Complementary feeding of children in the second year of life Alimentação complementar em crianças no segundo ano de vida Alimentación complementar en niños en el segundo año de vida

Objective: To study the eating habits of children in their second year of life, comparing these habits in children that received complementary breastfed to the ones who weaned before 12 months of life. Methods: Cross-sectional study involving children aged 12 to 24 months in a primary health care service in Belo Horizonte, in the state of Minas Gerais, Southeast Brazil. Mothers were interviewed about the feeding practices of their children. The eating habits of complementary breastfed children were compared to those who weaned before 12 months of life by the chi-square, Fisher, Student’s t, and Kruskal-Wallis tests, with a significance level of 5%. Results: 118 children were included with a mean age of 16.8±4.0 months. About 35% of them were still breastfed and only 15.3% kept exclusive breastfeeding for six months. In breastfed children the median duration of exclusive breastfeeding was four months and, for the weaned ones, two months (p=0.13). In both groups there was early introduction of complementary food, high intake of processed food, high daily consumption of oils and fats (90.7%), and low consumption of fruits (38.1%). Conclusions: The results point out the existence of inadequate feeding practices in infants regardless of the recommended intake of breast milk, therefore indicating the need for improvement and integration of actions to promote breastfeeding and healthy diets in primary attention services. Key-words: breastfeeding; supplementary feeding; weaning; food habits; infant nutrition.


Introduction
Exclusive breastfeeding (EB) in the first 6 months of life and continuation of breastfeeding with complementary foods (CMB) for 2 years or more are recommended practices by the World Health organization (WHO) and by the Brazilian Ministry of Health (MH) (1,2) .Nevertheless, in Brazil, studies identified inadequacies, such as low prevalence of exclusive breastfeeding (EBF) and early introduction of food in infants' diets (3)(4)(5)(6) .
The National Demographic Health Survey (NDHS), conducted in 2006, found that 62% of breastfed children in Brazil already consumed complementary foods in the 4th and 5th months of life (6) .Data from the Second Survey on Prevalence of Breastfeeding in Brazilian capitals and the federal District (2009) showed introduction of water, teas and other milks in the first month of life in 13.8, 15.3 and 17.8% of children, respectively (3) .It was also verified that 25% of infants between 3 and 6 months old consumed salty food and fruits, showing that, despite the progress made since the implementation of the National Policy to Encourage breastfeeding, Brazil is below the goals proposed by the WHO and the MH (3) .
It is noteworthy that the early introduction of foods may be associated with increased infant morbidity and mortality, due to risk of inadequate food hygiene, occurrence of allergic reactions, decreased absorption of nutrients and lack of intake of the protective factors present in breast milk (BM) (1) .On the other hand, the late introduction of the complementary foods is also unfavorable due to the possibility of reducing infant growth and, therefore, increasing malnutrition and micronutrient deficiencies, particularly from iron, zinc and vitamin A (2,7,8) .Despite the importance of complementary feeding and food supply at an appropriate age, studies on the process of introduction of foods and dietary practices in infants are scarce, compared to the number of publications on EBF (9,10) .This study aimed to study the eating habits of children in the 2nd year of life, in Belo Horizonte (MG), comparing those receiving CBF with those who had been weaned before 12 months of life (WBF), besides investigating the association of BM consumption for longer than 1 year with sociodemographic, economic, behavior, and health care variables.

