Healthy infants should receive complementary feeding near the end of the first half-year of life, when breast-feeding or infant formula alone cannot always secure an adequate nutrient supply. Since 2001, the World Health Organization recommendation is to introduce complementary foods from the seventh month of life, rather than in the fifth or sixth month as previously recommended (1,2). Although the World Health Organization recommendation addresses all countries, advisory bodies in industrialised countries continue to recommend an age range for introduction of complementary foods. The European Society of Paediatric Gastroenterology, Hepatology and Nutrition supports exclusive or full breast-feeding for about 6 months as a desirable goal and recommends the introduction of complementary feeding not before 17 weeks and not later than 26 weeks (3). The American Academy of Pediatrics recommends that solid foods should not be introduced before 4 to 6 months of age (4).
In Germany, the recommendation at the time of recruitment is exclusive breast-feeding for the first 4 to 6 months, with the introduction of Beikost from the age of 5 to 7 months (5,6) (the full text of the recommendation is available from the Nationale Stillkommission at http://www.bfr.bund.de/cd/922). In Belgium, recommendations of the L'Office de la Naissance et de l'Enfance, an institution that supports and assesses the well-being of children from 0 to 12 years outside of his or her family, are not to start with the introduction of complementary feeding before the age of 4 completed months (7). In Italy, the Italian Society of Neonatology recommended in 2002 that term healthy babies can continue breast-feeding exclusively for 6 months, whereas introduction of complementary foods can be started at 4 or 5 months depending on maternal and infant circumstances (8). In Poland as well as in Spain recommendations state that complementary feeding should not start before 4 completed months of age (9–11).
Cultural and sociodemographic characteristics of families may influence infant feeding patterns. In some studies earlier introduction of complementary foods were found to be associated with lower socioeconomic status (12–14) and education level (15,16), maternal smoking (16), and younger age of the mother (15,17). In some studies formula-fed (FF) infants started complementary foods earlier than breast-fed (BF) infants (12,13,15,17).
We analysed data of food protocols from 5 European countries with similar infant feeding recommendations, which were collected as part of the prospective European Childhood Obesity Project, with the aim of exploring whether type of milk feeding, country of residence, or other factors were associated with the introduction of solids.
MATERIALS AND METHODS
The data evaluated were collected as part of the European Childhood Obesity Project, a multicentre intervention trial in 5 European countries to investigate the relation of the infant diet and protein supply on early growth and later obesity risk (17a).
Eligible for study participation were apparently healthy, singleton, term infants who were born between October 1, 2002 and July 31, 2004 in Germany (Munich and Nuremberg), Belgium (Brussels and Liège), Italy (Milano), Spain (Tarragona and Reus), and Poland (Warsaw). Mothers were approached by trained study personnel in maternity hospitals before their discharge or contacted via midwives or paediatricians. Infants were enrolled during the first 8 weeks of life. Excluded were mothers with a hormonal or metabolic disease (eg, gestational diabetes) or intake of drugs during pregnancy that are known to influence infant growth (eg, thyroid hormones, antithyroid drugs, corticosteroids). Moving too far away from the study centre to come for visits was a further reason for exclusion.
Dietary Intervention, Study Formulae
Study groups included a BF reference group and 2 groups of study FF infants with exclusive FF starting between birth and week 8, which were randomized to 1 of our study formulae with either higher and lower protein content. Infant formulae had a protein content of 7.1% and 11.7% of energy (1.25 g protein/100 mL and 2.1 g protein/100 mL) and follow-on formulae 8.8% and 17.6% of energy (1.6 g protein/100 mL and 3.2 g protein/100 mL) for the lower and higher protein groups, respectively. Equal energy content was achieved by adapting fat content, whereas the intake of carbohydrates and other nutrients was not different (17a). The study formulae were provided by Bledina (Steenvorde, France) and supplied to families until the infant age of 1 year.
There was no further intervention addressing infant dietary intakes. Information on dietary intakes of infants were collected prospectively with 3-day-weighed food records which included 2 weekdays and 1 weekend day, at the age of each completed month from 1 to 9 months, and at 12 completed months. The 3-day food protocol collected information on all dietary intakes of infants, including breast milk, formula, and any other liquids or foods. The regular study visits and additional contacts by mailings and telephone calls between study personnel and parents allowed to clarify questions arising on filling out the dietary protocol, and to enhance compliance. Parents were provided with a study telephone hotline number for more information and to address any arising questions.
