RESEARCH ARTICLE
Socio-demographic determinants of low birth
weight: Evidence from the Kassena-Nankana
districts of the Upper East Region of Ghana
Isaiah Awintuen Agorinya 1,2,3ID *, Edmund Wedam Kanmiki4, Engelbert
Adamwaba Nonterah1,5, Fabrizio Tediosi3, James Akazili 1,2ID , Paul Welaga
1,
Daniel Azongo1, Abraham Rexford Oduro1
1 Navrongo Health Research Centre, Navrongo, Ghana, 2 INDEPTH-Network Secretariat, Accra, Ghana,
3 Swiss Tropical and Public Health Institute, University of Basel, Basel, Switzerland, 4 Regional Institute for
Population Studies, University of Ghana, Accra, Ghana, 5 Julius Centre for Health Sciences and Primary
a1111111111 Care, UMCU, Utrecht, The Netherlands
a1111111111
a1111111111 * iagorinya@gmail.com
a1111111111
a1111111111
Abstract
Objective
OPEN ACCESS To examine the social, economic and demographic factors that determine low birth weight in
Citation: Agorinya IA, Kanmiki EW, Nonterah EA, the two Kassena Nankana districts of the Upper East region of Ghana.
Tediosi F, Akazili J, Welaga P, et al. (2018) Socio-
demographic determinants of low birth weight: Methods
Evidence from the Kassena-Nankana districts of
the Upper East Region of Ghana. PLoS ONE 13 Cross-sectional data was collected from January 2009 to December 2011 using the Nav-
(11): e0206207. https://doi.org/10.1371/journal. rongo Health and Demographic Surveillance System which monitors routine health and
pone.0206207
demographic outcomes in the study area. Data on foetal characteristics such as birth
Editor: Sabine Rohrmann, University of Zurich, weight, and sex and maternal age, parity, maternal education, marital status, ethnicity, reli-
SWITZERLAND
gious affiliation and socio-economic characteristics were collected and described. Tests of
Received: February 28, 2018 means, proportions and Chi-squares are employed in bivariate analysis, and adjusted logis-
Accepted: October 9, 2018 tic regression models fitted to control for potential confounding variables. All tests were two-
sided and test of significance was set at p-value of < 0.05.
Published: November 14, 2018
Copyright: © 2018 Agorinya et al. This is an open
Results
access article distributed under the terms of the
Creative Commons Attribution License, which There were 8,263 live births (44.9% females) with an overall average birth weight of 2.85 kg
permits unrestricted use, distribution, and (2.9 kg for males and 2.8 kg for females). The average maternal age was 28 years, median
reproduction in any medium, provided the original
parity 2, maternal literacy rate was about 70% and 83% of mothers were married. The prev-
author and source are credited.
alence of low birth weight was 13.8% 95%CI [13.10, 14.6] and more in female babies than in
Data Availability Statement: All relevant data are
males (15.5% vs 12.2%; p<0.0001). Determinants of low birth-weight after controlling for
within the paper and its Supporting Information
files. confounding factors were sex of neonate (OR = 1.32, 95%CI [1.14,1.52]; p<0.0001), mater-
nal age (p = 0.004), and mothers who are not married (OR = 1.44 [1.19, 1.74]; p<0.0001).
Funding: The authors received no specific funding
for this work. The data used for this work is part of
routine data collection of the Navrongo Health and Conclusion
Demographic surveillance System which is funded
primarily by the Navrongo Health Research Centre. Female neonates in this population were likely to present with low birth weight and maternal
This study is primarily a student research work. factors such as younger age, lower socio-economic status and single parenthood were
PLOS ONE | https://doi.org/10.1371/journal.pone.0206207 November 14, 2018 1 / 10
Socio-demographic determinants low birth weight
Competing interests: The authors have declared major determinants of low birth weight. Effective and adequate antenatal care should there-
that no competing interests exist. fore target women with these risk factors.
Background
Most low and middle-income countries could not meet the target for Millennium Develop-
ment Goal 4 (MDG) which is reduction in childhood mortality by two-thirds between 1990
and 2015 [1]. This was in spite of global political, financial and social commitment to the
attainment of the MDGs [2]. This calls for more efforts especially in developing countries to
unearth context-specific factors that play critical roles in child survival and development.
