polygynous marriages were found to have higher risk
of infant and under-fve mortality, when compared with children from monogynous
families. Te fndings from previous studies supported the arguments that child
death is associated to marriage types (Arthi and Fenske 2018; Lawson and Gibson
Table 3 (continued)
Model I Crude model, Model II Adjusted model, n/a not estimated due to small sample, HR hazard ratio, CI Confdence
Interval
*Signifcant at p < 0.05
Ekholuenetale et al. ICEP (2020) 14:10 Page 12 of 15
Table 4 Household factors associated with under-fve mortality in SSA; DHS 2008–2017
Variable Model III Model IV
HR (95%CI) P HR (95%CI) P
Household headship
Male 1.00 1.00
Female 0.88 (0.85, 0.92) < 0.001* 0.90 (0.84, 0.96) 0.002*
Family type
Monogyny 1.00 1.00
Polygyny 1.51 (1.46, 1.57) < 0.001* 1.33 (1.28, 1.38) < 0.001*
Family size
1–2 1.00 1.00
3–5 1.29 (1.24, 1.35) < 0.001* 1.83 (1.73, 1.95) < 0.001*
≥6 1.78 (1.71, 1.86) < 0.001* 3.28 (3.03, 3.55) < 0.001*
Marital status for mothers
Never married 1.00 1.00
In union/living with a man 1.25 (1.16, 1.36) < 0.001* n/a
Formerly in union/living with a man 1.41 (1.28, 1.55) < 0.001* n/a
Number of unions for mothers
Once 1.00 1.00
More than once 1.33 (1.28, 1.39) < 0.001* 1.30 (1.24, 1.36) < 0.001*
Current residence of mother
Living with husband/partner 1.00 1.00
Staying elsewhere 0.94 (0.89, 0.99) 0.015* 1.04 (0.97, 1.11) 0.271
Duration of couple in union
1 month–7 years 1.00 1.00
> 7 years–14 years 1.04 (1.00, 1.09) 0.045* 0.78 (0.73, 0.82) < 0.001*
> 14 years–21 years 1.08 (1.03, 1.13) 0.001* 0.66 (0.61, 0.71) < 0.001*
> 21 years 1.27 (1.21, 1.35) < 0.001* 0.72 (0.65, 0.80) < 0.001*
Never in union 0.94 (0.39, 2.26) 0.890 0.87 (0.36, 2.09) 0.756
Household wealth quintiles
Poorest 1.00 1.00
Poorer 0.98 (0.94, 1.03) 0.465 1.02 (0.97, 1.07) 0.419
Middle 0.86 (0.82, 0.91) < 0.001* 0.93 (0.88, 0.98) 0.007*
Richer 0.77 (0.73, 0.81) < 0.001* 0.89 (0.0.84, 0.94) < 0.001*
Richest 0.56 (0.0.53, 0.60) < 0.001* 0.77 (0.71, 0.83) < 0.001*
Mother’s educational level
No formal education 1.00 1.00
Primary 0.74 (0.71, 0.77) < 0.001* 0.78 (0.75, 0.82) < 0.001*
Secondary 0.57 (0.54, 0.59) < 0.001* 0.74 (0.70, 0.79) < 0.001*
Higher 0.27 (0.23, 32) < 0.001* 0.52 (0.42, 0.64) < 0.001*
Place of residence
Urban 1.00 1.00
Rural 1.39 (1.33, 1.44) < 0.001* 1.07 (1.01, 1.12) 0.013*
Maternal age
15–19 1.00 1.00
20–24 0.75 (0.70, 0.81) < 0.001* 0.63 (0.58, 0.69) < 0.001*
25–29 0.73 (0.68, 0.79) < 0.001* 0.51 (0.46, 0.56) < 0.001*
30–34 0.71 (0.66, 0.76) < 0.001* 0.41 (0.37, 0.45) < 0.001*
35–39 0.74 (0.68, 0.80) < 0.001* 0.35 (0.31, 0.40) < 0.001*
40–44 0.78 (0.71, 0.85) < 0.001* 0.32 (0.28, 0.37) < 0.001*
45–49 0.97 (0.87, 1.09) 0.645 0.37 (0.32, 0.44) < 0.001*
Ekholuenetale et al. ICEP (2020) 14:10 Page 13 of 15
2018; Omariba and Boyle 2007; Smith-Greenaway and Trinitapoli 2014; Wagner and
Rieger, 2015). Similarly, history of mothers’ involvement in multiple union was found
to be associated with increased risk of child mortality. Also, this is consistent with
report from a previous study (Arntzen et al. 1996). In general, there is no doubt that
polygyny and history of involvement in multiple unions by mothers could be closely
connected to increased number of children ever born, which was also found to be
associated with higher risk of child mortality. Tis is in line with previous fndings
(Kabagenyi and Rutaremwa 2013; Sonneveldt et al. 2013). Te major pathway by
which polygyny is known to negatively infuence child survival is through resource
dilution, with the assumption that polygyny leads to a greater number of children
to support on a limited family budget. Based on this, there is a competing paradigm
that polygyny impairs the survival chances of children. Polygyny and high parity can
be linked with high child mortality via resource constraints, paternal investment
and selectivity. Te resource constraint is premised on the notion that usually large
households are associated with low resource per head which adversely impacts on
child health and survival.