The procedure was as follows: Dietary food records were checked by nutritionists at each study centre; to clarify any questions parents were contacted if necessary; data were introduced to the specifically developed nutrition software “NutrCalc” and transferred to a central database. Standard operating procedures were developed to harmonize the introduction of food intake data and the calculations of dieticians in the different study centres.
Complementary food items were categorised as food groups and according to their ingredients. For our analysis solids included food items such as beef, cereals or bread, egg, fat, fish, fruit, meat, milk or milk products, nuts or seeds, potatoes, poultry, pulses, sausages, soy or soy products, sweets or infant sweets, and vegetables. Each food item was classified either as a commercial infant product (CIP) if it was offered as baby food for the first year of life (excluding infant or follow-on formula), or as normal food (NF) offered for consumption not only by infants but also by children or adults.
We grouped mothers in 4 categories on the basis of age at birth: I: ≤25 years, II: >25 to 30 years, III: >30 to 35 years, and IV: >35 years. Maternal educational levels were categorised in 3 groups: no/low = prepreliminary to lower secondary, middle = upper secondary and postsecondary nontertiary, and high = first and second stage of tertiary education according to the standards of the International Standard Classification of Education.
For the formula group 1090 infants were enrolled, of whom 851 participated in the follow-up visits at 6 months and 767 at 12 months; for the observational group of BF infants 588 infants were enrolled, of whom 349 participated in the follow-up visits at 6 months, 327 at 12 months. Thus, 687 infants dropped out until the age of 12 months for the following reasons: refusal of parents/loss of contact (310 [45%]), no compliance with assigned milk feeding (250 [36%]), address unknown (51 [7%]), moving out of the study region (27 [4%]), exclusion of illness/medication (12 [2%]), and other reason/unknown (37 [5%]).
Stata 9.2, statistical package for statistical science software, version 16.0 (SPSS Inc, Chicago, IL), and Excel were used for data analyses. χ 2 and multiple logistic regression analysis was used to adjust differences in solid introduction at each month by confounders.
The study was approved by the ethics committees of all of the study centres. Written informed parental consent was obtained for each infant.
A total of 1678 infants were enrolled (1090 FF infants and 588 BF infants, 65% and 35%, respectively) in the study, and for 1366 (81.4%) infants (928 FF infants and 438 BF infants, 68% and 32%, respectively) at least 1 iterative 3 day-food protocol could be evaluated (Table 1). At the age of 1 month some 1045 three-day-food protocols were evaluated, at 2 months 1208, at 3 months 1184, at 4 months 1136, at 5 months 1128, at 6 months 1075, at 7 months 1008, at 8 months 984, at 9 months 967, and at 12 months 945 three-day-food protocols.
The sociodemographic characteristics of study participants are shown in Table 1. Except for country, we did not find differences in sociodemographic characteristics of those infants with food protocols and those without protocols. However, the proportion of filled-in protocols varied between countries (P < 0.001), with a lower participation rate in Belgium and a higher participation rate in Italy. Because the sociodemographic characteristics of our FF and BF infants differed in many aspects, we separated both groups in the analysis.
Solids had been introduced in 6% of FF infants at the age of 3 completed months, in 37.2% at 4, 96.2% at 6, and 99.3% at 7 completed months, respectively (Fig. 1). The median age of introduction was 19 weeks (interquartile range 17–21). Among the BF infants only 0.6% had received solids at the age of 3 completed months, whereas 17.3%, 87.1%, and 97.7% received solids at the ages of 4, 6, and 7 months, respectively, with a median age at introduction of 21 weeks (interquartile range 19–24) (Fig. 1).
A higher proportion of FF infants received CIPs than NFs during the first year of life. In contrast, fewer BF infants consumed CIP than NF (Table 2).
There were significant differences between countries at the time point of introduction of solids during the first 8 months of life (Table 3). We found an early introduction of solids in Belgium, with 15.8% and 55.6% at the ages of 3 and 4 completed months, respectively. At the age of 3 completed months, Italy and Poland had the lowest proportion of FF infants receiving solids, with only 2.4% and 3.1%, respectively. In all countries at the age of 6 completed months >90% of FF infants received some solids.
