Adolescent Pregnancy and Linear Growth of Infants: A Birth Cohort Study in Rural Ethiopia
Adolescent Pregnancy and Linear Growth of Infants: A Birth Cohort Study in Rural Ethiopia
Adolescent Pregnancy and Linear Growth of Infants: A Birth Cohort Study in Rural Ethiopia
Abstract
Background: Evidences indicate that the risk of linear growth faltering is higher among children born from young
mothers. Although such findings have been documented in various studies, they mainly originate from cross-
sectional data and demographic and health surveys which are not designed to capture the growth trajectories of
the same group of children. This study aimed to assess the association between young maternal age and linear
growth of infants using data from a birth cohort study in Ethiopia.
Methods: A total of 1423 mother-infant pairs, from a birth cohort study in rural Ethiopia were included in this
study. They were followed for five time points, with three months interval until the infants were 12 months old.
However, the analysis was based on 1378 subjects with at least one additional follow-up measurement to the
baseline. A team of data collectors including nurses collected questionnaire based data and anthropometric
measurements from the dyads. We fitted linear mixed-effects model with random intercept and random slope to
determine associations of young maternal age and linear growth of infants over the follow-up period after
adjusting for potential confounders.
Results: Overall, 27.2% of the mothers were adolescents (15–19 years) and the mean ± SD age of the mothers was
20 ± 2 years. Infant Length for Age Z score (LAZ) at birth was negatively associated with maternal age of 15–19
years (β = − 0.24, P = 0.032). However, young maternal age had no significant association with linear growth of the
infants over the follow-up time (P = 0.105). Linear growth of infants was associated positively with improved
maternal education and iron-folate intake during pregnancy and negatively with infant illness (P < 0.05).
Conclusion: Young maternal age had a significant negative association with LAZ score of infants at birth while its
association over time was not influential on their linear growth. The fact that wide spread socio economic and
environmental inequalities exist among mothers of all ages may have contributed to the non-significant association
between young maternal age and linear growth faltering of infants. This leaves an opportunity to develop
comprehensive interventions targeting for the infants to attain optimal catch-up growth.
Keywords: Maternal age, Adolescent pregnancy, Linear growth, Infant growth
* Correspondence: [email protected]
1
Department of Epidemiology, Institute of Health, Jimma University, P.O.Box
378, Jimma, Ethiopia
3
Department of Food Technology, Safety and Health, Ghent University,
Coupure links 653, B 9000 Ghent, Belgium
Full list of author information is available at the end of the article
© The Author(s). 2019 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0
International License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and
reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to
the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver
(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
Workicho et al. Nutrition Journal (2019) 18:22 Page 2 of 9
USA) to the nearest 0.1 cm with no shoes on and with the unstructured correlation matrix was chosen for the cor-
five points touching the vertical stand of the stadiometer. relation among repeated LAZ measurements per child
A wealth index variable was constructed using principal after comparison of models using other correlation
components analysis based on data on housing conditions, matrices. We considered the use of additional covariates
ownership of durable assets and availability of basic ser- of child growth to maternal age including maternal edu-
vices [24]. Infant illness was measured by maternal report- cation, wealth index, iron-folic acid intake and infant ill-
ing of symptoms like fever, cough, diarrhea or other ness. Model building was performed through several
symptoms. The outcome variable LAZ was generated steps and the selection of important covariates in the
using the WHO standards [25] over time, as a continuous final model was decided based on results of the regres-
variable. sion outputs and consideration of the literature. We also
assessed for effect modification by checking the interac-
Data quality control tions between the different covariates on child LAZ
Important precautions have been undertaken in order to whenever found to be relevant. We performed different
ensure quality of the data at various stages of the study. regression diagnostics assessing models goodness of fit
Enumerators and their supervisors were recruited based (normality and heteroscedasticity of the residuals at dif-
on their prior experiences of engaging in such large scale ferent levels), model specification and other numerical
surveys, fluency in speaking Afan Oromo, familiarity in problems like multicollinearity, and the sensitivity of the
using electronic data collection tools and their academic findings to potential influential observation. All the ana-
backgrounds. Afterwards, adequate training was given lysis was performed using STATA version 14 (StataCorp,
on each item of the data collection tool and how to take Texas, USA) and all the tests were two-sided with a stat-
all the measurements needed in the study using practical istical significance considered at p < 0.05. P values were
applications through role playing. A three days long pre- adjusted for multiple testing of hypothesis using
testing was also conducted in order to understand any Benjamini-Hochberg method [26].
