Baby-led weaning: what a systematic review of the literature adds on

The term weaning describes the time period in which a progressive reduction of breast-feeding or the feeding of infant-formula takes place while the infant is gradually introduced to complementary foods. The introduction of complementary foods during the weaning period is generally progressive, and leads the infant to reach the dietary pattern of an adult within the second year of life [1].

The World Health Organization (WHO) recommends exclusive breastfeeding for the first six months of age, and complementary breastfeeding at least until the second year of age. According to the WHO, the introduction of complementary foods should be safe, well-timed and adequate; it should start when exclusive breastfeeding can no longer provide enough nutrients and energy for the infant’s growth and development, and it should contain foods that offer these nutrients and energy [2]. The weaning period is a crucial time in an infant’s life since it not only involves a great deal of rapid change for the child, but it is also associated with the development of food preferences, eating behaviours and body weight in childhood, adolescence as well as in adulthood.

Babies are traditionally first introduced to solid foods using spoon-feeding of specially prepared thin purées. Later, following the infant’s age and developmental progress, the foods offered gradually shift towards family foods [3].

Over the last 10–15 years, an alternative approach known as “baby-led weaning” (BLW), or “auto-weaning”, has grown in popularity, particularly in the United Kingdom and New Zealand and more recently also in other countries in Europe [4, 5]. The term “baby-led” weaning was first coined by Gill Rapley in 2005 [6]. BLW is an alternative method of infant feeding which promotes infant self-feeding from six months of age, instead of conventional parent spoon-feeding. Although BLW is not especially mentioned in the WHO’s recommendations, it is becoming more popular. The key features of BLW are that infants participate in family mealtimes, and ‘whole’ (baby-fist size) pieces of food are offered to them, so they feed themselves from the beginning of complementary feeding, at around six months of age [7]. It means that although parents offer foods, the child himself controls the weaning process (thus the term “baby-led”): infants decide what, how much and how quickly to eat [8]. The term underlines the fact that the infant is an active partner in the feeding process, and not a passive recipient to fill with food [9].

BLW may also be defined as auto-weaning, which means offering chopped and minced family meals to the infants [10]. While in the traditional weaning infants are offered puréed infant foods that are often made up of several ingredients, in the baby-led weaning a variety of single picked foods is offered to the baby. In the former approach, the tastes of the single foods are mixed together and the child is not always able to distinguish them; conversely, BLW approach might provide an early and more stable learning about the satiating capacities of foods and therefore it may enable a better satiety-responsiveness [11].

BLW seems to be associated with other positive aspects, such as lower maternal anxiety and control during the weaning period, but this point is controversial [12].

A recent review has found that many mothers use food to influence infant growth, contentment and sleep and that they choose ease of feeding over infant feeding recommendations [13].

However, this approach also causes concerns in primary care paediatricians and in parents themselves, regarding main topics such as iron adequacy [14, 15], energy and nutrient intake and choking risk [10].

The European Society for Paediatric Gastroenterology, Hepatology and Nutrition (ESPGHAN 2017), in a recent position paper, stated that there is not enough evidence to draw conclusions about the BLW approach [16]. At the same time, a non-systematic review by Brown et al., considering the available literature on BLW up to December 2016, came to similar conclusions [4].

We aimed to address the following Key Questions:

To answer these Key Questions we searched, from 2000 up to March 1, 2018 the Cochrane Database of Systematic Reviews and DARE (Database of Abstract of Reviews of Effects), for terms “weaning” and “baby-led weaning”.

The details of search strategies in PubMed (MedLine) and EMBASE are listed in Table 1.

We planned hand-search for other relevant articles from selected papers.

Our inclusion criteria, with respect to the predefined aims of the review, were: evidence-based guidelines, systematic reviews, randomized controlled trials, controlled clinical trials and observational studies that compared outcomes related to growth and development, energy, macro- and micronutrient intake and feeding attitudes in children and families that are following BLW approach or standard/traditional complementary feeding; outcomes related to the impact of different weaning styles on mothers feelings and attitudes about their children are considered.

No exclusions were made for age nor restriction for language.

Practice guidelines not evidence-based, narrative reviews, editorials, other publication types and articles in which BLW was not clearly defined and/or quantified, were considered for discussion of items but excluded from formal analysis and from tables of evidence.

We planned to appraise the relevant guidelines with the AGREE II instrument [17], systematic reviews with the AMSTAR-2 tool (A MeaSurement Tool to Assess systematic Reviews) [18], Randomized Controlled Trials with the Assessment of Risk of Bias Cochrane tool [19], Observational studies with the Newcastle-Ottawa Scale, modified for cross-sectional studies [20].

ED and MB independently searched and selected the literature based on the inclusion/exclusion criteria, then extracted results and appraised the included articles.

We considered methodological lacks of each study to answer Key Questions.

We also planned the pooling of results, with appropriate methods, if the same outcome was present in a sufficient number of not heterogeneous studies.

