Share this article on:

Breast-Feeding and Weaning Practices in the DONALD Study: Age and Time Trends

Foterek, Kristina*; Hilbig, Annett; Alexy, Ute*

Journal of Pediatric Gastroenterology and Nutrition: March 2014 - Volume 58 - Issue 3 - p 361–367
doi: 10.1097/MPG.0000000000000202
Original Articles: Hepatology and Nutrition

Objectives: Besides influencing short- and long-term health status, infant feeding practices are known to have an effect on later food preferences. This study aimed to identify present trends in breast-feeding duration and weaning practices with special focus on preparation methods of complementary food (CF), that is, homemade and commercial CF.

Methods: In total, 1419 three-day weighed diet records collected between 2004 and 2012 from 366 children of the German DOrtmund Nutritional and Anthropometric Longitudinally Designed study ages 6 to 24 months were analysed. Full (n = 339) and total breast-feeding duration (n = 344) was collected by questionnaire. To investigate age and time trends, logistic regression and polynomial mixed regression models were used.

Results: Infants born between 2008 and 2012 were 3.3-fold less likely to be fully breast-fed for ≥4 months than those born before 2004 (P < 0.0001). Overall, 59.3% commercial, 21.1% homemade, and 19.6% combined CF was consumed by the study sample. Subjects with high commercial CF consumption (percentage of commercial CF > median 62%) were significantly older (P < 0.0001), showed shorter full and total breast-feeding duration (P < 0.0001), and were more likely to have mothers with a lower educational status (P = 0.01). Both commercial and homemade CF showed opposing, nonlinear age trends. No time trends could be found.

Conclusions: Decreasing duration of full breast-feeding should encourage health care providers to further promote longer breast-feeding duration. With the constantly high consumption of commercial CF at all ages, nutritional adequacy of both homemade and commercial CF needs to be investigated closer, as does their long-term influence on health and dietary habits, for example, fruit and vegetable intake.

*IEL-Nutritional Epidemiology, University of Bonn

Research Institute of Child Nutrition (FKE), Dortmund, Germany.

Address correspondence and reprint requests to Kristina Foterek, IEL-Nutritional Epidemiology, University of Bonn, Heinstueck 11, 44225 Dortmund, Germany (e-mail:

Received 5 October, 2013

Accepted 5 October, 2013

The DONALD Study is supported by the Ministry of Science and Research of North Rhine Westphalia, Germany.

The authors report no conflicts of interest.

Infant feeding practices not only affect short- and long-term health status (1,2) but are also known to influence later taste and food preferences (3). Hence, infants’ early food environment plays an important role in the development of their individual eating behaviour (4).

Breast-feeding is one of the first opportunities for an early exposure to different flavours and can hence affect later food preferences in childhood (5,6). Undoubtedly, breast-feeding is also associated with several health benefits in infancy and adulthood (1,7). Therefore, exclusive breast-feeding (ie, breast milk only with no other food or drink) or full breast-feeding (ie, breast milk in combination with water or water-based drinks) for 4 to 6 months is recommended in Europe (8); however, breast-feeding patterns have been changing during the last few decades, for example, breast-feeding initiation rates have increased over time, presently reaching approximately 80% (9) to 90% (10,11).

In the representative cross-sectional German Health Interview and Examination Survey for Children and Adolescents (KiGGS), breast-feeding duration was questioned retrospectively. The average duration of full breast-feeding for infants born between 1986 and 2005 was 4.6 months and therefore in line with present recommendations; however, both full and total breast-feeding duration decreased in KiGGS from 2001 onwards (9). Overall, in comparison with other European countries, German breast-feeding rates are ranked somewhere in the middle (12).

In accordance with the recommendations on breast-feeding, complementary food (CF)—in Europe, usually given as semisolid, spoonable foods—should be introduced not more than 6 months of age and not less than 4 months (13). With the introduction of CF, the exposure to different flavours continues (3), although infants’ flavour experiences vary with respect to the way in which the CF is prepared. The taste of jarred commercial CF differs from homemade CF because of the different cooking process during manufacturing.

