European Journal of Clinical Nutrition (2014) 68, 253–258 & 2014 Macmillan Publishers Limited All rights reserved 0954-3007/14 www.nature.com/ejcn
ORIGINAL ARTICLE
Relationship between socioeconomic status and anemia prevalence in adolescent girls based on the fourth and fifth Korea National Health and Nutrition Examination Surveys JY Kim1, S Shin2, K Han3, K-C Lee1, J-H Kim2, YS Choi4, DH Kim4, GE Nam4, HD Yeo1, HG Lee1 and B-J Ko4 BACKGROUND/OBJECTIVES: We studied the relationship between socioeconomic status (SES), represented by household income, and the prevalence of anemia and iron deficiency anemia (IDA) among adolescent girls in Korea. SUBJECTS/METHODS: The samples were based on the data from a four-year (2008–2011) collection for the Korea National Health and Nutrition Examination Survey (1312 girls, age 10–18 years). The survey included demographic, anthropometric, biochemical and nutritional parameters. A multiple regression analysis after adjusting for age, body mass index (BMI), red blood cell count, white blood cell count and red meat intake was performed. Anemia was defined as hemoglobin level lower than 11.5 g/dl for ages 10–11 years and 12.0 g/dl for ages 12–14 years. Iron deficiency was defined as serum ferritin level below 15 mg/l. RESULTS: The prevalences of anemia and IDA in Korean girls were 5.3 and 4.2%, respectively. Girls with anemia were older, taller, weighed more, had higher BMI, had higher portion of menarche experience and consumed less red meat than girls without anemia. Girls with higher income had lower anemia prevalence and consumed more iron and vitamins. Logistic regression analysis showed a decreasing trend in anemia prevalence as household income increased. Correlation analysis demonstrated that there is a relationship between household income and serum hemoglobin and ferritin levels (P ¼ 0.003 and P ¼ 0.026, respectively). CONCLUSIONS: Higher SES leads to lower prevalence of anemia and IDA in Korean adolescent girls. This may be due to the fact that higher SES individuals consume more iron and vitamin C. European Journal of Clinical Nutrition (2014) 68, 253–258; doi:10.1038/ejcn.2013.241; published online 4 December 2013 Keywords: social class; income; anemia; iron deficiency anemia; nutrition status; nutrition surveys
INTRODUCTION Anemia is a disease of great burden, especially in developing countries such as Asian countries.1 The estimated global prevalence of anemia is as high as 24.8%, affecting 1.62 billion people.2 The most common form of anemia is iron deficiency anemia (IDA), which is more prevalent in adolescent girls than in other age groups.3 Recent studies conducted in Asia showed the high prevalence of anemia and IDA in adolescent girls, from 41.1 to 96.5% for anemia4–6 and from 35.8 to 78.3% for IDA.5,7 This high prevalence of anemia represents a poor nutrition status and reduced health of a population.1,2 Iron deficiency has several adverse effects on pediatric populations. A recent study showed that iron deficiency impairs cognitive function in adolescents3 and has negative impact on social–emotional behavior of infants.8 Moreover, iron deficiency contributes to increased risk of infection by impairing innate immunity and cell-mediated immunity.9 Especially in adolescent girls of childbearing age, iron deficiency can lead to preterm delivery, low birth weight and inferior neonatal health.10 Iron deficiency is influenced by various host factors including age, sex and physiological, pathological, dietary and 1
socioeconomic conditions.1 Some examples of dietary factors are diets with low iron or low iron bioavailability. In addition, other nutrients necessary for hematopoiesis such as folic acid or vitamins A, B12 and C may also be deficient.11 Previous reports have shown that there is definite correlation between socioeconomic status (SES) and anemia.1,12 The percentage of anemia was higher in the lower socioeconomic strata among adolescent girls.13 However, these studies have been concentrated on limited populations such as developing countries or a specific age group such as children younger than 5 years of age. As Korea is a developed country (The World Bank. The world bank list of economies http://data.worldbank.org/about/country-classifications/countryand-lending-groups.), it is difficult to directly apply previous data collected in developing countries to Korea. One recent study conducted in Korea has shown that low maternal education level leads to higher prevalence of anemia, but household income does not; however, this study was not representative of the Korean pediatric population.14 The aim of this study was to reveal whether SES, represented by household income, has an effect on prevalence of anemia in nationally representative Korean adolescent girls.
College of Medicine, Korea University, Seoul, South Korea; 2Department of Public Health, Graduate School of Public Health, Seoul National University, Seoul, South Korea; Department of Biostatistics, College of Medicine, Catholic University, Seoul, South Korea and 4Department of Family Medicine, College of Medicine, Korea University, Seoul, South Korea. Correspondence: Dr B-J Ko, Department of Family Medicine, College of Medicine, Korea University, 126-1, 5-ka, Anam-Dong, Seongbuk-Gu, Seoul 136-705, South Korea. E-mail: [email protected] Received 4 April 2013; revised 12 August 2013; accepted 20 October 2013; published online 4 December 2013 3
Anemia and socioeconomic status in Korean girls JY Kim et al
254 SUBJECTS AND METHODS Study participants This study was based upon the data collected in the second and third years of Korea National Health and Nutrition Examination Surveys (KNHANES) IV (2008, 2009) and the first and second years of KNHANES V (2010, 2011).15 To assess the health and nutritional status of the civilian noninstitutionalized population of Korea, the target population, KNHANES has been performed periodically since 1998. KNHANES IV and V were crosssectional and nationally representative surveys conducted by the Division of Chronic Disease Surveillance, Korea Centers for Disease Control and Prevention. In KNHANES IV, rolling survey sampling was adopted and the sampling frame was defined based on the 2005 National Census Registry. A stratified, multistage, cluster probability sampling design according to geographic area, sex and age group was used to select the household unit. Two hundred primary sampling units was proportionally allocated to reflect the target population. A total of 9200 households were investigated in KNHANES IV (2008, 2009). In KNHANES V (2010, 2011), rolling survey sampling was also adopted and 7680 households were included. The selected samples were weighted to represent the entire Korean population. The surveys consisted of a health interview survey, a nutrition survey and a health examination survey. Data were collected by household interviews, and direct standardized physical examinations were performed at specially equipped mobile examination centers. A total of 37 753 individuals enrolled initially. From among the participants, we analyzed the data of 1312 girls aged 10–18 years. Exclusion criteria included cancer, smoking, pregnancy and menstruation at the time of the survey because of
Table 1.
the obvious effects of these factors on serum hemoglobin level. Subjects with acute infection, viral hepatitis, connective tissue disease, thyroid disease and a white blood cell (WBC) count 410 000 (cells/ul) were also excluded because of the possibility that can affect ferritin levels. Records without biochemistry data were also excluded. All the participants in the survey provided informed consent.
Demographic variables Age, residency area and menarche experience were obtained from a health interview survey. Experience of basic livelihood security system was defined as a participation of the program under the National Basic Livelihood Security System, which involves the households with income less than minimal cost of living. Household income was stratified into four categories according to quartile: low, middle low, middle high and high.
