Hostname: page-component-cd9895bd7-dzt6s Total loading time: 0 Render date: 2024-12-13T06:01:32.618Z Has data issue: false hasContentIssue false

The role of mother’s education in the nutritional status of children in Serbia

Published online by Cambridge University Press:  18 April 2016

Zeljka Stamenkovic*
Affiliation:
Institute of Social Medicine, Medical Faculty University of Belgrade, Dr Subotica 15, 11 000 Belgrade, Serbia
Bosiljka Djikanovic
Affiliation:
Institute of Social Medicine, Medical Faculty University of Belgrade, Dr Subotica 15, 11 000 Belgrade, Serbia
Ulrich Laaser
Affiliation:
Section of International Public Health (S-IPH), Faculty of Health Sciences University of Bielefeld, Bielefeld, Germany
Vesna Bjegovic-Mikanovic
Affiliation:
Institute of Social Medicine, Medical Faculty University of Belgrade, Dr Subotica 15, 11 000 Belgrade, Serbia
*
*Corresponding author: Email zeljka.nikolic@mfub.bg.ac.rs
Rights & Permissions [Opens in a new window]

Abstract

Objective

The present study aimed to identify the role of mother’s education in the nutritional status of children aged 2–5 years in Serbia.

Design

Nationally representative population-based study. Age- and gender-specific BMI percentiles of children were analysed. In accordance with the WHO growth reference, children with BMI less than the 5th percentile were considered undernourished. Logistic regression was used to calculate the association between mother’s education and other socio-economic determinants as possible confounders.

Setting

UNICEF’s fourth Multiple Indicator Cluster Survey, conducted in both Roma and non-Roma settlements in Serbia.

Subjects

Children (n 2603) aged 2–5 years (mean age 3·05 years).

Results

Less than 5 % of children aged 2–5 years were undernourished. There were significantly more undernourished children among the Roma population, in the capital of Serbia and among those whose mothers were less educated. There were statistically significant differences according to mother’s education in all socio-economic characteristics (ethnicity, area, region of living and wealth index). Mother’s level of education proved to be the most important factor for child’s nutritional status; place of living (region) was also associated.

Conclusions

Mother’s education is the most significant predictor of children’s undernutrition. It confirms that investment in females’ education will bring benefits and progress not only for women and their children, but also for society as a whole.

Type
Research Papers
Copyright
Copyright © The Authors 2016 

Child undernutrition is a significant global health problem and one of the leading risk factors for child mortality and morbidity worldwide( Reference Lim, Vos and Flaxman 1 ). It accounts for 45 % of all deaths in children under the age of 5 years( Reference Black, Victora and Walker 2 ). Since adequate nutrition is essential during childhood to ensure healthy growth, undernutrition during early childhood may substantially increase vulnerability to infection and disease and the risk of premature death( Reference Dewey and Begum 3 ). Undernutrition in children leads to permanent effects diminishing health capital later in life. It produces serious health, social and economic consequences in the course of life, as well as across generations( Reference Dewey and Begum 3 Reference Martorell and Zongrone 7 ).

The child undernutrition prevalence rate remains a significant public health problem in many developing countries, including Serbia. Based on data from 2005 and 2010, undernutrition has increased in Serbia and become more alarming. Multiple Indicator Cluster Surveys (MICS) have been conducted periodically in Serbia (2010 and 2005). These surveys showed that the prevalence of child undernutrition (under the age of 5 years) in 2010 was higher than that in 2005 in non-Roma and Roma children (11 % v. 13 % for non-Roma children, 31 % v. 28 % for Roma children)( 8 , 9 ).

To date, different studies have shown that various indicators of socio-economic status are associated with children’s nutritional status, such as educational level of parents, parental income and family assets( Reference Marmot, Allen and Bell 10 Reference Nguyen, Eriksson and Petzold 14 ). Poverty, which is very typical for low- and middle-income countries, is one of the main risk factors for child undernutrition. Low level of household wealth is also a background characteristic contributing to the prevalence of undernourished children( Reference Keino, Plasqui and Ettyang 13 , Reference Black, Allen and Bhutta 15 Reference Hong, Banta and Betancourt 18 ). While low household income has a negative effect on children’s nutritional status, education acts as a protective factor following the pattern that children whose mothers have higher or high education are more likely to be well-nourished( Reference Mohammad, Kassab and Gamble 12 Reference Nguyen, Eriksson and Petzold 14 ). Some studies have confirmed that mother’s education is one of the main determinants of undernutrition( Reference Mohammad, Kassab and Gamble 12 Reference Nguyen, Eriksson and Petzold 14 ). Also, it has been shown that children whose mother was more educated were more likely to consume the optimal amounts of protein, Ca and vitamins necessary for their proper physical and psychological growth( Reference Tornaritis, Philippou and Hadjigeorgiou 19 ). Apart from the socio-economic impact on child undernutrition, some recent studies have confirmed a negative influence of health factors such as diarrhoea or symptoms of respiratory disease on children’s nutritional status( Reference Tette, Sifah and Nartey 20 Reference Kinyoki, Berkley and Moloney 22 ).