Recent studies suggest that the method of preparation of CF is linked to diet quality and later eating behaviour (14,15); however, results are inconsistent: in a cross-sectional survey of families receiving governmental benefits, those infants fed commercial CF consumed a larger variety of fruits and vegetables (14). Within the prospective Avon Longitudinal Study of Parents and Children, participants who were fed home-cooked fruits and vegetables at 6 months were more likely to show an increased intake and variety of fruits and vegetables at 7 years (15). An association between consumption of commercial CF in infancy and lower intelligence quotient scores in childhood has also been discussed (16).

Although the American Academy of Pediatrics only recommends ensuring nutrient and energy sufficiency when preparing CF at home (17), UK infant feeding guidelines promote the use of homemade rather than commercial CF (18). There are no such recommendations in Germany; the German Research Institute of Child Nutrition (FKE) has stated that given their respective advantages and disadvantages, both homemade and commercial CF are equally acceptable alternatives for infant feeding (19). There is presently insufficient scientific evidence to make a more specific recommendation.

Extremely few studies exist on the consumption of commercial and homemade CF in industrialised countries. In Germany, the quantity of commercial CF products on sale increased between 2010 and 2012 from 276 to 309 jarred vegetable-potato-meat meals (Mesch et al, unpublished data). Moreover, there was a slight increase in sales, by 0.4%, for all commercial CF between 2010 and 2011 (20).

To determine existing trends in infant feeding practices, it is important to monitor these practices during this sensitive time frame. The aim of our study, therefore, was to identify present breast-feeding and weaning practices in German infants, with a special focus on the method of preparation of CF.

Back to Top | Article Outline


Study Sample

The present analysis is based on data from the German DOrtmund Nutritional and Anthropometric Longitudinally Designed (DONALD) study, an ongoing open cohort study. Details of the study protocol have been published previously (21). Briefly, since 1985, detailed information concerning diet, growth, development, and metabolism in healthy subjects between infancy and early adulthood has been collected. Every year, infants are newly recruited and first examined at the age of 3 months. Each child returns for up to 3 more visits in the first year, 2 in the second, and then annually until young adulthood. The noninvasive assessments include 3-day weighed diet records, interviews on lifestyle, anthropometric measurements, and a medical examination. The study was approved by the ethics committee of the University of Bonn, and all of the examinations were performed with parental written consent.

For the purpose of this analysis, we considered data from 366 infants (194 boys, 172 girls) ages 6 to 24 months collected between 2004 and 2012. Each infant had to have completed at least one 3-day diet record. Subsamples were used for specific analyses depending on individual data availability.

Back to Top | Article Outline

Diet Records

All of the foods and beverages consumed by the infant were weighed and recorded by the parents using electronic food scales (±1 g) on 3 consecutive days. Breast milk intakes were determined by weighing the infant before and after each breast-feeding. Recipes of homemade CF as well as a detailed description including type and brand of commercial CF were also requested. If weighing was not possible, semiquantitative recording (eg, number of glasses, cups, spoons) was allowed. At the end of the 3-day recording period, a dietician visited the family and checked the record for completeness and accuracy.

Any food or drink consumed by the DONALD study participants was entered into the continuously updated in-house food composition database LEBTAB, which incorporates information from standard nutrient tables, product labels, or recipe simulation based on the labelled ingredients and nutrients (22). Dietary supplements and pharmaceuticals were recorded, but excluded from this analysis.

For this analysis, CF was defined as all semisolid, pureed, or mashed foods fed with a spoon during weaning. Snack foods intended for infants and toddlers (eg, biscuits, cereal bars) were not included in the definition of CF because they usually have a solid texture. Commercial CF was defined as all industrially processed, prepackaged CF (from jars or packets). Homemade CF was defined as all home-prepared, semisolid, pureed, or mashed foods made from fresh ingredients. Combined CF was defined as homemade CF that was mixed with commercial CF products in a single meal, for example, homemade vegetable puree in combination with commercial jarred baby meat.

