Micronutrients are vitamins and minerals consumed in small quantities, but are essential for body biochemical processes(1). They are critical in producing enzymes, hormones and metabolites that are essential for growth and development.
According to the WHO, micronutrient deficiencies of iodine, vitamin A and Fe are of the greatest concern worldwide(1). Thirty-two countries reported a significant proportion of their population who were affected and classified as iodine-deficient in 2012(Reference Zimmermann and Andersson2). Iodine-deficiency disorder (IDD) is characterised by a range of symptoms/diseases which includes thyroid function abnormalities, goitre, impaired brain development and brain damage(Reference Hetzel3, Reference Delange4). However, excessive iodine intake was found in twenty-nine countries, potentially leading to iodine-induced thyroid dysfunction(5). Vitamin A deficiency (VAD) is one of the important causes of blindness in children. Pre-school age children were affected in 122 countries from 1995 to 2005(6). VAD was reported to account for 0·6 million deaths globally and constituted the largest disease burden (6 % of deaths in children under 5 years of age) among micronutrient deficiencies in 2004(Reference Black, Allen and Bhutta7). The prevalence of Fe-deficiency anaemia is usually indicated by the prevalence of anaemia(8). Anaemia was known to affect about 47 % of pre-school age children globally during 1993–2005(8, Reference McLean, Cogswell and Egli9). Fe deficiency can lead to premature death and impaired or delayed mental and physical development in children(10, Reference Brabin, Premji and Verhoeff11).
Vitamins B, D and K, Zn, Se and Ca are also considered vital for child health and development. Thiamin (vitamin B1) assists in the metabolism of carbohydrates and branched-chain amino acids(12). Beriberi is associated with those populations who have a low thiamin intake together with a low-fat but a high-carbohydrate diet(Reference Djoenaidi, Notermans and Verbeek13). Vitamin B12 is responsible for brain development during infancy and DNA synthesis(14). One review demonstrated the existence of low vitamin B12 levels in specific populations of children and adolescents in five countries(Reference McLean, de Benoist and Allen15). Vitamin D plays an important role in Ca uptake for bone growth, osteoporosis and rickets prevention(16). Vitamin D deficiency was suggested to be common despite the lack of information on global prevalence(17). Zn and vitamin K are essential for protein synthesis(18, 19). More than 25 % of the populations in Latin American, Asian and African countries were categorised as at high risk of Zn deficiency(Reference Brown and Rivera20). Vitamin K deficiency bleeding (VKDB) is commonly recognised in neonates with inadequate intake/levels of vitamin K(Reference van Hasselt, de Koning and Kvist21). Se acts as an antioxidant to help protect the body against free radicals(22). Keshan disease, a congestive cardiomyopathy, is recognised to be associated with Se deficiency(Reference Nève23) and is endemic in China(Reference Li, Liu and Hou24). Se deficiency is also implicated in Kashin–Beck disease (osteochondropathy)(Reference Moreno-Reyes, Mathieu and Boelaert25) although iodine deficiency is considered common in this disease(Reference Moreno-Reyes, Suetens and Mathieu26, Reference Suetens, Moreno-Reyes and Chasseur27). Ca is required to maintain the rigidity and strength of bones to prevent osteoporosis(28). It was suggested that low Ca intake was commonly seen in developing countries(17).
In order to address micronutrient deficiency among children aged 6–59 months, UNICEF collaborated with the WHO to develop a daily multiple micronutrient formula for supplementation(29). Further, the FAO and the WHO published guidelines for implementing the fortification of widely distributed and consumed foods such as wheat, maize and flour(30). In 1995, China initiated universal salt iodisation (USI) to prevent and control IDD(Reference Wang, Zhang and Ge31). The effectiveness of food-based approaches has also been explored(Reference Gu, Yin and Xu32–Reference Yin, Gu and Xu34). The objective of the present study was to review the literature on the deficiency status of micronutrients among children and the factors influencing various deficiency statuses in China. Information on deficiencies can guide policy makers to assess the potential long-term consequential impact on children and implement interventions accordingly.