So far, little is known about the contribution of maternal education to the occurrence of child undernutrition in Serbia, as well as the interrelatedness of other sociodemographic factors and new health symptoms. Therefore, the present study aimed to identify whether and how mother’s education is associated with the nutritional status of children aged 2–5 years in Serbia.

Method

Population and sampling

The present study was performed as a secondary data analysis of the fourth Multiple Indicator Cluster Survey (MICS4) that was conducted in Serbia in 2010, carried out by the Statistical Office of the Republic of Serbia with support from UNICEF( 9 ).

The MICS4 was conveyed in two separate population samples: (i) a national sample representative of the population of Serbia; and (ii) a national Roma sample representative of the population living in Roma settlements in Serbia. In the present study we were interested in the population of children aged 2–5 years. MICS4 was conducted at the national level for urban and rural areas, and for four regions composing Serbia: Belgrade, Vojvodina, Sumadija and Western Serbia, and Southern and Eastern Serbia. For both populations, the following sampling method was used to ensure representativeness and random selection of households. The urban and rural domains within twenty-five areas were identified as the main sampling strata, and the sample was selected in two stages. Within each stratum, a specified number of census enumeration areas was selected systematically with probability proportional to size. After a household listing was carried out within the selected enumeration areas, the listed households were divided into households with and without children under the age of 5 years, and then a separate systematic sample of households was selected for each group.

According to this recruitment method 6885 non-Roma households were randomly selected, and 6803 were found to be occupied during the survey and approached by interviewers. In 6392 non-Roma households the interview was completed (household response rate was 94 %). In the interviewed households 3398 non-Roma children under 5 years of age were identified, and 3374 of them completed the questionnaire (response rate within interviewed households was 99 %).

According to this recruitment method 1815 Roma households were randomly selected, and 1782 were found and successfully approached by interviewers. In 1711 Roma households the interview was completed (household response rate was 96 %). In the interviewed households 1618 Roma children under 5 years of age were identified, and 1604 of them completed the questionnaire (response rate within interviewed households was 99 %).

In the present study we selected children who fulfilled the following criteria: (i) they were 2–5 years old; and (ii) their height and weight were measured. The number of children who met these criteria and who therefore were included in the study was 2603 (1734 non-Roma and 869 Roma children).

Survey instrument

Information on sociodemographic characteristics of children was obtained from the household questionnaire (Household Questionnaire), while the information on new health symptoms that a child might have were obtained using the questionnaire for children under the age of 5 years (Questionnaire for Children Under Five). Both questionnaires were administered to mothers (or primary caregivers), filled in during a face-to-face interview conducted by trained interviewers and completed for each child up to the age of 5 years living in that household. Both questionnaires were based on the standard questionnaires that are used in MICS rounds( 9 ). After the questionnaires had been completed, anthropometric measurements (weight and height) were done by trained interviewers, using anthropometric equipment recommended by UNICEF. Children’s weight was measured with a Seca 874 U Scale. Height was measured with a measuring board placed on a stable surface against a wall, table, tree or staircase. Children were without footwear and should have removed all but minimal clothes (such as their underwear) to obtain an accurate weight.

Variables

We were interested in three groups of variables in the present study: (i) sociodemographic characteristics of the household and mother/caregiver; (ii) the recent presence of certain health symptoms in children; and (iii) anthropometric measures that were used to calculate BMI.

Sociodemographic characteristics employed in the current analysis include: ethnicity (non-Roma/Roma); gender (male/female); type of settlement (urban/rural); region (Belgrade, Vojvodina, Sumadija and Western Serbia, and Southern and Eastern Serbia); mother’s education (no school at all, primary school, secondary school and university degree); and wealth. Wealth of the household was measured by the Demographic and Health Survey Wealth Index (hereafter, ‘wealth index’), which in its calculation includes variables related to examinees’ assets (i.e. household facilities)( Reference Rutstein and Johnson 23 ). The statistical procedure of principal components analysis was used to assign the factor scores to each variable. The resulting asset scores were standardized in relation to a standard normal distribution with a mean of 0 and an sd of 1. These standardized scores were summed up by household and individuals were ranked according to the total score of the household. According to the wealth index, respondents were classified into five socio-economic groups or quintiles: poorest, poorer, middle, richer and richest, with the same number of individuals in each. Information related to the recent presence of certain health symptoms included symptoms of diarrhoea and cough. Mothers/caregivers of children were supposed to answer whether their child had had diarrhoea or a cough within the last 2 weeks preceding the interview.

