Zinc is an essential mineral that performs biochemical functions important for maintaining human healthReference Abdel-Mageed and Oehme1–Reference Mertz, Morris, Smith, Udomkesmalee, Fields, Levander, Munro and Danford6. Zinc deficiency may be widespread in developing countries but the true magnitude of mild and moderate zinc deficiency is unknown, in part because of the lack of a reliable and specific index of zinc statusReference Hambidge7, 8. Use of the Recommended Dietary AllowanceReference Briefel, Bialostosky, Kennedy-Stephenson, McDowell, Ervin and Wright9 and Estimated Average Requirement10, Reference Barr, Murphy and Poos11 is suggested for estimating the prevalence of nutrient inadequacy in population groups.
Zinc deficiency may arise from low dietary intake, low bioavailability and/or interaction with other nutrients, and losses through disease processesReference Cousins, Ziegler and Filer12–Reference Walsh, Sandstead, Prasad, Newberne and Fraker14. Inhibitors of zinc absorption are believed to be the more likely causative factorReference Lönnerdal15. Phytate, present in whole grains, cereals and legumes, is a strong inhibitor of zinc absorption. Phytate is high in plant-based diets, which is a major inhibitory factor for zinc absorption. The phytate/zinc molar ratio is used to predict the inhibitory effect of phytate on the bioavailability of zinc8, Reference Turnlund, King, Keyes, Gong and Michel16, Reference Rossander, Sandberg, Sandström, Schweizer and Edwards17. The World Health Organization (WHO) suggests that the assessment of dietary zinc status should take inhibitory factors into account, and phytate data must be available8. However, the inhibitory effect of phytate on the bioavailability of zinc has not been examined in China owing to a lack of information on the phytate content of foods in the China Food Composition Table Reference Yang, Wang and Pan18. In 2005, the phytate content of 60 food samples commonly consumed in China was analysedReference Ma, Jin, Piao, Kok, Bonnema and Jacobsen19 using an anion-exchange method, which made it possible to assess the dietary zinc intake of people in China taking the inhibitory effect of phytate into account. Thus the aims of the present study were to: (1) assess the prevalence of zinc intake inadequacy in relation to the effect of phytate in Chinese diets; (2) examine if zinc intake inadequacy differs by age, sex and region in China; and (3) describe the food sources of zinc in Chinese diets.
Subjects and methods
The 2002 China National Nutrition and Health Survey (CNNHS) is a nationally representative cross-sectional survey that covered 31 provinces, autonomous regions and the municipalities directly affiliated with the Central Government (Hong Kong, Macao and Taiwan were not included). A multi-stage cluster sampling method was used for subject selection. In stage 1, all 2860 counties/districts/cities of China were divided into six categories (big cities, medium and small cities, rural 1, 2, 3 and 4) based on type and level of economic development (from high to low). Twenty-two counties/districts/cities were randomly selected from each category, thus a total of 132 counties/districts/cities were randomly selected at this stage. In stage 2, three townships/sub-districts were randomly selected from each of the selected counties/districts/cities, resulting in 396 townships/sub-districts being randomly selected at this stage. In stage 3, two villages/neighbourhood committees were randomly selected from the selected townships/sub-districts, giving a total of 792 villages/neighbourhood committees. In stage 4, 90 households were randomly selected from each selected village/neighbourhood, which resulted in a final total of 71 971 households that were randomly selected to represent the national data.
Dietary intake assessment
The dietary survey was conducted among all members of 30 households that were randomly selected from the pre-selected 90 households. All family members above 2 years old from the selected households were invited for the dietary intake assessment. A total of 23 470 from 71 971 households participated in the dietary intake assessmentReference Wang20.
The protocol of the survey was approved by the Ethical Committee of the National Institute for Nutrition and Food Safety, Chinese Center for Disease Control and Prevention. A signed consent form was obtained from each subject or his/her parent or guardian.
Information on food intake was collected using the 24-hour dietary recall method for three consecutive days (two weekdays and one weekend day) by trained interviewers. The parent or guardian was interviewed for children aged 2–16 years.
Zinc intake was calculated using the data of dietary recall in conjunction with the China Food Composition Table Reference Yang, Wang and Pan18. The composition of infant formula is available in the tables and was included in the calculation. Zinc content of foods in the China Food Composition Table was determined by atomic absorption spectrophotometryReference Yang, Wang and Pan18. Duplicate food samples were analysed. Standard reference materials were obtained from the China National Center of Standard Material for quality control. The relative standard deviation was within 10%.