Method
The systematic review was conducted in accordance with the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) Statement, which was included to ensure the transparency and completeness of reporting systematic reviews and meta-analyses(Reference Moher, Liberati and Tetzlaff35). The search was performed in PubMed by using MeSH (Medical Subject Headings) terms and keywords. Search terms were used as follows: (Vitamin OR minerals OR nutrient) AND child AND (Hong Kong OR China) AND (deficiency OR malnutrition OR deficient). All databases were searched in June 2012. Titles, abstracts and the content of the articles were screened to determine the suitability for inclusion. Reference lists from retrieved studies were reviewed for the identification of potentially relevant studies.
Inclusion and exclusion criteria
The inclusion criteria were published studies investigating the distribution of micronutrient deficiencies and factors influencing deficiency status. Subjects included should live in China. Articles written in both English and Chinese were included. Conference proceedings, animal studies, studies reporting interventions only and studies examining macronutrients in subjects aged >18 years were excluded.
Data extraction
Extracted data included study site (regions, cities, provinces of China), sample size, study duration and age of subjects. Statistical data such as prevalence and incidence; mean serum, urinary or hair levels of different micronutrient indicators; and odds ratios were also extracted.
Results
Figure 1 summarises the literature search and study selection process. The electronic database search yielded 335 studies. A total of 300 records were screened and full texts of eighty-seven studies were retrieved for in-depth evaluation. Of these, thirty-five studies were excluded as micronutrient deficiency was not reported. Nine additional publications were identified in the reference lists of retrieved studies. As a result, sixty-one articles were included in the current review. A summary of the included studies and the serum, urinary or hair levels of different micronutrient indicators are presented in Supplemental Table 1 and Supplemental Table 2, respectively (online supplementary material).
Fig. 1 Summary of the literature search and selection of studies for inclusion
Table 1 Summary of the prevalence of VAD in different groups of Chinese children aged <18 years
VAD, vitamin A deficiency; NA, not applicable; SES, socio-economic status.
*Marginal VAD defined as serum retinol level >0·70–1·05 μmol/l or 20–30 μg/dl.
†Subclinical VAD defined as serum retinol level ≤0·70 μmol/l or ≤20 μg/dl.
Table 2 Summary of the prevalence of vitamin B deficiency and insufficiency in different groups of Chinese children aged <18 years
*Vitamin B12 deficiency defined as serum vitamin B12 level <200 pg/ml.
†Marginal vitamin B12 deficiency defined as serum vitamin B12 level of 200–300 pg/ml.
‡Thiamin deficiency defined as thiamin (μg)/creatinine (g) <120.
§Insufficiency in thiamin defined as thiamin (μg)/creatinine (g) between 85 and 120.
Vitamin A
The prevalence of subclinical VAD (serum retinol level ≤0·70 μmol/l or ≤20 μg/dl) decreased gradually from approximately 40 % to lower than 10 % from 1988 to 2009(Reference Lin, Liu and Ma36–Reference Chen, Zhang and Li46) (Table 1). However, there was little variation in the prevalence of marginal VAD (serum retinol level >0·70–1·05 μmol/l or 20–30 μg/dl), which ranged from 20 % to 45 % from 1989 to 2009(Reference Lin, Liu and Ma36–Reference Jiang, Toschke and von Kries39, Reference Jiang, Lin and Lian42, Reference Wang, Zhu and Huang44–Reference Hu, Tong and Oldenburg47).