variations in administering questions and taking mea-
surements among the enumerators before commencing
Ethical issues
actual data collection. Electronic data collection method
Ethical approval was granted from the Institutional Re-
was used to minimize errors in data collection and entry
view Board of Jimma University in Ethiopia (RPGC/264/
by using android tablet computers. The collected data
2013) and Tufts University in USA (Tufts Health Sci-
was checked by the supervisors in regular basis before
ences Campus IRB reference number:11088) before
the data was sent to a centrally located server. Add-
commencement of the study. Informed consent was ob-
itionally, a data manager regularly checked quality of
tained from the participants after a detailed explanation
the data and took back mistakes and incomplete data
of the objectives of the study. Data was registered and
to the field to be corrected. A research team from
stored in a secured server and access to the data was
Jimma University also closely followed up for tech-
upon permission of the principal investigators with per-
nical and administrative supports. Refreshment train-
sonal identifiers removed. During the study women or
ings on data quality were given on a regular basis
infants who had health problems were referred to a
during the two years of study period.
nearby health facility to seek proper medical care.
Statistical analysis
We used linear mixed-effects model with random inter- Results
cept child and random slope time to fit child linear A total of 1378 infants with at least one follow up meas-
growth curve (change in LAZ) over the study follow-up urement to the baseline were included in the analysis.
period. Fixed-effects in the model included LAZ at birth, Length was measured in all five rounds for 1184 (84%) of
maternal age, time of study follow-up, a quadratic term the infants and average number of measurements per
of time and the interaction term between maternal age child was 4.6. We have used the Strengthening the Report-
and time that compares maternal age categories on the ing of Observational Studies in Epidemiology-nutritional
evolution of child LAZ (linear growth) over time. A Epidemiology (Additional file 2: STROBE-nut) [27] guide-
quadratic term of time was considered in the model to line to report findings in this manuscript. Figure 1 shows
capture the nonlinear change in the growth curve. The the follow up of study participants throughout the study
use of other possible models using polynomial and period. Half (50.5%) of the infants were males and there
spline functions of time were also considered for a better was a significant variation by the level of maternal educa-
fitting model by comparing the AIC and BIC estimates tion among pregnant women who were 15–19 and 20–24
of model performance (Additional file 1: Estimates table years of age (P < 0.001). The results of the descriptive ana-
for the linear, quadratic and cubic-spline models). An lysis are presented in Table 1.
Workicho et al. Nutrition Journal (2019) 18:22 Page 4 of 9
Discussion
The present study examined the association between
young maternal age at delivery and linear growth of in-
fants. LAZ score at birth of infants was negatively asso-
ciated with young maternal age while infant illness is
negatively associated with infant linear growth during
the 12 month follow-up period. Studies of the effects of
young maternal age on infant linear growth generally re-
ported that the findings vary by context attributing rela-
tive roles of biological, socioeconomic and child care
factors for the differences [8, 10, 28]. In our analysis, we
observed that LAZ score of infants of young mothers
was lower compared to infants of older mothers at birth.
Nutritional and biological disadvantages are higher in
young mothers, which interfere with the development of
the fetus affecting fetal growth. Reports from the same
cohort and other studies have demonstrated that young
maternal age at delivery is associated with adverse birth
outcomes, including childhood stunting, smaller birth
length, small for gestational age and other adverse neo-
natal and infant health problems [9, 22, 29].