Anthropometric measurements Anthropometric data was measured by trained medical personnel.15 Standing height was measured on all participants aged 2 years and older. Participants were instructed to stand with the heels together and toes apart touching their back of the head, shoulder blades, buttocks and heels on the backboard. Body weight was measured after the readout on the digital measurement device becomes stable. Body weight was measured to the nearest 0.1 kg, and height was measured to the nearest 0.1 cm. While measuring body weight and height, participants were wearing typical indoor clothing without shoes. Body mass index (BMI) was
Clinical, nutritional and demographic characteristics according to the presence of anemia Anemia
Age (year) Height (cm) Weight (kg) BMI (kg/m2) WC (cm) SBP (mmHg) DBP (mmHg) Hb (g/dl) Hct (%) Ferritin (ng/ml)a RBC count (Mil/ul) WBC count (Thous/ul)a Energy intake (kcal/day) Fat intake (g/day) Protein intake (g/day) Carbohydrate intake (g/day) Fiber intake (g/day) Vitamin C intake (mg/day)a Iron intake (mg/day)a Calcium intake (mg/day)a Phosphorus intake (mg/day)a Sodium intake (mg/day)a Potassium intake (mg/day)a Vitamin A intake (mgRE/day)a Carotene intake (mg/day)a Retinol intake (mg/day)a Vitamin B1 intake (mg/day)a Vitamin B2 intake (mg/day)a Niacin intake (mg/day)a Red meat intake (g/day) Non-heme iron-rich foods intake (g/day) Urban residence (%) Experience of basic livelihood security system (%) Menarche (%)
Iron deficiency anemia
No (n ¼ 1243)
Yes (n ¼ 69)
P-valueb
No (n ¼ 1257)
Yes (n ¼ 55)
P-valueb
13.9±0.1 156.3±0.3 49.7±0.4 20.1±0.1 66.9±0.3 102.8±0.4 65.2±0.3 13.4±0.0 40.0±0.1 32.8±0.7 4.6±0.0 6.1±0.0 1807.6±25.6 21.9±0.3 14.1±0.1 64.0±0.4 5.0±0.1 85.8±3.3 10.9±0.3 446.8±10.7 1025.4±16.3 3622.3±85.9 2339.6±43.0 625.5±26.6 2915.4±155.4 134.2±5.4 1.2±0.0 1.2±0.0 13.9±0.3 101.6±5.8 12.6±1.0 82.6 (2.1) 8.4 (1.1) 72.8 (1.4)
15.7±0.2 160.1±0.8 53.8±1.1 21.0±0.4 68.0±0.9 104.8±1.3 67.0±1.0 11.2±0.1 34.9±0.3 11.8±2.3 4.3±0.0 5.3±0.2 1701.5±98.1 21.8±1.3 13.0±0.5 65.2±1.5 4.6±0.4 80.6±12.5 10.4±1.3 444.4±37.2 926.5±55.7 3349.7±342.2 2072.6±116.4 522.8±49.6 2275.4±207.6 227.9±96.2 1.1±0.1 1.1±0.1 11.8±0.8 59.1±9.3 11.8±3.3 77.3 (7.4) 24.3 (7.1) 99.5 (0.5)
o0.001 o0.001 o0.001 0.035 0.226 0.121 0.071 o0.001 o0.001 o0.001 o0.001 o0.001 0.299 0.978 0.047 0.446 0.223 0.747 0.424 0.547 0.136 0.251 0.130 0.498 0.845 0.375 0.118 0.382 0.035 o0.001 0.823 0.394 o0.001 o0.001
13.9±0.1 156.4±0.3 49.7±0.4 20.2±0.1 66.9±0.3 102.8±0.4 65.2±0.3 13.4±0.0 40.0±0.1 32.8±0.7 4.6±0.0 6.1±0.0 1806.9±25.3 21.9±0.3 14.2±0.1 64.0±0.4 5.1±0.1 85.9±3.3 10.9±0.3 447.5±10.7 1026.1±16.2 3619.3±84.3 2340.3±42.5 625.1±26.3 2910.7±153.9 140.3±8.7 1.2±0.0 1.2±0.0 13.9±0.3 101.6±5.8 12.6±0.9 82.6 (2.1) 8.7 (1.1) 73.1 (1.4)
15.6±0.3 160.3±1.0 53.9±1.3 21.0±0.5 67.7±1.1 106.0±1.4 68.1±1.2 11.0±0.2 34.7±0.4 5.6±0.4 4.3±0.0 5.3±0.2 1694.6±115.9 21.2±1.5 12.6±0.5 66.2±1.6 4.3±0.4 77.4±14.0 9.8±1.5 429.4±42.8 892.6±61.9 3357.4±394.2 2004.9±130.0 512.5±55.5 2246.6±238.9 121.6±21.4 1.0±0.1 1.0±0.1 11.2±0.8 50.2±7.6 11.8±3.9 76.7 (8.3) 21.7 (7.8) 99.4 (0.6)
o0.001 o0.001 0.002 0.091 0.486 0.027 0.016 o0.001 o0.001 o0.001 o0.001 o0.001 0.346 0.640 0.006 0.159 0.049 0.523 0.259 0.967 0.061 0.299 0.069 0.381 0.677 0.892 0.035 0.132 0.014 o0.001 0.838 0.426 0.014 o0.001
Results are shown as mean±s.e. or percentage (s.e.). Red meat intake was defined as consumption of beef, veal, pork and lamb (fresh, minced and frozen). Non-heme iron-rich foods included a top 10 list of grains and vegetables with non-heme iron-rich content among Korean common food. Abbreviations: BMI, body mass index; WC, waist circumference; SBP, systolic blood pressure; DBP, diastolic blood pressure; Hb, hemoglobin; Hct, hematocrit; RBC, red blood cell; WBC, white blood cell. Bold values are statistically significant. aValues for ferritin, WBC count and nutrients were log-transformed before analysis. bP-values were calculated by Student’s t-test or Chi-square test.
European Journal of Clinical Nutrition (2014) 253 – 258
& 2014 Macmillan Publishers Limited
Anemia and socioeconomic status in Korean girls JY Kim et al
255 calculated by dividing body weight (kg) by the square of height (m2). Waist circumference was measured using the midpoint between the lower margin of the rib cage and the iliac crest. Systolic blood pressure and diastolic blood pressure were measured at the right arm by a standard mercury sphygmomanometer (Baumanometer; W. A. Baum Co., Inc., Copiague, NY, USA). The average values of two measurements with 5 min intervals were used for statistical analysis of systolic blood pressure and diastolic blood pressure.
Biochemical measurements Blood samples were obtained from the antecubital vein after the participants had fasted for eight or more hours to determine the concentrations of hemoblobin (Hb), hematocrit (Hct), serum ferritin, red blood cell count and WBC count. Blood samples were processed appropriately, then immediately refrigerated and transported in cold storage to the Central Testing Institute in Seoul, Korea. Blood samples were analyzed within 24 h after transportation. Serum ferritin was measured with a Chemiluminescence Immunoassay (CLIA) Analyzer (ACS 180, Bayer Diagnostics, USA). The blood cell count was measured by laser flow cytometry using the XE-2100D (Sysmex, Kobe, Japan).
Dietary intake Dietary intake was assessed with the single-day 24 h recall method. A nutrition survey was conducted through in-person interviews at participant’s homes by trained dieticians. Dietary data was based on food database of KNHANES and the food composition table of the Rural Development Administration and the nutrient database of the Korea Health Industry Development Institute. A total of 17 nutrients (total energy, fat, protein, carbohydrate, fiber, vitamin C, iron, calcium, phosphorus, sodium, potassium, vitamin A, carotene, retinol, vitamin B1, vitamin B2 and niacin) was calculated. Red meat intake was defined as consumption of beef, veal, pork and lamb (fresh, minced and frozen) based on the definition of red meat from European Prospective Investigation into Cancer and Nutrition cohorts.16 We included top 10 grains and vegetables most rich in non-heme iron among common Korean food to build a variable ‘Non-heme iron-rich foods’.
Definitions of anemia, iron deficiency and IDA In children aged 10–11 years, anemia was defined as serum Hb level below 11.5 g/dl. The cutoff values for children aged 12–14 years and nonpregnant women aged 15 years or older were the same: serum Hb level
Table 2.
below 12 g/dl. Iron deficiency was defined as serum ferritin level below 15 mg/l. IDA was specified as an anemia with iron deficiency.1,17
Statistical analyses Statistical analyses were performed using the SAS (Version 9.2; SAS Institute, Cary, NC, USA) survey procedure to account for the complex sampling design and to provide nationally representative prevalence estimates. Two-sided P-value of o0.05 was regarded as statistically significant. Student’s t-test or ANOVA test was applied for analysis of demographic, anthropometric, biochemical and nutritional factors, and results were presented as mean and standard errors. Residency area, experience of basic livelihood security system and menarche experience were analyzed by Chi-square test, and results were presented as percentage (%) instead of mean. Serum ferritin level, WBC count and values for nutrients were log-transformed before analyses for their skewed distribution. Multiple logistic regression analyses were performed to evaluate the household income as a risk factor for anemia and IDA. The highest household income group served as the reference. We entered age and BMI as covariates in model 1. Separate model (model 2) additionally included red meat intake and WBC count plus all variables in model 1. We selected red meat intake only as a covariate instead of nutrients considering its statistically significance on anemia and to avoid the problem of multicollinearity that can occur if both red meat intake and several nutrients are included in the same model.
RESULTS Clinical, nutritional and demographic characteristics of the subjects according to the presence of anemia or IDA are shown in Table 1. Prevalence of anemia and IDA in Korean adolescent girls was 5.3 and 4.2%, respectively. Girls with anemia were older, taller, weighed more and had higher BMI, whereas serum ferritin level, WBC count and intake of protein, niacin and red meat were lower in anemic patients than in participants without anemia. More girls with anemia experienced menarche and had higher incidence of basic livelihood security system experience than girls without anemia. Residence did not differ between the two groups. The result of comparison between girls with and without IDA was similar: age, height, body weight, WBC count and intake of protein, niacin and red meat were different between the two groups.
