When investigating breastfeeding benefits either for mother or child, the readers should be aware that confounding is inevitable (Shamir, 2016). INSights on the topic are predominantly based on observational studies. This is because a randomized control trial for (exclusive) breastfeeding versus (exclusive) formula feeding is considered unethical: Mothers would be randomly assigned to one of the feeding modes instead of being given the choice. (Double)blinding is impossible (How would a mother not know she is breastfeeding?). Therefore, most studies tracking breastfeeding benefits are observational cohorts were participants are often asked to recall earlier events or (medical) records are sifted retrospectively for relevant output. Recall bias and incomplete information are not uncommon. Even distinction between exclusive, partial or any breastfeeding becomes a challenge. Furthermore, there are confounders which cannot be controlled such as the participants' socioeconomic status, parental education or their intelligence quotient, lifestyle choices such as alcohol consumption, smoking, diet, exposure to health hazards at work. All these factors and an overall high(er) quality of life are health promoting in breastfeeding women (Shamir, 2016).

Just as for the mother, skin-to-skin contact and breastfeeding release oxytocin. This hormone induces milk release from maternal breast tissue but is also involved in many social interactions even in adult life. The release at this time helps the child to bond to its primary caretaker, feeling safe and secure (Uvnäs Moberg et al., 2019). 

Milk is a highly variable secretion. It is influenced intra-individually by lactation stage and -duration, by the mother's health status, and changes within a day, within a feed, between feeds and inter-individually by ethnicities, genotype, climate, diet, and many other factors (Ballard & Morrow, 2013, Demmelmair & Koletzko, 2017, Jensen, 1995).

Milk stages include colostrum, which is high in immune factors (Berdi et al., 2019), transitional milk that provides more nutrients, and mature milk with the full range of macro- and micronutrients to support natural growth and immune protection (Czosnykowska-Łukacka et al., 2018, Floris et al., 2019). During a breastfeeding session, the watery and lactose-rich foremilk is replaced by the fat- and therefore energy-rich hindmilk that also contains most fat-soluble vitamins (Koletzko, 2016). Within the circadian rhythm, milk composition is steered by melatonin (Italianer et al., 2020), which helps to calm the child and supports its sleep pattern. Mothers' milk will also be rich in immunoactive cells and molecules (e.g. tumor necrosis factor-α) when the child is running a fever (Riskin et al., 2012).

It is established that ever or longer breastfeeding increases intelligence by at least 3 points (Anderson et al., 1999, Horta et al., 2018, Horta et al., 2015, Victora et al., 2016). This effect was more pronounced in low birth weight infants (Anderson et al., 1999) and one high quality randomised clinical trial even reported a difference of seven points (Victora et al., 2016). The effect is strengthened by breastfeeding duration. It was shown that each month of breastfeeding improved language skills at three years of age by 0.21 points and that at seven years of age each months of breastfeeding improved the Kaufman Brief Intelligence Test by 0.35 points (Belfort et al., 2013). Differences in grey and white matter development have been discussed (Shamir, 2016) and in a study that used two tests for adult intelligence, the authors did not report higher IQ in the breastfeeding group but that fewer adults in the breastfeeding group had an IQ too low to be socially independent (Mortensen et al., 2002, Shamir, 2016).

The different degrees of impact could be steered by formula composition (Shamir, 2016) that has changed over time. Previously, formula did generally not contain long-chain polyunsaturated fatty acids (LCPUFA) whereas breast milk contains a spectrum of LCPUFA including arachidonic acid (ARA), docosahexaenoic acid (DHA), and eicosapentaenoic acid (EPA). This absence was proposed to be the distinguishing factor between formula and breastfeeding as being responsible for the differences in cognitive development. However, such a link could not be established (Shamir, 2016). This could be explained by the insight that although LCPUFA, particularly docosahexaenoic acid (DHA), were not mandatory in infant and follow-on formula in Europe until 2020 (EU/2016/127), DHA and ARA were optionally permitted to be added (EC/2006/141, 2006). Some manufacturers claim to have done so from the early 1990s onwards at least in specialised product ranges (Nutricia Research, 2020-11-02). Other factors that could explain the breastfeeding advantage are human milk oligosaccharides and particularly acidic sialylated human milk oligosaccharide (→). These molecules are rare in formula and carry sialic acid, a component necessary for brain development that could contribute to the reported difference in cognitive function between breastfed and formula-fed children (Belfort et al., 2013, Cerdó et al., 2019). Since LCPUFA – or at least DHA – is now mandatory in Europe (EU/2016/127) and human milk oligosaccharides – including sialylated HMO – can be optionally and selectively added to infant formulas in Europe (→) , it will be exciting to notice if these ingredients will be able to diminish some of the disadvantages that formula-fed infants experience.