Fig. 1 Flow diagram of follow up of study participants
Unlike the association with the LAZ score at birth,
the interaction of young maternal age with time was
Figure 2 presents the linear growth curve of infants not statistically significant. Linear growth did not vary
over the 12 months follow-up period by maternal age significantly by maternal age among the infants during
category and adjusted for important covariates. Infants the follow up period. Maternal youth, through its effect
from adolescent mothers tend to have lower LAZ score on nutritional, socio economic and behavioral factors
compared to those born to mothers older than 20 years could impair fetal development and hence impair linear
during the first 8 months of follow up. These infants do growth of infants [10, 28, 29]. Although it is not a
Workicho et al. Nutrition Journal (2019) 18:22 Page 5 of 9
complete catch-up, where trans-generational catch up age and infant growth is linked through epigenetic ef-
is still required, these developmental insults can be fects. At the same time however, socioeconomic and
averted during early childhood if proper postnatal in- environmental differences affect optimal linear growth
fant care is implemented for the infants to reach their of infants. A large birth cohort study [29] indicated the
potential [30, 31]. In some cases [22], young maternal role of socioeconomic inequalities rather than maternal
Fig. 2 Linear growth of infants over 12 months from Mothers of 15–19 years old (---------) and Mothers of 20–24 years (______). Plotted values for the
difference from baseline of the repeated measurements are estimated from linear mixed-effects model using random intercept child and random slope
time with fixed effects including time, quadratic time, maternal age group, and timeXmaternal age group interactions adjusted for important covariates.
The P-value is for group difference on monthly changes of LAZ score over time. β coefficient on monthly changes in LAZ was 0.05; P = 0.179
Workicho et al. Nutrition Journal (2019) 18:22 Page 6 of 9
Fig. 3 linear growth of infants by maternal and infant characteristics: Linear growth and maternal educational status; no formal education (____),
primary (---------) and secondary and above (— — —), linear growth and iron and folic acid intake; < 90 days (--------) and ≥ 90 days (_____) and
linear growth and infant illness; No (____) and Yes (-------). P-values are for group difference on monthly changes of LAZ score over time. β (95%
CI) on monthly changes in LAZ were 0.03 (0.003,0.55); P = 0.024 for Maternal educational status of secondary and above, 0.02 (− 0.01,0.30); P =
0.051for primary maternal educational status, 0.15 (0.02,0.30); P = 0.018for iron-folic acid intake, − 0.15 (− 0.30, − 0.002); P = 0.001 for infant illness
Table 2 Parameter estimates from linear mixed effects model predicting linear growth of infants
Fixed effects (n = 1378) Coefficient (β) SE P
Intercept 1.38 0.143 < 0.001*
Infant age (Time) 0.14 0.029 < 0.001*
Quadratic time effect (Time*Time) −0.01 0.002 < 0.001*
Maternal age (15–19 years) − 0.24 0.128 0.032
Maternal age*Time 0.05 0.034 0.179
LAZ at birth 0.38 0.029 < 0.001*
Maternal education
(ref = no formal education)
Primary 0.05 0.077 0.518
Primary*Time 0.02 0.007 0.051
≥ Secondary 0.10 0.135 0.417
≥ Secondary*Time 0.03 0.013 0.024
Wealth index (ref = low)
Middle 0.21 0.084 0.004*
Middle*Time −0.05 0.008 0.485
High 0.13 0.084 0.018
High*Time 0.01 0.008 0.245
Iron-folate supplement (≥90 days) −0.07 0.074 0.305
Iron-folate*Time 0.15 0.007 0.018
Illness in 2 weeks (Yes) 0.05 0.070 0.495
Illness in 2 weeks *Time −0.15 0.006 0.001*
Random effects
Variance of random intercept 0.004 0.006
Variance of random slope 0.913 0.063
Covariance of random intercept and slope −0.033 0.005
Variance of residuals 0.402 0.011
Restricted maximum likelihood estimation (REML) was used to estimate the parameters, SE standard error, *Statistical significance after adjustment for multiple
testing using Benjamini-Hochberg method
adolescents have been reported [37]. Child bearing infant illness was associated with linear growth falter-
starts at early ages among Ethiopian women with 12% ing. The fact that wide spread socio economic and
of adolescent girls being pregnant or gave birth to environmental inequalities exist among mothers of all
their first child already [38]. In such settings, imple- ages may have contributed to the non-significant as-
menting interventions targeting adolescent girls widely sociation between young maternal age and linear
helps to break the intergenerational cycle of undernu- growth faltering of infants. This leaves an opportunity
trition. The strength of this study is the use of longi- to develop comprehensive interventions targeting for
tudinal data with a large sample size to show the the infants to attain optimal catch-up growth.