Clinical and demographic characteristics of the subjects by household income Household income
Age (year) Height (cm) Weight (kg) BMI (kg/m2) WC (cm) SBP (mmHg) DBP (mmHg) Hb (g/dl) Hct (%) Ferritin (ng/ml)a RBC count (Mil/ul) WBC count (Thous/ul)a Urban residence (%) Experience of basic livelihood security system (%) Menarche (%) Proportion of anemia (%) Proportion of IDA (%)
Low (n ¼ 145)
Middle low (n ¼ 365)
Middle high (n ¼ 413)
High (n ¼ 389)
P-valueb
14.8±0.2 157.9±0.8 51.6±1.2 20.5±0.4 67.6±1.0 103.2±1.2 65.9±1.1 13.0±0.1 39.2±0.3 28.1±2.1 4.5±0.0 5.9±0.1 81.6 (3.7) 45.8 (5.4) 86.8 (3.0) 13.8 (3.6) 10.8 (3.2)
13.9±0.2 155.1±0.6 49.9±0.8 20.5±0.3 67.3±0.7 103.6±0.6 65.2±0.6 13.2±0.1 39.6±0.2 30.1±1.4 4.6±0.0 6.2±0.1 78.4 (3.6) 9.1 (2.1) 73.2 (2.7) 5.3 (1.3) 4.6 (1.2)
13.7±0.2 156.0±0.5 48.7±0.7 19.8±0.2 66.1±0.5 102.2±0.6 64.9±0.5 13.3±0.0 39.9±0.1 32.1±1.3 4.6±0.0 6.1±0.1 81.9 (2.9) 1.4 (0.6) 67.9 (2.8) 4.0 (1.0) 3.4 (1.0)
14.0±0.1 158.0±0.5 50.5±0.6 20.1±0.2 67.0±0.5 102.7±0.6 65.7±0.5 13.4±0.1 40.1±0.1 34.8±1.2 4.6±0.0 6.1±0.1 87.5 (2.7) 0.9 (0.7) 76.4 (2.6) 2.6 (0.8) 1.5 (0.6)
0.002 o0.001 0.126 0.174 0.403 0.410 0.570 0.002c 0.008c 0.002c 0.073 0.185 0.096 o0.001c o0.001 o0.001c o0.001c
Results are shown as mean±s.e. or percentage (SE). Abbreviations: BMI, body mass index; WC, waist circumference; SBP, systolic blood pressure; DBP, diastolic blood pressure; Hb, hemoglobin; Hct, hematocrit; RBC, red blood cell; WBC, white blood cell; IDA, iron deficiency anemia. aValues for ferritin and WBC count were log-transformed before analysis. bP-values were calculated by ANOVA test. cIndicates that P for trend o0.05.
& 2014 Macmillan Publishers Limited
European Journal of Clinical Nutrition (2014) 253 – 258
Anemia and socioeconomic status in Korean girls JY Kim et al
256 Table 3.
Nutritional intakes of study participants according to household income Household income
Energy (kcal/day) Fat (g/day) Protein (g/day) Carbohydrate (g/day) Fiber (g/day) Vitamin C (mg/day)a Iron (mg/day)a Calcium (mg/day)a Phosphorus (mg/day)a Sodium (mg/day)a Potassium (mg/day)a Vitamin A (mgRE/day)a Carotene (mg/day)a Retinol (mg/day)a Vitamin B1 (mg/day)a Vitamin B2 (mg/day)a Niacin (mg/day)a Red meat (g/day) Non-heme iron-rich foods (g/day)
Low
Middle low
Middle high
High
P-valueb
1741.8±75.2 22.9±1.2 14.1±0.6 63.0±1.6 4.5±0.3 79.0±11.0 10.0±0.7 381.2±21.9 963.4±48.2 3630.0±238.2 2161.3±112.3 526.7±41.8 2410.3±225.2 125.7±15.8 1.2±0.1 1.1±0.1 13.0±0.8 94.0±12.6 13.5±3.7
1747.4±36.7 22.0±0.5 14.1±0.3 63.9±0.6 4.6±0.1 76.4±4.8 10.7±0.6 419.0±15.9 981.8±26.3 3387.6±103.1 2176.1±57.6 571.0±25.5 2646.5±144.2 125.8±6.3 1.2±0.0 1.1±0.0 13.3±0.4 83.9±7.3 11.0±1.2
1814.3±43.9 20.4±0.6 14.1±0.2 65.4±0.7 5.3±0.2 90.7±6.2 10.9±0.4 469.7±20.4 1047.2±29.3 3805.7±166.0 2434.5±80.9 653.8±56.8 3054.2±329.7 132.6±10.0 1.2±0.1 1.2±0.0 14.0±0.5 117.8±14.1 14.0±1.9
1884.5±50.7 22.8±0.6 13.9±0.2 63.2±0.7 5.4±0.2 94.0±5.6 11.4±0.4 487.4±21.5 1063.7±28.2 3638.1±164.7 2464.3±80.4 687.9±61.6 3203.6±365.5 171.1±28.9 1.3±0.1 1.3±0.0 14.4±0.5 99.3±9.6 12.2±1.5
0.204c 0.011 0.742 0.095 0.005c 0.002c 0.081c 0.017c 0.043c 0.542 0.007c 0.148c 0.299 0.630 0.170 0.009c 0.124c 0.180 0.468
Results are shown as mean±s.e. Red meat intake was defined as consumption of beef, veal, pork and lamb (fresh, minced and frozen). Non-heme iron-rich foods included a top 10 list of grains and vegetables with non-heme iron-rich content among Korean common food. aValues for nutrients were log-transformed before analysis. bP-values were calculated by ANOVA test. cIndicates that P for trend o0.05.
In addition, girls with IDA had higher systolic blood pressure and diastolic blood pressure, and consumed less vitamin B1 than those without IDA. The characteristics of the subjects according to household income are shown in Table 2. When household income increased from lowest to highest quartile, the prevalence of anemia and IDA decreased, and serum Hb, Hct and ferritin levels increased. Nutritional characteristics according to household income are exhibited in Table 3. When household income increased, daily intake of energy, fiber, vitamin C, iron, calcium, phosphorus, potassium, vitamin A, vitamin B2 and niacin increased. There was a significant increasing trend in daily iron intake according to income levels (P for trend ¼ 0.011). Red meat intake increased as household income increased, although statistically insignificant (P-value ¼ 0.180). Non-heme iron-rich foods intake did not show a difference between income groups. Table 4 shows the odds ratios (ORs) and 95% confidence intervals (CIs) of anemia and IDA across categories of household income. Model 1 was adjusted for age and BMI. Red meat intake and WBC count were additionally adjusted in model 2. In both models, when household income decreased, the ORs for anemia or IDA increased. Compared with the highest income group, the OR for anemia was higher in the lowest income group (in model 2, OR 7.10, 95% CI 2.49, 20.23). The same trend was also observed in the OR for IDA (in model 2, OR 5.83, 95% CI 1.81, 18.79). Figure 1 represents the correlations between household income and serum Hb and serum ferritin levels. Both serum Hb level and serum ferritin level showed a positive correlation with household income (r ¼ 0.11, P ¼ 0.003; r ¼ 0.08, P ¼ 0.026, respectively). DISCUSSION Lower SES is associated with the prevalence of anemia and IDA in Korean adolescent girls. Higher SES seems to have a protective effect on anemia and IDA. This was in concordance with results from previous studies. One study conducted in Greece showed that children living in an urban area and consuming more meat have a lower prevalence of anemia.12 Some studies performed in European Journal of Clinical Nutrition (2014) 253 – 258
Table 4. Odds ratios and 95% confidence intervals of anemia and iron deficiency anemia among Korean adolescent girls across categories of household income Model 1 OR Anemia household income High 1.00 Middle high 1.63 Middle low 2.12 Low 5.11 P for trend o0.001
Model 2
95% CI
OR
95% CI
Reference (0.73, 3.66) (0.95, 4.72) (2.23, 11.75)
1.00 2.48 3.35 7.10 o0.001
Reference (0.90, 6.86) (1.23, 9.12) (2.49, 20.23)
1.00 2.76 3.32 5.83 0.005
Reference (0.90, 8.46) (1.11, 9.97) (1.81, 18.79)
Iron deficiency anemia household income High 1.00 Reference Middle high 1.68 (0.69, 4.11) Middle low 1.86 (0.78, 4.44) Low 3.24 (1.26, 8.32) P for trend 0.022
Logistic regression analysis was performed. Model 1 was adjusted for age and BMI, and model 2 was adjusted for all variables in model 1 plus WBC count and red meat intake. Abbreviations: IDA, iron deficiency anemia; BMI, body mass index; WBC, white blood cell; OR, odds ratio; CI, confidence interval.
developing countries also showed that low SES is associated with high prevalence of anemia.4,18–20 The only study conducted in Korea focusing on this topic showed that low maternal education level affects anemia prevalence.14 Most of these studies on anemia, however, were performed in developing countries and were focused on limited populations or were not nationally representative. Our study is the first report to reveal the protective effect of high SES on anemia and IDA prevalence in a representative pediatric population in Korea. There are several factors that affect the onset of anemia and iron deficiency: age, sex and physiological, pathological and nutritional conditions.1 Nutritional components are very important & 2014 Macmillan Publishers Limited
Anemia and socioeconomic status in Korean girls JY Kim et al
16.00
5.00
14.00
4.00 Ferritin (ng/mL)*
Hb (g/dl)
257
12.00
10.00
3.00
2.00
1.00
8.00 r = 0.107 p -value = 0.003
6.00 0.00
2.00
4.00
6.00
8.00
10.00
r = 0.077 p -value = 0.026
0.00 0.00
2.00
Houshold income*
4.00
6.00
8.00
10.00
Household income*
Figure 1. Correlations between household income and hemoglobin and ferritin levels. Figure on the left shows the correlation between household income and Hb level. Figure on the right shows correlation between household income and ferritin level. Pearson’s correlation analysis was performed. *Household income and ferritin level were log-transformed. Abbreviation: Hb, hemoglobin.