Studies showed that the prevalence of VAD decreased with increasing age(Reference Lin, Liu and Ma36, Reference Mi, Lin and Ma38, Reference Jiang, Toschke and von Kries39, Reference Tian, Yang and Wang43). No significant difference was found in serum vitamin A level between genders in all age groups(Reference Lin, Liu and Ma36, Reference Mi, Lin and Ma38, Reference Yang, Li and Mao40, Reference Hu, Tong and Oldenburg47). Vitamin A status was associated with growth rates among children(Reference Hu, Tong and Oldenburg47). This implies that children with a high growth rate have high demand for vitamin A. However, a review suggested that children's diets generally contain small quantities of plant carotenoids(Reference Yin48). Four studies indicated that children living in rural areas had a higher risk of VAD than those in urban areas and over 50 % of vitamin A-related child deaths were in western provinces(Reference Lin, Liu and Ma36, Reference Tan, Ma and Lin37, Reference Yang, Li and Mao40, Reference Wang, Zhu and Huang44, Reference Ross, Chen and He49). A study showed that children with low socio-economic status had significantly lower mean serum retinol concentration than those with high socio-economic status in both urban (35·6 v. 37·5 μg/dl) and rural areas (26·5 v. 31·6 μg/dl)(Reference Jiang, Lin and Lian42). A short duration of breast-feeding and low consumption of vitamin A-rich foods were suggested as reasons for low serum retinol concentration among children with low socio-economic status(Reference Jiang, Lin and Lian42). However, breast-feeding was concluded to be a risk factor of subclinical VAD in Auhui since some lactating mothers might not provide adequate vitamin A in the breast milk(Reference Zhang, Tao and Yin41). Children living in livestock farming counties had generally higher serum vitamin A concentration than those in other farming counties due to a common diet including milk and meat products in livestock farming counties in Tibet(Reference Mi, Lin and Ma38).
Vitamin B
Two surveys conducted in Yunnan and Chongqing reported the prevalence of thiamin and vitamin B12 deficiency, respectively(Reference Li, Hotta and Shi50, Reference Gao and Li51) (Table 2). In 2001, the prevalence of thiamin deficiency was 10·5 % in Yunnan(Reference Li, Hotta and Shi50). Vitamin B12 deficiency and marginal vitamin B12 deficiency respectively were present in 4·5 % (8/177) and 10·7 % (19/177) of children aged 2–7 years in Chongqing(Reference Gao and Li51). The average serum vitamin B12 level was higher among urban children (615 pg/ml) than rural children (481 pg/ml)(Reference Gao and Li51). There were no significant differences in serum vitamin B12 level between genders(Reference Gao and Li51).
Vitamin D
Vitamin D deficiency rates were reported in different parts of China(Reference Zhu, Zhan and Shao52–Reference Liang, Qin and Li56) (Table 3). The prevalence of severe vitamin D deficiency was found to be 45·2 % in winter and 6·7 % in summer among adolescents in Beijing(Reference Du, Greenfield and Fraser55). In Yuci, the percentage of children with 25-hydroxyvitamin D (25(OH)D) level <12 ng/ml was higher in spring (65·3 %) than in autumn (2·8 %) in 2003(Reference Strand, Perry and Zhao53). Another study showed that the percentage of children with 25(OH)D level <25 nmol/l and <50 nmol/l ranged from 0·4 to 3·3 % and from 5·4 to 46·4 %, respectively(Reference Zhu, Zhan and Shao52).
Table 3 Summary of the prevalence of vitamin D deficiency in different groups of Chinese children aged <18 years
25(OH)D, 25-hydroxyvitamin D; NA, not applicable.
*Severe vitamin D deficiency defined as 25(OH)D level <12·5 nmol/l.
†Children with 25(OH)D level <12 ng/ml.
Age, gender and health status were the risk factors contributing to low serum 25(OH)D level. The mean serum 25(OH)D level decreased with age in 6000 children <16 years old in Zhejiang Province but no significant difference was found between genders(Reference Zhu, Zhan and Shao52). Among them, adolescents aged 12–16 years had the highest prevalence of low serum 25(OH)D. In Anhui Province, a significant seasonal difference between genders was found and the serum 25(OH)D level among children at age 15 years was maximum in summer. In general, boys had higher levels of 25(OH)D than girls. In non-summer periods, boys’ 25(OH)D level dropped with increasing age, but girls’ 25(OH)D level increased gradually(Reference Arguelles, Langman and Ariza57). A study in Nanjing reported that the mean serum 25(OH)D level among sick children (65·7 nmol/l) was less than that of healthy children (80·5 nmol/l). There was no significant difference between age groups(Reference Liang, Qin and Li56). Furthermore, a seasonal pattern of vitamin D deficiency was observed from studies conducted in Yuci and Beijing(Reference Strand, Perry and Zhao53–Reference Du, Greenfield and Fraser55). Such patterns were also observed in Anqing(Reference Arguelles, Langman and Ariza57) where vitamin D deficiency rates of nearly 50 % were found in spring and winter(Reference Strand, Perry and Zhao53–Reference Du, Greenfield and Fraser55).
