World Journal Of Pediatrics-articles

Jianghong Liu, Yue-Xian Ai, Alexandra Hanlon, Zumin Shi, Barbra Dickerman, Charlene Compher

Author Affiliations: University of Pennsylvania School of Nursing, 418 Curie Blvd., Philadelphia, Pennsylvania 19104, USA (Liu J, Hanlon A, Dickerman B, Compher C); Jintan People's Hospital, Jintan, Changzhou, China (Ai YX); Discipline of Medicine, University of Adelaide, 122 Frome Street, Adelaide SA 5005, Australia (Shi Z)

Corresponding Author: Jianghong Liu, PhD, University of Pennsylvania School of Nursing, 418 Curie Blvd., Philadelphia, Pennsylvania 19104, USA (Tel: (215) 573-7492; Email: jhliu@nursing.upenn.edu)

Background: Although the prevalence of malnutrition has decreased, micronutrient deficiency still exists among children. While iron level has been studied, limited information on serum levels of zinc in Chinese children is available. This study aims to describe the status of micronutrients and its association with sociodemographic factors, and to assess associated risk factors.

Methods: A cross-sectional study was performed on 1375 Chinese preschool children. Venous blood samples were collected and analyzed for zinc and iron by atomic absorption spectrophotometry. Sociodemographic information was obtained from self-administered questionnaires given to the parents.

Results: The high prevalence of low serum zinc (38.2%) and iron (24.3%) was found. Children from rural preschools had the lowest zinc levels, whereas those from urban preschools had the lowest iron levels. Children living in small houses had the highest prevalence of low serum zinc (47.1%), and children from small families (≤3 persons) had a higher prevalence of low serum zinc than those from large families (41.8% vs. 34.1%).

Conclusions: Our findings in the recent socioeconomic development of China suggest the need for continuous monitoring of nutritional factors and highlight the importance of public health implication in preschool children even in developed regions.

Key words: iron; micronutrient deficiency; socio-demographic factors; zinc

Iron and zinc play important roles in children's health. Zinc is a component of enzymes that affect growth in infancy and childhood, sexual maturation, neuromotor development, and immunity. Mental function is improved by zinc's promotion of normal brain development and physiology.^[1,2] Iron also has an important mental function by serving as a co-enzyme involving in the production and release of neurotransmitters,^[1,3] and by influencing cognitive function^[4] and behavioral disorders such as attention deficit hyperactivity disorder (ADHD).^[5] Iron deficiency anemia has been linked with long-term cardiovascular problems,^[6,7] obesity,^[8] and hyperglycemia.^[9] Zinc deficiency has been implicated in hyperglycemia,^[10] diabetes,^[11] and cardiovascular disease.^[12] Furthermore, we previously reported that early micronutrient deficiency (zinc and iron) was linked to later cognitive deficit^[13] and increased externalizing behavior outcomes in children.^[14]

In developing countries, iron and zinc deficiency is common.^[15,16] Over 90% of affected individuals live in developing countries and approximately one-tenth of the worldwide population suffers from iron deficiency.^[17] Zinc deficiency in children continues to be a nutritional problem in both developing and developed countries.^[18] Sociodemographic factors are related to food intake and infection risk as well as zinc and iron deficiency in developing countries.^[19]

Iron and zinc deficiency used to be common in China, but over the past two to three decades reports on iron and zinc status have been mixed due to the socioeconomic reform and rapid economic development that has taken place since 1979. The availability^[20] and affordability^[21] of foods have increased dramatically during the past three decades. As a result of this increased output and access to food, the prevalence of malnutrition has decreased, while over-nutrition has increased.^[22,23] Specific to micronutrients such as iron and zinc, studies show that iron deficiency is less prevalent than zinc deficiency among Chinese children.^[24,25] Although the prevalence of anemia has decreased in China, it still exists among children.^[22,26] While iron status has been studied, limited systematic information on serum levels of zinc in Chinese children is available.

Based on a large cohort study in an economically booming area in China, we described the status of serum iron and zinc and the associations with sociodemographic factors. The purposes of this study are to describe iron and zinc status among representative communities of Chinese children and to assess risk factors associated with low iron and zinc status.