factors, especially with respect to iron deficiency. Dietary iron is classified into two forms: heme iron and non-heme iron. Heme iron is found in meat and has high bioavailability, whereas non-heme iron is rich in plant and has lower bioavailability. The bioavailability of non-heme iron depends on the other nutrients that act as enhancers or inhibitors. Vitamin C and meat are enhancers, and calcium, fiber, tea and coffee are inhibitors of non-heme iron absorption.21,22 Our study showed that iron uptake increased as household income rose. Although the intake of non-heme iron-rich foods did not differ across SES strata, and despite statistically insignificant, there was a trend that higher income group consumed more red meat than lower income group (Table 3). In addition to the high iron uptake, more consumption of red meat and vitamin C (Table 3) may lead to higher absorption of iron and, as a consequence, lower the prevalence of anemia in the higher income group (Table 2). This hypothesis is supported by Table 1, which demonstrates that the anemia group consumed less red meat. As iron is lost by menstruating, girls who experienced menarche and were undergoing menstruation lose more iron than those without menarche. This resulted in higher prevalence of menarche experience among girls with anemia and IDA (Table 1). Regarding mean age at menarche in Korean girls born between 1980 and 1985 was 13.8,23 more girls aged 14–18 must have undergone menarche than those aged 10–13. This explains that girls with anemia were older than non-anemic girls (Table 1), which is in accordance with previous study.4 As adolescent girls are still under growth, anemic girls were taller and weighted more due to their older age (Table 1). Multiple logistic regression analysis revealed that the group with the lowest income had higher ORs for anemia and IDA compared with the group with the highest income after adjusting confounders including age, BMI, WBC count and red meat consumption. In addition, correlation analysis showed that there is a positive relationship between household income and Hb or ferritin level. These results suggest that household income independently influences anemia and IDA prevalence. This study had the following limitations: first, only girls were analyzed because of the very low prevalence of anemia in boys. This makes it difficult to apply the results directly to the general pediatric population. Second, some biochemical data were not included in the analysis. For example, total iron-binding capacity, which is an important parameter of IDA, was measured in too few adolescent girls to analyze. Third, confounders from previous studies such as maternal education level could not be analyzed as KNHANES did not survey these factors in children. Finally, this was a cross-sectional and therefore prospective study to reveal the & 2014 Macmillan Publishers Limited
relationship between SES and anemia prevalence with careful control of other confounders required. In conclusion, the results suggest that higher SES represented by household income leads to lower prevalence of anemia and IDA in Korean adolescent girls. This may be due to the fact that higher SES individuals consume more iron and vitamin C, which aids the absorption of iron.
CONFLICT OF INTEREST The authors declare no conflict of interest.
AUTHOR CONTRIBUTIONS JY Kim, KC Lee and BJ Ko designed the study; JH Kim, GE Nam, HD Yeo, HG Lee and K Han collected data; S Shin and K Han analyzed the data; JY Kim and BJ Ko wrote the manuscript; YS Choi and DH Kim provided significant advice and contributed to the discussion.
REFERENCES 1 World Health Organization. Iron Deficiency Anaemia: Assessment, Prevention And Control. A guide for programme managers. WHO: Geneva, 2001. 2 McLean E, Cogswell M, Egli I, Wojdyla D. de Benoist B. Worldwide prevalence of anaemia, WHO Vitamin and Mineral Nutrition Information System, 1993-2005. Public Health Nutr 2009; 12: 444–454. 3 Halterman JS, Kaczorowski JM, Aligne CA, Auinger P, Szilagyi PG. Iron deficiency and cognitive achievement among school-aged children and adolescents in the United States. Pediatrics 2001; 107: 1381–1386. 4 Shilpa S, Biradar SPB, Alatagi AC, Wantamutte AS, Malur PR. Prevalence of anaemia among adolescent girls: a one year cross-sectional study. J Clin Diag Res [serial online] 2012; 6: 372–377. 5 Htet MK, Dillon D, Akib A, Utomo B, Fahmida U, Thurnham DI. Microcytic anaemia predominates in adolescent school girls in the delta region of Myanmar. Asia Pac J Clin Nutr 2012; 21: 411–415. 6 Bulliyy G, Mallick G, Sethy GS, Kar SK. Hemoglobin status of non-school going adolescent girls in three districts of Orissa, India. Int J Adolesc Med Health 2007; 19: 395–406. 7 Baral KP, Onta SR. Prevalence of anemia amongst adolescents in Nepal: a community based study in rural and urban areas of Morang District. Nepal Med Coll J 2009; 11: 179–182. 8 Lozoff B, Clark KM, Jing Y, Armony-Sivan R, Angelilli ML, Jacobson SW. Dose-response relationships between iron deficiency with or without anemia and infant social-emotional behavior. J Pediatr 2008; 152: 696–702. 9 Kumar V, Choudhry VP. Iron deficiency and infection. Indian J Pediatr 2010; 77: 789–793. 10 Allen LH. Anemia and iron deficiency: effects on pregnancy outcome. Am J Clin Nutr 2000; 71: 1280S–1284S. 11 Samuelson G, Bratteby LE, Berggren K, Elverby JE, Kempe B. Dietary iron intake and iron status in adolescents. Acta Paediatr 1996; 85: 1033–1038.
European Journal of Clinical Nutrition (2014) 253 – 258
Anemia and socioeconomic status in Korean girls JY Kim et al
258 12 Gompakis N, Economou M, Tsantali C, Kouloulias V, Keramida M, Athanasiou-Metaxa M. The effect of dietary habits and socioeconomic status on the prevalence of iron deficiency in children of northern Greece. Acta Haematol 2007; 117: 200–204. 13 Chaudhary SM, Dhage VR. A study of anemia among adolescent females in the urban area of nagpur. Indian J Community Med 2008; 33: 243–245. 14 Choi HJ, Lee HJ, Jang HB, Park JY, Kang JH, Park KH et al. Effects of maternal education on diet, anemia, and iron deficiency in Korean school-aged children. BMC Public Health 2011; 11: 870. 15 Ministry of Health and Welfare, Korea Centers for Disease Control and Prevention. Korea National Health and Nutrition Examination Survey (KNHANES). Korea Centers for Disease Control and Prevention: Seoul, Korea, 2011. 16 Linseisen J, Kesse E, Slimani N, Bueno-De-Mesquita HB, Ocke MC, Skeie G et al. Meat consumption in the European Prospective Investigation into Cancer and Nutrition (EPIC) cohorts: results from 24-hour dietary recalls. Public Health Nutr 2002; 5: 1243–1258. 17 Expert Scientific Working Group. Summary of a report on assessment of the iron nutritional status of the United States population. Am J Clin Nutr 1985; 42: 1318–1330.
European Journal of Clinical Nutrition (2014) 253 – 258
18 Thankachan P, Muthayya S, Walczyk T, Kurpad AV, Hurrell RF. An analysis of the etiology of anemia and iron deficiency in young women of low socioeconomic status in Bangalore, India. Food Nutr Bull 2007; 28: 328–336. 19 Issaragrisil S, Kaufman DW, Anderson TE, Chansung K, Thamprasit T, Sirijirachai J et al. An association of aplastic anaemia in Thailand with low socioeconomic status. Aplastic Anemia Study Group. Br J Haematol 1995; 91: 80–84. 20 Animasahun BA, Temiye EO, Ogunkunle OO, Izuora AN, Njokanma OF. The influence of socioeconomic status on the hemoglobin level and anthropometry of sickle cell anemia patients in steady state at the Lagos University Teaching Hospital. Niger J Clin Pract 2011; 14: 422–427. 21 Conrad ME, Umbreit JN, Moore EG. Iron absorption and transport. Am J Med Sci 1999; 318: 213–229. 22 Andrews NC. Iron Deficiency and Related Disorders. In: Greer JP, Poerster J, Rodgers GM, Paraskevas F, Glader B, Arber DA et al. (eds) Wintrobe’s Clinical Hematology. 12 edn. (Wolters Kluwer Health/Lippincott Williams & Wilkins: Philadelphia) 2009, pp 810–834. 23 Cho GJ, Park HT, Shin JH, Hur JY, Kim YT, Kim SH et al. Age at menarche in a Korean population: secular trends and influencing factors. Eur J Pediatr 2010; 169: 89–94.