Vitamin K
One study reported the incidence of VKDB to be 3·3 % among 28 156 live newborns in Shandong from 1998 to 2001(Reference Zhou, He and Wang58) (Table 4). VKDB was more prevalent in rural (5·0 %) than in urban areas (1·2 %). Further, preterm infants with diarrhoea, pneumonia and jaundice had a higher incidence of VKDB than full-term infants(Reference Zhou, He and Wang58). Infants who were breast-fed had higher risk of having VKDB(Reference Zhou, He and Wang58). Although one study proposed that vitamin K deficiency was the major cause of haemorrhagic stroke in young infants, 73·6 % of patients with VKDB did not receive vitamin K after birth in Beijing(Reference Wang, Shi and Li59).
Table 4 Summary of the prevalence of vitamin K deficiency and incidence of VKDB in different groups of Chinese children aged <18 years
VKDB, vitamin K deficiency bleeding.
Iodine
According to the WHO, median urinary iodine concentration <0·79 μmol/l or <99 μg/l indicates mild iodine deficiency(60). Table 5 illustrates the prevalence of goitre and percentages of subjects with different urinary iodine concentrations. A review indicated that Hong Kong children had optimal iodine nutrition levels in 1995–1997 in two surveys but elevated cord blood thyroid-stimulating hormone (>10 mU/l), indicating iodine insufficiency, among 22 % of neonates in 1984(Reference Kung, Lao and Chau61). In contrast, another cross-sectional study conducted in Hong Kong reported that 45·3 % of 104 children aged 5–16 years had iodine deficiency(Reference Kung, Chan and Low62). There was an increase in the proportion of children with iodine deficiency from 11·4 % in 1999 (1476/12 984) to 15·6 % in 2005 (1707/10 939)(Reference Liu, Liu and Su63). A mean urinary iodine level of 100·7–110·0 μg/l was found in children aged 7–14 years in Guizhou Province(Reference Ma, Shi and Shi64) and only 29·9 % of 448 urinary samples in Zhejiang Province reached the normal urinary iodine level(Reference Wang, Tu and Huang65). In addition, 353 children aged 5–14 years from three areas with endemic Kashin–Beck disease and one non-endemic area in Yulin had mean urinary iodine values in the range of 43–89 μg/l(Reference Zhang, Neve and Xu66). A study conducted in Tibet with 557 children aged 5–15 years reported that 66 % had urinary iodine level <20 μg/l(Reference Moreno-Reyes, Suetens and Mathieu67). However, some studies reported that about 80 % of the children reached normal urinary iodine level in Gansu(Reference Wang, Zhang and Ge31, Reference Shi, Tian and Piao68). A median urinary iodine level of 209·8–223·7 μg/l among forty children aged 8–10 years was observed(Reference Wang, Wang and Zhang69). In contrast, excessive iodine levels were observed in various parts of China including Anhui, Shandong, Shanxi and Jiangsu(Reference Shen, Liu and Sun70). In provinces with median drinking-water iodine levels of 150 to ≥500 μg/l, the median urinary iodine level varied from 357·7 to 1150·4 μg/l in children aged 8–10 years(Reference Shen, Liu and Sun70). In Shangdong Province, the mean urinary iodine levels of children aged 6–12 years were 427·0–1194·5 μg/l before the termination of iodised salt supply(Reference Guo, Qin and Liu71). One study also reported a higher median urinary iodine level in rural areas (456–519 μg/l) compared with urban areas (231–354 μg/l)(Reference Li, Fan and Chen72).
Table 5 Summary of the prevalence of goitre and the percentage of subjects with different UIC in different groups of Chinese children aged <18 years
UIC, urinary iodine concentration; SAR, Special Administrative Region; NA, not applicable.
*Iodised salt, surgery and injection of iodine supplement.
†UIC < 0·79 μmol/l = mild iodine deficiency (WHO)(60).