The outcome of interest was age- and gender-specific BMI of children aged 2–5 years, whose percentiles were constructed according to the WHO Child Growth Standards, released in April 2006( 24 ). We used percentiles for assessing/indicating the relative position of the child’s BMI value among children of the same gender and age. An unfavourable nutritional status for children aged 2–5 years was determined using the cut-off for undernutrition: less than the 5th percentile( 25 ).

Statistical analysis

Data were analysed by the χ 2 test and logistic regression analyses. Chi-square analyses were used to test differences in the frequency of answers to each variable between these two samples and also to test the differences in the level of mother’s education and their sociodemographic characteristics. To assess the degree of correlation between variables, the Spearman correlation coefficient was calculated.

Univariate logistic regression analysis examined the association between children’s undernutrition, socio-economic characteristics and recent health status. Multicollinearity among independent variables was tested by the test of tolerance and variance inflation factor values. All variables whose P values in univariate logistic regression analysis were less than 0·05 were included in the multivariate logistic regression analysis. Three separate multivariate models were generated. The results of regression analyses were presented as odds ratios with 95 % confidence intervals, and considered significant if the P value was less than 0·05.

The analyses were done using the statistical software package IBM SPSS Statistics Version 20.

Results

Out of 2603 children aged 2–5 years (mean age 3·05 years), less than 5 % (116 or 4·45 %) were undernourished according to the WHO Child Growth Standard. Data on undernourished children were stratified for a broad range of socio-economic characteristic and new symptoms of diarrhoea or a cough in the last 2 weeks (Table 1). We found statistically significant differences in the number of undernourished children in the two ethnic groups (non-Roma and Roma population), regarding the region they lived in and mother’s level of education (Table 1). More than one-third of all children (37·3 %) had a cough within the 2 weeks preceding the interview. There were fewer undernourished children in the group who had a cough within the 2 weeks before the interview in comparison to those who did not have this symptom (3·5 % v. 5·0 %).

Table 1 Socio-economic characteristics and recent health status, according to nutritional status, among children aged 2–5 years from Roma and non-Roma settlements; UNICEF fourth Multiple Indicator Cluster Survey, Serbia, 2010

*According to the χ2 test.

The majority of mothers of children in our sample held a secondary education (41·8 %); less than every third women in the sample had just primary education (31·8 %) and no education at all was confirmed by 7·6 % of women (Table 1). When stratifying by the level of mother’s education, for all sociodemographic characteristics we found that there were statistically significant differences in the tested categories (Table 2). While 86·4 % of non-Roma mothers obtained a secondary or higher/high degree, only 8·8 % of Roma mothers had a secondary or higher/high level of education. On the other hand, 91·1 % Roma mothers finished only primary school or had no education at all. When looking at the regional distribution of mother’s (un)education (both samples together), we found that the largest percentage of mothers who had no education at all (13·0 %) lived in Southern and Eastern Serbia. This result is significantly higher than in other regions (Vojvodina, Belgrade, and Sumadija and Western Serbia) where the percentage of uneducated mothers was 8·4 %, 4·8 % and 2·0 %, respectively. Also, results showed that the highest percentage of mothers with higher/high level of education was found in Belgrade (24·5 %), in comparison to other regions where this percentage was significantly lower (Sumadija and Western Serbia, 21·9 %; Vojvodina, 19·3 %; Southern and Eastern Serbia, 11·6 %; Table 2).

Table 2 Socio-economic characteristics, according to level of mother’s education, among children aged 2–5 years from Roma and non-Roma settlements; UNICEF fourth Multiple Indicator Cluster Survey, Serbia, 2010

*According to the χ2 test.