& 2014 Macmillan Publishers Limited
ORIGINAL ARTICLE
Relationship between socioeconomic status and anemia prevalence in adolescent girls based on the fourth and fifth Korea National Health and Nutrition Examination Surveys JY Kim1, S Shin2, K Han3, K-C Lee1, J-H Kim2, YS Choi4, DH Kim4, GE Nam4, HD Yeo1, HG Lee1 and B-J Ko4 BACKGROUND/OBJECTIVES: We studied the relationship between socioeconomic status (SES), represented by household income, and the prevalence of anemia and iron deficiency anemia (IDA) among adolescent girls in Korea. SUBJECTS/METHODS: The samples were based on the data from a four-year (2008–2011) collection for the Korea National Health and Nutrition Examination Survey (1312 girls, age 10–18 years). The survey included demographic, anthropometric, biochemical and nutritional parameters. A multiple regression analysis after adjusting for age, body mass index (BMI), red blood cell count, white blood cell count and red meat intake was performed. Anemia was defined as hemoglobin level lower than 11.5 g/dl for ages 10–11 years and 12.0 g/dl for ages 12–14 years. Iron deficiency was defined as serum ferritin level below 15 mg/l. RESULTS: The prevalences of anemia and IDA in Korean girls were 5.3 and 4.2%, respectively. Girls with anemia were older, taller, weighed more, had higher BMI, had higher portion of menarche experience and consumed less red meat than girls without anemia. Girls with higher income had lower anemia prevalence and consumed more iron and vitamins. Logistic regression analysis showed a decreasing trend in anemia prevalence as household income increased. Correlation analysis demonstrated that there is a relationship between household income and serum hemoglobin and ferritin levels (P ¼ 0.003 and P ¼ 0.026, respectively). CONCLUSIONS: Higher SES leads to lower prevalence of anemia and IDA in Korean adolescent girls. This may be due to the fact that higher SES individuals consume more iron and vitamin C. European Journal of Clinical Nutrition (2014) 68, 253–258; doi:10.1038/ejcn.2013.241; published online 4 December 2013 Keywords: social class; income; anemia; iron deficiency anemia; nutrition status; nutrition surveys
INTRODUCTION Anemia is a disease of great burden, especially in developing countries such as Asian countries.1 The estimated global prevalence of anemia is as high as 24.8%, affecting 1.62 billion people.2 The most common form of anemia is iron deficiency anemia (IDA), which is more prevalent in adolescent girls than in other age groups.3 Recent studies conducted in Asia showed the high prevalence of anemia and IDA in adolescent girls, from 41.1 to 96.5% for anemia4–6 and from 35.8 to 78.3% for IDA.5,7 This high prevalence of anemia represents a poor nutrition status and reduced health of a population.1,2 Iron deficiency has several adverse effects on pediatric populations. A recent study showed that iron deficiency impairs cognitive function in adolescents3 and has negative impact on social–emotional behavior of infants.8 Moreover, iron deficiency contributes to increased risk of infection by impairing innate immunity and cell-mediated immunity.9 Especially in adolescent girls of childbearing age, iron deficiency can lead to preterm delivery, low birth weight and inferior neonatal health.10 Iron deficiency is influenced by various host factors including age, sex and physiological, pathological, dietary and 1
socioeconomic conditions.1 Some examples of dietary factors are diets with low iron or low iron bioavailability. In addition, other nutrients necessary for hematopoiesis such as folic acid or vitamins A, B12 and C may also be deficient.11 Previous reports have shown that there is definite correlation between socioeconomic status (SES) and anemia.1,12 The percentage of anemia was higher in the lower socioeconomic strata among adolescent girls.13 However, these studies have been concentrated on limited populations such as developing countries or a specific age group such as children younger than 5 years of age. As Korea is a developed country (The World Bank. The world bank list of economies http://data.worldbank.org/about/country-classifications/countryand-lending-groups.), it is difficult to directly apply previous data collected in developing countries to Korea. One recent study conducted in Korea has shown that low maternal education level leads to higher prevalence of anemia, but household income does not; however, this study was not representative of the Korean pediatric population.14 The aim of this study was to reveal whether SES, represented by household income, has an effect on prevalence of anemia in nationally representative Korean adolescent girls.
College of Medicine, Korea University, Seoul, South Korea; 2Department of Public Health, Graduate School of Public Health, Seoul National University, Seoul, South Korea; Department of Biostatistics, College of Medicine, Catholic University, Seoul, South Korea and 4Department of Family Medicine, College of Medicine, Korea University, Seoul, South Korea. Correspondence: Dr B-J Ko, Department of Family Medicine, College of Medicine, Korea University, 126-1, 5-ka, Anam-Dong, Seongbuk-Gu, Seoul 136-705, South Korea. E-mail: [email protected] Received 4 April 2013; revised 12 August 2013; accepted 20 October 2013; published online 4 December 2013 3
Anemia and socioeconomic status in Korean girls JY Kim et al
254 SUBJECTS AND METHODS Study participants This study was based upon the data collected in the second and third years of Korea National Health and Nutrition Examination Surveys (KNHANES) IV (2008, 2009) and the first and second years of KNHANES V (2010, 2011).15 To assess the health and nutritional status of the civilian noninstitutionalized population of Korea, the target population, KNHANES has been performed periodically since 1998. KNHANES IV and V were crosssectional and nationally representative surveys conducted by the Division of Chronic Disease Surveillance, Korea Centers for Disease Control and Prevention. In KNHANES IV, rolling survey sampling was adopted and the sampling frame was defined based on the 2005 National Census Registry. A stratified, multistage, cluster probability sampling design according to geographic area, sex and age group was used to select the household unit. Two hundred primary sampling units was proportionally allocated to reflect the target population. A total of 9200 households were investigated in KNHANES IV (2008, 2009). In KNHANES V (2010, 2011), rolling survey sampling was also adopted and 7680 households were included. The selected samples were weighted to represent the entire Korean population. The surveys consisted of a health interview survey, a nutrition survey and a health examination survey. Data were collected by household interviews, and direct standardized physical examinations were performed at specially equipped mobile examination centers. A total of 37 753 individuals enrolled initially. From among the participants, we analyzed the data of 1312 girls aged 10–18 years. Exclusion criteria included cancer, smoking, pregnancy and menstruation at the time of the survey because of
Table 1.
the obvious effects of these factors on serum hemoglobin level. Subjects with acute infection, viral hepatitis, connective tissue disease, thyroid disease and a white blood cell (WBC) count 410 000 (cells/ul) were also excluded because of the possibility that can affect ferritin levels. Records without biochemistry data were also excluded. All the participants in the survey provided informed consent.
Demographic variables Age, residency area and menarche experience were obtained from a health interview survey. Experience of basic livelihood security system was defined as a participation of the program under the National Basic Livelihood Security System, which involves the households with income less than minimal cost of living. Household income was stratified into four categories according to quartile: low, middle low, middle high and high.