Goitre rates varied in different parts of China. In Hong Kong, 3·5 % of 2439 adolescents aged 12–18 years had goitre(Reference Wong, Lam and Kwok73). The goitre rate was much higher in Jiangxi Province (7·6 % in 1991)(Reference Yan, Zhao and Wan74), Tibet (46 % in 1995)(Reference Moreno-Reyes, Suetens and Mathieu26) and Gansu Province (13·5 % in 2005)(Reference Wang, Zhang and Ge31) and the rate increased with age(Reference Wang, Zhang and Ge31).
Iron
The prevalence of anaemia ranged from 8 % to 40 % in different regions(Reference Gao and Li51, Reference Leung, Lee and Sung75–Reference Ma, Jin and Li78) (Table 6). A survey reported the variation in the prevalence of anaemia to be 7·5–32·9 % among students aged 9–11 years in Shaanxi Province(Reference Luo, Kleiman-Weiner and Rozelle79). The Fe deficiency rate in 540 children aged 6–14 years was reported to be even higher (55·7 %) in Jiangxi Province(Reference Wu, Shen and Zhou80) and a rate of 24·3 % was found in Jintan among 1656 children aged 3–5 years(Reference Liu, McCauley and Zhao81).
Table 6 Summary of the prevalence of anaemia in different groups of Chinese children aged <18 years
SAR, Special Administrative Region.
*Anaemia defined as serum Hb level <110 g/l for children aged 3–6 years, serum Hb level <120 g/l for children aged 7–12 years, serum Hb level <130 g/l for males aged 13 years or more and serum Hb level <120 g/l for females aged 13 years or more.
A difference of up to 10 % in Fe deficiency rates was reported between urban and rural areas in 1992–2005(Reference Chang, He and Jia76) but another study reported similar deficiency rates in these areas(Reference Ma, Jin and Li78). A cross-sectional study reported that the incidence of Fe deficiency in 1012 children was 26·5 % in rural Beijing(Reference Lin, Wang and Shen82). In Beijing, the highest anaemia rate (48·8 %) was found among the 211 exclusively breast-fed infant boys at 4 months in 2003–2004(Reference Gong, Ji and Zheng83). Another study also reported lower blood Hb levels among infants when compared with other age groups(Reference Song, Xu and Li84). A study suggested that possible reasons to explain the high rate of anaemia among children in China(Reference Chang, He and Jia76) was that mothers lacked Fe during pregnancy. Moreover, Chinese infants and children mainly consumed vegetables, which have low Fe content, in their diets(Reference Chang, He and Jia76).
Zinc
Seven studies reported Zn deficiency rates(Reference Liu, McCauley and Zhao81, Reference Song, Xu and Li84–Reference Yang, Chen and Feng90) (Table 7). A study involving 13 929 children showed that the prevalence of Zn deficiency in Beijing was 13·7 % in 2007 and these rates decreased with age(Reference Song, Xu and Li84). A study conducted in Jintan reported the Zn deficiency rate among children aged 3–5 years to be 38·4 %(Reference Liu, McCauley and Zhao81). Other studies showed even higher Zn deficiency rates which were over 50 % in the age of 0–8 years(Reference Rao, Wang and Cheng85–Reference Yang and Guo89). However, two studies showed that Zn deficiency rates increased with age(Reference Rao, Wang and Cheng85, Reference Yang and Guo89).
Table 7 Summary of the prevalence of zinc deficiency in different groups of Chinese children aged <18 years
NA, not applicable.
*Hair Zn level normal reference value: children aged 1–6 years, 100 μg/g; children aged 7–14 years, 135 μg/g.
†Hair Zn level criterion: age 0 years, 136 μg/g; children, 119 μg/g.
‡Peripheral blood Zn level normal reference range: children aged 0–1 year, 58–100 μmol/l; children aged 1–2 years, 62–110 μmol/l; children aged 2–3 years, 66–120 μmol/l; children aged 3–5 years, 66–130 μmol/l; children aged >5 years, 76·5–140 μmol/l.