Younger mothers (younger than or equal to 25 years) of FF infants introduced solids significantly earlier during the first 4 months of life to their infant's diet (at 1, 3, and 4 completed months: P < 0.05). Logistic regression showed a 2.9-fold higher probability that younger mothers (younger than or equal to 25 years) start to introduce solids at the age of 3 months than mothers older than 35 years. A low maternal education had a 1.8 higher risk and maternal smoking 3 months before and during early pregnancy had a 1.5 higher risk to introduce some solids at the age of 4 months.
Applying multiple regression analysis including the effects of maternal age, education level, smoking behaviour, and country of residence on the introduction of solids in FF infants, we found a 4-fold higher probability for introduction of solids at the age of 3 completed months in Belgian FF infants (odds ratio [OR] 3.96), and a 3-fold higher risk (OR 2.86) in the youngest group of mothers (younger than or equal to 25 years).
The probability of having solids already introduced at the age of 4 completed months was 3-fold higher in Belgium (OR 3.28), 2-fold higher for low maternal educational level (OR 1.87), and 1.4-fold higher for mothers who smoked (OR 1.40) (Table 4).
The different protein and fat levels of the study formulae did not influence the time of introduction of solids. Neither were the birth weight, birth order, or sex of the FF infants related to the time point of introduction of any solid.
Between the age of 4 and 6 completed months, introduction of solids differed significantly by country. Up to 3 completed months, solids were not given to >1.4% of BF infants in any country. At the age of 4 and 5 completed months Belgium had the highest proportion (43.2% and 84.8%) and Germany (4.9% and 25.0%) and Poland (6.7% and 36.2%) the lowest proportion of BF infants receiving solids.
At 6 completed months of age at least 84% of BF infants in every country received some solids except in Germany (69.5%), and at 7 completed months >93% of all BF infants in each country had been introduced to solids (Table 5).
Low maternal education level was significantly associated with the introduction of solids at 3 and 4 completed months (3 months: P < 0.001, 4 months: P < 0.05) in univariate analysis, whereas multiple regression analysis showed an effect only at 4 months. Maternal age and smoking behaviour was unrelated to timing of introduction of solids.
Applying multiple regression analysis, including the effects of maternal age, education level, smoking behaviour, and country of residence, receiving solids at 4 completed months was 16 times more likely (OR 15.93) in Belgium and 7 times more likely in Spain (OR 6.60) compared with German BF infants. Low- and middle-level maternal education had a 3-fold higher association (OR 3.13 and 2.68, respectively) with an earlier introduction of solids compared with a high educational level (Table 4). Birth weight, birth order, and sex were unrelated to the time of introducing solids in BF infants.
In this sample some 37.2% of FF infants and 17.2% of BF infants already received solids at the age of 4 completed months, although the European Society of Paediatric Gastroenterology, Hepatology, and Nutrition and national recommendations in all participating countries advise not to introduce complementary foods before 4 months of age. Some 6.0% of FF infants, but only 0.6% of the BF infants, had introduced solids already at the age of 3 completed months. Although introduction of solids is recommended at the same age for FF and BF infants, FF infants received solids much earlier than BF infants, which is consistent with previous findings (13,15,17). Our findings show that higher parental socioeconomic status and educational level, as well as exclusive breast-feeding during the first months of life, are associated with later introduction of complementary foods. There are only a few infants in whom solids are introduced later than recommended; 0.7% of FF infants and 2.3% of BF infants, respectively.
Our study involved 5 European countries with different cultural traditions and food patterns. Even though guidelines for the introduction of complementary foods are similar in these countries, there are significant differences in infant feeding practice between the countries, both in FF and BF infants (Tables 3 and 5). We found much higher percentages of FF and BF infants with intake of solids at the age of 4 months in Belgium (Tables 3 and 5). This earlier introduction, relative to other countries, is not because of different recommendations in Belgium and remains unexplained.
Giovannini et al (18) studied infant feeding practices in Italy through the first year of life and found that 5.6% and 34.2% of the infants had introduced solids before the age of 3 and 4 months, respectively. Stated factors explaining an early introduction of solid foods were the infants' body weight at 1 month of age and maternal smoking during pregnancy. We found similar proportions for our Italian FF infants (2.4% and 30.5%, Table 3) although they had the lowest proportions of FF infants in our study with introduction of solids in the first 3 months of life.