Method
The present study was a cross-sectional analysis involving children from 12 to 24 months old, living in the area of a Basic Health Unit (BHU) in Belo Horizonte, state of Minas Gerais, who present medium to high risk of illness and death, according to the Health Vulnerability Risk (HVR), developed by the Municipality of Belo Horizonte (11) .
Children aged 12-24 months were considered eligible for the study.Based on the register of community health agents (CHA), it was estimated that there were approximately 200 children in this age group in the beginning of 2009.Mothers were invited to participate in the study through home visits and interviews at the BHU from February to April 2009.We evaluated 118 children, representing 59% of all children identified.Sixteen mothers refused to participate and the remaining could not be contacted after the printed invitations, home visits, and telephone contacts.
Interviews were performed by students attending their final year in the Nutrition graduate program of Universidade Federal de Minas Gerais (UFMG), through a structured questionnaire previously tested in a pilot study.The instrument consisted on 73 questions, addressing sociodemographic and economic aspects of the mother (age, education, occupation, participation in stable union (marital status), and per capita income) and of the infant (sex and age), obstetric history (type of delivery, parity and number of children), use of health services for health control, and number of visits in the last year, as well as questions on the access of information about infant feeding, breastfeeding (BF) patterns, frequency of consumption of foods, and about the process of introduction of complementary foods.
The patterns of BF were assessed according to criteria proposed by the WHO (1) .The length of EBF was regarded as the period in which the children received only breast milk (BM) without any intake of other liquid or solid, except for vitamins and minerals and/or medicine.CBF was considered as the period in which children received BM associated with any kind of supplement: liquid, semi-solid, solid, or other milks (1) .
Dietary intake was assessed with a qualitative food frequency questionnaire, designed specifically for the study, in which the mother had to report the frequency of the infant's consumption by food groups: "breads, cereals and tubers"; "fruits and vegetables"; "milk and dairy products"; "meat, poultry, and eggs"; "legume"; "oils and fats"; and "sugar and sweets".
To evaluate the process of introduction of complementary foods we used information regarding the age of introduction of foods such as porridge vegetables, juices, fruits, meats, eggs, snacks, processed foods, teas, milks, and sweets.The obtained data were evaluated according to the Brazilian Ministry of Health (2) .
Data were analyzed with SPSS version 12.0 (12) .We conducted a descriptive analysis and applied the chi-square or Fisher exact, Student t and Kruskal-Wallis tests, considering 5% as level of significance (13) .
Children were distributed in two groups: Those in CBF and those not receiving maternal milk (WBF).We tested the association between consumption of BM for over 1 year with sociodemographic, economic, behavioral, and health care variables.
The study was approved by the Research Ethics Committee of the Municipal Secretary of Health of Belo Horizonte (CAE 0010.0.410.203-09A).Mothers received information on the objectives and procedures of the study and signed an informed consent, as required by resolution 196 of the National Health Institute.

Results
The 118 children had, in average, 16.8±4.0months of age, 50.8% were female, and 44% were only children (Table 1).Mothers had, on average, 28±6.4 years, and 6% were teenagers; 68% had studied for more than 8 years, and less than half worked outside home at the time.The per capita income of 51% of families was below the minimum wage/month (Table 1).Most children were monitored routinely at the BHU and almost 85% of them went through seven or more consultations in the last 12 months preceding the interview (Table1).When asked about routine consultations at the BHU, it was verified that 90.7% of mothers received guidance about breastfeeding, introduction of foods, and supplementary feeding during control visits; and 89% reported having had the opportunity to answer their questions during these visits.
About 35% of children still received BM, corresponding to the group of children in CBF.Among these, 36.6% received BM on demand and 63.4% received it at pre-determined times.The median duration of breastfeeding in the case of children already weaned was of 7 months.
Only 15.3% of children kept EBF until 6 months of age.In children on CBF, the median duration of EBF was 4 months and in those weaned, 2 months (p=0.13,Kruskal-Wallis).
Table 2 presents minimal and maximum (range) ages and the median age at which foods were introduced comparing the CBF group and the WBF group.There was statistical difference between the medians of age of introduction of water (p=0.032),teas (p=0.046),natural juices (p=0.002) and other milks (p=0.007), the latter introduced later in the group under CBF, when compared to the WBF group.Considering the general behavior of both groups, practically  all kinds of foods studied (except for coffee and snacks such as chips) were offered at least to one child before six months of age.The study highlights the frequent introduction of foods like soft drinks, snacks and sweets before 12 months of age.
Table 3 shows the frequency of consumption from the food groups by the children assessed.It appears that the food groups with the highest daily consumption were the "bread, cereals and tubers" (95.8%), followed by the groups of "oils and fats" (90.7%), and "dairy products" (86.4%), which contrasts with the low daily intake of "fruits" (38.1%), "meat and poultry", (55.9%) and "vegetables" (69.5%).Furthermore, 44% of children consumed "eggs" only occasionally, 25.4% had meat once or fewer times per day (5 to 7 times a week) and 26.3% had fruit only 1 to 4 times a week.
The analysis of the association between explanatory variables and the children's feeding patterns is presented in Table 4.There was no statistically significant difference between the groups regarding the number of offspring, education, maternal working condition and/or education, and per capita income.No association between the CBF with the reaction of the child to the introduction of new foods was demonstrated (p=0.710).It was found that the group of breastfed children had higher frequency of daily consumption of legumes than those weaned (p=0.03).The consumption of other kinds of milk was higher among weaned children (p=0.01).Both groups of children showed similar results regarding the consumption of other types of food.