Birth weight is known to influence the survival and development of neonates. It refers to
the first weight measurement taken of a new-born immediately after birth. Normal birth
weight is essential for child survival, development and health later in life. According to the
World Health Organization (WHO), Low Birth-Weight (LBW) is weight at birth that is less
than 2.5 kg [2]. A LBW infant can be born too small (small for gestational age), too early (pre-
term) or both. Several epidemiological studies have shown that neonates with LBW have a dis-
proportionately higher mortality compared to those with weight higher than 2.5 kg [3]. Even
though recent evidences point to declining mortalities among LBW infants, adverse develop-
mental outcomes have still been observed in surviving children [4–6].
A number of studies have reported an association between LBW and pulmonary dysfunc-
tion, impaired physical growth, adverse neurological outcome, psycho-social development and
social disadvantages [6–9]. An association between LBW and increased respiratory, cognitive,
neurological and psychological deficits or dysfunction have also been reported by Kelly et al
and Gissler et al [4,8]. Other scholars have observed low school performance, delayed psycho-
motor development, adverse emotional well-being, as well as increased conduct disorders
among children and adolescents with prior LBW [9]. Evidence from life course epidemiologi-
cal studies indicates that LBW is a major contributor to later development of cardiometabolic
diseases such as obesity and atherosclerosis [10]. Further to this, it has been identified as an
important contributor to infant and childhood mortality and morbidity [11].
Several studies have reported on the determinants of LBW but these have often not focussed
on socio-demographic factors [4]. Non-seasonal factors such as increasing maternal age,
socio-economic factors, racial and ethnic differences and availability of health care services
have been associated with LBW by several researchers [5,12,13].
Interventions to reduce LBW will require an understanding of the associated foetal, mater-
nal, prevailing environmental and socio-demographic factors. There is however, limited infor-
mation on this from Northern Ghana. This study examined biological and environmental
factors that influence low birth weight in the coverage area of the Navrongo Health and Demo-
graphic Surveillance Site (NHDSS) in Upper East region of Ghana.
Methods
Study setting
The study was conducted in the Navrongo Health and Demographic Surveillance site. The
area lies between latitude 10.300 and 11.100 North and longitude 1.10 West [14]. This area is
predominantly Guinea Savannah vegetation with two seasons in a year. It has a short rainy
season from April to September and a prolonged dry season from October to March with tem-
peratures ranging from 22.88C to 34.48C [15]. The Navrongo Health and Demographic
PLOS ONE | https://doi.org/10.1371/journal.pone.0206207 November 14, 2018 2 / 10
Socio-demographic determinants low birth weight
Surveillance System (NHDSS) is the routine Data collection arm of the Navrongo Health
Research Centre which has been in operation in the last 20 years. Ethical clearance was
obtained from the Navrongo Health Research Centre Institutional Review Board (NHRC IRB)
for the operations of the NHDSS and signed informed consent obtained from all participating
households within the study area before data collection. The surveillance area is divided into 5
zone within which are 247 clusters and currently monitors 160, 000 people living in 18,000
compounds (an average of 70 per clusters) and with 32,000 households [14]. The area has one
referral district hospital, seven health centres and several primary health-care clinics and 37
Community based Health and Planning Services (CHPS) compounds with resident commu-
nity health officers providing door-to-door health services to the community.
The study area has a crude birth rate of 25/1000, crude death rate of 10/1000, total fertility
rate of 3.8, neonatal mortality rate of 13.4 per 1000 live births, and infant mortality rate of 32.1.
Under-five mortality rate per 1000 live births is report as 60.8; life expectancy at birth is 56.4
years for males and 67 years for females [14].
Study population and sample
The study involved all mothers and their babies born at term (37 weeks and above) from Janu-
ary 2009 to December 2011 and whose parents were resident or natives of the study area. The
data was collected through the NHDSS which collects and updates health and socio-demo-
graphic characteristics of all persons and households in the two Kassena-Nankana districts of
the study area. This involves the regular visits to households by trained fieldworkers every 4
months to interview heads of households or an adult member of the household. Routinely col-
lected data included pregnancies, live and stillbirths, morbidity, deaths, in-migration and out-
migrations, childhood vaccinations and verbal autopsy (VAs) on all reported deaths within its
operational area. In addition, information on household socio-economic characteristics is also
collected.