Back to Top | Article Outline

Breast-Feeding Duration

In addition to the diet records, duration of breast-feeding from birth was questioned by the study paediatricians at each of the first visits at 3, 6, 9, 12, and 18 months until the infant was fully weaned. Moreover, the consumption of any additional liquids, formula, or food was questioned to distinguish between full and partial breast-feeding. For this analysis, breast-feeding definitions as provided by the World Health Organisation were used (23). Thus, full breast-feeding duration was defined as the number of weeks of both exclusive (breast milk only with no other food or drink) and predominant (breast milk in combination with water or water-based drinks) breast-feeding. Total breast-feeding duration was defined as the number of weeks of full and partial (breast milk in combination with formula or CF) breast-feeding. Data on total breast-feeding duration were available for a subsample of 344 participants, as well as on full breast-feeding duration for 339 participants.

Back to Top | Article Outline

Potentially Confounding Factors

On their child's admission to the study, parents were interviewed by the study paediatrician about family characteristics, and were weighed and measured by the study nurses. Information on the child's birth characteristics were abstracted from the “Mutterpass,” a standardised document given to all the pregnant women in Germany.

For this analysis, the following characteristics were considered as potential covariates: total energy intake (kilocalories per day), maternal employment (yes/no), ongoing breast-feeding (yes/no), high maternal educational status (≥12 years of schooling), household size, maternal age at birth (years), parental overweight (body mass index [BMI] ≥25 kg/m2), and birth weight (g). For missing values (n = 2 for maternal employment, n = 20 for household size) the respective median of the total sample was used. Parental overweight was defined as a BMI >25 kg/m2, calculated with maternal height and weight. If maternal data were missing (n = 45), paternal data were used. For the remaining missing values (n = 3), the median of the total sample was used.

Back to Top | Article Outline

Statistical Analyses

SAS procedures (version 9.1.3; SAS Institute, Cary, NC) were used for data analyses. Food and energy intake was calculated as the individual means of 3 recorded days using LEBTAB. Descriptive data were presented as frequencies, medians with an interquartile range (in the case of skewed data), or means with a standard deviation. P < 0.05 was considered as significant. Because there was no significant sex interaction, data from girls and boys were pooled for all of the analyses.

Because of the number of zero values and the resulting skewness of breast-feeding data, time trends in breast-feeding duration were analysed using logistic regression for ordinal-dependent variables (PROC LOGISTIC in SAS). For this purpose, the duration of full breast-feeding was categorised into 3 groups (<4 months, 4–<6 months, and ≥6 months), and the duration of total breast-feeding into 4 groups (<4 months, 4–<8 months, 8–<12 months, and ≥12 months). The birth years of the sample (2001–2012) were categorised into 3 strata (Table 2).



To determine differences in the characteristics of low and high commercial CF consumers, the study participants were split into 2 groups according to whether their average commercial CF intake was lower or higher than the median commercial CF consumption of 62% (median of all records where CF was consumed). Tests for differences were performed using the t test for normally distributed variables, Wilcoxon rank-sum test for non-normally distributed variables, and χ2 test for categorical variables.

To analyse time and age trends in the methods of CF preparation (calculated as percentage of total CF intake), a polynomial mixed-effects regression model including both fixed and random effects was used (PROC MIXED in SAS). A repeated statement was used to account for the lack of independence that exists between repeated observations on the same person; a random statement accounted for the nested nature of our data (children within families).

Models were also adjusted for potential confounding factors. Only those variables that significantly modified regression coefficients in the basic models by ≥10%, had a significant, independent effect on the outcome variable, or led to an improvement of the Akaike Information Criterion (AIC) were considered in the final models (24).