Selenium
Table 8 illustrates the prevalence and incidence of Se deficiency. A cross-sectional study showed that the blood Se level was lower than the normal reference value in 84·7 % of 532 subjects in Qinghai Province(Reference Shu, Yuan and Yan91). A study mentioned the increase in blood Se levels among patients with Keshan disease aged <18 years living in rural areas from 1995 to 2005(Reference Cai, Deng and Ouyang92). The mean serum Se level increased steadily to a non-endemic level in 2000 and attained the highest level in 2005. Two studies conducted in Lhasa Prefecture, an Se- and iodine-deficient area in Tibet, reported that 49 % of 280 subjects had Kashin–Beck disease(Reference Moreno-Reyes, Suetens and Mathieu67) and all had severe Se deficiency(Reference Suetens, Moreno-Reyes and Chasseur27). There was also high Se deficiency rate (83 %) in 120 children aged 6–14 years in Shaanxi Province(Reference Peng, Lingxia and Schrauzer93). Low hair Se concentration was found to be associated with an increased risk of Kashin–Beck disease in a study conducted in Shaanxi Province(Reference Zhang, Neve and Xu66). Another study conducted in Shaanxi also showed that the Se levels in hair samples of children living in a Kashin–Beck disease-endemic area were lower than the normal range(Reference Fang, Wu and Hu94).
Table 8 Summary of the prevalence and incidence of selenium deficiency, Kashin–Beck disease and Keshan disease in different groups of Chinese children aged <18 years
NA, not applicable.
Calcium
Table 9 demonstrates the prevalence of Ca deficiency. The prevalence of coexisting Ca and vitamin D deficiency was reported to be 9·4 % in winter in 1995–1996 among 1248 adolescent girls(Reference Du, Greenfield and Fraser55) in Beijing. In 2000, 116 children (19·6 %) had decreasing Ca level in hair with age in Jiangxi Province(Reference Rao, Wang and Cheng85). Another study also reported similar findings(Reference Song, Xu and Li84). Ca deficiency was also profound in Jintan (34·3 %) among children aged 3–5 years(Reference Liu, McCauley and Zhao81).
Table 9 Summary of the prevalence of calcium deficiency in different groups of Chinese children aged <18 years
*Hair Ca level criterion: for age 0 years, 992 μg/g; children, 801 μg/g.
Discussion
This is the first systematic review summarising the available literature regarding micronutrient deficiency status and factors influencing micronutrient status among children in China. All studied micronutrients of interest (except iodine) were deficient among some children in China.
Public health implications of vitamin A, vitamin D, iron and iodine deficiencies
The WHO has established certain criteria for assessment of the severity of micronutrient deficiency. The prevalence of subclinical VAD among children aged <6 years in 1995–2009 indicated that VAD was a moderate public health problem in many parts of China(95). However, the trend of VAD appeared to decrease throughout the study years. This may be due to administration of vitamin A in deficient areas which is based on the national programme of action for child development in China during 2001–2010(96). With respect to Fe-deficiency anaemia, China could also be classified as a country in which this is of moderate public health significance as the reported prevalence ranged from 20 % to 40 % in 2007–2011(97). For iodine, IDD in Hong Kong may be considered to be less severe than in Gansu Province, Jiangxi Province and Tibet as the total goitre rate in school-aged children was below 5 %(60, Reference Wong, Lam and Kwok73). According to a consensus statement of an Expert Panel Group in 2003, 35·8 % of pregnant women may have mild iodine deficiency(Reference But, Chan and Chan98, Reference Kung, Lao and Low99). Since iodised salt was not widely available in Hong Kong(Reference Lin100), IDD is still a concern. In contrast, some provinces had excess iodine supply after the implementation of USI including Shandong. Hence, termination of the iodised salt supply was considered. In addition to vitamin A, Fe and iodine, vitamin D deficiency was also a public health concern among adolescents in China especially in spring and winter. Further research should be conducted to monitor the latest deficiency prevalences of these micronutrients.
