Anemia in adolescents remains a public health problem in many developing countries, including Indonesia. In Indonesia through 2005, the prevalence of anemia in women of child bearing age was 33.1% and according to World Health Organization (WHO) anemia in Indonesia is a moderate public health problem1. The prevalence of anemia differ between urban and rural areas. Basic public health research showed that in Indonesia in 2013 the prevalence of anemia in women of child bearing age was slightly higher for women living in rural areas compared to those in urban areas (23 vs. 22.4%) but the difference was more significant for men (Rural: 18.5%; Urban: 14.5%)2. Despite various interventions to lower the rate of anemia in adolescents, the persistence of high anemia rates continues to be a major public health concern in Indonesia.
Several studies highlighted that anemia in adolescents is associated with reduced ability to concentrate in a school setting, decreased fitness levels and growth disorders3. An overall decrease in the level of productivity by adolescents with anemia may hamper social and economic development of the country3-4. Moreover, anemia in pregnant adolescents increases the likelihood of low birth weight and premature birth5. Thus, anemia can affect not only the present generation but also several subsequent generations.
An important factor that is significantly related to anemia is the amount of family food expenditure6. Anemia is more prevalent in families that have lower levels of expenditure for food purchasing and these families frequently live in rural areas. A low level of expenditure for food purchasing indicates low purchasing power and inadequate food consumption, which can lead to malnutrition and anemia in adolescents7. However, data concerning how family expenditures relate to anemia are limited. This study aimed to examine whether family expenditure patterns for iron-rich foods as well as differences in food-related expenditures by families living in rural and urban areas is related to the prevalence of anemia in adolescents. This study provides information concerning how expenditures for iron-rich foods made by urban and rural families could guide the development of public health policy to address food insecurity associated with anemia.
MATERIALS AND METHODS
This cross-sectional study used data from the Indonesian Family Life Survey (IFLS) conducted by the RAND Corporation. The data were collected in 2007 for several selected provinces on Sumatera Island. The research population was 860 adolescents aged 15 and 19 years-old who lived in the targeted provinces during the data collection period. For this study, 731 adolescents were included whose complete data for hemoglobin levels were available.
Independent variables that were considered included: individual characteristics (sex, age, nutritional status, BMI for age, living area), socio-demographic characteristics (job status of the head of household),consumption frequency of iron and vitamin C rich foods [meat, green vegetables, fruits (papaya, mangos, bananas), eggs, fish] and expenditures for food purchasing (meat, vegetables, fruits, poultry, fresh fish, preserved fish, preserved meat, tofu and tempeh, eggs and milk). Standardized procedures were applied to measure body weight and height to obtain BMI8. Data for individual and socio-demographic characteristics were collected through interviews using the questionnaire “US1” and “BUKU K”9. Food consumption frequency was assessed through interview using the Food Frequency Questionnaire. Data regarding expenditure for food purchasing were obtained through interviews using the questionnaire “BUKU I”.
Anemia was determined based on hemoglobin levels (Hb) measured by trained personnel using Hemocue. Respondents having Hb <12 g dL1 were declared to have anemia. Nutrition levels were categorized into “normal”, “wasted and severely wasted” and “overweight and obese” based on the Anthropometric Standard of The Ministry of Health of Republic Indonesia10. The age of adolescent was categorized as “middle adolescents” (15-16 years old) and “late adolescents” (17-19 years old). The job status of household head was categorized into “not-working” and “working”. Living area was classified into “urban” and “rural”. Consumption frequency within the previous week was categorized as “not often” (<3 days/week) and “often” (≥3 days/week). Expenditure levels for food were categorized as “low expenditure” (<median) and “high expenditure” (≥median). Family purchasing expenditures encompassed all expenses of families for food purchasing during the week prior to data collection and the values are given in Indonesian Rupiah (IDR).