Back to Top | Article Outline


Study Sample Characteristics

Between 1 (n = 31, 8.5% of total sample) and 5 (n = 183, 50.0%) 3-day weighed diet records were available per subject (mean 3.9), a total of 1419 records. Descriptive data on the study sample are presented in Table 1, stratified by age. Total energy intake increased with age, whereas the percentage of breast milk consumers and total breast milk intake decreased. More than two-thirds of the study sample was still being breast-fed at the age of 6 months, and one-fifth at the age of 12 months.



Back to Top | Article Outline

Breast-Feeding Practice

In the subsample with available questionnaire data on breast-feeding duration, the majority of mothers (96.2%) had breast-fed their infants for at least 1 week. Afterwards, breast-feeding rates declined continuously (Fig. 1). At 4 months of age, 60.5% of the infants were fully breast-fed, with a total breast-feeding rate of 79.1%. At 6 months of age, the rate of full breast-feeding was 24.5%, whereas the rate of total breast-feeding remained high (71.2%). The median duration of full breast-feeding in the DONALD sample was 21 weeks (interquartile range 7–25 weeks), and the median duration of total breast-feeding was 35 weeks (21–50 weeks, data not shown).



Table 2 shows time trends in the duration of full and total breast-feeding. The odds ratio for a short duration of full breast-feeding (<4 months) was higher for later-born infants in both the unadjusted (P = 0.0004) and the adjusted (P < 0.0001) models. Infants born between 2008 and 2012 had a 3.3-fold higher chance for a short duration of full breast-feeding than those born between 2001 and 2003; however, there was no significant difference in the likelihood for a long duration of total breast-feeding (≥12 months) for different birth year cohorts (P = 0.140). Breast-feeding duration did not differ between boys and girls.

Back to Top | Article Outline

Weaning Practice

In the present sample, CF was introduced in 29.4% of infants before the age of 4 months, and in 2.2% after 6 months (Fig. 1). Thus, two-thirds of the study sample (68.4%) was introduced to CF between the recommended ages of 4 to 6 months. The median age of introduction was 23 weeks (interquartile range 21–26, data not shown).

The number of CF consumers (99% of records) and CF intake (548 g/day) was highest at 9 months of age. At the age of 12 months, nearly all infants still consumed CF (96%), but intake was lower (428 g/day). Thereafter, consumer frequency and CF intake declined, but at the age of 18 months, almost two-thirds of the study sample still received CF, at least in small amounts (Table 1).

Back to Top | Article Outline

Preparation Method of CF

Overall, 59.3% of CF was commercially prepared, 21.1% was homemade, and 19.6% was a combination of both. (For detailed percentages of different methods of preparation by age, see Table 1.) Of all of the records in which CF was consumed (n = 1080), 354 records (32.8%) contained exclusively commercial CF, whereas homemade CF was eaten exclusively in 61 records (5.6%, data not shown).

The differences between high (percentage of commercial CF > median) and low (percentage of commercial CF < median) commercial CF consumers are shown in Table 3. Subjects with high commercial CF consumption were significantly older (P < 0.0001) and showed a shorter duration of full (P < 0.0001) and total (P < 0.0001) breast-feeding. Those infants with lower commercial CF consumption were introduced to CF later than those with a high consumption (P = 0.002). Total CF (P = 0.001) and total food intake (P < 0.0001), as well as total energy intake (P < 0.0001), was higher in subjects with high commercial CF consumption. The differences persisted when calculated per kilogram body weight (P < 0.0010). Mothers of low commercial CF consumers had a significantly higher educational status (P = 0.01). No differences between low and high commercial CF consumers were seen with respect to sex, maternal age, birth weight, breast milk intake, and parental BMI.



Back to Top | Article Outline

Age and Time Trends in the Consumption of CF

Table 4 shows age and time trends for the different methods of preparation of CF, calculated as percentage of total CF intake. No significant time trends could be found for any method of preparation during the study period, neither in the unadjusted nor in the adjusted models.



