Asthma is a major public health issue predominantly facing developed countries, including Australia [1
]. It is the most common chronic disease among children [2
]. The prevalence of asthma in Australian children is amongst the highest in the world [3
]. The worldwide prevalence of childhood asthma has been increasing considerably in the last few years [4
]. Explanations for this increase are not yet clear, thus limiting the opportunities to develop targeted primary prevention measures [4
According to hygiene hypothesis, changes of lifestyle in industrialized and western countries, in combination with limited exposure to microorganisms such as viruses and bacteria during early childhood, increase the susceptibility to allergic diseases including asthma [5
]. Children raised in developed cities like those in Australia are exposed to indoor and outdoor allergens and irritants during their early years of life. This may have tipped the balance of the child’s immune system from a TH1 (no asthmatic) to a TH2 (asthmatic) phenotype response [6
Clinically, asthma is defined as a chronic inflammatory disorder of the lungs. It is characterized by recurrent episodes of wheezing, shortness of breath, tightness of chest, and coughing, associated with limitation of airflow in the respiratory system [7
]. Asthma and other allergic diseases are often linked to the immune system and inflammation response. This is because the immune system is thought to be a regulator of asthma and airways inflammations by producing immune factors in response to stimuli and foreign bodies [8
Defining asthma through testing lung function is challenging before the age of 5 [9
]. Nonetheless, several studies monitored early signs of asthma during the first 3 years of the infant’s life to test for the impact of various factors on development of asthma in later life [10
]. Several studies documented a number of factors that may increase the likelihood of developing asthma, like maternal asthma [13
], young maternal age [14
], low level of mother’s education [15
], Aboriginality [16
], child sex [17
], maternal smoking during pregnancy [18
], early gestational age [19
], caesarean section [20
], prematurity [21
], low birth weight [22
], the existence of eczema [23
], and food allergies [24
Infant feeding is an important early-life exposure that may influence the development of asthma [7
]. The components of breast-milk stimulate the infant’s own immune system, safeguarding them against the development of allergic disease [7
]. Although breastfeeding provides health benefits against respiratory infections and maturation of the infant’s immunity [25
], findings of many studies assessing the role of breastfeeding in decreasing the risk of asthma development have been controversial [7
]. Moreover, Milk Other than Breast Milk (MOTBM) is usually introduced early in infant life for many reasons [29
], and the link between the early introduction of MOTBM and the development of Asthma has yet to be discovered.
Most infant formulas are derived from standard cow’s milk, which contains powerful food allergens [30
]. There is some evidence of association between infant consumption of cow’s milk in the initial few months of life and the development of asthma [30
]. Numerous studies were conducted on MOTBM feeding and its effect on development of Asthma [30
]. These studies, however, were focusing on comparing the effects of various types of MOTBM, such as cow milk, goat milk, soymilk or hydrolyzed milk, on increasing or decreasing the risk of development of Asthma relative to each other. Some results showed that certain types of these MOTBM, such as hydrolyzed formula, reduce the risk of developing asthma [30
]. However, the effect of any type of MOTBM in increasing the risk of development of Asthma relative to breast milk was not studied.
Hence, the aim of our study is to assess whether the introduction of any type of MOTBM during the first 6 months after birth increases the risk of later development of asthma among 3-year old children.
All screened studies compared the effects of different types of MOTBM on increasing or decreasing the risk of asthma; however, all of these studies compared the effects of different types of MOTBM relative to each other. To our knowledge, this research is the first to assess whether MOTBM, regardless of its source, increases the risk of development of asthma.
There are three major types of infant formulas: standard cow’s milk-based formulas, soy-based formulas, and hypoallergenic formulas (hydrolysed cow’s milk formulas) [41
]. Previous studies assessing standard cow-based formulas found that these formulas were not protective against asthma during infancy or childhood [30
]. Similarly, soy-based formula did not reduce the risk of asthma when compared with standard cows-based milk formula in infants and children [30
]. The literature evidence supporting the preventive effects of hydrolysed infant formulas for asthma is inconsistent and insufficient. Some studies suggest that certain partially- or extensively-hydrolysed formulas may reduce the risk of asthma compared to non-hydrolysed formulas for children with a family history of atopic disease [30
]. In contrast, it was reported by others that infants who received hydrolysed cow’s milk formula did not have a lower risk of asthma compared with those who received human breast milk or standard cow’s milk formula [31
Although several studies examined the association between different infant formulas and asthma [30
], this study uniquely distinguished between breast milk and any type of MOTBM. Our results revealed that in comparison with infants who received breast milk only, those who received MOTBM during the first 6 months after birth had a 71% increased risk of persistent asthma at the age of 3 (ARR: 1.71, 95% CI: 1.03–2.83, p
= 0.038). This association was independent of established maternal and environmental risk factors.
Previous findings regarding duration of breastfeeding and the risk of asthma are contradictory. Although several studies showed that breastfeeding prevents asthma [45
], there were also several disagreeing studies [26
]. This study found no significant association between duration of breastfeeding up to 6 months and development of persistent asthma in the first 3 years of life (RR: 1.13, 95% CI: 0.70–1.83). This finding is consistent with previous studies carried out in Denmark and Sweden, which have shown that breastfeeding does not lower the risk of childhood asthma (OR: 1.08, 95% CI: 0.93–1.25) and (OR: 0.99, 95% CI: 0.96–1.02) respectively [47
Gender appears to be a factor in the development of asthma, i.e., boys in general are reported to have more severe asthma than girls [49
]. This was suggested to be the result of smaller airway diameters relative to lung volume in boys compared to girls [50
]. Consistent to what was observed in these studies [49
] more males (10.4%) than females (3.4%) in our study were diagnosed with persistent asthma with strong association between sex differences and development of asthma (p
-value < 0.001).