association of young maternal age with linear growth
of infants. On the other hand, the restriction of the Additional files
follow up period only until 12 months can be men-
tioned as a limitation since we may not be able to Additional file 1: STROBE-nut check list. (DOCX 66 kb)
capture the full spectrum of growth trajectories. Additional file 2: Estimates table for the linear, quadratic and cubic-
spline models. (DOCX 16 kb)
Conclusion
We observed that young maternal age had a signifi- Abbreviations
cant negative association with LAZ score of infants at AIC: Akaike Information Criterion; BIC: Bayesian Information Criterion;
ENGINE: Empowering the New Generation In Nutrition and Economic
birth while its association over time was not influen- opportunities; LAZ: Length for Age Z-score; LMIC: Low and Middle Income
tial on the linear growth of infants. Furthermore, Countries
Workicho et al. Nutrition Journal (2019) 18:22 Page 8 of 9
Acknowledgments 7. Ulijaszek SJ. The international growth standard for children and adolescents
We would like to thank the data collection team and study participants for project: environmental influences on preadolescent and adolescent growth
their invaluable contributions. We are also thankful for VLIR-OUS program in weight and height. Food Nutr Bull. 2006;27(4_suppl5):S279–94.
(Belgium). 8. Gibbs CM, Wendt A, Peters S, Hogue CJ. The impact of early age at first
childbirth on maternal and infant health: impact of early age at childbirth
Funding on maternal and infant health. Paediatr Perinat Epidemiol. 2012;26:259–84.
This research work is made possible by the support of the American people 9. Ganchimeg T, Ota E, Morisaki N, Laopaiboon M, Lumbiganon P, Zhang J, et
through the United States Agency for International Development (USAID) al. Pregnancy and childbirth outcomes among adolescent mothers: a World
under Agreement No. AID-663-A-11–00017. The contents of this document Health Organization multicountry study. BJOG Int J Obstet Gynaecol. 2014;
are the sole responsibility of the researchers & do not necessarily reflect the 121:40–8.
views of USAID or the United States Government. 10. Finlay JE, Ozaltin E, Canning D. The association of maternal age with infant
mortality, child anthropometric failure, diarrhoea and anaemia for first
Availability of data and materials births: evidence from 55 low- and middle-income countries. BMJ Open
The datasets used and/or analyzed during the current study are available 2011;1(2):e000226–e000226.
from the corresponding author on reasonable request. 11. Fall CHD, Sachdev HS, Osmond C, Restrepo-Mendez MC, Victora C, Martorell
R, et al. Association between maternal age at childbirth and child and adult
Authors’ contributions outcomes in the offspring: a prospective study in five low-income and
AW: Conceived and designed the protocol, analysed the data, interpreted middle-income countries (COHORTS collaboration). Lancet Glob Health.
the results and wrote the manuscript. TB, PK, AA, SG, MK & CL assisted in 2015;3(7):e366–77.
data analyses, the interpretation of the results and reviewing the manuscript. 12. Markovitz BP, Cook R, Flick LH, Leet TL. Socioeconomic factors and
All authors have seen and approved the final version of the manuscript. adolescent pregnancy outcomes: distinctions between neonatal and post-
neonatal deaths? BMC Public Health [Internet]. 2005 Dec [cited 2017 Oct
Ethics approval and consent to participate 11];5(1). Available from: http://bmcpublichealth.biomedcentral.com/articles/
Ethical approval was granted from the Institutional Review Board of Jimma 10.1186/1471-2458-5-79.