Significant age trends were found for both commercial and homemade CF. For commercial CF, the model described a negative linear with a positive, U-shaped quadratic age trend. Thus, the percentage of commercial CF consumed showed a decrease, followed by an increase with increasing age. In contrast, a negative, U-shaped quadratic age trend could be found for homemade and combined CF, showing an initial increase followed by a decrease in the percentage of homemade and combined CF consumed with increasing age. For homemade CF, there was an additional positive linear age trend.

Back to Top | Article Outline


Breast-feeding and weaning practices are important for short- and long-term health, as well as for the development of dietary habits and preferences. New data from the DONALD study illustrate a negative time trend in full breast-feeding duration in German infants born between 2001 and 2012. In addition, we could show the relevance of commercial CF in today's infant nutrition, especially in infants who were breast-fed for a shorter duration and whose mothers had a lower educational status. Both commercial and homemade CF showed opposing, nonlinear age trends; however, no time trends could be found during the observed study period, illustrating that, in the DONALD study, the majority of CF was consistently commercially prepared.

The strengths of our analysis include the well-characterised cohort, the detailed dietary assessment method with recurring 3-day weighed diet records, and the food composition database, which allows differentiation between commercial and homemade CF. In addition to the detailed dietary data, numerous early life and socioeconomic family characteristics were assessed and could be adjusted for.

The longitudinal design of the DONALD study allows analyses on age and time trends based on individual intake data from close-meshed measurements using consistent survey methods for several years; however, the way in which CF is prepared has only been recorded since 2004. The time period considered in this analysis, therefore, may be too short to detect any underlying time trends, compared with previous trend analyses using DONALD study data (25,26). A comparison of data from the last 2 cross-sectional Infant Feeding Surveys in the United Kingdom, carried out in 2005 and 2010, showed that the proportion of mothers giving homemade CF remained about the same, whereas fewer infants were given commercial CF in 2010 than in 2005 (27,28).

The generalisability of our results may be limited because the participants in the DONALD study are characterised by a relatively high educational and socioeconomic status (21). Consequential differences in food expenses, health consciousness, or maternal employment could influence the choice of CF preparation methods (29,30). Thus, women who are better educated are more likely to go back to work sooner after birth (31), which could lead to an overestimation of the importance of commercial CF in this population, it being the more convenient and time-saving option.

Our finding of a slight downward trend in the rate of full breast-feeding is in accordance with the findings of the KiGGS survey, which also showed a decrease in full and total breast-feeding duration for infants born after 2001 (9). The reason for this observed decline in full breast-feeding duration is unknown. Although continuous breast-feeding promotion in Germany has led to a higher initial breast-feeding prevalence, our data show no success with respect to breast-feeding duration. Therefore, more effort should be made to achieve at least 4 months of exclusive or full breast-feeding in Germany.

Regarding the method of preparation of CF, our study sample illustrates the importance of commercial CF in Germany, given that 94.4% of the 3-day diet records included at least 1 commercial CF product. There are extremely few studies on the consumption of commercial versus homemade CF, possibly because of inconsistent or missing information on the different methods of preparation. In line with our data, the Feeding Infants and Toddlers study showed that 73% to 95% of US infants between the ages of 4 and 11 months had consumed commercially prepared baby foods during the previous 24 hours (32). In a cross-sectional telephone survey of US mothers receiving benefits as part of the Special Supplemental Nutrition Program for Women, Infants, and Children, a governmental nutrition programme, 54% of infants younger than 6 months and 98% of infants ages 6 to 12 months had consumed baby foods in the previous 24 hours, 60% and 81%, respectively, as commercial baby food (14). In the UK Infant Feeding Survey from 2010, mothers were more likely to give their 4- to 6-month-old infants commercial CF (38%) than homemade food (28%). Yet, at 8 to 10 months, homemade food (70%) was more common than commercial CF (44%) (27). Taken together, these studies illustrate the generally high, but also age-dependent, usage of commercially produced CF in industrialised countries.