After the inclusion/exclusion process, we included 12 articles, 10 observational cross-sectional studies and 2RCT (from the same population) for analysis of evidence.

In Fig. 1 we show the PRISMA Flow Diagram [21] of search and selection.

The characteristics of the included studies are listed in Table 2 and described in the text answers to Key Questions.

Trials excluded from formal analysis after full-text screening are listed in Appendix 1, with motivations for exclusion.

Pooling of results from very different outcomes in the studies included was not possible.

One of the critical aspects of the BLW approach is that no formal definition exists. In its purest form, BLW should not include any spoon-feeding and the child himself should put foods into his mouth. As a limitation, most existing studies on the baby-led approach include participant families who self-identify as following a BLW. In some studies, participants were asked to estimate the use of spoon-feeding opposed to self-feeding and the amount of puréed foods given during the weaning period in percent [11, 12, 30, 34]. In others, they were just asked to identify themselves as followers of BLW approach [14, 24, 31]. It is unclear whether BLW can include limited use (less than 10%) of purées and spoon-feeding, or if it is ruled by a more strict definition, where exclusively finger foods are provided. Actually, both views exist among parents who believe in a baby-led style infant feeding [33]. Concerning the method of recruitment, most studies recruited via internet sites, that can represent a selection bias [41]. A recent research conducted in the UK found that mothers following a baby-led approach were more likely to have higher levels of education, and a professional occupation, compared to mothers choosing traditional weaning. This may be due to the fact that the population group with higher education is more likely to have increased internet access, which is also one of the main information sources about BLW [25, 30].

Overall, the evidence about BLW is mostly derived from observational studies, of which only comparative cross-sectional studies were formally analyzed. The methodological quality of the included studies is generally very low, except for two studies, (Table 3).

Therefore, the observed results need to be treated with caution. As regards the major concerns about the BLW approach, e.g. inadequate energy intake and choking risk, there are very few data on the role of the paediatrician’s support on this practice. Although the randomized studies by Taylor and Fangupo specifically addressed these issues, in the same cohort, both of them have methodological weaknesses that call into questions the results (Table 4).

Most of the outcomes were self-reported, the intention to treat analysis was not performed and both had a high rate of drop-out. Moreover, the evaluation of nutrient intakes has not been included in the secondary outcomes of the study, but has been described only in an unpublished doctoral dissertation by Erickson, that examines initial findings of the BLISS Study, considering adherence to the weaning approach and nutrient intake, respectively [35]. Therefore, this issue needs a careful monitoring by the paediatrician in order to ensure adequate growth.

Even with the above mentioned limitations, the BLISS Study suggests that this kind of approach, employing resources and methods to educate participants, may be suitable to be tested in further randomized controlled trials.

Anyway, parents willing to follow BLW approach need a careful and deeper nutritional education in order to avoid any risk to their infants. In particular, they should be given advice on how to prepare foods in such a way they result safe, healthy and nutritious.

The role of different approaches to weaning in the development of the early-life gut microbiome is an additional interesting aspect, which has not yet been investigated. It is known that changes in the diet composition can alter the prevalence and types of microbial species living in the gut, as certain species are better equipped to use specific substrates [42].

The first solid foods introduced could therefore play a key role in shaping the infant’s gut microbiome. A recent study, assessing the effect of iron supplementation in infants, found that children fed iron-only fortified cereals, or puréed meats, iron- and zinc-fortified cereals, as the primary complementary foods until 9–10 months of age, had different prevalence in the gut microbial species [43]. The former had a relative decrease with time in the genera Bifidobacterium and Rothia and in the order Lactobacillales, and an abundance of the order Bacteroidales, compared to the latter. Further investigations are needed to determine if the BLW approach could shape the microbiome in a different way compared to the traditional weaning.

Another relevant issue is that in the BLW approach children are exposed to a wide range of family foods in a relatively early and mixed manner and sometimes also to packaged foods, which can contain multiple food allergens. Therefore, in case of reaction, it can be difficult to identify the specific food allergen involved.

This potential risk has not been formally investigated, even though the weaning period requires special attention, particularly in children at high risk of allergy.

Currently, there is still insufficient evidence to draw conclusions about the BLW approach, in terms of adequacy of energy and nutrient intakes, due to the low quality of the evidence. In fact, concerns persist since some previous observational studies indicated that mothers using the BLW approach estimated that their babies ate more milk feeds and less solid foods compared to those following a traditional weaning, focussing attention on inadequate nutrient intakes for infants from 6 months of age onwards. Nevertheless, other evidence from more recent randomized studies suggest that a modified BLW approach (BLISS Study), including recommendations about the introduction of selected iron-rich foods, as well as avoiding foods at risk of choking, might have positive preventive effects on the risk of choking and nutrients deficiency. Thus, these issues require further investigation in larger randomized studies.

In summary, there are still major unresolved issues in BLW that require answers from research, which should be considered when advices are requested from health professionals by parents willing to follow the BLW approach (Table 5).