Before conducting the univariate and multivariate logistic regression analyses, preliminary analyses which tested the assumptions of normality, linearity, multicollinearity and homogeneity of variance were done and demonstrated that they were not disturbed. The correlation coefficient between ethnicity and mother’s education (0·712) showed that high correlation exists, but the value of this correlation was not high enough to conduct analyses for the Roma and non-Roma samples separately. Univariate logistic regression analysis was performed to determine the associations of socio-economic characteristics and recent health status with undernutrition (Table 3). The results indicated that Roma children were more often undernourished than non-Roma children (OR=1·79; 95 % CI 1·23, 2·60). In comparison to Belgrade, children who lived in all other regions had lower chance to be undernourished (Vojvodina: OR=0·48; 95 % CI 0·30, 0·76; Sumadija and Western Serbia: OR=0·22; 95 % CI 0·11, 0·42; Southern and Eastern Serbia: OR=0·40; 95 % CI 0·24, 0·65). Children whose mothers had finished only primary school had more than two times greater risk to be undernourished in comparison to children whose mothers had finished higher or high school (OR=2·29; 95 % CI 1·23, 4·26). The presence of certain health symptoms (diarrhoea or a cough) showed no association with undernutrition (Table 3).

Table 3 Univariate logistic regression analysis: associations of socio-economic characteristics and recent health status with undernutrition of children aged 2–5 years from Roma and non-Roma settlements; UNICEF fourth Multiple Indicator Cluster Survey, Serbia, 2010

To control for the possible impact of confounding variables in the observed associations with undernutrition, three multivariate-adjusted logistic regression analyses were undertaken (Table 4). In the first multivariate logistic regression model, which included ethnicity, wealth index and region, the role of ethnicity remained almost equally (OR=1·79) and significantly (P=0·005) associated with undernutrition as in the univariate logistic regression analysis, as did the wealth index and region, although slightly attenuated. However, after adding mother’s education in the second model, the significance of the association of undernutrition with ethnicity completely disappeared (P=0·162), but secondary education (relative to higher education) became a predictor of children’s undernutrition (OR=1·92; 95 % CI 1·00, 3·67). The third multivariate logistic regression model was performed after removing ethnicity, and showed stronger relationships between the primary and secondary level of mother’s education and undernutrition (respectively: OR=2·98; 95 % CI 1·50, 5·94 and OR=2·02; 95 % CI 1·06, 3·86).

Table 4 Univariate and multivariate logistic regression analyses: associations of socio-economic characteristics with undernutrition among children aged 2–5 years from Roma and non-Roma settlements; UNICEF fourth Multiple Indicator Cluster Survey, Serbia, 2010

In two out of three models, the wealth index did not change its relationship significantly. This result confirmed that only children belonging to the second poorest socio-economic group were less likely to be undernourished in comparison to those who belonged to richest socio-economic category (model 1: OR=0·46; CI 95 % 0·22, 0·94; model 2: OR=0·39; CI 95 % 0·19, 0·82). Belgrade was the region with a significantly higher proportion of undernourished children, and this association remained significant and almost unchanged even after controlling for all other independent variables in all multivariate models.

Discussion

In the present study we investigated the associations between children’s undernutrition and a wide range of socio-economic determinants, based on a nationwide representative and random sample of children aged 2–5 years and their families, by applying the well-known MICS methodology developed by UNICEF( 9 ). Results of our study indicated that a lower level of mother’s education increased the chance for children’s undernutrition. Besides education, ethnicity (being Roma) was associated with higher odds for children’s undernutrition, but it seemed that this association was mediated by maternal education and vice versa. It was shown that Roma children had almost double the chance to be undernourished compared with non-Roma children (model 1). This association was mediated by the degree of mother’s education. The association between undernutrition and ethnicity disappeared after controlling for mother’s education (model 2), and only secondary education was a predictor of child undernutrition (OR=1·92; 95 % CI 1·00, 3·67). When removing ethnicity from the analysis (model 3), a significant change was noted in the strength of the association between the level of mother’s education and undernutrition (mother’s primary education v. high education: OR=2·98; 95 % CI 1·50, 5·94; mother’s secondary education v. high education: OR=2·02; 95 % CI 1·06, 3·86). Having in mind the strong correlation between education and ethnicity and the fact that ethnicity completely lost its significance after controlling for education and other socio-economic variables in the model, we realized that the role of education is crucial. This observation is understandable because, in general, the Roma population, especially Roma women, is less educated than the non-Roma population( Reference Vukmirovic, Djordjevic and Lakcevic 26 ). It is important to emphasize that the Roma population is the largest disadvantaged population group in Serbia( Reference Semba, de Pee and Sun 27 Reference Bobak, Dejmek and Solansky 30 ) and the second largest ethnic group, accounting for 2·05 % (147 604) of the total population in Serbia( 31 ). According to the data from a pilot census conducted among the Roma population, their educational status is very low: just 30·4 % of the Roma population completed primary school (adults older than 15 years), while the proportion of those with high or higher education is 7·41 %( Reference Vukmirovic, Djordjevic and Lakcevic 26 ). In our sample, almost 70 % of Roma women completed primary school, which might be a consequence of the younger age of women in our sample as compared with those covered by the pilot census. This observation indicates that new generations of Roma women are attending school more frequently than their ancestors.