Anthropometric measurements Anthropometric data was measured by trained medical personnel.15 Standing height was measured on all participants aged 2 years and older. Participants were instructed to stand with the heels together and toes apart touching their back of the head, shoulder blades, buttocks and heels on the backboard. Body weight was measured after the readout on the digital measurement device becomes stable. Body weight was measured to the nearest 0.1 kg, and height was measured to the nearest 0.1 cm. While measuring body weight and height, participants were wearing typical indoor clothing without shoes. Body mass index (BMI) was
Clinical, nutritional and demographic characteristics according to the presence of anemia Anemia
Age (year) Height (cm) Weight (kg) BMI (kg/m2) WC (cm) SBP (mmHg) DBP (mmHg) Hb (g/dl) Hct (%) Ferritin (ng/ml)a RBC count (Mil/ul) WBC count (Thous/ul)a Energy intake (kcal/day) Fat intake (g/day) Protein intake (g/day) Carbohydrate intake (g/day) Fiber intake (g/day) Vitamin C intake (mg/day)a Iron intake (mg/day)a Calcium intake (mg/day)a Phosphorus intake (mg/day)a Sodium intake (mg/day)a Potassium intake (mg/day)a Vitamin A intake (mgRE/day)a Carotene intake (mg/day)a Retinol intake (mg/day)a Vitamin B1 intake (mg/day)a Vitamin B2 intake (mg/day)a Niacin intake (mg/day)a Red meat intake (g/day) Non-heme iron-rich foods intake (g/day) Urban residence (%) Experience of basic livelihood security system (%) Menarche (%)
Iron deficiency anemia
No (n ¼ 1243)
Yes (n ¼ 69)
P-valueb
No (n ¼ 1257)
Yes (n ¼ 55)
P-valueb
13.9±0.1 156.3±0.3 49.7±0.4 20.1±0.1 66.9±0.3 102.8±0.4 65.2±0.3 13.4±0.0 40.0±0.1 32.8±0.7 4.6±0.0 6.1±0.0 1807.6±25.6 21.9±0.3 14.1±0.1 64.0±0.4 5.0±0.1 85.8±3.3 10.9±0.3 446.8±10.7 1025.4±16.3 3622.3±85.9 2339.6±43.0 625.5±26.6 2915.4±155.4 134.2±5.4 1.2±0.0 1.2±0.0 13.9±0.3 101.6±5.8 12.6±1.0 82.6 (2.1) 8.4 (1.1) 72.8 (1.4)
15.7±0.2 160.1±0.8 53.8±1.1 21.0±0.4 68.0±0.9 104.8±1.3 67.0±1.0 11.2±0.1 34.9±0.3 11.8±2.3 4.3±0.0 5.3±0.2 1701.5±98.1 21.8±1.3 13.0±0.5 65.2±1.5 4.6±0.4 80.6±12.5 10.4±1.3 444.4±37.2 926.5±55.7 3349.7±342.2 2072.6±116.4 522.8±49.6 2275.4±207.6 227.9±96.2 1.1±0.1 1.1±0.1 11.8±0.8 59.1±9.3 11.8±3.3 77.3 (7.4) 24.3 (7.1) 99.5 (0.5)
o0.001 o0.001 o0.001 0.035 0.226 0.121 0.071 o0.001 o0.001 o0.001 o0.001 o0.001 0.299 0.978 0.047 0.446 0.223 0.747 0.424 0.547 0.136 0.251 0.130 0.498 0.845 0.375 0.118 0.382 0.035 o0.001 0.823 0.394 o0.001 o0.001
13.9±0.1 156.4±0.3 49.7±0.4 20.2±0.1 66.9±0.3 102.8±0.4 65.2±0.3 13.4±0.0 40.0±0.1 32.8±0.7 4.6±0.0 6.1±0.0 1806.9±25.3 21.9±0.3 14.2±0.1 64.0±0.4 5.1±0.1 85.9±3.3 10.9±0.3 447.5±10.7 1026.1±16.2 3619.3±84.3 2340.3±42.5 625.1±26.3 2910.7±153.9 140.3±8.7 1.2±0.0 1.2±0.0 13.9±0.3 101.6±5.8 12.6±0.9 82.6 (2.1) 8.7 (1.1) 73.1 (1.4)
15.6±0.3 160.3±1.0 53.9±1.3 21.0±0.5 67.7±1.1 106.0±1.4 68.1±1.2 11.0±0.2 34.7±0.4 5.6±0.4 4.3±0.0 5.3±0.2 1694.6±115.9 21.2±1.5 12.6±0.5 66.2±1.6 4.3±0.4 77.4±14.0 9.8±1.5 429.4±42.8 892.6±61.9 3357.4±394.2 2004.9±130.0 512.5±55.5 2246.6±238.9 121.6±21.4 1.0±0.1 1.0±0.1 11.2±0.8 50.2±7.6 11.8±3.9 76.7 (8.3) 21.7 (7.8) 99.4 (0.6)
o0.001 o0.001 0.002 0.091 0.486 0.027 0.016 o0.001 o0.001 o0.001 o0.001 o0.001 0.346 0.640 0.006 0.159 0.049 0.523 0.259 0.967 0.061 0.299 0.069 0.381 0.677 0.892 0.035 0.132 0.014 o0.001 0.838 0.426 0.014 o0.001
Results are shown as mean±s.e. or percentage (s.e.). Red meat intake was defined as consumption of beef, veal, pork and lamb (fresh, minced and frozen). Non-heme iron-rich foods included a top 10 list of grains and vegetables with non-heme iron-rich content among Korean common food. Abbreviations: BMI, body mass index; WC, waist circumference; SBP, systolic blood pressure; DBP, diastolic blood pressure; Hb, hemoglobin; Hct, hematocrit; RBC, red blood cell; WBC, white blood cell. Bold values are statistically significant. aValues for ferritin, WBC count and nutrients were log-transformed before analysis. bP-values were calculated by Student’s t-test or Chi-square test.
European Journal of Clinical Nutrition (2014) 253 – 258
& 2014 Macmillan Publishers Limited
Anemia and socioeconomic status in Korean girls JY Kim et al
255 calculated by dividing body weight (kg) by the square of height (m2). Waist circumference was measured using the midpoint between the lower margin of the rib cage and the iliac crest. Systolic blood pressure and diastolic blood pressure were measured at the right arm by a standard mercury sphygmomanometer (Baumanometer; W. A. Baum Co., Inc., Copiague, NY, USA). The average values of two measurements with 5 min intervals were used for statistical analysis of systolic blood pressure and diastolic blood pressure.
Biochemical measurements Blood samples were obtained from the antecubital vein after the participants had fasted for eight or more hours to determine the concentrations of hemoblobin (Hb), hematocrit (Hct), serum ferritin, red blood cell count and WBC count. Blood samples were processed appropriately, then immediately refrigerated and transported in cold storage to the Central Testing Institute in Seoul, Korea. Blood samples were analyzed within 24 h after transportation. Serum ferritin was measured with a Chemiluminescence Immunoassay (CLIA) Analyzer (ACS 180, Bayer Diagnostics, USA). The blood cell count was measured by laser flow cytometry using the XE-2100D (Sysmex, Kobe, Japan).
Dietary intake Dietary intake was assessed with the single-day 24 h recall method. A nutrition survey was conducted through in-person interviews at participant’s homes by trained dieticians. Dietary data was based on food database of KNHANES and the food composition table of the Rural Development Administration and the nutrient database of the Korea Health Industry Development Institute. A total of 17 nutrients (total energy, fat, protein, carbohydrate, fiber, vitamin C, iron, calcium, phosphorus, sodium, potassium, vitamin A, carotene, retinol, vitamin B1, vitamin B2 and niacin) was calculated. Red meat intake was defined as consumption of beef, veal, pork and lamb (fresh, minced and frozen) based on the definition of red meat from European Prospective Investigation into Cancer and Nutrition cohorts.16 We included top 10 grains and vegetables most rich in non-heme iron among common Korean food to build a variable ‘Non-heme iron-rich foods’.
Definitions of anemia, iron deficiency and IDA In children aged 10–11 years, anemia was defined as serum Hb level below 11.5 g/dl. The cutoff values for children aged 12–14 years and nonpregnant women aged 15 years or older were the same: serum Hb level
Table 2.
below 12 g/dl. Iron deficiency was defined as serum ferritin level below 15 mg/l. IDA was specified as an anemia with iron deficiency.1,17
Statistical analyses Statistical analyses were performed using the SAS (Version 9.2; SAS Institute, Cary, NC, USA) survey procedure to account for the complex sampling design and to provide nationally representative prevalence estimates. Two-sided P-value of o0.05 was regarded as statistically significant. Student’s t-test or ANOVA test was applied for analysis of demographic, anthropometric, biochemical and nutritional factors, and results were presented as mean and standard errors. Residency area, experience of basic livelihood security system and menarche experience were analyzed by Chi-square test, and results were presented as percentage (%) instead of mean. Serum ferritin level, WBC count and values for nutrients were log-transformed before analyses for their skewed distribution. Multiple logistic regression analyses were performed to evaluate the household income as a risk factor for anemia and IDA. The highest household income group served as the reference. We entered age and BMI as covariates in model 1. Separate model (model 2) additionally included red meat intake and WBC count plus all variables in model 1. We selected red meat intake only as a covariate instead of nutrients considering its statistically significance on anemia and to avoid the problem of multicollinearity that can occur if both red meat intake and several nutrients are included in the same model.
RESULTS Clinical, nutritional and demographic characteristics of the subjects according to the presence of anemia or IDA are shown in Table 1. Prevalence of anemia and IDA in Korean adolescent girls was 5.3 and 4.2%, respectively. Girls with anemia were older, taller, weighed more and had higher BMI, whereas serum ferritin level, WBC count and intake of protein, niacin and red meat were lower in anemic patients than in participants without anemia. More girls with anemia experienced menarche and had higher incidence of basic livelihood security system experience than girls without anemia. Residence did not differ between the two groups. The result of comparison between girls with and without IDA was similar: age, height, body weight, WBC count and intake of protein, niacin and red meat were different between the two groups.