Furthermore, the resource constraint in family resources for the upkeep of children may lead to poor living conditions which could possibly increase the vulnerability of children to diseases and subsequently death. Te lack of resources can also limit
access to modern healthcare particularly with the cash-and-carry system operated in
several SSA countries. Te large family size associated with the foregoing could practically reduce parent–child emotional attachments, which is crucial in promoting active
childcare. Te implication is that children from large family household’s size may be less
catered for, more vulnerable and consequently exposed to higher risk of death. In addition, rural place of residence was found to be associated with higher under-fve mortality. Te urban–rural diferentials have previously been identifed in child mortality
(Gruebner et al. 2015; van De Poel et al. 2009; Yaya et al. 2017). Tis could be explained
in terms of family poverty and consequently inaccessibility of paediatric healthcare services. Conventionally, rural children have poor access to healthcare services utilisation;
as such early detection of abnormalities for appropriate management is unlikely. Furthermore, higher child mortality rates in rural areas are mainly derived from the socioeconomically disadvantaged household characteristics. Safe source of drinking water
and improved sanitation system which are key contributors of healthy living are commonly lacking in rural areas.
Te fndings from this study have unravelled the relationship between household structure and child mortality in SSA. Te policy implication of the fndings is that concerted
eforts of social welfare intervention towards household structure improvement could
serve as a panacea for child mortality by operating through either socioeconomic status or
fertility-related behaviour. For instance, household socioeconomic status determined the
children nutrition patterns (Ekholuenetale et al. 2020). No doubt, the time and resources
available for a child’s care are assumed to be afected by household structure or family
Table 4 (continued)
Model III Crude model, Model IV Adjusted model, n/a not estimated due to small sample, HR hazard ratio; CI Confdence
Interval
*Signifcant at p < 0.05
Ekholuenetale et al. ICEP (2020) 14:10 Page 14 of 15
characteristics (Akinyemi et al. 2013). Terefore, positive changes in household structure
can largely be promoted by family involvement in social welfare intervention.
Strength and limitation
Tis study utilised nationally representative multi-country data collected via standardised
questionnaires to ensure similarities across geographies in SSA countries and to strengthen
the evidence base. A major limitation for this study is the cross-sectional design, which
makes exploration of the pathway of household characteristics on child survival difcult.
Furthermore, the categories used for household relationship structure from which household type was derived did not permit exploration of the efect of inter-generational households. Caution must be taken in making comparisons of mortality across countries as the
data were collected in diferent years.
Conclusion
In this study, our focus was to examine the impact of the household structure on infant
and under-fve mortality in SSA countries. Te fndings suggest that polygyny, large family
size or increased number of children ever born, history of mothers’ involvement in multiple union and rural residence were associated with the risk of child mortality. Conversely,
female household headship, long duration in union, maternal education, and improved
household wealth status were associated with reduction in the risk of infant and under-fve
mortality. In light of the above, we suggest that healthcare programmes and policies should
be designed specifcally to encourage healthy family structure. In addition, policies to support strong healthy families would help to intervene in the areas which likely connect family structure to children’s outcomes, including parental life style and practices. In addition,
the allocation of funds under parenthood, marriage and family life should focus on programmes to improve social support networks.