There is evidence suggests the existence of a complex relationship between eczema and asthma. Research found that the majority of infants with early eczema develop asthma in childhood; an example of the so-called “atopic march” [23
]. Researchers suggested that understanding the relationship between eczema and asthma might help in preventing asthma developing in these vulnerable children, i.e., to prevent the atopic march. Demehri and his colleagues reported the existence of many theories about the link between eczema and asthma. One theory suggests that the impairment of skin, which is body’s protective external barrier, might stimulate the immune system to over-react to any potential allergen present in the body, including the surface of the airways in the lungs [53
]. In line with several previous studies [23
], we found that eczema in infancy was strongly associated with the development of asthma in children at the age of 3. Multivariable analyses showed that existence of eczema at baseline increased the risk of developing asthma by about two-fold (RR: 2.15, 95% CI: 1.15–4.04, p
= 0.017) relative to children without eczema at the beginning of the study.
Maternal smoking during pregnancy has been consistently associated with early childhood asthma [56
]. Passive fetal exposure to tobacco smoke during pregnancy has been reported by numerous studies to result in reduced lung function in the early months of infant’s life, with a dose–response relationship between exposure and decreased airway calibre in early life [58
]. In our study, there was a strong association between smoking during pregnancy and the development of childhood asthma, indicative of a dose response relationship. Children who were exposed to high level of tobacco smoke of more than 10 cigarettes/day had higher risk for development of asthma (ARR: 4.23, 95% CI: 1.91–9.37, p
< 0.001 for test for trend). These results are consistent with what was previously reported by Tariq et al. [57
], where the authors found that smoking during pregnancy was associated with the development of childhood asthma at two years of age (RR: 2.2, 95% CI: 1.5–3.4, p
In our study, young maternal age at delivery was also associated with an increased risk of asthma in three-year old children. This finding is consistent with previous studies conducted in United Kingdom and Finland [7
]. Additionally, the 31–35 age group was also associated with an increased risk of asthma in children. Including age in its continuous form in the final model gave a RR = 0.95 with 95% CI: 0.91–0.99 (p
= 0.036). This agrees with previous studies relating young maternal age with increased risk of asthma irrespective of the weak effect size.
As demonstrated by several studies [13
], maternal asthma was also associated with an increased risk of childhood asthma. Inconsistently, our univariate analysis did not show any such association in three-year old children. It is worth mentioning that maternal asthma was not included in our multivariable analysis as the variable regarding this issue was asked only to QLD state participants. It has been hypothesized that caesarean section might increase the risk for developing asthma compared to vaginal delivery, as a result of depriving the newborn of exposure to maternal microflora [60
]. A number of studies found that caesarean section might have a positive association with development of asthma [60
]. In the present study, we did not observe similar associations. Researchers argue that mothers with lower levels of education receive limited social support and may have decreased access to health information. Overall, this increases the exposure to risk factors that affects mothers, and consequently, their children’s health negatively [15
]. Our built multivariable model did not include maternal education as risk factor for development of asthma although it was a risk factor in the univariate analysis.
There are several limitations of this study that are consistent with biases likely to arise from cohort studies. Firstly, the recruitment was through three public maternity hospitals only in the participating districts (Logan, Gold Coast and the Tweed Hospitals) limiting the generalizability in other populations. As children included in the study were offspring of women who planned to give birth in one of these hospitals, there is a potential for selection bias. Secondly, this study has a relative low follow-up rate (49%) naturally stemming from the original EFHL study. The systematic differences between the characteristics of children remaining in the study and those lost to follow-up are likely to introduce bias. Thirdly, the use of self-report questionnaires to collect the information from primary carers may lead to recall bias as evidenced in other cohort studies [27
]. Lastly, the outcome, medically-diagnosed asthma was addressed by a question asked at both waves 1 and 2. As the question was based on information reported by the carer, there is possibility that it does not accurately predict medically-diagnosed asthma in children. Moreover, asthma was medically diagnosed among children less than 3 years of age; this study acknowledges the challenges of such diagnosis before the age of 5 [9
]. Similarly, there is a possibility of misclassification of the exposure variable, MOTBM at less than six months after birth. All these limitations may lead to imprecise risk estimation.
The strengths of this study lie in its large sample size, its prospective design, and the use of carefully constructed questionnaires to maximize accuracy and completeness through asking specific questions, which decreased recall bias of results. In addition, asthma was diagnosed by a physician and not by parental self-report of symptoms.
In summary, the current study indicates that the introduction of MOTBM during the first six months after birth is a risk factor for asthma in three-year old children. Although more studies and further analyses are needed to confirm these findings and to better understand the underlying mechanisms, public health interventions promoting the risk of early introduction of MOTBM and encouraging mothers to refrain from using it during the first six months after birth may help in reducing the morbidity and prevalence of childhood asthma. In support of previous research, the current data did not show a correlation between breastfeeding and protection against asthma.