Sixty food samples, including rice, wheat flour, corn and soybean products commonly consumed in China, were selected based on the frequency of foods consumed in the 2002 CNNHSReference Ma, Jin, Piao, Kok, Bonnema and Jacobsen19. The phytate content of these food samples was determined using an anion-exchange methodReference Harland and Oberleas21. Phytate intake was calculated using the phytate content in conjunction with the dietary recall data. The average daily dietary intakes of zinc and phytate were calculated using the mean value of the three days' intakes.
The contributions of individual foods to zinc were calculated by summing the amount of zinc consumed from each food by all subjects in each age group and dividing by the total intake from all foods for all subjects in the respective age group.
Phytate/zinc molar ratio was calculated as the phytate intake per day in mmol divided by the zinc intake per day in mmol. The proportions of subjects with different phytate/zinc molar ratios ( < 5, 5–15, >15) were calculated. The percentage of people with zinc intake below the WHO normative requirement8 at different phytate/zinc molar ratios was calculated. The average individual normative requirements for zinc from the diet are developed according to zinc availability, the three (high, moderate and low) bioavailability levels corresponding to 50%, 30% and 15% absorption8. The normative requirements of the first trimester for pregnant women, and of first three months for lactating women, were used to assess the inadequacy of zinc intake for these groups.
Values of dietary zinc intake are expressed as median and interquartile range. Zinc intake is expressed as mg day− 1 and zinc density as mg day− 1 (1000 kcal)− 1. Normal probability plots and Kolmogorov–Smirnov tests were used to determine whether variables followed a normal distribution. To analyse the associations of sex, age, region and their interactions with dietary zinc intake, a general linear model factorial analysis was applied with Tukey’s post hoc comparisons. The results are presented by sex, age and region for the significant interactions found between sex and age, and between age and region. Differences in zinc intake of subjects by sex and region were compared using Wilcoxon's signed rank sum test. The proportion of subjects with different phytate/zinc molar ratios was calculated, while the proportion of zinc intakes less than the WHO suggested normative requirement8 was calculated by region and phytate/zinc molar ratio. Differences in the above-mentioned proportions between urban and rural areas were compared using the χ2 test. All statistical analyses were conducted with the SAS Statistical Package (8.2e for Windows; SAS Institute Inc.). Statistical significance was set at 0.05.
Characteristics of the study population
A total of 68 962 subjects were included in the present analysis. Of these, 21 103 (males 10 027, females 11 076) were from urban areas, while 47 859 (males 23 524, females 24 335) were from rural areas. There were 310 pregnant women and 470 lactating women. The means of body mass index and dietary energy intake are presented by age group and region in Table 1.
BMI – body mass index; SD – standard deviation.
Dietary zinc intake of people in China
The median zinc intakes by age, sex and region are presented in Table 2. There were significant effects of age (F(6, 68 956) = 372.8; P < 0.001) and sex (F(1, 68 961) = 113.8; P < 0.001), and first-level interactions between region and age (F(6, 68 956) = 22.5; P < 0.001) and between age and sex (F(6, 68 956) = 15.5; P < 0.001), for dietary zinc intakes. Median zinc intake ranged from 4.9 mg day− 1 (urban girls, 2–3 years) to 11.9 mg day− 1 (rural males, 19+ years). The zinc intake of urban boys aged 2–3 years was higher than that of their rural counterparts (6.2 vs. 5.1 mg day− 1). In contrast, the zinc intake of rural adults (19+ years) was significantly higher than that of their urban counterparts (males 11.9 vs. 10.5 mg day− 1; females 10.2 vs. 8.9 mg day− 1). The younger the age group, the less the zinc intake. Rural lactating women consumed more zinc than their urban counterparts (11.0 vs. 9.8 mg day− 1) whereas urban pregnant women consumed comparable amounts of zinc to their rural counterparts (10.7 vs. 10.9 mg day− 1).
A general linear model was performed with Tukey post hoc analysis to compare the effects of age (by sex and by region). Values not sharing the same superscript letter (a–g) denote a significant difference between age groups of the same sex, P < 0.05.
* Significant difference between urban and rural residents within the same age and sex group, Wilcoxon's signed rank sum test: P < 0.05.
† Sex difference within the same age and region, Wilcoxon's signed rank sum test: P < 0.05.