Outcome and exposure variables
The main outcome variable for this study is birth weight which was recoded into two mutually
exclusive binary outcome: low birth weight defined as birth weight < 2.5 kg and normal birth
defined as birth weight� 2.5 kg according to WHO recommendations [16]. Covariates was
extracted from the NHDSS data and they included; maternal characteristics such as age, reli-
gion, educational status, marital status, parity, and ethnicity. New-born characteristics
included date of birth, birth weight and sex at birth. These variables were selected based on the
available literature and the potential to influence birth weight. Socio-economic status which is
estimated from the wealth index of the household (used as a proxy for household income) was
constructed in quintiles (1 = poorest, 2 = poor, 3 = average, 4 = rich, 5 = richest) using Princi-
pal Component Analysis (PCA).
Maternal ages at delivery were calculated using the mothers’ and babies’ birthdates. Educa-
tional status was defined as either not having any form of formal education or the highest level
of formal education attained. Parity refers to the number of births by a mother including the
index baby and ethnicity referred to self-reported ethno-linguistic grouping the mother identi-
fies with.
Data management and analysis
All data are routinely collected using compound registration system, checked for inconsisten-
cies and corrected before they are entered into a Foxpro database. Five per cent of all house-
hold data are randomly selected and re-entered as a quality control measure.
PLOS ONE | https://doi.org/10.1371/journal.pone.0206207 November 14, 2018 3 / 10
Socio-demographic determinants low birth weight
The variables were extracted and analysed using the statistical software STATA version 12.0
SE (Stata corp Texas, USA).
Descriptive statistics was used to describe the baseline maternal and new born characteris-
tics. Categorical data are presented in proportions and the differences in socio-demographic
characteristics by birth weight status examined using Pearson’s Chi squared test.
Determinants of LBW were examined using logistic regression analysis. Reference catego-
ries were defined as those usually associated with the lowest birth weight. Odds ratios with
95% confidence intervals are presented and factors associate with LBW are those with a
p<0.05. For categorical variables with more than two categories, a post-estimation analysis
was done to examine the overall significance of the variable and one p-value obtained. The var-
iance inflation test was employed to test for multi-collinearity between related variables and
variables that were found to be collinear were those with a VIF of more than 10 and these
excluded from the final logistic regression model.
Results
Presented in Table 1 are the basic characteristics of neonates and mothers. A total of 8,263
women with singleton babies at term were analysed in this study. More than half (53.3%) of
the women were from the Kassena ethnic group. There was a near equal number of female
(49.9%) and male (50.1%) neonates. The total average birth weight was 2.85 ±0.52 kg; 2.90 kg
for males and 2.80 kg for females. The proportion of LBW was 13.8% 95%CI (13.1, 14.6) with a
mean LBW of 1.94 kg (±0.51SD). The proportion of female neonates with LBW was 15.5%
95%CI with an average LBW of 1.95kg (±0.34 SD) and the proportion of LBW among the
male neonates was 12.2% (9.4, 15.3) with an overall mean LBW of 1.94 kg (± 0.51SD).
The average maternal age was 28 years (±7.4 SD), about 11% of the mothers were teenagers,
about 31% had not attained any formal education, about 23% of the mothers were in the poor-
est quintile and 18% in the rich quintile. 83% of the mothers were married.
A large fraction of the LBW outcomes 19.8% (17.3, 22.5) were from mothers in the age
group 15–19 years and the least from mothers aged 35 plus 11.7% (10.2, 13.3). Large propor-
tion (20.5%) of mothers, who previously had one child, had higher LBW children compared to
those who had between 2–3 children (20.5% vs 11.0%). The proportion was still high when
compared with their counterparts who previously had 4 or more children (20.5% vs 11.1%).
Mothers who belong to a household in the upper socio-economic class (Rich, 12.8% & Richest,
13.7%) and lower socio-economic (Poor, 12.2% & Poorest, 14.7%) class had comparable pro-
portion of birth weights, but those who were in the middle class (Average) had large propor-
tion (16%) of low weight babies (Pearson chi2(4) = 12.2861 P-value = 0.015).
Bivariable and multivariable logistic regression analysis
An initial unadjusted bivariable analysis of the maternal and socio-demographic characteris-
tics showed that sex of neonate, maternal age, household socio-economic status and marital
status significantly influenced the birth weight of a neonate as shown in Table 2.
Multicollinearity was detected between maternal age and parity after performing a variance
inflation test (VIF) and thus parity was not included in the adjusted logistic regression model.