Another possible explanation for the association and interplay between mother’s education, ethnicity and undernutrition of children is the fact that Roma women and men are initiating intimate partner relationships in their early age, which are often ending in pregnancy. According to the pilot census conducted in Roma households, more than one-third of Roma women (38 %) have their first child before the age of 18 years, and 10 % of them even before the age of 16 years( Reference Vukmirovic, Djordjevic and Lakcevic 26 ). Generally speaking, undereducated mothers (those who completed just primary school) are usually not able to understand birth control options( Reference El-Ibiary and Youmans 32 , Reference Ishikawa, Yano and Fujimori 33 ), so they get pregnant and deliver the baby very early as teenagers, before completing secondary school. Most often they do not get back to school due to the obligations related to motherhood. It is a pity that the data set we worked on did not provide information on mother’s age, so we cannot firmly support these assumptions. Still, our assumptions are based on the fact that teenage pregnancies are associated with low birth weight, which leads to undernutrition further in childhood( 34 Reference Harville, Madkour and Xie 38 ).

Surprisingly, we found that there was no strong relationship between wealth index and children’s undernutrition, except the finding that poor children (second quintile) had fewer chances to be undernourished in comparison to the richest (fifth quintile). A possible explanation might be that food and nutrition are a priority for low-income families: ‘basic needs first’, according to Maslow’s hierarchy of human needs( Reference Maslow 39 ). It could also be due to natural selection which is present from early childhood, so that children in low-income families who are less capable of living or undernourished simply do not survive( Reference Janevic, Petrovic and Bjelic 40 , Reference Sepkowitz 41 ).

Children living in all regions other than Belgrade had less chance to be undernourished, independently of other variables. Unfavourable environmental factors such as air pollution, noise and high density of population in Belgrade might be responsible for adverse effects on the nutritional status of children. Except environment, a possible explanation could be a great migration (from rural areas to Belgrade) which is mostly caused by unemployment due to the lack of qualifications and low level of education. In Belgrade, people have very low income and disproportionately high rents, so they can barely feed the family. With the need to earn more, they are forced to have more than one job or to spend many hours at work, and almost to neglect children and their developmental needs. During their parents’ working hours, most children spend time at day-care centres. The Statistical Office of the Republic of Serbia has reported that Belgrade is the region with the highest percentage of day-care centre attendance and also the highest percentage of children who spend more than 9 h/d there, in comparison to other Serbian regions( 42 ). Although day-care centre attendance could serve as a protective factor against malnutrition( Reference Nakahara, Poudel and Lopchan 43 , Reference Silva, Miranda and Puccini 44 ), a study conducted among Dutch toddlers (1–3 years old), both at childcare and at home, showed that children consumed more or less equal amounts of energy at home and childcare( Reference Gubbels, Raaijmakers and Gerards 45 ). In Belgrade, we have quite the opposite situation. The duration of children’s stay in day-care centres can be very variable. Children can stay at day-care centres for between 6 and 12 h/d. During that time they are supposed to receive a certain number of meals, according to the official nutritional recommendations for their age. While we can only speculate about children’s compliance to the food intake in day-care centres, the above-mentioned patterns of parental long working hours and less time spent at home are usually associated with a lack of time to prepare meals. Therefore, parents most often rely only on the nutrition children receive at day-care centres, while they provide not more than some snacks at home. These results have been confirmed in one study conducted in North America, where attending day-care centres did not show any favourable effects on child nutritional status( Reference Bruening, Gilbride and Passannante 46 ).

The present study has certain limitations. The cross-sectional study design used provides just a ‘snapshot’ of the prevalence rate of undernourished children in Serbia and the socio-economic characteristics associated with it. Future longitudinal studies can solve this disadvantage. Also, the data were self-reported, so there is a possibility that some answers were not credible.

On the other hand, using a well-acknowledged method (MICS4) and data based on a nationally representative sample ensure the relevance of the obtained results and their use for future planning and generalization. Despite the limitations, the present study is among the first in Serbia which examines the role of mother’s education in child undernutrition in Serbia. Therefore, it can serve as a benchmark for further studies that will analyse factors associated with children’s nutritional status.