Although the percentage of commercial CF consumed decreased initially and then rose from 12 months onwards, the percentage of homemade and combined CF consumed showed an opposing trend in our data. A possible explanation is that parents become more experienced and confident in the preparation of CF toward the end of their child's first year of life. Nevertheless, at every age between 6 and 24 months, commercially produced CF accounted for a larger proportion of the CF consumed than homemade CF. When considering the noticeably higher proportion of commercial CF consumed in the second year of life, one should keep in mind the smaller number of consumers and therefore the lower absolute amount of CF consumed at this age. In addition, the ever increasing quantity and variety of commercial CF and toddler foods on sale could also influence the dietary habits described above.

Given these results, the question arises as to whether commercial or homemade CF is nutritionally more favourable for infants. Although the nutrient content and composition of commercial CF are regulated by an EC directive (38), the composition of homemade CF or the extent to which parents abide by the recommended recipes for cooking homemade CF remains unclear. Two older studies on the composition of homemade and commercial CF in Spain and the United Kingdom showed that homemade CF contained on average a broader nutrient range and a less favourable nutrient composition (39,40). In 2 recent studies, Zand et al showed that the macronutrient content and composition of commercial “ready-to-feed” CF in the United Kingdom were within the regulatory requirements, whereas the content of some essential and trace elements did not meet the recommended intakes (41,42). Another cross-sectional UK survey found infant commercial weaning foods to be generally less nutrient dense than homemade family weaning foods (43). Dietary pattern analysis from the ALSPAC study also points to some contrary differences in nutrient intake between homemade and commercial CF, although the nutrient profiles of both patterns were not consistent in themselves between 6 to 8 and 15 to 18 months (44). A European focus group discussion showed that taste is an important deciding factor for parents when preparing infant food at home (33). Although reliable data on possible differences between commercial and homemade CF with regards to taste and texture are missing, some infant feeding guidelines encourage the use of homemade CF because this offers the infant a wider range of different tastes and textures (34,35), which is known to increase food acceptance later in life (36,37).

In addition to those short-term effects, the preparation method of CF may have an influence on the later development of eating behaviour and diet quality (14,15). In view of the fact that in Germany commercial CF does not incorporate a great variety of vegetables (Mesch et al, unpublished data), a possible association between commercial CF and fruit and vegetable variety later in life needs to be considered in future studies.

Back to Top | Article Outline


The development of healthy dietary habits should start early in life with favourable breast-feeding and weaning practices. With respect to breast-feeding, the decreasing trend in full breast-feeding duration observed here should encourage health care providers and public health authorities to further promote adequate full breast-feeding duration through greater maternal encouragement as well as extended professional information and education. In addition, national breast-feeding monitoring needs to be established to provide representative data periodically. Although the constantly high and widespread consumption of commercial CF may have some advantages given that its composition is closely monitored in the EU, there are still concerns related to its long-term influence on dietary habits, in particular on fruit and vegetable intake and variety. Parents choosing to prepare CF at home need to be advised carefully on appropriate ingredients and recipes.