In this cross-sectional study, the subjects were drawn from The Jintan Child Study which consists of a preschool cohort of 1656 children accounting for 24.3% of all children in this age-range in Jintan city. The cohort comprised 55.5% boys, 44.5% girls, and 99.8% Han ethnicity, which was a representative sample based on data from a recent population census.[27] Between Fall 2004 and Spring 2005, all children (aged 3-5 years old) attending the following four preschools in Jintan consented to participate in our study: Jianshe (city), Huacheng (suburb), Xuebu, and Huashan (rural). Among the original group of 1656 children, complete data were available for 1375 children. We did not find significant difference in demographic factors between the two groups with completed and non-completed data (for example, no gender difference in the two groups with χ2=0.77, P=0.381). We recently published the Jintan Cohort Profile, which contains detailed information on the subjects, recruitment, and setting.[27] Institutional Review Board (IRB) approval was obtained from the University of Pennsylvania and the ethical committee for research at Jintan Hospital in China. Furthermore, we held several focus group meetings, involving the Jintan research advisory board (pediatricians, school teachers, principals, representative parents of subjects, and researchers), in order to prepare the instruments of measurement and to formulate culturally relevant explanations of the data obtained.

Blood specimens were collected in Autum 2004 and Spring 2005 by trained pediatric nurses using a strict research protocol to avoid lead contamination. Approximately 0.5 mL of venous blood was collected in a lead-free EDTA tube for zinc and iron analysis. Samples were frozen and shipped to the Child Development Center, Nanjing Medical University, Nanjing, China for analysis. Specimens remained frozen at -20ºC until analysis. Whole blood concentrations of iron and zinc were determined by atomic absorption spectrophotometry (BH model 5.100 manufactured by Beijing Bohu Innovative Electronic Technology Corporation), with duplicate readings taken with an integration time of two seconds. The reliability and validity of the analysis and the detailed analytic procedure have been described previously.^[28]

Sociodemographic information obtained from the questionnaire included gender, age, parental education, parental occupation, living conditions, house size, family size, and home location (city, suburb, rural). Parents filled out the sociodemographic questionnaire during their meeting at the pre-schools at the end of the school year when blood samples were collected, including the parents' occupation that they had held for the longest period in their lives. While this questionnaire was self-administered, research assistants were on-site to assist parents filling out the forms.

We did not ask for data on household income because it is often not the best indicator of socioeconomic status (SES) since the use of household income does not address the fact that family members may have unequal access to household income^[29,30] and it may not adequately represent the standard of living of retired persons.^[31] In addition, since we speculated that current income may be associated with recent health, a comparison of income and health status may therefore be subject to issues of reverse causation. In particular, China is experiencing rapid socioeconomic changes which have resulted in the instability of income and salary distribution. In a focus group held in preparation for this research, the Jintan research advisor board suggested that data on income not be collected because reported annual income may not reflect actual income, which may include additional financial bonuses and benefits. Furthermore, people in China still consider income to be private information and are reluctant to report it.^[32]

Therefore, we asked for information on house size to indirectly indicate socioeconomic status. We also gathered data on family size. Although more than half of the families in our data set contained 3 persons (56.5%) (2 parents and 1 child), the rest contained more than 3 family members, including grandparents or additional children. Although the majority of families had only one child due to the one-child policy that still exists in China, there were a few families that had more than one child.

Descriptive statistics (means, standard deviations) were provided for Zn and Fe levels according to sociodemographic factors. Additionally, frequencies and percentages were provided for low zinc and iron status defined by levels of serum zinc <76.5 μg/dL and serum iron <7.5 μg/dL, respectively, within the various sociodemographic subgroups. These cutoffs were taken from the middle of the normal range. The Chi-square test and two-sample t test or ANOVA as appropriate was used to compare zinc and iron distributions by subgroup levels defined by sociodemographic factors. Multiple logistic regression models were used to regress dichotomized outcomes (low Zn and low Fe, separately) on sociodemographic factors, including family size, gender, education of parents, current living place (city, suburb, rural), and house size. Parameters for these subsets are listed in Table 1. The logistic regression models were adjusted for school cluster. The association between Zn and Fe levels was examined using a linear regression model. Statistical significance was taken at the two-sided P<0.05 level. All of the analyses were performed using STATA software.^[33]