University and Tufts University before commencement of the study. 13. Paranjothy S, Broughton H, Adappa R, Fone D. Teenage pregnancy: who
Informed consent was obtained from the participants after a detailed suffers? Arch Dis Child. 2009;94(3):239–45.
explanation of the objectives of the study. During the study women who 14. LeGrand TK, Mbacké CS. Teenage pregnancy and child health in the urban
had health problems were referred to a nearby health facility to seek proper Sahel. Stud Fam Plan. 1993;24(3):137–49.
medical care. 15. Wambach KA, Cole C. Breastfeeding and adolescents. J Obstet Gynecol
Neonatal Nurs JOGNN. 2000;29(3):282–94.
Consent for publication 16. Alam N. Teenage motherhood and infant mortality in Bangladesh: maternal
Not applicable. age-dependent effect of parity one. J Biosoc Sci. 2000;32(2):229–36.
17. Holmboe-Ottesen G, Mascarenhas O, Wandel M, Tomkins A, Watson F.
NUTRITIONAL TRENDS. [cited 2017 Aug 22]; Available from: http://www.nzdl.
Competing interests
org/gsdlmod?e=d-00000-00%2D%2D-off-0fnl2.2%2D%2D00-
The authors declare that they have no competing interests.
0%2D%2D%2D%2D0-10-0%2D%2D-0%2D%2D-0direct-10%2D%2D-
4%2D%2D%2D%2D%2D%2D-0-1l%2D%2D11-en-50%2D%2D-20-
Publisher’s Note about%2D%2D-00-0-1-00%2D%2D-4%2D%2D%2D%2D0-0-11-1-0utfZz-8-
Springer Nature remains neutral with regard to jurisdictional claims in 00&cl=CL3.65&d=HASHc5369a287949bd9c630e5d.1>=2
published maps and institutional affiliations. 18. Ramakrishnan U, Martorell R, Schroeder DG, Flores R. Role of
intergenerational effects on linear growth. J Nutr. 1999;129(2):544S–9S.
Author details 19. de Vienne CM, Creveuil C. Dreyfus M. Does young maternal age increase
1
Department of Epidemiology, Institute of Health, Jimma University, P.O.Box the risk of adverse obstetric, fetal and neonatal outcomes: a cohort study.
378, Jimma, Ethiopia. 2Department of Population and Family Health, Institute Eur J Obstet Gynecol Reprod Biol. 2009;147(2):151–6.
of Health, Jimma University, P.O.Box 378, Jimma, Ethiopia. 3Department of 20. Borja JB, Adair LS. Assessing the net effect of young maternal age on
Food Technology, Safety and Health, Ghent University, Coupure links 653, B birthweight. Am J Hum Biol. 2003;15(6):733–40.
9000 Ghent, Belgium. 4Friedman School of Nutrition Science and Policy, Tufts 21. Gomez R, Santolaya J. Being mothers too early. Am J Obstet Gynecol. 2005;
University, 150 Harrison Avenue, Boston, MA, USA. 192(2):340–1.
22. Yu SH, Mason J, Crum J, Cappa C, Hotchkiss DR. Differential effects of young
Received: 1 October 2018 Accepted: 22 March 2019 maternal age on child growth. Glob Health Action. 2016;9(1):31171.
23. Fitzmaurice G, Laird N, Ware J. Applied longitudinal analysis. In: 2nd ed.
New Jersey: Jhon Wiley and sons. p. 2011.
References 24. Filmer D, Pritchett LH. Estimating wealth effects without expenditure data--
1. Victora CG, de Onis M, Hallal PC, Blossner M, Shrimpton R. Worldwide or tears: an application to educational enrollments in states of India.
timing of growth faltering: revisiting implications for interventions. Demography. 2001;38(1):115–32.