Factors associated with LBW from the multivariable analyses included sex of neonate,
maternal age, and marital status. Female neonates had about 32% increased odds of being
born with low birth weights compared to their male colleagues; (OR = 1.32 95%CI [1.14, 1.52];
p< 0.0001).
Mothers in the 20-34-year age group were less likely to have neonates with LBW compared
to those below 20 years (OR = 0.69 95% CI 0.55, 0.87, p = 0.004). Similarly, the odd of having
PLOS ONE | https://doi.org/10.1371/journal.pone.0206207 November 14, 2018 4 / 10
Socio-demographic determinants low birth weight
Table 1. Descriptive statistics of maternal and neonatal characteristics.
Socio-demographic parameters LBW (<2500g) (%) Total P-value
Sex of neonate
Male 502 (12.2) 4133 <0.0001
Female 641 (15.5) 4130
Total 1143 (13.8) 8263
Maternal age (yrs)
15–19 186 (19.8) 939 <0.0001
20–34 759 (13.5) 5628
35+ 198 (11.7) 1696
Total 1143 (13.8) 8263
Maternal parity <0.0001
1 498 (20.5) 2429
2–3 360 (11.0) 3265
4+ 285 (11.1) 2569
Total 1143 (13.8) 8263
Maternal Education
No formal education 293 (12.8) 2293 0.080
Primary 311 (12.5) 2483
JHS 265 (15.1) 1761
SHS 105 (14.7) 716
Tertiary 41 (15.4) 266
��Total 1015 (13.5) 7519
Socio-economic status
Poor 218 (12.2) 1794 0.015
Next poor 214 (14.7) 1454
Average 222 (16.1) 1383
Next rich 232 (13.7) 1698
Rich 180 (12.8) 1408
Missing 77 (14.6) 526
Marital Status
Married 711 (12.2) 5845 <0.0001
Not Married 224 (19.2) 1170
��Total 935 (13.3) 7015
Ethnicity 0.197
Kasena 590 (14.3) 4118(100)
Kasem 590 (14.3) 4118
Nankana 412 (12.9) 3196
Builsa 34 (16.8) 203
Others 30 (13.6) 221
��Total 1066 (13.8) 7738
Religion
Traditional 395 (13.7) 2885
Catholic 271 (14.3) 1898
Other Christians 304 (13.3) 2282 0.833
Islam 94 14.4) 651
Other 2 (9.1) 22
��Total 1066 (13.8) 7738
�� Total not 8263 due to missing values in the variables
https://doi.org/10.1371/journal.pone.0206207.t001
PLOS ONE | https://doi.org/10.1371/journal.pone.0206207 November 14, 2018 5 / 10
Socio-demographic determinants low birth weight
Table 2. Maternal, demographic and socio-economic factors associated with LBW in the Kassena-Nankana districts.
Determinants Unadjusted OR P-value Adjusted OR P-value
(95% CI) (95% CI)
Sex of Child
Male 1 1
Female 1.33 (1.17, 1.51) <0.000 1.32 (1.14, 1.52) <0.001
Mothers age
15–19 yrs 1 1
20-34yrs 0.63 (0.53, 0.75) <0.000 0.69 (0.55, 0.87) 0.004
35+ yrs 0.54 (0.43, 0.67) 0.65 (0.49, 0.85)
Mother education
No education 1 1
Primary 0.98 (0.82, 1.16) 0.001 0.88 (0.73, 1.06) 0.190
JHS 1.21 (1.01, 1.45) 1.08 (0.88, 1.33)
SHS 1.17 (0.92, 1.49) 1.16 (0.87, 1.54)
Tertiary 1.24 (0.87, 1.77) 1.49 (0.95, 2.34)
Socio-economic status
Poorest 1 1
Poor 0.80 (0.65, 0.98 0.016 0.81 (0.65, 1.02) 0.002
Average 1.11(0.90, 1.36) 1.13 (0.91, 1.42)
Rich 0.85 (0.69, 1.05) 0.65 (0.49, 0.87)
Richest 0.92 (0.75, 1.12) 0.86 (0.69, 1.09)
Marital status
Married 1 1
Not married 1.71 (1.45, 2.02) <0.001 1.44 (1.19, 1.74) <0.01
Ethnicity
Kasenna 1 1
Nankana 0.88 (0.77, 1.01) 0.198 0.86 (0.73, 1.01) 0.139
Builsa 1.20 (0.82, 1.76) 1.19 (0.79, 1.82)
Others 0.94 (0.63, 1.39) 0.21 (0.41, 1.30)
Religion
Traditional 1 1
Catholic 1.05 (0.89, 1.24) 0.843 0.99 (0.82, 1.22) 0.711
Other Christian 0.97 (0.82, 1.14) 0.93 (0.77, 1.13)
Islam 1.06 (0.83, 1.36) 1.11 (0.81, 1.53)
Others 0.63 (0.15, 2.71) 1.05 (0.24, 4.69)
https://doi.org/10.1371/journal.pone.0206207.t002
LBW neonate was 35% (OR = 0.65, 95% CI 0.49, 0.85, p-value = 0.002) lower among women
who were 35 years and above when compared to the under 20 years group. Socio-economic
condition as measured in this study was an important risk factor for LBW as identified by
other studies in similar settings [17,18]. Our results also suggest that more non-married
women are prone to having babies with LBW than married women.