Breastmilk contains many antibodies and molecules that either act in a protective way themselves or that stimulate factors of the immune system (Gila-Diaz et al., 2019). This interplay of bioactive compounds leads to a strong overall protection against mortality from infectious disease especially for children six months old or younger (Sankar et al., 2015, Victora et al., 2016). Depending on the degree of breastfeeding, the odds ratio not to die due to infectious diseases can be as high as 88% (exclusive breastfeeding compared to none) (Victora et al., 2016).The incidence of gastroenteritis is reduced by 64%. This protection lasts for two months after breastfeeding has ceased (AAP, 2012). Risk for respiratory infections can be reduced through breastfeeding in the first two years of life (Horta BL, 2013b, Victora et al., 2016). For a breastfeeding duration of more than six months, the risk of upper respiratory tract infections reduced by 63% (AAP, 2012); at least four months of breastfeeding reduced the risk for lower respiratory tract infections by 72% (AAP, 2012).

Victora and colleagues showed consistent evidence for the reduction of 33% of acute otitis media through a longer breastfeeding duration for children up to 24 months (Victora et al., 2016). An earlier report described that a breastfeeding duration for at least three months reduced the risk for acute otitis media by 50% (AAP, 2012).

A 36 ‑ 60% risk reduction for the sudden infant death syndrome (SIDS) is also associated with any breastfeeding for more than one month (AAP, 2012, Moon & AAP, 2016). This effect was independent of the child's sleep position (AAP, 2012) and is further strengthened with exclusivity of breastfeeding (Moon & AAP, 2016). The hypothesised mechanism by which breastfeeding exerts this protection is its immune strengthening effect and therefore risk reduction of upper and lower respiratory infection, gastroenteritis, diarrhoea and infections in general (Moon & AAP, 2016). Other factors and practices that can help prevent sudden infant death have been outlined by the American Academies of Pediatrics and others (AAP, 2016, Horne, 2019).

An effect of breastfeeding on allergic rhinitis for children under five years of age was deemed possible with a presumable risk reduction of 21% but was hampered by a low number of studies. For older children evidence was lacking (Victora et al., 2016). Evidence for an association between breastfeeding and asthma or wheezing in children 5 ‑ 18 years remains inconclusive (Shamir, 2016, Victora et al., 2016). The reported effects were not significant in this age group despite a large number of studies (n=16) and cohorts (n=13) and for conclusions in adults data was lacking. Earlier analyses reported that at least three months of breastfeeding reduced the risk of asthma by 26%; with a family history for atopic dermatitis the risk could be reduced by 40% (AAP, 2012). Little or no evidence supported the association of breastfeeding to eczema and food allergies below or above two years of age despite large number of studies being available (Victora et al., 2016). Confounding factors for such research questions are likely the diversity in maternal diet, type and time of introduction for complementary foods and that infants with a family history of allergic disease are more likely to be breastfed (reverse causality) (Shamir, 2016).

Children aged below five years or below six months of age are protected against diarrhoea through breastfeeding. This strong protection has been demonstrated in many studies (Horta BL, 2013b, Victora et al., 2016). Definitions of breastfeeding varied and always compared "more versus less" scenarios. These were "exclusive versus non-exclusive breastfeeding", "predominant versus partial breastfeeding", "partial versus no breastfeeding" and "any versus no breastfeeding". Incidence was reported most often and hospital admissions were also captured (Horta BL, 2013b, Victora et al., 2016). Infections of the gastrointestinal tract occurred in breastfed infants (any breastfeeding) with an odds ratio of 0.36 (95% CI 0.32-0.40) (AAP, 2012).

Feeding with mother's milk or formula shapes the gastrointestinal micobiota well into child- if not into adulthood (Cioffi et al., 2020). This association with the microbiome contributes to the immuno-protective effects of breastfeeding (Cebra, 1999, Fernández et al., 2013). Breastfed children in their first year of life have a lower intrapersonal colonic diversity (α-diversity) of bacterial species compared to formula-fed infants; they also express a different β-diversity (samples within a community/population) (Brink et al., 2020, Walters & Martiny, 2020). The breastfed gut is colonised predominantly with Bifidobacteria and Bacteroidetes but other species are present (Brink et al., 2020, Fehr et al., 2020) and even though in adults a high diversity is considered most favourably, in infants, the example of breastfed children – a low diversity - is leading (Brink et al., 2020). The gut microbiota is affecting the immune system function of the gut favourably through multiple mechanisms and secondary fermentation products like short-chain fatty acids (SCFA) (Koh et al., 2016); components such as human milk oligosaccharides may serve as prebiotic (→) and seem to affect microbiota composition; the gut-microbiota-brain axis also affects other central organ systems including the brain (Cryan et al., 2019).

The effect on anthropometric and metabolic parameters remains controversial. It had been long suspected that breastfeeding affects weight and length favourably. However, insights from 15 randomized trials and 17 studies in total did not support the idea that breastfeeding affected length at six month of age. Similarly, the 16 studies on weight gain showed similar developments compared to the control grou(s) (Victora et al., 2016). Breastfeeding could possibly affect body composition (BMI) or weight for length in the age range from three to 24 months as indicated by 11 studies (Victora et al., 2016).