Univariate analysis was performed for all variables and bivariate analyses concerned relationships between variables and anemia and food expenditures with living areas using chi-square. Associations having a p-value ≤0.05 were considered significant.
Prevalence of anemia: The average Hb levels for adolescents aged 15-19 years-old living in selected provinces of Sumatera were 13.96±1.86 g dL1 (Girls 12.7±1.48 g dL1, Boys 15.1±1.56 g dL1).
||Data distribution of Anemia, BMI and other variable
|*IDR: Indonesian Rupiah, SD: Standard deviation, Min: Minimum, Max: Maximum
The overall prevalence of anemia in these adolescents was 11.7% (Girls 26.1%, Boys 2.4%) (Table 1).
Individual characteristics: Among the individual characteristics considered, the prevalence of anemia was higher for female than male adolescents (OR 14.55 (7.21-29.37; p<0.0005) (Table 2). Although not significantly related, anemia tended to occur most frequently in “middle adolescents” (14.3%) and adolescents who had a nutritional status of “overweight and obese” (21.5%). The Z score for the average BMI of the respondents was -0.43 (-4.38-4.67).
Socio-economic status: A bivariate test did not show a significant relationship between job status of head of households and anemia (p>0.05). The anemia rate was higher for adolescents whose family head had the job status “employed” (12.2%).
Food consumption frequency: Adolescents experiencing anemia consumed meat less frequently than did adolescents without anemia (p<0.05, OR = 2.438 (1.112-5.345)). However, there was no significant relationship between frequencies of green vegetable, fruit, eggs and fish consumption and adolescent anemia.
Expenditures for food purchasing: Levels of expenditures for food purchasing varied depending on food type and area of residence. The median for the highest expenditure was for fresh fish (15,000.00 IDR; range 0-31,765.38), whereas the median for the lowest expenditure was for purchasing of legumes and nuts, meat, poultry, preserved meat and milk (0 IDR). Bivariate analyses showed no significant relationship between levels of expenditure for purchasing of meat, vegetables, fruits, poultry, fresh fish, preserved fish, preserved meat, tofu and tempeh, eggs and milk and the rate of anemia in adolescents in Sumatera (p>0.05).
Living areas: The prevalence of anemia was higher for adolescents living in rural areas compared to those living in urban areas (16.6 vs. 10.4%; p<0.05). Families living in rural areas had lower levels of expenditure for purchasing of meat, poultry, vegetables, fruit, eggs and milk than did families living in urban areas, whereas rural families had higher expenditures for fresh and preserved fish than did urban families (p<0.05; Table 3).
The overall average of Hb levels among adolescents living in Sumatera in 2007 was classified as normal (13.96±1.86 g dL1), as was the average based on sex (Girls: 12.7±1.48 g dL1, Boys: 5.1±1.56 g dL1). The prevalence of anemia in 2007 for Sumatera Island was 11.7% (Girls: 26.1, Boys: 2.4%). Based on WHO1 classification, this overall rate of anemia can be categorized as a mild public health problem, however, for girls anemia is a moderate public health problem. Notably, the prevalence of anemia in adolescents on Sumatera Island did not significantly differ from that on Java Island (Overall: 16.6%, Girls: 26.4%; Boys: 5.9%).
This study found a significant relationship between anemia and sex but not age or nutrition. Girls had a 14.55 times higher risk of anemia, which is due in part to physiological factors such as menstruation5. The prevalence of anemia in middle adolescence (14.3%) and late adolescence (13.5%) was similar and likely reflects the high rate of growth during both periods when demands for nutrients including iron and protein are relatively high.
||Proportion differences of anemia based on food consumption frequency, food purchasing expenditures and other variables
|*Significantly correlated (p<0.05)
||Proportion differences of food purchasing expenditure levels among urban and rural adolescents
Anemia prevalence based on nutritional status did not significantly differ, although anemia was more common in “overweight and obese” adolescents (21.5%) compared to “normal” adolescents (6.4%) and “wasted and severely wasted” adolescents (13.6%). This proportion difference could be due to inflammatory activity of adipose tissue and subsequent reduction in iron usage by the body11.