Discussions
In this study, we have examined how socio-demographic, foetal and maternal factors predict
low birth weights of new-born babies in the Kassena-Nankana districts in northern Ghana.
We found that the mean birth weight of babies in the study area was 2.85 ±1.0 kg with males
significantly heavier than females by 0.1 kg. Our sample size was powered enough to detect
small but significant difference in the birth weight of male and female babies. Some studies
PLOS ONE | https://doi.org/10.1371/journal.pone.0206207 November 14, 2018 6 / 10
Socio-demographic determinants low birth weight
[17,18] also found similar results where males were found to weigh heavier than boys at birth.
Voldner et al. in their study accessing the determinants of birth weight for boys and girls at
birth, found boys to be heavier than girls. They explained that, paternal birth weight signifi-
cantly influenced the weight of boys but not for girls and further suggested that there is a
genetic regulation along the male line [19]. We also found out that the prevalence of low birth
weight in the study area (13.8%) was higher than the national prevalence of 10.7% [20].
Though the national prevalence was reported as 10.7%, the administrative regional prevalence
varied considerably. Both our study and that of Manyeh et al, 2016 had prevalence comparable
to the estimates of the administrative regions where the study sites are located. That is, 13.8%
in our study site vs 14.5% in the Upper East region and 7.9% in Dodowa HDSS vs 9.9% in the
Greater Accra region. However, Abubakari (2015) in their study in Northern region, a similar
setting as ours, found prevalence of LBW nearly twice (26% vs 11.9%) when compared to the
regional estimate. The large difference observed in Abubakari [17] and the regional estimate
compared to our study and Manyeh et al, 2016 could be attributed to methodological differ-
ences. Both our study and Manyeh et al, 2016 utilized data from health and demographic sur-
veillance sites where data is collected at the population level whilst Abubakari (2015) relied-on
data from some selected health facilities using systematic sampling approach. The northern
part of Ghana is classified as the poorest regions [21], while Accra located in the southern part
classified as the richest. This might explain why the proportions reported by Manyeh (2016)
for the southern part of Ghana are considerable lower than what we found in our study in
northern Ghana since socio-economic status influence birth weight.
Determinants of LBW in this population were maternal age, lower socio-economic status
and single parenthood. Among these, maternal age constituted an important risk factor with
women less than 20 years of age recording the highest proportion (19.8%) of LBW born in this
study. The strong association between maternal age and birth weight is consistent with a study
conducted by MacLeod and Kielyin 1988 where they found a significant progression of birth
weight with advancing age of mother [22]. Similar results were obtained by Dičkute et al, 2004
who in their study in Lithuania identified a U-shaped relationship between maternal age and
LBW risk consistent with the under 20 year group of this study but differed in the 35 year group
of this study where higher proportions of LBW were recorded for under 20 years and 20–34
years but lowest for the above 35 years group [23,24]. Some studies have argued that the negative
effect of a higher maternal age can be weakened by maternal education [23,25] which is likely to
be the case with the low proportion of LBW recorded for the 35 year age group for this study.