Status of intake and its relationship to micronutrient deficiency
Apart from deficiency rates, insufficient micronutrient intake could be the proxy to indicate the severity of certain micronutrient deficiencies which were rarely reported in the available literature. The intakes of vitamin B, Zn, Se and Ca are discussed below to complement our findings. Most of the studies reported low intakes of these micronutrients, which supported our results showing the deficiency problems of these micronutrients. One study reported that vitamin B12 intake in 6·8 % of 177 children was lower than the recommendation from the WHO and the mean vitamin B12 intake was 5·1 μg/d(Reference Gao and Li51). For Zn, 150 children (87·2 %) did not meet the estimated average requirement in Yunnan Province(Reference Willows, Barbarich and Wang77). A cross-sectional study reported that the mean Zn intake (8–10 mg/d) in thirty-one provinces in 2002 was lower than the recommended nutrient intake(Reference Qin, Melse-Boonstra and Shi101). The inadequate Zn intake rate was shown to be higher in the rural area than in the urban area(Reference Ma, Jin and Li78). High prevalence of Se deficiency was found in our review. A study reported that 36·9 % of 385 children aged 11–17 years in Jiangsu Province had insufficient Se intake in 2002 with a daily intake of <45–50 mg/d(Reference Qin, Melse-Boonstra and Shi101). However, a national nutrition survey found that 80–90 % of the recommended daily intake was achieved in 1992(Reference Ge and Chang102). In line with our findings, a low Ca intake was found among Chinese children, especially in rural areas, in many studies(Reference Willows, Barbarich and Wang77, Reference Ge and Chang102–Reference Yin, Su and Liu104). One study reported that all pre-school children (172/172) from Yunnan did not have adequate Ca intake in 2004(Reference Willows, Barbarich and Wang77). This may be due to the lack of milk and dairy products intake in the traditional Chinese diet(Reference Yin48, Reference Ge and Chang102). For vitamin K, incidence of VKDB could give some insights to assess vitamin K deficiency and our result showed that it was fairly common among newborns. Due to the limited information on the deficiency status of these micronutrients, an effective national surveillance system should be developed to assess micronutrient deficiency in China.
Other factors influencing micronutrient deficiencies
Factors that influenced vitamin A status such as poor family economic status and living in rural areas may be associated with a poor socio-economic status. Other micronutrients such as Fe and Zn also had higher deficiency rates in rural areas as compared with urban areas(Reference Chang, He and Jia76, Reference Ma, Jin and Li78). It is suggested that strong seasonal patterns may be observed with vitamin D deficiency. This may be due to the amount of sunlight exposure and lack of outdoor activities among Chinese children(Reference Strand, Perry and Zhao53, Reference Arguelles, Langman and Ariza57). Few studies in the literature reported the impact of dietary intake on vitamin D deficiency, but one proposed that the amount of sunlight exposure was the main determinant of vitamin D status compared with dietary intake(Reference Fraser54). Furthermore, another study suggested that the lack of a formal recommendation on vitamin D daily requirement for children aged >2 years led to lower intakes of vitamin D supplements among this age group(Reference Zhu, Zhan and Shao52). Infants with pre-existing disease or those from rural areas are at higher risk of developing VKDB. More attention should be focused on this vulnerable group in order to prevent vitamin K deficiency. Zn deficiency may also result from reduced Zn intake rates and the reduction in bioavailability of Zn. Consumption of foods with high phytate content (e.g. rice and wheat) in the rural population may result in low Zn absorption secondary to reduced bioavailability(Reference Ma, Li and Jin105). Low Ca intake rates were observed(Reference Ge and Chang102, Reference Xia, Zhou and Sun106) and can result in Ca deficiency.
Interventions to tackle micronutrient deficiencies
Supplementation should be provided to those vulnerable groups on a short-term basis, while food fortification is recommended in a long-term strategy. A study showed that the cost of supplementation per disability-adjusted life year was higher than that of food fortification(Reference Ma, Jin and Li78).