Table 1 provides mean values for zinc and iron among preschool children, aged 3 to 5 years, within sociodemographic subgroups. The mean serum zinc levels were 82.2 (SD 13.4) µg/dL in boys and 82.9 (SD 13.4) µg/dL in girls. Both boys and girls had the same level of serum iron (8.1 µg/dL). Overall, the prevalence of low zinc and iron was 38.2% and 24.3%, respectively. There was a significant difference in the mean serum zinc and iron levels by preschool location. Children from rural preschools had the lowest level of serum zinc, but those from the city preschools had the lowest level of serum iron. Family size was positively associated with serum zinc level, but not with iron. Children living in a small house (<80 m²) had the highest prevalence of low serum zinc (47.1%), and children from small families (≤3 persons) had a higher prevalence of low serum zinc than those from large families (41.8% vs. 34.1%, P=0.006). Furthermore, the education level of parents was classified into three categories: low education in those who had only completed junior high school or less (48.1% of mothers, 39.0% of fathers), medium education in those who had some high school or other vocational training (31.6% of mothers, 33.6% of fathers), and high education in those who had finished high school or beyond (20.3% of mothers, 27.4% of fathers).

The coexistence of low serum zinc and iron was seen in 17.0% of the children (Fig. 1). While the low serum zinc was more common than the low serum iron, over 7% of the children had a low iron but normal zinc level. Only about half of the children had both normal zinc and iron levels. No gender differences were observed in the prevalence of low zinc and iron levels. Serum zinc was positively correlated with serum iron (r=0.29) but after excluding extreme zinc values of <50 μg/dL or >150 μg/dL, r=0.33 (P<0.001, Fig. 2).

The multiple logistic regression adjusted model shows that a small family size was associated with a higher risk of low zinc and iron levels (Table 2). Compared with large families, small families had an odds ratio of low zinc level of 1.36 (95% CI 1.15-1.60). The odds of low iron level was 1.17 (95% CI 1.01-1.37). Mother's education was positively associated with the risk of low zinc level, as children of mothers with high education had a 33% increased risk of low zinc level, compared to children of mothers with less education. However, an association between mother's education and low iron level was not demonstrated. Living in a house larger than 80 m² was associated with a lower risk of low zinc level, but not low iron level.

In multivariate logistic regression adjusting for sociodemographic factors including family size, gender, mother's education, location, and house size, low zinc level was positively associated with low iron level (OR 5.61, 95% CI 3.85-8.16; data not shown). After adjusting for zinc level, none of the above sociodemographic factors was associated with the low level of iron. Furthermore, additional adjustment for children's age while adjusting for other covariates did not change the association in the model for zinc (P=0.777) or iron (P=0.587) deficiency.

In this cross-sectional study, we found a significant prevalence of low serum zinc and iron in Chinese children. While low zinc was more common than low iron, the coexistence of low levels of both zinc and iron was high. The low zinc level was associated with sociodemographic factors. Living in a small family or a small house and having a mother with high education were positively associated with low zinc levels. The low zinc level was highly associated with low iron level. There was no clear association between sociodemographic factors and low iron level.

The high prevalence of poor iron level among children was consistent with findings in adults and adolescents.^[34,35] However, in contrast to findings in adults, we did not find an association between sociodemographic factors and low iron level. In adults, income has been positively associated with anemia.^[34] Results from Chinese national surveys show that this region together with Shanghai has the highest prevalence of iron-deficiency anemia in adults despite the highest economic resources.^[36] Anemia has been associated with a high prevalence of Helicobacter pylori infection (60%), suggesting that infection and inflammation may partially contribute to anemia in adults.^[34,37,38] The current study has provided information on iron level in children. The lack of association between sociodemographic factors and low iron level suggests that there is a common environmental factor inducing anemia in this region. Generally speaking, there are several possible nutritional reasons for low iron level in the Chinese population. First, Asians in general consume less red meat, which is a well-recognized source of iron, contributing to around 20% of iron intake, on average, in developed countries.^[39] In addition, it is customary for the Chinese to prepare pots of tea from which the entire family drinks, and the tannic acid in this tea can inhibit iron absorption.^[40] Phytates can also inhibit iron absorption and are found in vegetables, an important part of the Chinese diet.^[41] Finally, the Chinese typically lack ascorbic acid in their diet that can enhance iron absorption.^[40] Our study also indicates a higher prevalence of low iron level in city children. While the exact etiology is unknown, we speculate that city children have more access to processed food and consequently consume more food additives and preservatives in soft drinks, etc than do rural children.^[19] These food additives and preservatives, such as phosphate, inhibit iron absorption.^[42]