Both bivariable and multivariable analyses did not show any significant association between
birth weight and educational status of the mother, this contrast studies conducted in two dif-
ferent political and social systems, West and East Germany. They found that mothers among
the lowest category of education in both West and East Germany had an unadjusted relative
risk of 2.5 for delivering an small for gestational age child compared to those of the highest
education category [26]. Both the chi square test of association and the multivariable logistic
regression model study found a significant association between marital status and birth weight
of a neonate. Being married was found to be protective against low birth weight in our findings
and those who were not married had 71% increased odds (OR = 1.71) of having a neonate
with low birth weight. This is consistent with a study by Foix L’Helias and their work where
they found risk factors including being single [27] to low birth weight among neonates. How-
ever, a study from the Dodowa HDSS found no relationship between birth weight and marital
status, this could be attributed to methodological difference in the classification of the marital
status variable (i.e. two variables vs four variables) [18]. Also difference in socio-cultural prac-
tices in terms of marriages between the northern and southern parts of Ghana could account
for these differences.
PLOS ONE | https://doi.org/10.1371/journal.pone.0206207 November 14, 2018 7 / 10
Socio-demographic determinants low birth weight
Both regression analyses and chi squared test did not show any association between reli-
gious affiliation of the mother and birth weight. This is contrary to Burdette et al. in their
examination of maternal religious attendance and low birth weight in the United States where
they identified maternal religious attendance to be protective against low birth weight and
stated that lower rates of cigarette use help to mediate 11% of the association between religious
affiliation and low birth weight. Their results further suggested that the health benefits of reli-
gious involvement have the potency to extend across generations [28]. However, this was not
the case in our study, this can be explained by the fact that smoking is not a common practice
and in most cases a taboo for women in rural Ghanaian setting. Household socio-economic
status was significantly associated with low birth weight at both the bivariable and multivari-
able analyses. LBW could be due to poverty which is a consequence of poor socio-economic
status. Our findings are consistent with some previous studies conducted elsewhere [29,30].
Similar findings were reported in a study from southern rural Ghana [18]. Spencer et al in
1999 concluded in their study that, a substantial proportion of births below 2.5 kg and below
1.5 kg are attributable to social inequality, their results demonstrated the likelihood of being
born weighing more than 3.5 kg if one comes from socially advantaged group [30]. Ethnicity
was not associated with birth weight. This is not surprising since there is not much difference
in the socio-cultural practices among the major ethnic groups in the study area.
Neonatal sex is a very important risk factor identified by this study as it was highly associ-
ated in all levels of the analysis. Female neonates had 33% increased odds (OR = 1.33 95%CI
[1.17, 1.51]) of being born with LBW than their male counterparts. Manyeh (2016), found that
being born female was protective against LBW in southern Ghana while Abubakari (2015) on
the contrast found similar results with our study where being a male neonate was rather pro-
tective against LBW in rural Northern Ghana [17,18]. The southern and northern parts of
Ghana are geographically and ethnically different and this might explain the difference
observed. Further studies will be needed in this regard to establish the variation of neonatal sex
as a risk factor of LBW in the northern and southern parts of Ghana.
LBW continues to be a significant public health problem and as multiple factors are associ-
ated with it, it requires a more holistic and multipronged approach for its reduction. The con-
cept of high-risk approach needs to be implemented which means better health care services to
all antenatal subjects with special attention to those who are found to be at high risk. Early reg-
istration of pregnancy should be promoted so as to detect the presence of any high-risk factors
at the earliest. Importance of regular ANC visits should be explained to each of the high-risk
women so that any untoward consequences can be averted. Strengthening Information-Educa-
tion-Counselling activities at health centres and/or CHPS compounds and in the community
would help to a great extent. Such education must address issues like harms of early marriage,
teenage pregnancy and proper nutrition during pregnancy.
Conclusion
Female neonates in this population were likely to present with low birth weight and maternal
factors such as younger age, lower socio-economic status and single parenthood were major
determinants of low birth weight. Effective and adequate antenatal care should therefore target
women with these risk factors.
Supporting information
S1 Data. De-identified dataset.
(XLS)
PLOS ONE | https://doi.org/10.1371/journal.pone.0206207 November 14, 2018 8 / 10
Socio-demographic determinants low birth weight
S1 Table. Generalized estimating equation for birth weight adjusting for cluster effects.
(DOCX)
Acknowledgments
We wish to express our gratitude to the Chiefs and people of the Kasena-Nankana districts for
their support for health research over years. Our gratitude also goes to data collectors for
doing due diligence in collecting the data and finally to the INDEPTH-Network and Swiss
Tropical and Public Health Institute for their technical support through the young scientist
programme and PhD fellowship.
Author Contributions
Conceptualization: Isaiah Awintuen Agorinya, Paul Welaga.
Data curation: Isaiah Awintuen Agorinya, Paul Welaga.