Clinical and demographic characteristics of the subjects by household income Household income
Age (year) Height (cm) Weight (kg) BMI (kg/m2) WC (cm) SBP (mmHg) DBP (mmHg) Hb (g/dl) Hct (%) Ferritin (ng/ml)a RBC count (Mil/ul) WBC count (Thous/ul)a Urban residence (%) Experience of basic livelihood security system (%) Menarche (%) Proportion of anemia (%) Proportion of IDA (%)
Low (n ¼ 145)
Middle low (n ¼ 365)
Middle high (n ¼ 413)
High (n ¼ 389)
P-valueb
14.8±0.2 157.9±0.8 51.6±1.2 20.5±0.4 67.6±1.0 103.2±1.2 65.9±1.1 13.0±0.1 39.2±0.3 28.1±2.1 4.5±0.0 5.9±0.1 81.6 (3.7) 45.8 (5.4) 86.8 (3.0) 13.8 (3.6) 10.8 (3.2)
13.9±0.2 155.1±0.6 49.9±0.8 20.5±0.3 67.3±0.7 103.6±0.6 65.2±0.6 13.2±0.1 39.6±0.2 30.1±1.4 4.6±0.0 6.2±0.1 78.4 (3.6) 9.1 (2.1) 73.2 (2.7) 5.3 (1.3) 4.6 (1.2)
13.7±0.2 156.0±0.5 48.7±0.7 19.8±0.2 66.1±0.5 102.2±0.6 64.9±0.5 13.3±0.0 39.9±0.1 32.1±1.3 4.6±0.0 6.1±0.1 81.9 (2.9) 1.4 (0.6) 67.9 (2.8) 4.0 (1.0) 3.4 (1.0)
14.0±0.1 158.0±0.5 50.5±0.6 20.1±0.2 67.0±0.5 102.7±0.6 65.7±0.5 13.4±0.1 40.1±0.1 34.8±1.2 4.6±0.0 6.1±0.1 87.5 (2.7) 0.9 (0.7) 76.4 (2.6) 2.6 (0.8) 1.5 (0.6)
0.002 o0.001 0.126 0.174 0.403 0.410 0.570 0.002c 0.008c 0.002c 0.073 0.185 0.096 o0.001c o0.001 o0.001c o0.001c
Results are shown as mean±s.e. or percentage (SE). Abbreviations: BMI, body mass index; WC, waist circumference; SBP, systolic blood pressure; DBP, diastolic blood pressure; Hb, hemoglobin; Hct, hematocrit; RBC, red blood cell; WBC, white blood cell; IDA, iron deficiency anemia. aValues for ferritin and WBC count were log-transformed before analysis. bP-values were calculated by ANOVA test. cIndicates that P for trend o0.05.
& 2014 Macmillan Publishers Limited
European Journal of Clinical Nutrition (2014) 253 – 258
Anemia and socioeconomic status in Korean girls JY Kim et al
256 Table 3.
Nutritional intakes of study participants according to household income Household income
Energy (kcal/day) Fat (g/day) Protein (g/day) Carbohydrate (g/day) Fiber (g/day) Vitamin C (mg/day)a Iron (mg/day)a Calcium (mg/day)a Phosphorus (mg/day)a Sodium (mg/day)a Potassium (mg/day)a Vitamin A (mgRE/day)a Carotene (mg/day)a Retinol (mg/day)a Vitamin B1 (mg/day)a Vitamin B2 (mg/day)a Niacin (mg/day)a Red meat (g/day) Non-heme iron-rich foods (g/day)
Low
Middle low
Middle high
High
P-valueb
1741.8±75.2 22.9±1.2 14.1±0.6 63.0±1.6 4.5±0.3 79.0±11.0 10.0±0.7 381.2±21.9 963.4±48.2 3630.0±238.2 2161.3±112.3 526.7±41.8 2410.3±225.2 125.7±15.8 1.2±0.1 1.1±0.1 13.0±0.8 94.0±12.6 13.5±3.7
1747.4±36.7 22.0±0.5 14.1±0.3 63.9±0.6 4.6±0.1 76.4±4.8 10.7±0.6 419.0±15.9 981.8±26.3 3387.6±103.1 2176.1±57.6 571.0±25.5 2646.5±144.2 125.8±6.3 1.2±0.0 1.1±0.0 13.3±0.4 83.9±7.3 11.0±1.2
1814.3±43.9 20.4±0.6 14.1±0.2 65.4±0.7 5.3±0.2 90.7±6.2 10.9±0.4 469.7±20.4 1047.2±29.3 3805.7±166.0 2434.5±80.9 653.8±56.8 3054.2±329.7 132.6±10.0 1.2±0.1 1.2±0.0 14.0±0.5 117.8±14.1 14.0±1.9
1884.5±50.7 22.8±0.6 13.9±0.2 63.2±0.7 5.4±0.2 94.0±5.6 11.4±0.4 487.4±21.5 1063.7±28.2 3638.1±164.7 2464.3±80.4 687.9±61.6 3203.6±365.5 171.1±28.9 1.3±0.1 1.3±0.0 14.4±0.5 99.3±9.6 12.2±1.5
0.204c 0.011 0.742 0.095 0.005c 0.002c 0.081c 0.017c 0.043c 0.542 0.007c 0.148c 0.299 0.630 0.170 0.009c 0.124c 0.180 0.468
Results are shown as mean±s.e. Red meat intake was defined as consumption of beef, veal, pork and lamb (fresh, minced and frozen). Non-heme iron-rich foods included a top 10 list of grains and vegetables with non-heme iron-rich content among Korean common food. aValues for nutrients were log-transformed before analysis. bP-values were calculated by ANOVA test. cIndicates that P for trend o0.05.
In addition, girls with IDA had higher systolic blood pressure and diastolic blood pressure, and consumed less vitamin B1 than those without IDA. The characteristics of the subjects according to household income are shown in Table 2. When household income increased from lowest to highest quartile, the prevalence of anemia and IDA decreased, and serum Hb, Hct and ferritin levels increased. Nutritional characteristics according to household income are exhibited in Table 3. When household income increased, daily intake of energy, fiber, vitamin C, iron, calcium, phosphorus, potassium, vitamin A, vitamin B2 and niacin increased. There was a significant increasing trend in daily iron intake according to income levels (P for trend ¼ 0.011). Red meat intake increased as household income increased, although statistically insignificant (P-value ¼ 0.180). Non-heme iron-rich foods intake did not show a difference between income groups. Table 4 shows the odds ratios (ORs) and 95% confidence intervals (CIs) of anemia and IDA across categories of household income. Model 1 was adjusted for age and BMI. Red meat intake and WBC count were additionally adjusted in model 2. In both models, when household income decreased, the ORs for anemia or IDA increased. Compared with the highest income group, the OR for anemia was higher in the lowest income group (in model 2, OR 7.10, 95% CI 2.49, 20.23). The same trend was also observed in the OR for IDA (in model 2, OR 5.83, 95% CI 1.81, 18.79). Figure 1 represents the correlations between household income and serum Hb and serum ferritin levels. Both serum Hb level and serum ferritin level showed a positive correlation with household income (r ¼ 0.11, P ¼ 0.003; r ¼ 0.08, P ¼ 0.026, respectively). DISCUSSION Lower SES is associated with the prevalence of anemia and IDA in Korean adolescent girls. Higher SES seems to have a protective effect on anemia and IDA. This was in concordance with results from previous studies. One study conducted in Greece showed that children living in an urban area and consuming more meat have a lower prevalence of anemia.12 Some studies performed in European Journal of Clinical Nutrition (2014) 253 – 258
Table 4. Odds ratios and 95% confidence intervals of anemia and iron deficiency anemia among Korean adolescent girls across categories of household income Model 1 OR Anemia household income High 1.00 Middle high 1.63 Middle low 2.12 Low 5.11 P for trend o0.001
Model 2
95% CI
OR
95% CI
Reference (0.73, 3.66) (0.95, 4.72) (2.23, 11.75)
1.00 2.48 3.35 7.10 o0.001
Reference (0.90, 6.86) (1.23, 9.12) (2.49, 20.23)
1.00 2.76 3.32 5.83 0.005
Reference (0.90, 8.46) (1.11, 9.97) (1.81, 18.79)
Iron deficiency anemia household income High 1.00 Reference Middle high 1.68 (0.69, 4.11) Middle low 1.86 (0.78, 4.44) Low 3.24 (1.26, 8.32) P for trend 0.022
Logistic regression analysis was performed. Model 1 was adjusted for age and BMI, and model 2 was adjusted for all variables in model 1 plus WBC count and red meat intake. Abbreviations: IDA, iron deficiency anemia; BMI, body mass index; WBC, white blood cell; OR, odds ratio; CI, confidence interval.
developing countries also showed that low SES is associated with high prevalence of anemia.4,18–20 The only study conducted in Korea focusing on this topic showed that low maternal education level affects anemia prevalence.14 Most of these studies on anemia, however, were performed in developing countries and were focused on limited populations or were not nationally representative. Our study is the first report to reveal the protective effect of high SES on anemia and IDA prevalence in a representative pediatric population in Korea. There are several factors that affect the onset of anemia and iron deficiency: age, sex and physiological, pathological and nutritional conditions.1 Nutritional components are very important & 2014 Macmillan Publishers Limited
Anemia and socioeconomic status in Korean girls JY Kim et al
16.00
5.00
14.00
4.00 Ferritin (ng/mL)*
Hb (g/dl)
257
12.00
10.00
3.00
2.00
1.00
8.00 r = 0.107 p -value = 0.003
6.00 0.00
2.00
4.00
6.00
8.00
10.00
r = 0.077 p -value = 0.026
0.00 0.00
2.00
Houshold income*
4.00
6.00
8.00
10.00
Household income*
Figure 1. Correlations between household income and hemoglobin and ferritin levels. Figure on the left shows the correlation between household income and Hb level. Figure on the right shows correlation between household income and ferritin level. Pearson’s correlation analysis was performed. *Household income and ferritin level were log-transformed. Abbreviation: Hb, hemoglobin.
