Study population and design
A cross-sectional study on malnutrition and anemia prevalence among infants aged 0–18 months was conducted in 2 rural counties in central Shaanxi province in northwestern China. A sample of 280 infants was estimated to be able to detect a 25% anemia rate and 8% acceptable variability, assuming α = 0.05, β = 0.20, and a design effect = 2.5. The sample size was expanded to 336 infants to account for a 20% non-response rate. We sampled 336 infants from 5 stratified age groups of 0–3 months, 4–6 months, 7–9 months, 10–12 months and 13–18 months. 11 infants in each age group were selected randomly for balanced age distribution in the subsample with size of 55 for further measuring. The subsample of 55 was selected to detect difference in serum micronutrients concentration between anemic and non-anemic group, assuming a two tailed test, with α = 0.05 and β = 0.20. These 5 age groups included the key growth periods and weaning time.
Singleton infants aged 0–18 months whose gestational age was 38–42 weeks at delivery and who were free of serious disease (including serious diseases of heart and brain, digestive system and urinary system, serious infective diseases) or congenital malformations were selected through a multistage sampling technique. We excluded infants who experienced diarrhea or upper respiratory tract infections in the last month or received antibiotics or micronutrient supplement therapy in the last 2 months. A selection of primary sampling units (28 villages in 2 counties) was based on a probability proportional to population size, and we randomly selected 12 households with 0–18 month-old singleton infants from each village. In all, 336 infants were enrolled in the study between March and October of 2010 and a stratified subsample of 55 infants was selected from all 336 infants (n = 11/group) for further laboratory studies.
Data collection and blood sampling
Well-trained researchers used a structured questionnaire to collect information about household socio-demographic characteristics and infant feeding practices during in-home interviews. Infant’s birth date was obtained according to the immunization card presented during the interview.
Infant weight and height were measured using standardized methods described by the World Health Organization (WHO, 1995). Weight was measured to the nearest 10 g on an electronic scale. Recumbent length was measured to the nearest mm on a length board .
Weight-for-age Z-score (WAZ), height-for-age Z-score (HAZ), and weight-for-length Z-score (WHZ) was calculated with 2006 WHO Anthro and WHO Child Growth Standards (WHO, 2006). Malnutrition including underweight, stunting and wasting were defined from WAZ, HAZ, and WHZ, respectively, as <2 standard deviations than the mean .
Three consecutive days of food records, were used to assess the infant dietary intake. Complementary foods were recorded and calculated, and the amount of each complementary food was divided into its ingredients. The total nutrient intake was calculated from the sum of the amount of foods consumed multiplied by the nutrient content. For complementary foods, nutrient content was derived from the electronic version of the 2002 China Food Composition Tables or food ingredient labels.
We performed venipuncture in a peripheral vein and collected 2-mL blood samples among 55 infants of subsample. Every blood sample was divided into 2 tubes; one anticoagulated for whole blood and the other was left to clot and centrifuged to have serum for measurement of vitamin A (retinol), iron and zinc. To avoid photo-oxidation of vitamin A (retinol), tubes were wrapped in aluminum foil and were placed in a box away from light. For anticoagulated tubes, blood samples were labeled and analyzed for hematocrit and hemoglobin. Full blood counts were performed with an automated cell counter (F820, Hitachi Inc., Tokyo, Japan). For the other tubes, the serum were stored and frozen until analysis for vitamin A (retinol), iron, and zinc. Retinol concentration was measured in serum and the experimentation was done in faint light. It was measured by microfluorometric determination (RF-1501, Shimadzu Inc., Tokyo, Japan). Serum iron and zinc concentrations were measured by atomic absorption spectrophotometry (Z-8000, Hitachi Inc., Tokyo, Japan). Only 1-mL blood samples were collected in anticoagulated tubes among the rest 281 infants (except 55 subsamples from 336 infants), which were just analyzed for hematocrit and hemoglobin.
Anemia was defined as the WHO standard for infants aged 6–18 months  and the Chinese standard for infants aged 0–5 months  (Hb level is below 145 g/L for infants aged 0–29 days, below 90 g/L for infants aged 30–119 days, below 100 g/L for infants aged 120–179 days, and below 110 g/L for infants aged 180 days to 18 months). Anemia included mild, moderate and severe anemia and the definitions of them by age were as follows. Mild anemia: Hb level is between 90–144 g/L for infants aged 0–29 days, between 90–99 g/L for infants aged 120–179 days, between 90–109 g/L for infants aged 180 days to 18 months; Moderate anemia: Hb level is between 60–89 g/L for all age group infants; Severe anemia: Hb level is between 30–59 g/L for all age group infants.
Breast-feeding was defined including exclusive breast-feeding and predominant breast-feeding during the first 4 months of life. Exclusive breast-feeding: The infant had received only breast milk from his/her mother or a wet nurse, or expressed breast milk, and no other liquids or solids, with the exception of drops or syrups consisting of vitamins, mineral supplements or medicines. Predominant breast-feeding: The infant’s predominant source of nourishment had been breast milk. However, the infant might also have received water and water-based drinks (sweetened and flavored water, teas, infusions etc.); fruit juice; oral rehydration salts solution; drop and syrup forms of vitamins, minerals and medicines; and ritual fluids (in limited quantities). With the exception of fruit juice and sugar-water, no food-based fluid was allowed under this definition.
Low income family was defined as monthly family income less than 800 yuan.