Formal analysis: Isaiah Awintuen Agorinya, Paul Welaga.
Investigation: Isaiah Awintuen Agorinya.
Methodology: Isaiah Awintuen Agorinya, Paul Welaga.
Project administration: Isaiah Awintuen Agorinya.
Supervision: Isaiah Awintuen Agorinya, James Akazili.
Validation: Isaiah Awintuen Agorinya.
Writing – original draft: Isaiah Awintuen Agorinya, Edmund Wedam Kanmiki.
Writing – review & editing: Isaiah Awintuen Agorinya, Edmund Wedam Kanmiki, Engelbert
Adamwaba Nonterah, Fabrizio Tediosi, James Akazili, Paul Welaga, Daniel Azongo, Abra-
ham Rexford Oduro.
References
1. United Nations. The Millennium Development Goals Report 2015 [Internet]. [cited 2018 Jul 15]. Avail-
able from: http://www.un.org/millenniumgoals/2015_MDG_Report/pdf/MDG%202015%20rev%20(July
%201).pdf
2. Wang H, Bhutta ZA, Coates MM, Coggeshall M, Dandona L, Diallo K, et al. Global, regional, national,
and selected subnational levels of stillbirths, neonatal, infant, and under-5 mortality, 1980–2015: a
systematic analysis for the Global Burden of Disease Study 2015. The Lancet. 2016 Oct 8; 388
(10053):1725–74.
3. Nayeri F, Dalili H, Nili F, Amini E, Ardehali A, Mansoori BK, et al. Risk Factors for Neonatal Mortality
Among Very Low Birth Weight Neonates. Acta Med Iran. 2013 May 30; 51(5):6.
4. Kelly YJ, Nazroo JY, McMunn A, Boreham R, Marmot M. Birthweight and behavioural problems in chil-
dren: a modifiable effect? Int J Epidemiol. 2001 Feb 1; 30(1):88–94. PMID: 11171863
5. Saigal S, Ouden L den, Wolke D, Hoult L, Paneth N, Streiner DL, et al. School-Age Outcomes in Chil-
dren Who Were Extremely Low Birth Weight From Four International Population-Based Cohorts. Pedi-
atrics. 2003 Oct 1; 112(4):943–50. PMID: 14523190
6. Walther FJ, den Ouden AL, Verloove-Vanhorick SP. Looking back in time: outcome of a national cohort
of very preterm infants born in The Netherlands in 1983. Early Hum Dev. 2000 Sep 1; 59(3):175–91.
PMID: 10996273
7. Avchen RN. Birth Weight and School-age Disabilities: A Population-based Study. Am J Epidemiol.
2001 Nov 15; 154(10):895–901. PMID: 11700243
8. Gissler M, Järvelin M-R, Louhiala P, Hemminki E. Boys have more health problems in childhood than
girls: follow-up of the 1987 Finnish birth cohort. Acta Paediatr. 88(3):310–4. PMID: 10229043
PLOS ONE | https://doi.org/10.1371/journal.pone.0206207 November 14, 2018 9 / 10
Socio-demographic determinants low birth weight
9. Cheung YB, Khoo KS, Karlberg J, Machin D. Association between psychological symptoms in adults
and growth in early life: longitudinal follow up study. BMJ. 2002 Oct 5; 325(7367):749. PMID: 12364303
10. Fagerberg B, Bondjers L, Nilsson P. Low birth weight in combination with catch-up growth predicts the
occurrence of the metabolic syndrome in men at late middle age: the Atherosclerosis and Insulin Resis-
tance study. J Intern Med. 256(3):254–9. https://doi.org/10.1111/j.1365-2796.2004.01361.x PMID:
15324369
11. McCormick MC. The Contribution of Low Birth Weight to Infant Mortality and Childhood Morbidity. N
Engl J Med. 1985 Jan 10; 312(2):82–90. https://doi.org/10.1056/NEJM198501103120204 PMID:
3880598
12. Sims M, Sims TL, Bruce MA. Race, Ethnicity, Concentrated Poverty, and Low Birth Weight Disparities.
J Natl Black Nurses Assoc JNBNA. 2008 Jul; 19(1):12–8. PMID: 18807774
13. Beydoun H, Itani M, Tamim H, Aaraj A, Khogali M, Yunis K. Impact of Maternal Age on Preterm Delivery
and Low Birthweight: A Hospital-Based Collaborative Study of Nulliparous Lebanese Women in Greater
Beirut. J Perinatol. 2004 Apr; 24(4):228–35. https://doi.org/10.1038/sj.jp.7211064 PMID: 15029216
14. Oduro AR, Wak G, Azongo D, Debpuur C, Wontuo P, Kondayire F, et al. Profile of the Navrongo Health
and Demographic Surveillance System. Int J Epidemiol. 2012 Aug 1; 41(4):968–76. https://doi.org/10.