factors, especially with respect to iron deficiency. Dietary iron is classified into two forms: heme iron and non-heme iron. Heme iron is found in meat and has high bioavailability, whereas non-heme iron is rich in plant and has lower bioavailability. The bioavailability of non-heme iron depends on the other nutrients that act as enhancers or inhibitors. Vitamin C and meat are enhancers, and calcium, fiber, tea and coffee are inhibitors of non-heme iron absorption.21,22 Our study showed that iron uptake increased as household income rose. Although the intake of non-heme iron-rich foods did not differ across SES strata, and despite statistically insignificant, there was a trend that higher income group consumed more red meat than lower income group (Table 3). In addition to the high iron uptake, more consumption of red meat and vitamin C (Table 3) may lead to higher absorption of iron and, as a consequence, lower the prevalence of anemia in the higher income group (Table 2). This hypothesis is supported by Table 1, which demonstrates that the anemia group consumed less red meat. As iron is lost by menstruating, girls who experienced menarche and were undergoing menstruation lose more iron than those without menarche. This resulted in higher prevalence of menarche experience among girls with anemia and IDA (Table 1). Regarding mean age at menarche in Korean girls born between 1980 and 1985 was 13.8,23 more girls aged 14–18 must have undergone menarche than those aged 10–13. This explains that girls with anemia were older than non-anemic girls (Table 1), which is in accordance with previous study.4 As adolescent girls are still under growth, anemic girls were taller and weighted more due to their older age (Table 1). Multiple logistic regression analysis revealed that the group with the lowest income had higher ORs for anemia and IDA compared with the group with the highest income after adjusting confounders including age, BMI, WBC count and red meat consumption. In addition, correlation analysis showed that there is a positive relationship between household income and Hb or ferritin level. These results suggest that household income independently influences anemia and IDA prevalence. This study had the following limitations: first, only girls were analyzed because of the very low prevalence of anemia in boys. This makes it difficult to apply the results directly to the general pediatric population. Second, some biochemical data were not included in the analysis. For example, total iron-binding capacity, which is an important parameter of IDA, was measured in too few adolescent girls to analyze. Third, confounders from previous studies such as maternal education level could not be analyzed as KNHANES did not survey these factors in children. Finally, this was a cross-sectional and therefore prospective study to reveal the & 2014 Macmillan Publishers Limited
relationship between SES and anemia prevalence with careful control of other confounders required. In conclusion, the results suggest that higher SES represented by household income leads to lower prevalence of anemia and IDA in Korean adolescent girls. This may be due to the fact that higher SES individuals consume more iron and vitamin C, which aids the absorption of iron.
CONFLICT OF INTEREST The authors declare no conflict of interest.
AUTHOR CONTRIBUTIONS JY Kim, KC Lee and BJ Ko designed the study; JH Kim, GE Nam, HD Yeo, HG Lee and K Han collected data; S Shin and K Han analyzed the data; JY Kim and BJ Ko wrote the manuscript; YS Choi and DH Kim provided significant advice and contributed to the discussion.
REFERENCES 1 World Health Organization. Iron Deficiency Anaemia: Assessment, Prevention And Control. A guide for programme managers. WHO: Geneva, 2001. 2 McLean E, Cogswell M, Egli I, Wojdyla D. de Benoist B. Worldwide prevalence of anaemia, WHO Vitamin and Mineral Nutrition Information System, 1993-2005. Public Health Nutr 2009; 12: 444–454. 3 Halterman JS, Kaczorowski JM, Aligne CA, Auinger P, Szilagyi PG. Iron deficiency and cognitive achievement among school-aged children and adolescents in the United States. Pediatrics 2001; 107: 1381–1386. 4 Shilpa S, Biradar SPB, Alatagi AC, Wantamutte AS, Malur PR. Prevalence of anaemia among adolescent girls: a one year cross-sectional study. J Clin Diag Res [serial online] 2012; 6: 372–377. 5 Htet MK, Dillon D, Akib A, Utomo B, Fahmida U, Thurnham DI. Microcytic anaemia predominates in adolescent school girls in the delta region of Myanmar. Asia Pac J Clin Nutr 2012; 21: 411–415. 6 Bulliyy G, Mallick G, Sethy GS, Kar SK. Hemoglobin status of non-school going adolescent girls in three districts of Orissa, India. Int J Adolesc Med Health 2007; 19: 395–406. 7 Baral KP, Onta SR. Prevalence of anemia amongst adolescents in Nepal: a community based study in rural and urban areas of Morang District. Nepal Med Coll J 2009; 11: 179–182. 8 Lozoff B, Clark KM, Jing Y, Armony-Sivan R, Angelilli ML, Jacobson SW. Dose-response relationships between iron deficiency with or without anemia and infant social-emotional behavior. J Pediatr 2008; 152: 696–702. 9 Kumar V, Choudhry VP. Iron deficiency and infection. Indian J Pediatr 2010; 77: 789–793. 10 Allen LH. Anemia and iron deficiency: effects on pregnancy outcome. Am J Clin Nutr 2000; 71: 1280S–1284S. 11 Samuelson G, Bratteby LE, Berggren K, Elverby JE, Kempe B. Dietary iron intake and iron status in adolescents. Acta Paediatr 1996; 85: 1033–1038.
European Journal of Clinical Nutrition (2014) 253 – 258
Anemia and socioeconomic status in Korean girls JY Kim et al
258 12 Gompakis N, Economou M, Tsantali C, Kouloulias V, Keramida M, Athanasiou-Metaxa M. The effect of dietary habits and socioeconomic status on the prevalence of iron deficiency in children of northern Greece. Acta Haematol 2007; 117: 200–204. 13 Chaudhary SM, Dhage VR. A study of anemia among adolescent females in the urban area of nagpur. Indian J Community Med 2008; 33: 243–245. 14 Choi HJ, Lee HJ, Jang HB, Park JY, Kang JH, Park KH et al. Effects of maternal education on diet, anemia, and iron deficiency in Korean school-aged children. BMC Public Health 2011; 11: 870. 15 Ministry of Health and Welfare, Korea Centers for Disease Control and Prevention. Korea National Health and Nutrition Examination Survey (KNHANES). Korea Centers for Disease Control and Prevention: Seoul, Korea, 2011. 16 Linseisen J, Kesse E, Slimani N, Bueno-De-Mesquita HB, Ocke MC, Skeie G et al. Meat consumption in the European Prospective Investigation into Cancer and Nutrition (EPIC) cohorts: results from 24-hour dietary recalls. Public Health Nutr 2002; 5: 1243–1258. 17 Expert Scientific Working Group. Summary of a report on assessment of the iron nutritional status of the United States population. Am J Clin Nutr 1985; 42: 1318–1330.
European Journal of Clinical Nutrition (2014) 253 – 258
18 Thankachan P, Muthayya S, Walczyk T, Kurpad AV, Hurrell RF. An analysis of the etiology of anemia and iron deficiency in young women of low socioeconomic status in Bangalore, India. Food Nutr Bull 2007; 28: 328–336. 19 Issaragrisil S, Kaufman DW, Anderson TE, Chansung K, Thamprasit T, Sirijirachai J et al. An association of aplastic anaemia in Thailand with low socioeconomic status. Aplastic Anemia Study Group. Br J Haematol 1995; 91: 80–84. 20 Animasahun BA, Temiye EO, Ogunkunle OO, Izuora AN, Njokanma OF. The influence of socioeconomic status on the hemoglobin level and anthropometry of sickle cell anemia patients in steady state at the Lagos University Teaching Hospital. Niger J Clin Pract 2011; 14: 422–427. 21 Conrad ME, Umbreit JN, Moore EG. Iron absorption and transport. Am J Med Sci 1999; 318: 213–229. 22 Andrews NC. Iron Deficiency and Related Disorders. In: Greer JP, Poerster J, Rodgers GM, Paraskevas F, Glader B, Arber DA et al. (eds) Wintrobe’s Clinical Hematology. 12 edn. (Wolters Kluwer Health/Lippincott Williams & Wilkins: Philadelphia) 2009, pp 810–834. 23 Cho GJ, Park HT, Shin JH, Hur JY, Kim YT, Kim SH et al. Age at menarche in a Korean population: secular trends and influencing factors. Eur J Pediatr 2010; 169: 89–94.
& 2014 Macmillan Publishers Limited