Previous studies have reported various interventions to control nutrient deficiencies. For vitamin A, supplementation was shown to be effective among children aged 2–7 years(Reference Lin, Song and Yao107). The daily administration of fortified biscuits is also suggested to be given to schoolchildren since it was shown to be effective in reducing VAD(Reference Zhang, Chen and Qu108) and easily administered in schools. Other countries have used sugar and margarine as vehicles for vitamin A fortification(17, Reference Krause, Delisle and Solomons109). Fortification vehicles should be carefully chosen based on their distribution and consumption patterns among the target group in China. Vitamin D supplements should be given to the adolescents, especially in winter. Milk powder can also be used to increase vitamin D and Ca intake(Reference Lau, Lynn and Chan110). However, Chinese children in general have low milk consumption, so nutrition education is needed to supplement the implementation of fortified milk. For iodine, there is evidence of an association between the 10-year implementation of USI and the reduction of iodine deficiency and goitre rate(Reference Wang, Shi and Li59). This finding is validated by other studies conducted in several countries(Reference Ashraf, Amini and Hovsepian111–Reference Marwaha, Tandon and Ganie114). However, in some places where the iodine concentration in water is high, the implementation of USI may result in excess iodine intake(Reference Guo, Qin and Liu71, Reference Sui, Li and Mao115). Further studies are required to determine the need for USI in these areas. For Fe deficiency, soya sauce fortified with NaFeEDTA was shown to be effective in Guizhou Province in reducing the anaemia rates in children(Reference Chen, Zhao and Zhang116). In Venezuela, wheat and flour have been fortified with Fe but there are limited data demonstrating the efficacy and effectiveness of this addition(Reference Hertrampf117). For Zn, infant formula, milk and cereals could be considered as fortification vehicles but there is limited evidence of effectiveness(17). For Se, supplements were introduced in Tibet and showed a reduction in incidence of Keshan disease(Reference Cai, Deng and Ouyang92). More studies are needed to evaluate the effectiveness of these fortified vehicles on the nutritional status of Chinese children.
In contrast to supplementation and food fortification, increasing dietary diversity is a more sustainable option to control malnutrition(17). However, regulatory policies and government support are needed for this type of intervention. Breast-feeding is recommended for infants to prevent micronutrient deficiencies, but supplementation of vitamins A, B and K and Fe should be given to improve the micronutrient intakes of lactating mothers before, during and after pregnancy as well as breast-fed neonates. A comprehensive review is strongly recommended to evaluate the cost-effectiveness and cost–benefit of the interventions specifically for the Chinese.
Limitations
There are several limitations with the current review. First, there is little published literature relating to some of the micronutrients, namely vitamin B, vitamin K and Ca. Second, a large range of micronutrient deficiency rates was observed in the included studies. This is likely due to a wide variation in sample sizes and different study locations, which makes the interpretation of results less applicable. Third, various blood sampling methods were used in the different studies – for example, samples collected from finger prick, an antecubital vein and the jugular vein – contributing to outcome measures that were not comparable. Fourth, subjects in some of the cross-sectional studies were recruited in hospitals or clinics resulting in higher deficiency rates that might result from selection bias. Moreover, most studies examined the deficiency rates in regions, provinces or counties, levels which covered only a small portion of the population in China. Hence generalisability is limited, especially considering the different cultures and food consumption habits in different regions of China.
Conclusion
The included studies reported that all micronutrients of interest were deficient among children in China to varying extents, except iodine. Poor socio-economic status, environmental factors and the Chinese diet inevitably contribute to micronutrient status. However, few studies relating to some micronutrients were found in the literature; therefore, it is hard to draw a conclusion of overall deficiency status for these micronutrients. Both supplementation and food fortification are effective ways to reduce the prevalence of deficiency, but even when supplements are given out free of charge, adoption would require a significant investment on attracting public engagement. Public health policies should consider implementing strategies such as increasing dietary diversity and fostering nutrition education to address these deficiencies among Chinese children.
Acknowledgements
Sources of funding: This research received a research grant from Sci-Pharm Ltd. Sci-Pharm Ltd had no role in the design, analysis or writing of this article. Conflicts of interest: The authors have no conflicts of interest to declare. Ethics: Ethical approval was not required. Authors’ contributions: I.C.K.W., E.C.W. and A.Y.S.W. were responsible for the study's conception and design. C.S.L.C. and A.Y.S.W. conducted the literature search and screening for eligible studies. A.Y.S.W. and E.C.W. drafted the article. A.Y.S.W., E.C.W., C.S.L.C., A.G.S. and I.C.K.W. contributed to the interpretation of the results and gave final approval of the version to be published.