The study results indicate that one of the main determinants of undernutrition among children in Serbia is rooted in mothers’ education. This result has been confirmed in other countries too( Reference Mohammad, Kassab and Gamble 12 Reference Nguyen, Eriksson and Petzold 14 , Reference Tornaritis, Philippou and Hadjigeorgiou 19 , 34 Reference Harville, Madkour and Xie 38 ). This finding is independent of ethnicity, area of living and the wealth index, with the exception of Belgrade, the capital of Serbia, where child undernutrition is more frequent than in the other Serbian regions, and surprisingly it is least in the second poorest layer of the wealth index. It implies that decision makers and public health authorities need to initiate actions to enhance women’s educational promotion and to address social determinants to achieve optimal health outcomes as highlighted by the Commission on the Social Determinants of Health( 47 ).

Acknowledgements

Acknowledgements: The authors would like to thank UNICEF for the access to the MICS4 database. Financial support: This work was supported by the Ministry of Science and Technological Development of Serbia (project number 175025). Conflict of interest: None. Authorship: All authors took part in this research from the very early beginning until the final approval of the manuscript. Z.S. contributed to concept and design, acquisition of data, analysis and interpretation of data, drafting the article and final approval of the manuscript. B.D. contributed to concept and design, analysis of data, revising the article draft critically for important intellectual content and final approval of the manuscript. U.L. contributed to concept and design, revising the article draft critically for important intellectual content and final approval of the manuscript. V.B.-M. contributed to concept and design, revising the article draft critically for important intellectual content and final approval of the manuscript. All authors agree to take responsibilities for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. Also, all authors accept responsibility for the paper as published. Ethics of human subject participation: After receiving information on the study and before completing the questionnaire, all participants gave an informed consent in compliance with the Helsinki Declaration.