With respect to zinc, in our data 38% of the children had low zinc levels. Ours is the first report from this region showing a low level of zinc among children of this age group. The finding is supported by another study from this region, which shows that 64.6% of children aged <11 years old had insufficient zinc intake.^[24] The association between low zinc level and small family size as well as high education level of mother could be due to these groups having developed a less traditional lifestyle that includes eating fast food, skipping breakfast, and not being breastfed.

The finding that living in a large house is related to a better zinc level could be due to the fact that a big house could reflect a better economic status and thus more intake of animal foods.^[19] A mother who has a higher education level is more likely to have a demanding job that prevents her from spending time selecting food and cooking at home for her family.^[43-45] In our Jintan focus group, the local advisory board suggests that smaller families are also more likely to eat out more as opposed to larger families who are more likely to cook at home. These two factors might contribute to the lifestyle of a family and thereby to the low zinc level of children. Studies show that the nutrient quality of fast food is low, and it is especially low in zinc.^[46] Fast food is becoming an increasingly popular part of children's diets in China. A study of children in Beijing shows that roughly 97.5% of children ate at least one type of junk food in the month prior to the study and 15.9% had eaten all types of them (deep fried food, processed food, fast food, and more).^[47] While a high SES is traditionally associated with better zinc level, children with higher SES backgrounds can afford the convenience of fast food, and studies show that children living in high-SES areas had more than twice as many outlets where takeaway or fast food could be purchased en route to school.^[48]

Additionally, there are several environmental and dietary factors that are very likely contributors to low zinc level in China. Modern farming technology and crop intensification result in reduced soil zinc, particularly paddy soil: the main type of soil in the study location. Reduced soil zinc consequently reduces the zinc content of rice―a dietary staple in China.^[49]In addition to the zinc-poor soils, low zinc level may be due in part to the use of low-zinc starchy roots and tubers as food staples, and an intake of phytate that is sufficient to negatively affect the bioavailability of dietary zinc.^[50] Furthermore, as previously discussed, the Chinese consume less red meat.^[39] The zinc in red meat is in a highly bioavailable form, making this food a very valuable source of this mineral.^[39]

Finally, our results showed that serum zinc was highly correlated with serum iron, consistent with other published reports.^[35,51] A strong positive association between low zinc levels and low iron levels as well as the much higher prevalence of low zinc levels points out the importance of considering low zinc level when dealing with iron status.

There are several limitations to the study that should be acknowledged. First, we used serum iron instead of using serum ferritin and serum transferrin receptor as measures of iron deficiency, which may be better indicators of iron stores and therefore true iron level. Nevertheless, serum iron is an adequate measure of iron level. Second, we did not adjust for inflammation, which may have given rise to a false positive result for low iron level in some subjects. The prevalence of zinc deficiency may also be overestimated because of unknown inflammation and infections in children, which depress serum zinc concentrations. In future research, indicators of infection (C reactive protein, alpha 1-acid glycoprotein) should be measured and children with elevated levels of the infection status indicators should be excluded or serum zinc concentrations should be adjusted based on the effect of infection status indicators on serum zinc in the sample. Other limitations of the study includes that there were no data on family income or fast food consumption, both of which are variables which can contribute to better interpretation of the data and add to the overall values of the study. The reasons for not collecting family income data are mentioned earlier in the text, and a variable to measure fast food consumption can be potentially added to future follow-ups or studies of the same nature and format. Lastly, we acknowledge that there were other potential confounders that were not controlled in the logistic regression, e.g., dietary intakes (food frequency questionnaires or 7-day recall), hookworm, infection, anthropometry measures (BMI), and child age. We have addressed this and collected this information in other follow-up studies from this cohort.

