• Complementary food should ‘complement’ breastfeeding when nutritionally needed and when the infant is ready for it.
• Too early [before 12 weeks of age] and too late introduction [beyond 26 weeks] can have undesirable health consequences.
• Continuation of breastfeeding after the introduction of complementary food is beneficial.
Introduction
A discussion of the consequences of the timing of the introduction of complementary food into the diet of infants on their health should neither be misunderstood as a discussion of the ‘optimal’ or desirable duration of breastfeeding nor as an assessment of the evidence for an appropriate age for the introduction of complementary feeding, both have been done already: the former in systematic reviews [1,2] with subsequent recommendations for a desirable length of exclusive [6 and 4–6 months] and partial breastfeeding [up to 2 years] [3,4], and the latter in several reviews [5,6,7] concluding that some infants may need complementary food before the age of 6 months, preferably in addition to breastfeeding.
This article only considers the available studies on the health effects of age at which complementary food has been introduced regardless of breastfeeding or formula feeding. Such studies are scarce and are mostly observational; furthermore, these studies have often been performed retrospectively and, with a few exceptions, for short follow-up periods, i.e. only into early childhood.
In this article, complementary food is defined as any food, solid or [semi-] liquid, besides breast milk or its substitutes, i.e. infant [or follow-on] formula. This definition of complementary food was chosen because not all infants are breastfed or are only breastfed for short periods [8]. This is different from the World Health Organization [WHO] definition [9]: any food or liquid given along with breast milk. Nevertheless, it has been shown that the introduction of complementary food before the age of 4 months is inversely related to the level of maternal education, maternal age, socioeconomic status, maternal smoking, duration of breastfeeding and information on health care [5,10,11,12], and these factors by themselves may have an impact on health consequences in later life. When such consequences are assessed, the nature and composition of the complementary food cannot be disregarded, because this varies in different regions of the world due to tradition, food availability and socioeconomic status of the parents. Moreover, the pattern of introduction of other food than human milk [or formula], as shown in figure 1 [adapted from ref. [13]], is so complex that it is impossible to find meaningful associations with health consequences for each food item separately. Furthermore, each pattern may be associated with health consequences by itself: this should be investigated in more detail in the future. In low- and middle-income countries, infants are at the greatest risk of malnutrition and stunting during the period when breast milk is complemented or replaced by other food [14].
Fig. 1
Making sense of complex feeding practices. Prospective birth cohort study throughout Bavaria, Germany: cumulative percentage of consumption of other foods than human milk in 3,092 infants aged 1–9 months [modified from ref. 13]. Observed differences in growth and infection rates from feeding practice patterns or interventions should be further investigated.
Nutritional Inadequacy of Prolonged Exclusive Breastfeeding
The time span that exclusive breastfeeding is adequate to meet the infants’ requirements for proteins, most vitamins and essential minerals has been found to be 6 months in the case of healthy term infants of well-nourished mothers [15,16].
There are few data on health effects of exclusive breastfeeding prolonged beyond the first 6 months. In one prospective longitudinal study on 193 healthy term infants born to non-smoking mothers, 116 were exclusively breastfed beyond 6 months of life [7 infants for >9 months]. Length velocity lagged slightly but progressively behind, and the weight for length was higher for infants exclusively breastfed more than 6 months compared to infants receiving both breast milk and complementary food [6–9 months]. This may indicate stunting related to insufficient intake of energy and some nutrients [17]. Both serum iron and serum ferritin were significantly lower in infants exclusively breastfed beyond 6 months than in weaned infants [18].
The iron, zinc and vitamin D requirements of young infants cannot be provided by human milk alone. There is a higher risk of iron deficiency in infants exclusively breastfed for 6 months compared to infants exclusively breastfed for 3–4 months [2]. The risk for iron deficiency anemia at 6 months of age is increased by male sex, birth weight 2,500–2,999 g and weight gain above the reference value since birth [19]. Iron deficiency anemia is a risk factor for long-term adverse effects on motor, mental and social development [20,21]. The risk of zinc deficiency was found to be increased after 6 months of exclusive breastfeeding, and zinc deficiency may contribute to a deceleration in growth of some fully breastfed infants [22,23]. Rickets can be a consequence of prolonged breastfeeding without vitamin D supplements and has been observed in 1 of the 3 infants exclusively breastfed up to the age of 9 months who participated in Clara Davis’ 1928 historical experiment on nutritional adequacy of self-selected complementary food in late infancy [24]. However, vitamin D insufficiency of human milk should be compensated by supplements and not by the early introduction of complementary food.