Discussion
The present study identified an inappropriate dietary pattern in children, with short duration of EBF, early introduction of complementary foods and consumption of an unbalanced diet.This occurred despite the mothers having received counseling from health professionals on various aspects of infant feeding and their children having been assisted regularly during the first year of life.Studies conducted in São Paulo with children younger than 2 years found similar results with low duration of EBF and consumption of inadequate foods for the age (14,15) .Other authors, in a research conducted with children from 24 to 72 months of age, in Minas Gerais, also showed that the most consumed food groups were cereals, legumes, dairy products, sweets and fats (32) .It was also observed that most children do not have food intake as recommended by the Ministry of Health (2) .The low frequency of egg consumption may be related to food taboos that it would be inappropriate for consumption in childhood, rising blood cholesterol levels.The report of consumption of sweets, snacks, fried foods, and industrialized foods was common, even before the first year of life, which is a diet of low nutritional value.We emphasize the need to discourage the consumption of these kinds of food and encourage the consumption of fruits and vegetables.
Although a large difference in food groups consumption was not observed between breastfed children and the others, it is possible that those mothers still breastfeeding are better informed or aware of the importance of certain foods in the diet of their children, such as legumes, regarded as important sources of vitamins and minerals.
This study draws the attention to the importance of interventions aimed at improving the nutrition of infants.Santos et al conducted an educational intervention based on the training of professionals for advice on feeding according to the guidelines of the Integrated Management for Childhood Illnesses (IMCI), and observed quantitative and qualitative improvement of feeding practices of mothers of malnourished children from 2 months to 5 years old.The training also increased the knowledge of trained professionals who became more skilled in making infant feeding counseling (33) .Caldeira et al showed that the training of Family Health Teams was effective in increasing the prevalence of EBF (34) .
This study had as a limitation the fact that it was performed within only one Basic Health Unit in Belo Horizonte and assessed only 59% of eligible children.It is noteworthy, therefore, the need for prospective studies with larger numbers of participants, addressing perceptions, feelings, and conceptions of mothers to better understand the process of introduction of complementary foods to breastfeeding in our area.The results do not exhaust the subject, but they fill gaps and lead to the reorganization of actions to encourage the practice of BF and strategies to promote healthy eating at the appropriate time, incorporating, thus, the guidelines from the National Policy for Food and Nutrition (34) , at all levels of health care and within the community.
It can be concluded that the practice of breastfeeding for more than 12 months was not accompanied by satisfactory complementary feeding when the process of introducing these foods and the nutritional quality of the diet of these children was analyzed.These findings suggest that mothers are not sufficiently instructed yet regarding the offer of complementary foods or have not yet incorporated important concepts for the practice of a healthy diet for their children.
Health care professionals should be aware of this fact and be able to adapt promotion actions to the socio-demographic and cultural context of the population assisted in order to provide mothers with opportunities to acquire knowledge and skills on infant feeding.It is hoped, therefore, that this study can contribute to the reorientation of the actions of nutrition education in health services, besides strengthen families in positive attitudes and habits to ensure a healthier infant feeding.
It should also be emphasized the need for improvement and integration of actions that promote breastfeeding and healthy eating in the services of primary health care, in order to prevent nutritional abnormalities, such as micronutrient deficiency, malnutrition, and obesity.

Table 1 -
Sociodemographic characteristics of mothers and children from 12 to 24 months at a primary healthcare service, BeloHorizonte, 2009 *Median values±standard deviation

Table 2 -
Comparative analysis of median age of introduction of complementary foods in children from 12 to 24 months of age according to consumption of breast milk, at a primary healthcare service, BeloHorizonte, 2009 *Range of age of introduction of foods (in months); CBF: consumption of breast milk

Table 3 -
Frequency of consumption of food groups among chilren aged 12 to 24 months at a primary healthcare service, Belo