Nursing implications exist at several different levels. First, nurses must recognize that a high prevalence of micronutrient deficiency still exists despite recent socioeconomic reform that has been taking place in developing countries. Low zinc and low iron levels have been related to many adverse health outcomes in children, including negative physical,^[1,51] cognitive,^[13] and behavioral outcomes.^[14] Second, nurses can apply this knowledge to their practice by educating both children and families on the importance of eating well. Since small family size has been related to low serum zinc, perhaps due to an increased frequency of eating outside the home, it is particularly important for nurses to target this population. More specifically, nurses can educate children on avoiding fast food and eating a balanced diet. If children are aware of which foods are rich in iron (beef and liver) and zinc (wheat germ/bran, eggs, shellfish), they can make selections of healthier food when eating both inside and outside of the home.

In conclusion, this large cross-sectional study shows that poor iron and zinc levels are common in Chinese pre-school children despite recent socioeconomic reform in China. Low levels of zinc are significantly correlated with low levels of iron, and low zinc levels are associated with sociodemographic factors. These findings during the socioeconomic development of the region suggest the need for continued monitoring of nutritional factors and highlight the importance of this public health implication for preschool children.

Thanks are extended to the participating children and their families from Jintan City, and to the Jintan Cohort Study Group. This group consists of the Jintan Maternal Child Health Center, Jintan Hospital (Dr Liping Zhang, Yue Xian Ai and Liudi Han), Shanghai Jiaotong University (Dr Xiaoming Shen and Dr Chonghai Yan), China Southern Eastern University (Dr Guiju Sun and Honglei Peng), Nanjing Brain Hospital (Dr Tunong Chen) and The Chinese University of Hong Kong (Patrick W.L. Leung). We are very grateful to the Jintan city government and the Jintan Hospital for their support and assistance.

Funding: This study was funded by the National Institute of Environment Health Sciences (NIEHS, K01-ES015 877; R01-ES018858) US, UPenn CEET P30 ES013508, The Wacker Foundation US, Jintan City Government and Jintan Hospital, China.

Ethical approval: Institutional Review Board approval was obtained from the University of Pennsylvania and ethical committee for research at Jintan Hospital in China.

Contributors: Liu J contributed to project conception, data collection, data interpretation, and manuscript preparation and revision. Ai YX contributed to data collection and interpretation.Aanlou H contrixbuted to data analysis and interpretation. Shi Z contributed to data analysis, interpretation and manuscript preparation. Dickerman B contributed to manuscript preparation. Compher C contributed to data interpretation and manuscript preparation.

1 Singh M. Role of micronutrients for physical growth and mental development. Indian J Pediatr 2004;71:59-62.

2 Bitanihirwe BK, Cunningham MG. Zinc: the brain's dark horse. Synapse 2009;63:1029-1049.

3 Liu J, Raine A. The effect of childhood malnutrition on externalizing behavior. Curr Opin Pediatr 2006;18:565-570.

4 Grantham-McGregor S, Ani C. A review of studies on the effect of iron deficiency on cognitive development in children. J Nutr 2001;131:649S-668S.

5 Konofal E, Lecendreux M, Deron J, Marchand M, Cortese S, Zaïm M, et al. Effects of iron supplementation on attention deficit hyperactivity disorder in children. Pediatr Neurol 2008;38:20-26.

6 WHO. Iron deficiency anemia: assessment, prevention, and control: a guide for program manager. Geneva, 2001.

7 Naito Y, Tsujino T, Matsumoto M, Sakoda T, Ohyanagi M, Masuyama T. Adaptive response of the heart to long-term anemia induced by iron deficiency. Am J Physiol Heart Circ Physiol 2009;296:H585-H593.

8 McClung JP, Karl JP. Iron deficiency and obesity: the contribution of inflammation and diminished iron absorption. Nutr Rev 2009;67:100-104.

9 Shi Z, Zhou M, Yuan B, Qi L, Dai Y, Luo Y, et al. Iron intake and body iron stores, anemia and risk of hyperglycaemia among Chinese adults: the prospective Jiangsu Nutrition Study (JIN). Public Health Nutr 2010;13:1319-1327.