References

1. Lim, SS, Vos, T, Flaxman, AD et al. (2012) A comparative risk assessment of burden of disease and injury attributable to 67 risk factors and risk factor clusters in 21 regions 1990–2010: a systematic analysis for the Global Burden of Disease Study 2010. Lancet 380, 22242260.CrossRefGoogle ScholarPubMed
2. Black, RE, Victora, CG, Walker, SP et al. (2013) Maternal and child undernutrition and overweight in low-income and middle-income countries. Lancet 382, 427451.CrossRefGoogle ScholarPubMed
3. Dewey, KG & Begum, K (2011) Long-term consequences of stunting in early life. Matern Child Nutr 7, Suppl. 3, 518.CrossRefGoogle ScholarPubMed
4. UNICEF, World Health Organization & The World Bank (2012) UNICEF–WHO–World Bank Joint Child Malnutrition Estimates. New York/Geneva/Washington, DC: UNICEF/WHO/World Bank; available at http://www.who.int/nutgrowthdb/jme_unicef_who_wb.pdf Google Scholar
5. UNICEF, WHO, The World Bank et al . (2011) Levels and Trends in Child Mortality: Report 2011, p. 19. New York: Oxford University Press.Google Scholar
6. Caulfield, LE, Richard, SA, Rivera, JA et al. (2006) Stunting, wasting and micronutrient deficiency disorders. In Disease Control Priorities in Developing Countries, pp. 551567 [DT Jamison, JG Breman, AR Measham et al., editors]. New York: Oxford University Press.Google Scholar
7. Martorell, R & Zongrone, A (2012) Intergenerational influences on child growth and undernutrition. Paediatr Perinat Epidemiol 26, Suppl. 1, 302314.CrossRefGoogle ScholarPubMed
8. Statistical Office of the Republic of Serbia & Strategic Marketing Research Agency (2006) Republic of Serbia Multiple Indicator Cluster Survey 2005, Final Report. Belgrade: Statistical Office of the Republic of Serbia and Strategic Marketing Research Agency.Google Scholar
9. Republic Institute for Statistics (2012) Multiple Indicator Cluster Survey in the Republic of Serbia, 2010. Final Report. Belgrade: Republic Institute of Statistics.Google Scholar
10. Marmot, M, Allen, J, Bell, R et al.; Consortium for the European Review of Social Determinants of Health and the Health Divide (2012) WHO European review of social determinants of health and the health divide. Lancet 380, 10111029.CrossRefGoogle ScholarPubMed
11. Victora, CG, Adair, L, Fall, C et al.; Maternal and Child Undernutrition Study Group (2008) Maternal and child undernutrition: consequences for adult health and human capital. Lancet 371, 340357.CrossRefGoogle ScholarPubMed
12. Mohammad, K, Kassab, M, Gamble, J et al. (2014) Factors associated with birth weight inequalities in Jordan. Int Nurs Rev 61, 435440.CrossRefGoogle ScholarPubMed
13. Keino, S, Plasqui, G, Ettyang, G et al. (2014) Determinants of stunting and overweight among young children and adolescents in sub-Saharan Africa. Food Nutr Bull 35, 167178.CrossRefGoogle ScholarPubMed
14. Nguyen, HT, Eriksson, B, Petzold, M et al. (2013) Factors associated with physical growth of children during the first two years of life in rural and urban areas of Vietnam. BMC Pediatr 13, 149.CrossRefGoogle Scholar
15. Black, RE, Allen, LH, Bhutta, ZA et al.; Maternal and Child Undernutrition Study Group (2008) Maternal and child undernutrition: global and regional exposures and health consequences. Lancet 371, 243260.CrossRefGoogle ScholarPubMed
16. Chopra, M (2003) Risk factors for undernutrition of young children in a rural area of South Africa. Public Health Nutr 6, 645652.CrossRefGoogle Scholar
17. Hien, NN & Kam, S (2008) Nutritional status and the characteristics related to malnutrition in children under five years of age in Nghean, Vietnam. J Prev Med Public Health 41, 232240.CrossRefGoogle ScholarPubMed
18. Hong, R, Banta, JE & Betancourt, JA (2006) Relationship between household wealth inequality and chronic childhood under-nutrition in Bangladesh. Int J Equity Health 5, 15.CrossRefGoogle ScholarPubMed
19. Tornaritis, MJ, Philippou, E, Hadjigeorgiou, C et al. (2014) A study of the dietary intake of Cypriot children and adolescents aged 6–18 years and the association of mother’s educational status and children’s weight status on adherence to nutritional recommendations. BMC Public Health 14, 13.CrossRefGoogle ScholarPubMed
20. Tette, EM, Sifah, EK & Nartey, ET (2015) Factors affecting malnutrition in children and the uptake of interventions to prevent the condition. BMC Pediatr 15, 189.CrossRefGoogle ScholarPubMed
21. Fekadu, Y, Mesfin, A, Haile, D et al. (2015) Factors associated with nutritional status of infants and young children in Somali Region, Ethiopia: a cross-sectional study. BMC Public Health 15, 846.CrossRefGoogle ScholarPubMed
22. Kinyoki, DK, Berkley, JA, Moloney, GM et al. (2015) Predictors of the risk of malnutrition among children under the age of 5 years in Somalia. Public Health Nutr 18, 31253133.CrossRefGoogle ScholarPubMed
23. Rutstein, SO & Johnson, K (2004) The DHS Wealth Index. DHS Comparative Reports no. 6. Calverton, MD: ORC Macro; available at http://www.measuredhs.com/pubs/pdf/CR6/CR6.pdf Google Scholar
24. WHO Multicentre Growth Reference Study Group (2006) WHO Child Growth Standards: Length/Height-for-Age, Weight-for-Age, Weight-for-Length, Weight-for-Height and Body Mass Index-for-Age: Methods and Development. Geneva: WHO.Google Scholar
25. World Health Organization (1995) Physical Status: The Use and Interpretation of Anthropometry. Report of a WHO Expert Committee. WHO Technical Report Series no. 854. Geneva: WHO.Google Scholar