1093/ije/dys111 PMID: 22933645
15. Binka FN, Anto FK, Awini EA, Nazzar AK, Armah GE. Incidence and risk factors of paediatric rotavirus
diarrhoea in northern Ghana. Trop Med Int Health. 8(9):840–6. PMID: 12950670
16. ICD-10: International statistical classification of diseases and related health problems. Geneva: World
Health Organization; 2011.
17. Abubakari A, Kynast-Wolf G, Jahn A. Maternal Determinants of Birth Weight in Northern Ghana. PLOS
ONE. 2015 Aug 17; 10(8):e0135641. https://doi.org/10.1371/journal.pone.0135641 PMID: 26281013
18. Manyeh AK, Kukula V, Odonkor G, Ekey RA, Adjei A, Narh-Bana S, et al. Socioeconomic and demo-
graphic determinants of birth weight in southern rural Ghana: evidence from Dodowa Health and Demo-
graphic Surveillance System. BMC Pregnancy Childbirth. 2016; 16:160. https://doi.org/10.1186/
s12884-016-0956-2 PMID: 27417308
19. Voldner N, Frøslie KF, Godang K, Bollerslev J, Henriksen T. Determinants of birth weight in boys and
girls. human_ontogenetics. 3(1):7–12.
20. Ghana Statistical Service. Ghana Multiple Indicator Cluster Survey with an Enhanced Malaria Module
and Biomarker. In. Final Report. Accra, Ghana: Ghana Statistical Service; 2011 [Internet]. [cited 2016
Oct 15]. Available from: http://www.unicef.org/ghana/Ghana_MICS_Final.pdf
21. Ghana Living Standards Survey 6 Report [Internet]. [cited 2018 Jul 16]. Available from: http://www.
statsghana.gov.gh/docfiles/glss6/GLSS6_Main%20Report.pdf
22. MacLeod S, Kiely JL. The effects of maternal age and parity on birthweight: a population-based study in
New York City. Int J Gynaecol Obstet Off Organ Int Fed Gynaecol Obstet. 1988 Feb; 26(1):11–9.
23. Basso O, Olsen J, Johansen AM, Christensen K. Change in social status and risk of low birth weight in
Denmark: population based cohort study. BMJ. 1997 Dec 6; 315(7121):1498–502. PMID: 9420490
24. Dičkutė J, Padaiga Ž, Grabauskas V, NadišauskienėRJ, Basys V, GaižauskienėA. Maternal socio-eco-
nomic factors and the risk of low birthweight in Lithuania. Med Kaunas. 2004; 40(5):475–482
25. Astolfi P, Zonta LA. Delayed maternity and risk at delivery. Paediatr Perinat Epidemiol. 16(1):67–72.
PMID: 11856456
26. Raum E, Arabin B, Schlaud M, Walter U, Schwartz FW. The impact of maternal education on intrauter-
ine growth: a comparison of former West and East Germany. Int J Epidemiol. 2001 Feb 1; 30(1):81–7.
PMID: 11171862
27. Foix-L’Hélias L, Blondel B. Changes in risk factors of preterm delivery in France between 1981 and
1995. Paediatr Perinat Epidemiol. 14(4):314–23. PMID: 11101018
28. Burdette AM, Weeks J, Hill TD, Eberstein IW. Maternal religious attendance and low birth weight. Soc
Sci Med 1982. 2012 Jun; 74(12):1961–7.
29. OA K, OF N. Parental socioeconomic status and birth weight distribution of Nigerian term newborn
babies. Niger J Paed 2013; 40 (3): 299–302.
30. Spencer N, Bambang S, Logan S, Gill L. Socioeconomic status and birth weight: comparison of an
area-based measure with the Registrar General’s social class. J Epidemiol Community Health. 1999
Aug 1; 53(8):495–8. PMID: 10562868
PLOS ONE | https://doi.org/10.1371/journal.pone.0206207 November 14, 2018 10 / 10