10 Shi Z, Yuan B, Qi L, Dai Y, Zuo H, Zhou M. Zinc intake and the risk of hyperglycemia among Chinese adults: the prospective Jiangsu Nutrition Study (JIN). J Nutr Health Aging 2010;14:332-335.

11 Singh RB, Niaz MA, Rastogi SS, Bajaj S, Gaoli Z, Shoumin Z. Current zinc intake and risk of diabetes and coronary artery disease and factors associated with insulin resistance in rural and urban populations of North India. J Am Coll Nutr 1998;17:564-570.

12 Beattie JH, Gordon MJ, Rucklidge GJ, Reid MD, Duncan GJ, Horgan GW, et al. Aorta protein networks in marginal and acute zinc deficiency. Proteomics 2008;8:2126-2135.

13 Liu J, Raine A, Venables PH, Dalais C, Mednick SA. Malnutrition at age 3 years and lower cognitive ability at age 11 years. Arch Pediatr Adolesc Med 2003;157:593-600.

14 Liu J, Raine A, Venables PH, Mednick SA. Malnutrition at age 3 years predisposes to externalizing behavior problems at ages 8, 11 and 17 years. Am J Psychiatry 2004;161:2005-2013.

15 Zimmermann MB, Hurrell RF. Nutritional iron deficiency. Lancet 2007;370:511-520.

16 Gibson R. Zinc nutrition in developing countries. Nutr Res Rev 1994;7:151-173.

17 López MA, Martos FC. Iron availability: an updated review. Int J Food Sci Nutr 2004;55:597-606.

18 Sandstead HH. Is zinc deficiency a public health problem? Nutrition 1995;11:87-92.

19 Shi Z, Lien N, Kumar BN, Dalen I, Holmboe-Ottesen G. The sociodemographic correlates of nutritional status of school adolescents in Jiangsu Province, China. J Adolesc Health 2005;37:313-322.

20 Lin JY. Rural reforms and agricultural growth in China. Am Econ Rev 1992;82:33-51.

21 Chai JCH. Consumption and living standards in China. China Quarterly 1992;131:721-749.

22 Li LM, Rao KQ, Kong LZ, Yao CH, Xiang HD, Zhai FY, et al. A description on the Chinese national nutrition and health survey in 2002. Zhonghua Liu Xing Bing Xue Za Zhi 2005;26:474-484. [in Chinese]

23 Zhai F, Wang H, Du S, He Y, Wang Z, Ge K, et al. Lifespan nutrition and changing socio-economic conditions in China. Asia Pac J Clin Nutr 2007;16 Suppl:374-382.

24 Qin Y, Melse-Boonstra A, Shi Z, Pan X, Yuan B, Dai Y, et al. Dietary intake of zinc in the population of Jiangsu Province, China. Asia Pac J Clin Nutr 2009;18:193-199.

25 Ma G, Li Y, Jin Y, Du S, Kok FJ, Yang X. Assessment of intake inadequacy and food sources of zinc of people in China. Public Health Nutr 2007;10:848-854.

26 Zhu YP, Liao QK. Prevalence of iron deficiency in children aged 7 months to 7 years in China. Zhonghua Er Ke Za Zhi 2004;42:886-891. [in Chinese]

27 Liu J, McCauley LA, Zhao Y, Zhang H, Pinto-Martin J; Jintan Cohort Study Group. Cohort profile: the China Jintan child cohort study. Int J Epidemiol 2010;39:668-674.

28 Dong S, Zhu Z, Liu W, Zhao W. Determination of Ca, Mg, Fe, Mn, Cu, and Zn in lixin pill by atomic absorption spectrophotometry. Guang Pu Xue Yu Guang Pu Fen Xi 2001;21:391-392. [in Chinese]

29 Volger C, Pahl J. Money, power and inequality within marriage. Sociol Rev 1994;42:263-288.

30 Pahl J. Household spending, personal spending, and the control of money in marriage. Sociology 1990;24:119-138.

31 Robert S, House JS. SES differentials in health by age and alternative indicators of SES. J Aging Health 1996;8:359-388.