PEDIATRICS. 2010 Mar 1;125(3):e473–80. 25. WHO Multicentre Growth Reference Study Group. WHO child growth
2. Fenske N, Burns J, Hothorn T, Rehfuess EA. Understanding child stunting in standards: length/height-for-age, weightfor-age, weight-for-length, weight-
India: a comprehensive analysis of socio-economic, nutritional and for-height and body mass index-for-age: methods and development.
environmental determinants using additive quantile regression. Bhutta ZA, Geneva: World Health Organizationt [Internet]; 2006.
editor PLoS ONE. 2013;8(11):e78692. 26. Benjamini Y, Hochberg Y. Controlling the false discovery rate: a practical and
3. Caulfield LE, de Onis M, Blössner M, Black RE. Undernutrition as an powerful approach to MultipleTesting. JR Statist Soc B. 1995;57:289–300.
underlying cause of child deaths associated with diarrhea, pneumonia, 27. Lachat C, Hawwash D, Ocké MC, Berg C, Forsum E, Hörnell A, et al.
malaria, and measles. Am J Clin Nutr. 2004;80(1):193–8. Strengthening the reporting of observational studies in
4. Black RE, Morris SS, Bryce J. Where and why are 10 million children dying epidemiology—nutritional epidemiology (STROBE-nut): an extension of the
every year? Lancet. 2003;361(9376):2226–34. STROBE statement. PLoS Med. 2016;13(6):e1002036.
5. Bhutta ZA, Ahmed T, Black RE, Cousens S, Dewey K, Giugliani E, et al. What 28. Kramer KL, Lancaster JB. Teen motherhood in cross-cultural perspective.
works? Interventions for maternal and child undernutrition and survival. Ann Hum Biol. 2010;37(5):613–28.
Lancet. 2008;371(9610):417–40. 29. Lawlor DA, Mortensen L, Nybo Andersen A-M. Mechanisms underlying the
6. Black RE, Allen LH, Bhutta ZA, Caulfield LE, de Onis M, Ezzati M, et al. associations of maternal age with adverse perinatal outcomes: a sibling
Maternal and child undernutrition: global and regional exposures and study of 264 695 Danish women and their firstborn offspring. Int J
health consequences. Lancet. 2008;371(9608):243–60. Epidemiol. 2011;40(5):1205–14.
Workicho et al. Nutrition Journal (2019) 18:22 Page 9 of 9
30. Van IJzendoorn MH, Juffer F. The Emanuel miller memorial lecture 2006:
adoption as intervention. Meta-analytic evidence for massive catch-up and
plasticity in physical, socio-emotional, and cognitive development: the
Emanuel miller memorial lecture 2006: adoption as intervention. J Child
Psychol Psychiatry. 2006;47(12):1228–45.
31. Martorell R, Khan L, Schroeder D. Reversibility of stunting: epidemiological
findings in children from developing countries. Eur J Clin Nutr. 1994;
48(Suppl 1):S45–57.
32. Checkley W, Epstein LD, Gilman RH, Cabrera L, Black RE. Effects of acute
diarrhea on linear growth in Peruvian children. Am J Epidemiol. 2003;157(2):
166–75.
33. Wondimagegn ZT. Magnitude and determinants of stunting among
children in Africa: a systematic review. Curr Res Nutr Food Sci J. 2014;2(2):
88–93.
34. Dewey KG, Begum K. Long-term consequences of stunting in early life.
Matern Child Nutr. 2011;7:5–18.
35. Central Statistical Agency. Ethiopia Mini Demographic and Health Survey
[Internet]. 2014. Available from: https://www.unicef.org/ethiopia/Mini_DHS_
2014__Final_Report.pdf
36. World Food Programme (WFP). The cost of hunger in Ethiopia: implications
for the growth and transformation of Ethiopia [internet]. 2013. Available
from: https://documents.wfp.org/stellent/groups/public/documents/
newsroom/wfp263405.pdf?_ga=2.253741577.1208567362.1553585069-
1503076002.1553585069.
37. Compact2025 team. ETHIOPIA: ENDING HUNGER & UNDERNUTRITION
CHALLENGES & OPPORTUNITIES"SCOPING REPORT FOR ROUNDTABLE
DISCUSSION [Internet]. 2016. Available from: https://www.ifpri.org/
publication/ethioipia-ending-hunger-and-undernutrition-challenges-and-
opportunities
38. Central Statistical Agency Addis Ababa, Ethiopia ICF International Calverton,
Maryland, USA. Ethiopia Demographic and Health Survey 2011 [Internet].
2012. Available from: https://dhsprogram.com/pubs/pdf/FR255/FR255.pdf