Because obesity has a complex ethiology and many factors contribute to confounding (Shamir, 2016), the data from more than one hundred studies on overweight or obesity in childhood, adolescence, and adulthood has been described as "being suggestive". Yet despite residual confounding a pool of 23 high-quality studies showed a risk reduction of 13% (95% CI 6-19) and prevalence reduction of 20% through breastfeeding and breastfeeding duration (AAP, 2012, Victora et al., 2016). For the assessment whether breastfeeding and -duration reduced the risk for later type 2 diabetis mellitus only eleven studies were available – these showed a risk reduction of 14% (Horta & Lima, 2019, Horta BL, 2013a, Victora et al., 2016). However, others point out that the data is sparse, hardly allowing firm conclusions (Güngör et al., 2019b). Blood pressure – be it systolic or diastolic – was not influenced by breastfeeding in childhood, adolescence, or adulthood nor were total cholesterol concentrations (Victora et al., 2016).

Taken together, some meta-analyses suggest that breastfeeding does not reduce the risk of factors contributing to the metabolic syndrome, leaving hypertension unaffected. The "suggestive evidence" on weight reduction deserves a better understanding. If breastfeeding and breastfeeding duration are not contributing to weight control, a closer look could be directed towards other nutritional practices in the first years of life. These could include the type and timing of complementary foods as well as life-style factors, activity and sleep behaviour.

Breastfeeding provides many immune modulating molecules and affects colonic microbiota species that contribute to these effects indirectly. This mechanism was proposed to explain the 31% risk reduction for inflammatory bowel diesase through at least two months of breastfeeding (AAP, 2012).

The evidence for an association of breastfeeding and -duration on coeliac disease remains controversial. The AAP suggested a 52% risk reduction and proposed that gluten exposure should be started while still breastfeeding but not in parallel to formula or bovine milk (AAP, 2012). This hypothesis has been challenged (Chmielewska et al., 2017). Instead, the available evidence is contradictory since some studies report a sensitisation to gluten from breastfeeding wheras others report the opposite (Chmielewska et al., 2017). The challenge lies in the absence of randomized controlled trials and an observational study with more than 3700 participants reported that early gluten exposure reduced the risk for gluten sensitization at five years of age (Chmielewska et al., 2017).

Type 1 diabetes mellitus is an autoimmune disease that seems to have a similar ethiology than coelic disease (Goodwin, 2019). Some reported a 30% risk reduction though at least three months of breastfeeding. The same paper reported a 40% risk reduction for type 2 diabetes mellitus (AAP, 2012), which could not be confirmed because of data paucity (Victora et al., 2016) that seems to be persistent (Güngör et al., 2019b). The few studies reporting on type 1 diabetes mellitus development in relation to earlier breastfeeding seem to indicate that there could possibly be a protection from breastfeeding. A relation to breastfeeding duration could not be established (yet?) (Güngör et al., 2019b).

Misalignments of the jaw can lead to dysfunction and facial misformation such as severe overjet, can impact efforts to speak, transition to family foods and disordered breathing during sleep (D'Onofrio, 2019). Breastfed infants are at lower risk for misalignement of the jaws (primary dentition malocclusion) (Doğramacı et al., 2017). The risk for the development of anterior open bite was reduced by two thirds through prolonged breastfeeding beyond one year (Doğramacı et al., 2017, Victora et al., 2016). The reason for this protection is that breastfeeding requires effort and coordination from the infant. The mammary gland is drawn deep into the child's mouth where it expands and pushes agains the palate which then hardens over time. Additionally, jaw compression is needed and masseter muscles develop as consequence (D'Onofrio, 2019). Thus, breastfeeding affects facial formation and coordination.

Generally, data is sparse; only seven studies were identified for meta-analysis investigating breastfeeding and risk for development of dental misalignments (Doğramacı et al., 2017). Research for a strong correlation is hampered by inconsistent terminology and methodology (Peres et al., 2018). The authors point out that little alignment exists between definitions on breastfeeding practices e.g. night feeding or dental markers that could help improve insights into the subject.

Breastfeeding mothers are less likely to get sick from cancers in the breasts or ovaries (Victora et al., 2016) (→), a similar effect on different cancer types could not be confirmed for formerly breastfed adults. Some reports state a 15 ‑ 20% risk reduction for childhood leukemia associated with a breastfeeding duration of six months or longer (AAP, 2012). Other analyses are limited to case studies and retrospective reports: These indicate that a breastfeeding duration for six months or longer may provide slight protection against childhood leukemia. Firm conclusion could not be drawn because the evidence was deemed insufficient (Güngör et al., 2019a).

Synonyms: trophic stimuli, diabetes mellitus type 2, T2DM, T1DM, DMT1, DMT2, IBD