There were significant effects of age (F(6, 68 956) = 4.9; P < 0.001) and region (F(1, 68 961) = 278.8; P < 0.001), first-level interactions between region and age (F(6, 68 956) = 4.2; P < 0.001) and between age and sex (F(6, 68 956) = 3.3; P < 0.001), as well as second-level interactions between sex, region and age (F(4, 68 958) = 3.9; P < 0.001) for dietary zinc density. The median zinc density of all age groups was between 4.7 and 5.7 mg day− 1 (1000 kcal)− 1 (Table 2). The zinc density of urban residents (2–3 to 19+ years) was 5.0–5.3 mg day− 1 (1000 kcal)− 1, significantly higher than that of their rural counterparts (4.7–4.8 mg day− 1 (1000 kcal)− 1). No significant difference was found between different age groups. Variation in median zinc intake was found between males and females within the same region and age group. Zinc density of the diets of urban boys (2–3 years) was higher than that of their female counterparts; in contrast, the rural boys (2–3 years) and rural adults consumed diets of lower zinc density than their female counterparts. Urban pregnant women had higher dietary zinc density than their rural counterparts (5.7 vs. 4.7 mg day− 1 (1000 kcal)− 1). No difference in zinc density was found between urban and rural lactating women.
Food sources of zinc
Cereal grains including rice, wheat flour and its products contributed 27.4–63.2% to dietary zinc intake, while 16.8–54.8% zinc came from animal foods. Vegetables and fruits provided 8.2–13.8% of dietary zinc, while legumes contributed only 1.3–3.4% of dietary zinc. Large discrepancies were found in the food sources of zinc from staple and animal foods between urban and rural residents within the same age group. Staple foods provided 27.4–45.1% zinc for urban residents, 51.6–63.2% for rural residents. Animal foods contributed 28.4–54.8% of dietary zinc for urban residents, much higher than that for their respective rural counterparts (16.8–30.6%). Vegetables, fruits and legumes provided comparable zinc for both urban and rural residents. Detailed information on the food sources of dietary zinc intake stratified by age, sex and region is shown in Table 3.
Proportions of subjects with different phytate/zinc molar ratios
Variations in the proportion of subjects with the same phytate/zinc molar ratio were found between the different age and/or sex subgroups (Table 4). Some 28.6–46.8% of urban residents and 3.9–8.0% of rural residents had phytate/zinc molar ratio < 5, while 48.7–65.7% of residents had phytate/zinc molar ratio between 5 and 15, and 4.5–18.2% of urban residents and 27.8–36.4% of rural residents had phytate/zinc molar ratio >15. The proportion of rural residents with phytate/zinc molar ratio >15 was significantly higher than that of their urban counterparts. A significant effect of age on the proportion of subjects with different phytate/zinc molar ratios was found in all subgroups. Significant sex difference was only found in rural children aged 9–13 years (Table 4).
* Significant difference between urban and rural residents, χ2 test: P < 0.05.
† Sex difference within the same age and region, χ2 test: P < 0.05.
Prevalence of zinc intake inadequacy
Table 5 presents the prevalence of zinc intake inadequacy by age, sex and region. The proportion of zinc intake inadequacy ranged between 2.8% (urban females, 19+ years) and 29.4% (rural lactating women). The distribution was much higher when phytate/zinc molar ratio >15 was considered: 10.3% for rural females, 19+ years and 100% for urban lactating women. The proportion of zinc intake inadequacy among urban residents was 2.8–16.4%, significantly lower than that of their rural counterparts (3.6–29.4%). Prevalence of zinc intake inadequacy was significantly higher among male adults (19+ years) than their female counterparts. The zinc intake of 5.5% of pregnant women and 29.4% of lactating women was inadequate.
* Significant difference between urban and rural residents, χ2 test: P < 0.05.
† Sex difference within the same age and region, χ2 test: P < 0.05.
Information on zinc intake in developing countries is limited because of the paucity of data on the zinc content of local staple foodsReference Gibson22. China developed food composition tables in the 1960s and keeps them updated using reliable and valid determination methodsReference Yang, Wang and Pan18. The composition of fortified foods is also available in the China Food Composition Table. The strength of the present study is the use of three days of 24-hour dietary recall data of a national representative sample population in conjunction with the China Food Composition Table. One limitation of this study is that the zinc from dietary supplements was not taken into account. As the use of dietary supplements in ChinaReference Ma, Cui, Li, Hu, Wang and Yang23 is low (urban 10.3%, rural 2.9%) compared with that (39.5%) in the USAReference Briefel, Bialostosky, Kennedy-Stephenson, McDowell, Ervin and Wright9, we assumed that zinc intake was underestimated to a small extent. This limitation suggests that the use of dietary supplements and their contribution to total nutrient intake in the Chinese population, especially for urban persons, should be monitored in future surveys.