German FF and BF infants had the lowest percentage in intake of solids at the age of 4 to 6 months. The proportion of infants at the age of 3 completed (FF infants 5% and BF infants 0.0%) and 6 completed months (FF infants 91.2% and BF infants 69.5%) were lower compared with data from an earlier prospective study in German infants by Koletzko et al (19) with 16% and 97% of German infants consuming some Beikost (solid and liquid complementary food) at 3 and 6 months, respectively, whereas the German Dortmund Nutritional and Anthropometric Longitudinally Designed study reported 5% and 97% of the BF and 33% and 80% of the non-BF infants received solids at the ages of 3 and 6 months, respectively (20).
Other studies also reported discrepancies between feeding recommendations and practice. In the Euro-Growth Study, 50%, 67%, and 95% of infants were fed some solid food at the ages of 3, 4, and 5 months, respectively (21). In the United Kingdom the Department of Health recommendations suggests starting with complementary food between 4 to 6 months of age. Similarly, the Scientific Advisory Committee on Nutrition commented that complementary food should not be introduced before the end of the fourth month (22). However, many infants receive complementary feeding earlier (13,21,23,24).
In this study, the strongest risk factors for early introduction of solids in FF infants at 3 completed months were country of residence and young maternal age, and at 4 months the country of residence, low maternal education, and maternal smoking. In BF infants, the country of residence and lower maternal education level were associated with introduction of solids at the age of 4 completed months. These findings are consistent with other studies also finding earlier introduction of complementary feeding in children of lower parental educational level (25), lower socioeconomic status (14), maternal smoking (16,26), and younger maternal age (12).
Although we found no differences in the timing of introducing complementary foods between our infants from the intervention group-fed formulae with different protein and fat contents, there were significant differences between countries in the timing of introducing complementary foods. This observation suggests far stronger effects of cultural, social, and parental factors on the time of introducing complementary foods rather than of the dietary macronutrient composition.
More NFs than CIPs were given to BF infants during the first year of life, whereas a higher percentage of FF infants consumed CIP than NF. It seems that breast-feeding mothers prefer NFs to commercial prepared or semiprepared products; however, mothers of infants who feed formula and, hence, are already using a commercial product may tend to have a lower threshold for introducing other commercial products into the infants' diet.
Infants given formula milk started solids significantly earlier than BF infants. There are marked differences between the 5 European countries of our study in the timing of introducing solids. The macronutrient composition of our formula milk groups did not influence the time point of introduction of complementary foods. Only a few infants start complementary feeding later than recommended. Given the increasing evidence that early nutrition and growth has marked effects both on short-term and long-term health, further studies should evaluate strategies to improve complementary feeding practices in populations of healthy infants.
We are extremely grateful to the parents and their infants who participated in this study, and to the physicians, midwives, and nurses for their helpful support in recruiting mothers during pregnancy or after giving birth. We want to thank the CHOP team for their effort and dedication, the dieticians for the patience to complete the nutritional data, and Fabio Confalionieri for support in providing the NutrCalc software.
1. WHO 54th World Health Assembly. Infant and Young Child Nutrition
. WHA54.2, Agenda item 13.1. 2001.
2. WHO. Complementary Feeding of Young Children in Developing Countries: A Review of Current Scientific Knowledge
. Geneva: WHO 1998;WHO/NUT/98.1 [Abstract].
3. Agostoni C, Decsi T, Fewtrell M, et al. Complementary feeding: a commentary by the ESPGHAN Committee on Nutrition. J Pediatr Gastroenterol Nutr
4. Greer FR, Sicherer SH, Burks AW, Committee on Nutrition and Section on Allergy and Immunology. Effects of early nutritional interventions on the development of atopic disease in infants and children: the role of maternal dietary restriction, breastfeeding, timing of introduction of complementary foods, and hydrolyzed formulas. Pediatrics 2008; 121:183–191.
5. Kersting M. Ernährung des gesunden Säuglings. Monatsschrift Kinderheilkunde 2001; 149:4–10.
6. Wirth S, Böhles H, Höger P, et al., eds. Leitlinien Kinder- und Jugendmediozin der Deutschen Gesellschaft für Kinder- und Jugendmeidizin
. München: Urban & Fischer; 2007.
8. Società Italiana di Neonatologia. Raccomandazioni sull'allattamento materno per i nati a termine, di peso appropriato, sani. Medico e Bambino
9. Ksiazyk J, Rudzka-Kantoch Z, Weker H. Feeding plan for breast-fed infants and non-breast-fed infants. Medycyna Praktyczna Pediatria 2001; 5:1–2.