Developmental Aspects
The initiation of feeding food with a spoon or cup involves a number of important changes, including oral motor development, new tastes, new textures and new interaction between the infant and the caregiver. This occurs in parallel to greater stability and strength of the trunk, shoulder and neck musculature, which allows the infant to sit up and control his head position [25]. Some authors suggest that there is a critical window for introducing lumpy solid food into an infant’s diet and that introduction after the age of 10 months becomes more difficult [26], particularly in infants who had been tube fed or only fed purees throughout the first year of life [27].
Food Acceptance and Feeding Problems
Early exposure to a variety of flavors with complementary food in addition to the flavors provided by breast milk has a positive effect on the acceptance of new food [28]. The effect of age at the introduction of lumpy food on subsequent feeding difficulties was assessed in the large cohort of the Avon Longitudinal Study of Pregnancy and Childhood [ALSPAC]. An introduction later than 9 months of age resulted in a greater incidence of feeding problems at 15 months [29] and in a dislike of fruit and vegetables and being more choosy with food at 7 years of age compared to the group that had been introduced to lumpy food between 6 and 9 months. An introduction below the age of 6 months had no detrimental effects and, on the contrary, increased the likelihood of consumption of more varied vegetables more often [30].
Growth [Weight Gain]
Rapid weight gain during the first months of life may have negative implications on risk markers of other non-communicable diseases in later life [i.e. high blood pressure, obesity, non-insulin-dependent diabetes or coronary heart disease], but no associations with growth during the period from 6 to 12 months have been found [7,31]. Compared to formula-fed infants, breastfed infants gain weight more rapidly during the first 2 months and more slowly thereafter, i.e. between 3 and 12 months [32,33].
This observation has been proposed to be the consequence of the higher protein content of most infant formulae than of human milk, but the evidence from both observational and interventional studies is inconsistent for the first months. A lower protein intake in infancy was found to be associated with lower linear growth during the first 2 years of life and in children and adolescents [[34] and references therein]. The introduction of complementary food may lead to an increase in the total protein intake, but this is not necessarily the case and depends on the food selected. The composition of the diet of infants early in life may have long-lasting implications on body fat, for example, but this does not imply an effect of the timing of complementary food introduction. In the WHO Multicenter Growth Study [35],the prevalence of consumption of various food groups, i.e. cereals, legumes and nuts, tubers, milk/milk products, meat [fish], eggs, vitamin A-rich and other fruits and vegetables, fat/oil, juices, sweet beverages and soups, at different ages up to 24 months in the 903 children participating is also revealing: cereals were most frequently part of the first complementary food, followed by dairy products, whilst meat, poultry and fish, which are good protein sources, were introduced relatively late, and only more than half of the infants received them between 9 and 12 months of age. Total breastfeeding duration was 14.3 ± 7.9 months and mean age at the introduction of complementary [[semi]solid] food was 5.4 ± 0.7 months [mean ± SD] [35]. Notably, the anthropometric data of this growth study group form the basis of the WHO growth reference standard [36].
The available studies [9,37,38,39,40,41,42,43,44]that provide data on the effect of age at the introduction of complementary food on growth are listed in table 1. The demonstration of an independent effect of the timing of the introduction of complementary feeding on growth [and other health outcomes] should include adjustment for factors which might influence growth, which is seldom done. Body weight was found to be positively associated with maternal height, birth weight and male sex from 8 to 104 weeks of age [37].
Table 1
Effect of age at the introduction of complementary food on growth
In developed countries, age at the introduction of complementary food does not seem to have a significant impact on body weight during infancy and in the second year of life either in breastfed or formula-fed infants; in developing countries, however, it may prevent growth faltering related to the transition of exclusive breastfeeding to mixed feeding, provided that breastfeeding is continued and hygiene and composition of the complementary food are adequate. On the contrary, Peruvian infants breastfed >12 months whose complementary food was deficient both in amount and nutrient density showed growth faltering in the second half of the first year of life when compared to US infants of similar breastfeeding duration [45]. According to one study in 94 white US infants, gender explained 10% of the variance in weight gain during the period from 2 to 8 months of life, and weight prior to 12 months predicted 54% of the variance of weight gain from 12 to 24 months of life, whilst the timing of supplementary feeding was not statistically associated [46]. However, due to the lack of longitudinal studies of longer duration, the long-term effects on body weight and composition of this temporary acceleration in weight gain during the first year of life, associated in some studies with the introduction of complementary food at ages