Back to Top | Article Outline


1. Robinson S, Fall C. Infant nutrition and later health: a review of current evidence. Nutrients 2012; 4:859–874.
2. Przyrembel H. Timing of introduction of complementary food: short- and long-term health consequences. Ann Nutr Metabol 2012; 60 (suppl 2):8–20.
3. Mennella JA, Trabulsi JC. Complementary foods and flavor experiences: setting the foundation. Ann Nutr Metab 2012; 60 (suppl 2):40–50.
4. Scaglioni S, Arrizza C, Vecchi F, et al. Determinants of children's eating behavior. Am J Clin Nutr 2012; 94 (6 suppl):S2006–S2011.
5. Mennella JA, Jagnow CP, Beauchamp GK. Prenatal and postnatal flavor learning by human infants. Pediatrics 2001; 107:e88.
6. Forestell CA, Mennella JA. Early determinants of fruit and vegetable acceptance. Pediatrics 2007; 120:1247–1254.
7. World Health OrganizationInfant and Young Child Feeding. Geneva:WHO; 2009.
8. Agostoni C, Braegger C, Decsi T, et al. Breast-feeding: a commentary by the ESPGHAN Committee on Nutrition. J Pediatr Gastroenterol Nutr 2009; 49:112–125.
9. Lange C, Schenk L, Bergmann R. Verbreitung, Dauer und zeitlicher Trend des Stillens in Deutschland. Ergebnisse des Kinder- und Jugendgesundheitssurveys (KiGGS) [Distribution, duration and temporal trend of breastfeeding in Germany. Results of the German Health Interview and Examination Survey for Children and Adolescents (KiGGS)]. Bundesgesundheitsbl Gesundheitsforsch Gesundheitsschutz 2007; 50:624–633.
10. Kohlhuber M, Rebhan B, Schwegler U, et al. Breastfeeding rates and duration in Germany: a Bavarian cohort study. Br J Nutr 2008; 99:1127–1132.
11. Kersting M, Dulon M. Assessment of breast-feeding promotion in hospitals and follow-up survey of mother-infant pairs in Germany: the SuSe Study. Public Health Nutr 2002; 5:547–552.
12. Cattaneo A, Burmaz T, Arendt M, et al. Protection, promotion and support of breast-feeding in Europe: progress from 2002 to 2007. Public Health Nutr 2010; 13:751–759.
13. Agostoni C, Decsi T, Fewtrell M, et al. Complementary feeding: a commentary by the ESPGHAN Committee on Nutrition. J Pediatr Gastroenterol Nutr 2008; 46:99–110.
14. Hurley KM, Black MM. Commercial baby food consumption and dietary variety in a statewide sample of infants receiving benefits from the special supplemental nutrition program for women, infants, and children. J Am Diet Assoc 2010; 110:1537–1541.
15. Coulthard H, Harris G, Emmett P. Long-term consequences of early fruit and vegetable feeding practices in the United Kingdom. Public Health Nutr 2010; 13:2044–2051.
16. Smithers LG, Golley RK, Mittinty MN, et al. Dietary patterns at 6, 15 and 24 months of age are associated with IQ at 8 years of age. Eur J Epidemiol 2012; 27:525–535.
17. Kleinman RE. Pediatric Nutrition Handbook. 6th ed.Chicago:American Academy of Pediatrics; 2009.
18. Department of Health. Introducing solid foods: giving your baby a better start in life. Published April 5, 2011. Accessed September 20, 2012.
19. Alexy U. Die Ernährung des gesunden Säuglings nach dem “Ernährungsplan für das 1. Lebensjahr” [Feeding healthy infants according to the “Dietary schedule for the first year of life”]. Ern Umschau 2007; 54:588–594.
20. Manns K-J. Marktentwicklung Babynahrung: Wieder gewachsen [Market development baby food: grown again]. Lebensmittelzeitung 2011; 47:36.
21. Kroke A, Manz F, Kersting M, et al. The DONALD StudyHistory, current status and future perspectives. Eur J Nutr 2004; 43:45–54.
22. Sichert-Hellert W, Kersting M, Chahda C, et al. German food composition database for dietary evaluations in children and adolescents. J Food Compos Anal 2007; 20:63–70.
23. World Health Organization, Division of Child Health and Development Indicators for assessing breastfeeding practices. Published 1991. Accessed February 15, 2013.