10. Ksiazyk J, Rudzka-Kantoch Z, Weker H. Zalecenia zywienia niemowlat. Standardy Medyczne
12. Savage SA, Reilly JJ, Edwards CA, et al
. Weaning practice in the Glasgow longitudinal infant growth study. Arch Dis Child 1998; 79:153–156.
13. Wright CM, Parkinson KN, Drewett RF. Why are babies weaned early? Data from a prospective population based cohort study. Arch Dis Child 2004; 89:813–816.
14. Alder EM, Williams FL, Anderson AS, et al
. What influence the timing of the introduction of solid food to infants? Br J Nutr 2004; 92:527–531.
15. Lande B, Andersen LF, Baerug A, et al
. Infant feeding practices and associated factors in the first six months of life: the Norwegian infant nutrition survey. Acta Paediatr 2003; 92:152–161.
16. Ford RP, Schluter PJ, Mitchell EA. Factors associated with the age of introduction of solids into the diet of New Zealand infants. New Zealnd Cot Death Study Group. J Paediatr Child Health 1995; 31:469–472.
17. van den Boom SAM, Kimber AC, Morgan JB. Weaning practices in children up to 19 months of age in Madrid. Acta Paediatr 1995; 84:853–858.
17a. Koletzko B, von Kries R, Closa Monasterolo R, et al
. Lower protein in infant formula is associated with lower weight up to age two years: a randomized clinical trial. European Childhood Obesity Trial Study Group. Am J Clin Nutr 2009; 89:1836–1845.
18. Giovannini M, Riva E, Banderali G, et al
. Feeding practices of infants through the first year of life in Italy. Acta Paediatr 2004; 93:492–497.
19. Koletzko B, Dokoupil K, Reitmayr S, et al
. Dietary fat intakes in infants and primary school children in Germany. Am J Clin Nutr 2000; 72:1392S–1398.
20. Hilbig A, Kersting M. Effects of age and time on energy and macronutrient intake in German infants and young children: results of the DONALD study. J Pediatr Gastroenterol Nutr 2006; 43:518–524.
21. Freeman V, Van't Hof M, Haschke F. Patterns of milk and food intake in infants from birth to age 36 months: the Euro-Growth Study. J Pediatr Gastroenterol Nutr 2000; 31(Suppl 1):S76–S85.
22. Foote KD, Marriott LD. Weaning of infants. Arch Dis Child 2003; 88:488–492.
23. Anderson AS, Guthrie CA, Alder EM, et al
. Rattling the plate–reasons and rationales for early weaning. Health Educ Res 2001; 16:471–479.
24. Reilly JJ, Wells JCK. Duration of exclusive breast-feeding: introduction of complementary feeding may be necessary before 6 months of age. Br J Nutr
25. Hendricks K, Briefel R, Novak T, et al
. Maternal and Child Characteristics Associated with Infant and Toddler Feeding Practices. J Am Diet Assoc 2006; 106:135–148.
STUDY TEAM MEMBERS
Belgium (ULB Bruxelles and CHC St Vincent Liège): Carlier C, Goyens P, Hoyos J, Langhendries J-P, Martin F, Van Hees J-N, Xhonneux A. Germany (Division of Nutritional Medicine and Metabolism, Dr von Hauner Children's Hospital, and Division of Pediatric Epidemiology, Institute of Social Pediatrics and Adolescent Medicine, Ludwig-Maximilians-University of Munich): Beyer J, Demmelmair H, Fritsch M, Handel U, Hannibal I, Kreichauf S, von Kries R, Pawellek I, Verwied-Jorky S. Italy (University of Milan): Giovannini M, Agostoni C, Confalonieri F, Scaglioni S, Tedeschi S, Vecchi F, Verduci E. Spain (Universitat Rovira i Virgili): Closa Monasterolo R, Escribano Subias J, Méndez Riera G, Luque Moreno V. Poland (Children's Memorial Health Institute): Socha J, Dobrzańska A, Gruszfeld D, Socha P, Stolarczyk A, Janas R, Pietraszek E, Kowalik A.
26. Ratner PA, Johnson JL, Bottorff JL. Smoking relapse and early weaning among postpartum women: is there an association? Birth 1999; 26:76–82.