Acknowledgements
The authors appreciate the MEASURE DHS project for the approval and access to the original data.
Authors’ contributions
ME conceived and designed the study, performed data analysis and wrote the results; AIW, GT, and AO contributed to
the review of literature, discussion of the fndings and critically reviewed the manuscript for its intellectual content. ME
had the responsibility to submit the manuscript. All authors read and approved the fnal manuscript.
Funding
This research received no grant from any funding agency in the public, commercial or not-for-proft sectors.
Availability of data and materials
Data for this study were sourced from Demographic and Health surveys (DHS) and available in http://dhsprogram.com/
data/available-datasets.cfm.
Competing interests
The authors declare that the research was conducted in the absence of any commercial or fnancial relationships that
could be construed as a potential confict of interest.
Author details
1 Department of Epidemiology and Medical Statistics, Faculty of Public Health, College of Medicine, University of Ibadan,
Ibadan, Nigeria. 2 Department of Mathematics, Rivers State University, Port Harcourt, Nigeria. 3 School of Medicine, College of Medical Sciences, University of Benin, Benin City, Nigeria. 4 Program Management Unit, Management Sciences
for Health, Abuja, Nigeria.
Received: 8 May 2019 Accepted: 23 July 2020
Ekholuenetale et al. ICEP (2020) 14:10 Page 15 of 15
References
Adebowale, A. S., Yusuf, B. O., & Fagbamigbe, A. F. (2012). Survival probability and predictors for woman experience childhood death in Nigeria: “Analysis of north–south diferentials”. BMC Public Health, 12, 430. https://doi.
org/10.1186/1471-2458-12-430.
Adhikari, R., & Podhisita, C. (2010). Household headship and child death: evidence from Nepal. BMC International Health and
Human Rights, 10, 13.
Ajao, K., Ojofeitimi, E., Adebayo, A., Fatusi, A., & Afolabi, O. (2010). Infuence of family size, household food security status,
and child care practices on the nutritional status of under-fve children in Ile-Ife, Nigeria. African Journal of Reproductive
Health, 14(4 Spec no.), 117–126.
Akinyemi, J. O., Bamgboye, E. A., & Ayeni, O. (2013). New trends in under-fve mortality determinants and their efects on
child survival in Nigeria: A review of childhood mortality data from 1990-2008. African Population Stud. https://doi.
org/10.11564/27-1-5.
Akinyemi, J. O., Chisumpa, V. H., & Odimegwu, C. O. (2016). Household structure, maternal characteristics and childhood
mortality in rural sub-Saharan Africa. Rural Remote Health, 16(2), 3737.
Arntzen, A., Moum, T., Magnus, P., & Bakketeig, L. S. (1996). Marital status as a risk factor for fetal and infant mortality. Scand J
Social Med, 24(1), 36–42.
Arthi, V., & Fenske, J. (2018). Polygamy and child mortality: historical and modern evidence from Nigeria’s Igbo. Rev Econ
Household, 16(1), 97–141. https://doi.org/10.1007/s11150-016-9353-x.
Buor, D. (2003). Mothers’ education and childhood mortality in Ghana. Health Policy, 64(3), 297–309.
Burke, M., Heft-Neal, S., & Bendavid, E. (2016). Sources of variation in under-5 mortality across sub-Saharan Africa: a spatial
analysis. Lancet Global Health, 4(12), e936–e945. https://doi.org/10.1016/S2214-109X(16)30212-1.
Corsi, D. J., Neuman, M., Finlay, J. E., & Subramanian, S. (2012). Demographic and health surveys: a profle. Int J Epidemiol, 41(6),
1602–1613. https://doi.org/10.1093/ije/dys184.
Ekholuenetale, M., Tudeme, G., Onikan, A., & Ekholuenetale, C. E. (2020). Socioeconomic inequalities in hidden hunger, undernutrition, and overweight among under-fve children in 35 sub-Saharan Africa countries. J Egyptian Public Health Assoc,
95(1), 9. https://doi.org/10.1186/s42506-019-0034-5.