32 Shenkar O. The People's Republic of China. Int Stud Manag Organ 1994;24:9-34.

33 StataCorp, Stata statistical software release 5.0. , StataCorp, College Station, TX (1997).

34 Shi Z, Hu X, Yuan B, Pan X, Dai Y, Holmboe-Ottesen G, et al. Strong negative association between intake of tofu and anemia among Chinese Adults in Jiangsu, China. J Am Diet Assoc 2008;108:1146-1153.

35 Shi Z, Lien N, Kumar BN, Holmboe-Ottesen G. Socio-demographic differences in food habits and preferences of school adolescents in Jiangsu Province, China. Eur J Clin Nutr 2005;59:1439-1148.

36 Chang S, Ge K, Zhai F, Jia F, Xu X. The analysis of nutritional factors on anemia in Chinese adults. Acta Nutr Sin 1998;20:132-137.

37 Shi Z, Hu X, He K, Yuan B, Garg M. Joint association of magnesium and iron intake with anemia among Chinese adults. Nutrition 2008;24:977-984.

38 Shi Z, Hu X, Yuan B, Pan X, Dai Y, Holmboe-Ottesen G. Association between dietary patterns and anemia in adults from Jiangsu Province in China. Br J Nutr 2006;96:906-912.

39 Williamson CS, Foster RK, Stanner SA, Buttriss JL. Red meat in the diet. Nutr Bull 2005;30:323-355.

40 Jin F, Frohman C, Thannhauser TW, Welch RM, Glahn RP. Effects of ascorbic acid, phytic acid and tannic acid on iron bioavailability from reconstituted ferritin measured by an in vitro digestion-Caco-2 cell model. Br J Nutr 2009;101:972-981.

41 Ma G, Jin Y, Piao J, Kok F, Guusje B, Jacobsen E. Phytate, calcium, iron, and zinc contents and their molar ratios in foods commonly consumed in China. J Agric Food Chem 2005;53:10285-10290.

42 Sandberg AS, Brune M, Carlsson NG, Hallberg L, Skoglund E, Rossander-Hulthén L. Inositol phosphates with different numbers of phosphate groups influence iron absorption in humans. Am J Clin Nutr 1999;70:240-246.

43 Ortiz B, MacDonald M, Ackerman N, Goebel K. The effect of homemakers' employment on meal preparation time, meals at home, and meals away from home. Family Consumer Sci Res J 1981;9:200-206.

44 Neumark-Sztainer D, Hannan PJ, Story M, Croll J, Perry C. Family meal patterns: associations with sociodemographic characteristics and improved dietary intake among adolescents. J Am Diet Assoc 2003;103:317-322.

45 Fertig A, Glomm G, Tchernis R. The connection between maternal employment and childhood obesity: inspecting the mechanisms. Rev Econ Household 2009;7:1569-5239.

46 El-Seweidy MM, Hashem RM, Abo-El-matty DM, Mohamed RH. Frequent inadequate supply of micronutirents in fast food induces oxidative stress and inflammation in testicular tissues of weanling rats. J Pharm Pharmacol 2008;60:1237-1242.

47 Zhu SP, Ding YJ, Lu XF, Wang HW, Yang M, Wang JX, et al. Study on factors related to top 10 junk food consumption at 8 to 16 years of age, in Haidian District of Beijing. Zhonghua Liu Xing Bing Xue Za Zhi 2008;29:757-762. [in Chinese]

48 Timperio AF, Ball K, Roberts R, Andrianopoulos N, Crawford DA. Children's takeaway and fast-food intakes: associations with the neighbourhood food environment. Public Health Nutr 2009;12:1960-1964.

49 Wang XC, Yan WD, An Z, Lu Q, Shi WM, Cao ZH, et al. Status of trace elements in paddy soil and sediment in Taihu Lake region. Chemosphere 2003;50:707-771.

50 Hambidge KM. Zinc deficiency in young children. Am J Clin Nutr 1997;65:160-161.

51 Shi Z, Hu X, Yuan B, Pan X, Meyer HE, Holmboe-Ottesen G. Association between serum ferritin, hemoglobin, iron intake, and diabetes in adults in Jiangsu, China. Diabetes Care 2006;29:1878-1883.