The present study found that the zinc intake of people in China is lower than that of the US populationReference Briefel, Bialostosky, Kennedy-Stephenson, McDowell, Ervin and Wright9, Reference Ervin and Kennedy-Stephenson24, Reference Arsenault and Brown25, but comparable to that in the UKReference Bailey, Maisey, Southon, Wright, Finglas and Fulcher26, Reference Thane, Bates and Prentice27, GermanyReference Schulze, Linseisen, Krobe and Boening28 and New ZealandReference Jong de, Gibson, Thomson, Ferguson, McKenzie and Green29. People in China consume more zinc than their counterparts in ChileReference Olivares, Pizarro, de Pablo, Araya and Uauy30, IndiaReference Ram, Mohammad, Shanti, Sarita, Zhang and Zhu31 and KoreaReference Joung, Nam, Yoon, Lee, Shim and Paik32. However, the adequacy of zinc intake depends not only on the amount, but also its bioavailability. The inhibitory factors that may interact with zinc should be examined when assessing the dietary intake of zinc8, Reference Oberleas and Harland33. People consuming a diet that provides marginal zinc intake may not absorb an adequate amount of zinc if they are also consuming foods high in phytate. The average phytate intake of people in China (1186 mg day− 1)Reference Ma, Li, Jin, Zhai, Kok and Yang34 is higher than that in Western countriesReference Plaami and Kumpulainen35, Reference Carnovale, Lombardi-Boccia and Lugaro36 and less than that of KoreanReference Kwun and Kwon37 and Indian peopleReference Khokhar, Pushpanjali and Fenwick38. The proportion of people with inadequate zinc intake was considerably high in the phytate/zinc molar ratio >15 category, indicating that zinc in the diets of these populations would be affected by their high phytate intake. The vulnerable groups of zinc deficiency in China are children, adolescents and lactating women, especially those in rural areas. As zinc plays an important role in children's growthReference Brown, Peerson, Rivera and Allen39 and during pregnancyReference Caulfield, Zavaleta, Shankar and Merialdi40, Reference Shah and Sachdev41, zinc deficiency will impact severely on human healthReference Prasad42, with adverse effects that include growth retardation, abnormal immune function, impaired cognitive function, increased maternal morbidity, prolonged gestation and increased risk to the foetus.
Information on the inadequacy of dietary zinc intakes should be interpreted together with data derived from other assessment methods. The national prevalence of stunting among children < 5 years old can be used as an indirect indicator of a population's risk of zinc deficiencyReference Hotz, Lowe, Araya and Brown43, Reference Hotz and Brown44. The national prevalence of stunting in China is 3.3–8.6% for urban children < 5 years and 17.3–20.9% for their rural counterpartsReference Yang, Wang, He, Yu, Hu and Zhai45. It can be concluded that the level of risk for zinc deficiency is low for the urban population and moderate for the rural population in China. Further population assessment of zinc status by serum zinc concentrations in a representative sample of the population should be made in order to identify segments of the population at high risk. Appropriate intervention strategies including supplementation, food fortification, dietary diversification and biofortification should be developed to target different populations.
In conclusion, urban people in China are at low risk of zinc deficiency, whereas the rural population is at a moderate level of risk for zinc deficiency. Lactating women and rural children are vulnerable populations in China. Effective strategies correcting these problems should be developed and implemented in targeted risk groups.
Sources of funding: The 2002 China National Nutrition and Health Survey was funded by the Ministry of Health and Ministry of Science and Technology, China (2001DEA30035, 2003DIA6N008).
Conflict of interest declaration: There is no interest conflict.
Authorship responsibilities: G.M. was responsible for study design, project management, data analysis, manuscript writing and interpretation of results. Y.L. was responsible for data cleaning and analysis, manuscript writing and interpretation of results. Y.J. was responsible for determinig the phytate content of foods, manuscript writing and interpretation of results. S.D. was responsible for manuscript writing and interpretation of results. F.J.K. was responsible for manuscript writing and interpretation of results. X.Y. was responsible for study design, project management and interpretation of results.
Acknowledgements: We thank all team members and all participants from the 31 provinces. We are very appreciative of the support of UNICEF, WHO, Unilever China and the Danone Nutrition Institute China.