This study found a significant relationship between living area and anemia12,13. Anemia was more prevalent for adolescents living in rural areas (16.6%) compared to those living in urban areas (10.4%), although both rates constitute a moderate public health problem as defined by WHO1. The higher prevalence of anemia in rural areas is associated with a low frequency of meat consumption in these areas that could be related to lower family income and consequent purchasing power relative to urban families14. Other factors that could be related to the higher proportion of anemia in rural adolescents are lower socio-economic status, limited access to health services and lower levels of knowledge15.
This study considered the relationships between frequency of consumption of iron- and vitamin C-rich foods by adolescents in the week prior to survey data collection. Low rates of meat consumption constituted an anemia risk factor, because meat is a significant source of iron and vitamin B1212. The groups who responded “not often” for meat consumption were 2.4-fold more likely to suffer anemia than those who responded that they “often” consumed meat. In contrast, there was no significant relationship between anemia and consumption of other iron-rich foods, although those study subjects who responded “not often” to questions concerning frequency of green vegetables, fruit, eggs and fish consumption tended to have higher proportion of anemia.
The highest median food expenditure was for fresh fish and vegetables, followed by eggs (Table 1). This result was consistent with survey results obtained for the Expenditure for Consumption of Indonesian Population16, which showed that, after grains, average expenditures were highest for fish, vegetables and eggs, respectively. The average expenditures of respondents for vegetables, fruits, fish, eggs and milk were higher than the national per capita rates for Indonesia in 2007, whereas expenditures for meat slightly differed only and those for nuts were lower16. The lowest median expenditure (Rp 0) was for purchasing of legumes, meat, poultry, preserved meat and milk, although this low value could be because purchasing in only the previous week was considered and thus families may not have consumed those foods in that particular week or were consuming foods that were available from their own harvest.
A significant difference in food expenditures was seen between respondents living in urban and rural areas. Urban families had a higher prevalence of “high expenditure” for purchasing of meat vegetables, fruit, poultry, eggs and milk. This result was in accordance with results of the National Socio Economics Survey of Indonesia conducted in 2017 that was related to the Expenditure for Indonesian Population Consumption study, which showed that urban families had higher food expenditure in the form of animal foods and vegetables17. In developing countries, disparities in expenditures for food purchasing were related to higher food prices and income levels in urban areas18. The lower average expenditure for families in rural areas could be due to consideration of only total money spent on food purchases, since families in rural communities can obtain food from their own harvests. Relative to other food sources, rural families were more likely to respond “high expenditure” for fresh fish and preserved fish since fish is an affordable alternative protein source in rural areas compared to urban areas.
In this study, food purchasing expenditures were not significantly related to anemia occurrence in adolescents, likely because family food preferences are not a significant factor for adolescents19. Another factor that could influence the relationship between anemia and both variables was uneven distribution of iron-rich food among the families.
According to a survey of adolescents living in rural and urban areas of Sumatera Province in Indonesia, the highest household expenditure was for purchasing of fresh fish, whereas the lowest was for meat, poultry, legumes and nuts, preserved meat and milk. Families living in urban areas had higher food expenditures for purchasing of meat, vegetables, fruits, poultry, eggs and milk than did rural families, whereas expenditures for purchasing of fresh and dried fish were higher for rural families. The prevalence of anemia was significantly higher in rural adolescents compared to urban adolescents. However, there was no significant association between expenditure levels for iron-rich foods and anemia. The results of this study indicate the importance of education about iron-rich foods and encourage exploration of alternative sources of protein, to meet the needs for iron-rich foods for rural families.
The authors would like to thank RAND Corporation for authorizing the authors to conduct research using Indonesian Family Life Survey (IFLS) Data 2007.