26. Vukmirovic, D, Djordjevic, LJ & Lakcevic, S (editors) (2010) Roma in Census. Pilot Census of Population, Households and Dwellings from 1 to 15 November 2009. Belgrade: Republic Institute for Statistics.Google Scholar
27. Semba, RD, de Pee, S, Sun, K et al. (2008) Effect of parental formal education on risk of child stunting in Indonesia and Bangladesh: a cross sectional study. Lancet 371, 322328.CrossRefGoogle ScholarPubMed
28. Stojanovski, K, McWeeney, G, Emiroglu, N et al. (2012) Risk factors for low vaccination coverage among Roma children in disadvantaged settlements in Belgrade, Serbia. Vaccine 30, 54595463.CrossRefGoogle ScholarPubMed
29. Hajioff, S & McKee, M (2000) The health of the Roma people: a review of the published literature. J Epidemiol Community Health 54, 864869.CrossRefGoogle ScholarPubMed
30. Bobak, M, Dejmek, J, Solansky, I et al. (2005) Unfavourable birth outcomes of the Roma women in the Czech Republic and the potential explanations: a population-based study. BMC Public Health 5, 106.CrossRefGoogle ScholarPubMed
31. Republic Institute for Statistics (2012) Census 2011 in the Republic Serbia. Book 1: Ethnicity. http://media.popis2011.stat.rs/2013/publikacije/Saopstenje29novembarKnjiga1.pdf (accessed November 2014).Google Scholar
32. El-Ibiary, SY & Youmans, SL (2007) Health literacy and contraception: a readability evaluation of contraceptive instructions for condoms, spermicides and emergency contraception in the USA. Eur J Contracept Reprod Health Care 12, 5862.CrossRefGoogle ScholarPubMed
33. Ishikawa, H, Yano, E, Fujimori, S et al. (2009) Patient health literacy and patient-physician information exchange during a visit. Fam Pract 26, 517523.CrossRefGoogle ScholarPubMed
34. World Health Organization (2014) Adolescent pregnancy fact sheet. http://www.who.int/mediacentre/factsheets/fs364/en/ (accessed November 2014).Google Scholar
35. Santos, GH, Martins, G & Sousa, S (2008) Teenage pregnancy and factors associated with low birth weight. Rev Bras Ginecol Obstet 30, 224231.Google ScholarPubMed
36. Guimarães, AM, Bettiol, H, Souza, Ld et al. (2013) Is adolescent pregnancy a risk factor for low birth weight? Rev Saude Publica 47, 1119.CrossRefGoogle ScholarPubMed
37. Chen, XK, Wen, SW, Fleming, N et al. (2007) Teenage pregnancy and adverse birth outcomes: a large population based retrospective cohort study. Int J Epidemiol 36, 368373.CrossRefGoogle ScholarPubMed
38. Harville, EW, Madkour, AS & Xie, Y (2012) Predictors of birth weight and gestational age among adolescents. Am J Epidemiol 176, Suppl., S150S163.CrossRefGoogle ScholarPubMed
39. Maslow, AH (1943) A theory of human motivation. Psychol Rev 50, 370396.CrossRefGoogle Scholar
40. Janevic, T, Petrovic, O, Bjelic, I et al. (2010) Risk factors for childhood undernutrition in Roma settlements in Serbia. BMC Public Health 10, 509.CrossRefGoogle ScholarPubMed
41. Sepkowitz, KA (2006) Health of the world’s Roma population. Lancet 367, 17071708.CrossRefGoogle ScholarPubMed
42. Statistical Office of the Republic of Serbia (2013) Announcement No. 104, Year LXIII, Statistical Office of the Republic of Serbia. http://webrzs.stat.gov.rs/WebSite/repository/documents/00/00/99/66/dd20042013.pdf (accessed May 2015).Google Scholar
43. Nakahara, S, Poudel, KC, Lopchan, M et al. (2010) Differential effects of out-of-home day care in improving child nutrition and augmenting maternal income among those with and without childcare support: a prospective before-after comparison study in Pokhara, Nepal. Health Policy 97, 1625.CrossRefGoogle ScholarPubMed
44. Silva, EM, Miranda, CT, Puccini, RF et al. (2000) Day care centres as an institution for health promotion among needy children: an analytical study in São Paulo, Brazil. Public Health 114, 385388.Google ScholarPubMed
45. Gubbels, JS, Raaijmakers, LG, Gerards, SM et al. (2014) Dietary intake by Dutch 1- to 3-year-old children at childcare and at home. Nutrients 6, 304318.CrossRefGoogle ScholarPubMed
46. Bruening, KS, Gilbride, JA, Passannante, MR et al. (1999) Dietary intake and health outcomes among young children attending 2 urban day-care centers. J Am Diet Assoc 99, 15291535.CrossRefGoogle ScholarPubMed
47. World Health Organization (2008) Closing the Gap in a Generation: Health Equity through Action on the Social Determinants of Health. Geneva: WHO.Google Scholar
Figure 0

Table 1 Socio-economic characteristics and recent health status, according to nutritional status, among children aged 2–5 years from Roma and non-Roma settlements; UNICEF fourth Multiple Indicator Cluster Survey, Serbia, 2010

Figure 1

Table 2 Socio-economic characteristics, according to level of mother’s education, among children aged 2–5 years from Roma and non-Roma settlements; UNICEF fourth Multiple Indicator Cluster Survey, Serbia, 2010

Figure 2

Table 3 Univariate logistic regression analysis: associations of socio-economic characteristics and recent health status with undernutrition of children aged 2–5 years from Roma and non-Roma settlements; UNICEF fourth Multiple Indicator Cluster Survey, Serbia, 2010

Figure 3

Table 4 Univariate and multivariate logistic regression analyses: associations of socio-economic characteristics with undernutrition among children aged 2–5 years from Roma and non-Roma settlements; UNICEF fourth Multiple Indicator Cluster Survey, Serbia, 2010