Ester, P. V., Torres, A., Freire, J. M., Hernández, V., & Gil, Á. (2011). Factors associated to infant mortality in Sub-Saharan Africa. J
Public Health Afr. https://doi.org/10.4081/jphia.2011.e27.
Freeman, L. L., & Brewer, M. (2013). Family Matters: Links Between Family Structure and Early Child Health. Family Matters, 25.
Gruebner, O., Lautenbach, S., Khan, M. M. H., Kipruto, S., Epprecht, M., & Galea, S. (2015). Place of Residence Moderates the
Risk of Infant Death in Kenya: evidence from the Most Recent Census 2009. PLoS ONE, 10(10), e0139545. https://doi.
org/10.1371/journal.pone.0139545.
Guillot, M., Gerland, P., Pelletier, F., & Saabneh, A. (2012). Child mortality estimation: a global overview of infant and child
mortality age patterns in light of new empirical data. Plos Medicine, 9(8), e1001299. https://doi.org/10.1371/journ
al.pmed.1001299.
Islam, R., Rahman, M., & Rahman, O. (2013). Estimation of childhood mortality in Bangladesh: indirect approach. Appl Math,
3(2), 61–69.
Kabagenyi, A., & Rutaremwa, G. (2013). The efect of household characteristics on child mortality in Uganda. Am J Sociol Res,
3(1), 1–5.
Kazembe, L., Clarke, A., & Kandala, N.-B. (2012). Childhood mortality in sub-Saharan Africa: cross-sectional insight into smallscale geographical inequalities from Census data. Br Med J Open, 2(5), e001421. https://doi.org/10.1136/bmjopen-2012-
001421.
Lawson, D., & Gibson, M. A. (2018). Polygynous marriage and child health in sub-Saharan Africa: what is the evidence for
harm? Demographic Res, 39, 177–208. https://doi.org/10.4054/DemRes.2018.39.6.
Midi, H., Sarkar, S. K., & Rana, S. (2010). Collinearity diagnostics of binary logistic regression model. J Interdiscip Math, 13(3),
253–267. https://doi.org/10.1080/09720502.2010.10700699.
Ntoimo, L. F., & Odimegwu, C. O. (2014). Health efects of single motherhood on children in sub-Saharan Africa: A crosssectional study. BMC Public Health. https://doi.org/10.1186/1471-2458-14-1145.
Omariba, D. W. R., & Boyle, M. H. (2007). Family structure and child mortality in sub-saharan Africa: cross-national efects of
polygyny. J Marriage Family, 69(2), 528–543. https://doi.org/10.1111/j.1741-3737.2007.00381.x.
Rosa, W. (Ed.). (2017). Transforming Our World: The 2030 Agenda for Sustainable Development. In A New Era in Global Health.
Springer Publishing Company. https://doi.org/10.1891/9780826190123.ap02
Rutstein, S. O., & Staveteig, S. (2014). Making the demographic and health surveys wealth index comparable. DHS methodological reports No. 9. Rockville, Maryland, USA: ICF International.
Smith-Greenaway, E., & Trinitapoli, J. (2014). Polygynous contexts, family structure, and infant mortality in sub-saharan Africa.
Demography, 51(2), 341–366. https://doi.org/10.1007/s13524-013-0262-9.
Sonneveldt, E., DeCormier Plosky, W., & Stover, J. (2013). Linking high parity and maternal and child mortality: what is the
impact of lower health services coverage among higher order births? BMC Public Health, 13(Suppl 3), S7. https://doi.
org/10.1186/1471-2458-13-S3-S7.
Van De Poel, E., O’donnell, O., & Van Doorslaer, E. (2009). What explains the rural–urban gap in infant mortality: household or
community characteristics? Demography, 46(4), 827–850.
Wagner, N., & Rieger, M. (2015). Polygyny and child growth: evidence from twenty-six African Countries. Feminist Econ, 21(2),
105–130. https://doi.org/10.1080/13545701.2014.927953.
Yaya, S., Ekholuenetale, M., Tudeme, G., Vaibhav, S., Bishwajit, G., & Kadio, B. (2017). Prevalence and determinants of childhood
mortality in Nigeria. BMC Public Health. https://doi.org/10.1186/s12889-017-4420-7.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional afliations.