24. Maldonado G, Greenland S. Simulation study of confounder-selection strategies. Am J Epidemiol 1993; 138:923–936.
25. 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.
26. Alexy U, Kersting M. Time trends in the consumption of dairy foods in German children and adolescents. Eur J Clin Nutr 2003; 57:1331–1337.
27. McAndrew F, Thompson J, Fellows L, et al.Infant Feeding Survey 2010. Published November 30, 2012. Accessed April 19, 2013.
28. Bolling K, Grant C, Hamlyn B, et al.Infant Feeding Survey 2005. Published May 14, 2007. Accessed April 23, 2013.
29. Robinson S, Marriott L, Poole J, et al. Dietary patterns in infancy: the importance of maternal and family influences on feeding practice. Br J Nutr 2007; 98:1029–1037.
30. Northstone K, Emmett P. Multivariate analysis of diet in children at four and seven years of age and associations with socio-demographic characteristics. Eur J Clin Nutr 2005; 59:751–760.
31. Hanel B, Riphahn RT. The Employment of Mothers: Recent Developments and Their Determinants in East and West Germany. Discussion Paper No. 5752. Bonn:Institute for the Study of Labor; 2011.
32. Briefel RR, Reidy K, Karwe V, et al. Feeding Infants and Toddlers Study: improvements needed in meeting infant feeding recommendations. J Am Diet Assoc 2004; 104 (1 suppl):S31–S37.
33. Synnott K, Bogue J, Edwards CA, et al. Parental perceptions of feeding practices in five European countries: an exploratory study. Eur J Clin Nutr 2007; 61:946–956.
34. NHS Survey. Surrey Infant and Child Feeding Guidelines. Published August 2010. Accessed July 7, 2013.
35. Koletzko B, Bauer CP, Brönstrup A, et al. Säuglingsernährung und Ernährung der stillenden Mutter. Aktualisierte Handlungsempfehlungen des Netzwerk Gesund ins Leben - Netzwerk Junge Familie, ein Projekt von IN FORM [Infant Nutrition and nutrition for breastfeeding mothers. Updated practice recommendations of the network “Healthy Start - Young Family Network” a project of IN FORM]. Monatsschr Kinderheilkd 2013; 137:1309–1314.
36. Maier AS, Chabanet C, Schaal B, et al. Breastfeeding and experience with variety early in weaning increase infants’ acceptance of new foods for up to two months. Clin Nutr 2008; 27:849–857.
37. Coulthard H, Harris G, Emmett P. Delayed introduction of lumpy foods to children during the complementary feeding period affects child's food acceptance and feeding at 7 years of age. Matern Child Nutr 2009; 5:75–85.
38. European CommissionCommission Directive 2006/125/EC of 5 December 2006 on processed cereal-based foods and baby foods for infants and young children. Off J Eur Union 2006; 339:16–35.
39. Stordy BJ, Redfern AM, Morgan JB. Healthy eating for infants–mothers’ actions. Acta Paediatr 1995; 84:733–741.
40. van den Boom S, Kimber AC, Morgan JB. Nutritional composition of home-prepared baby meals in Madrid. Comparison with commercial products in Spain and home-made meals in England. Acta Paediatr 1997; 86:57–62.
41. Zand N, Chowdhry BZ, Pollard LV, et al. Commercial ‘ready-to-feed’ infant foods in the UK: macro-nutrient content and composition. Matern Child Nutr 2012; October 1 [Epub ahead of print].
42. Zand N, Chowdhry BZ, Wray DS, et al. Elemental content of commercial ‘ready to-feed’ poultry and fish based infant foods in the UK. Food Chem 2012; 135:2796–2801.
43. Garcia AL, Raza S, Parrett A, et al. Nutritional content of infant commercial weaning foods in the UK. Arch Dis Child 2013; 98:793–797.
44. Smithers LG, Golley RK, Brazionis L, et al. Dietary patterns of infants and toddlers are associated with nutrient intakes. Nutrients 2012; 4:935–948.

breast-feeding duration; commercial complementary food; homemade complementary food; infant feeding

© 2014 by European Society for Pediatric Gastroenterology, Hepatology, and Nutrition and North American Society for Pediatric Gastroenterology,