Early nutrition and overweight
A number of studies have examined early nutrition, including breastfeeding, and associations with obesity with conflicting results( Reference Watson 1 – Reference Auld 9 ). Differences in findings are likely due to a lack of control for confounding variables, small sample size, socioeconomic status and other differences such as subject age at time of measurement and family structure( 10 ). Some studies have found a positive, dose-dependent association between breastfeeding and weight status( Reference Kramer 11 – Reference Harder, Bergmann and Kallischnigg 13 ), while others have reported a nonlinear positive response( Reference von Kries, Doletzko and Sauerwald 14 – Reference Gillman, Rifas-Shiman and Camargo 16 ). On the contrary, other studies have demonstrated little or no protective association between breastfeeding and overweight; including a longitudinal study of infant feeding and obesity which found that exclusive breastfeeding was not related to being overweight or obese during adult life( Reference Toschke, Martin and von Kries 17 – Reference Michels, Willett and Graubard 21 ). Similarly, a systematic analysis conducted by the WHO( Reference Horta, Bahl and Martines 22 ) and a systematic review of the literature found that after adjustment for maternal overweight the effect of breastfeeding was eliminated( Reference Owen, Martin and Whincup 23 ). It has been suggested that breastfeeding alone does not protect children from overweight but rather other healthy lifestyle habits or other social advantages of mothers who chose to breastfeed( 10 ). While the relationship between breastfeeding and obesity has not been fully elucidated, it is recommended by the WHO and Cochrane Review that children are exclusively breastfed for 6 months due to other known benefits and this is supported by the ESPGHAN Committee on Nutrition, which recommended children be exclusively breastfed for about 6 months( Reference Agostoni, Braegger and Decsi 24 – Reference Kramer and Kakuma 26 ).
Recently, our group examined exposure to exclusive breastfeeding and overweight in the Identification and prevention of Dietary- and lifestyle-induced health EFfects In Children and infantS (IDEFICS) study. IDEFICS is a multi-centre European study involving eight countries: Italy, Estonia, Cyprus, Belgium, Sweden, Hungary, Germany and Spain. Community-based survey centres in each of the participating countries collected information when children were age 2–9 years from September 2007 until June 2008 at survey centres that cannot be considered country representative. In total, over 16 000 children participated in IDEFICS( Reference Ahrens, Bammann and Siani 27 ). From the IDEFICS population, children were included in the first study if their exposure to exclusive breastfeeding could be defined( Reference Hunsberger, Lanfer and Reeske 28 ). After exclusions, the total sample size in our exclusive breastfeeding study was 14 726 children with sample sizes across countries ranging from 1399 in Spain to 2431 in Hungary.
Characteristics by survey centre have been previously published in detail( Reference Hunsberger, Lanfer and Reeske 28 ). In brief, the prevalence of overweight was 19·4 %, ranging from a high of 42·3 % in Italy to a low of 8·3 % in Belgium at baseline data collection in IDEFICS. Children were aged 2–9 years at the baseline survey with a mean age of 6 years and were analysed in two age groups of similar size, pre-school age (2 to <6 years) and school age (6–9 years). The gender distribution is nearly equal across survey centres.
When all centres are combined, <20 % of IDEFICS children were never exclusively breastfed for at least some period although there are distinct differences across the countries. For example, in Belgium 46·1 % of the children were never breastfed whereas in Estonia only 9·9 % of the children were never breastfed( Reference Hunsberger, Lanfer and Reeske 28 ). The mean duration of exclusive breastfeeding was 3·2 months across all survey centres.
Exclusive breastfeeding and overweight/obesity combined were examined with adjustment for survey country, child age, sex and birth weight, included household income, maternal education, maternal overweight/obesity, single or dual parent family structure, the presence of one or more foreign born parents, and tobacco use during pregnancy. In our fully adjusted model, breastfeeding exclusively for 4–6 months was protective of overweight (including obesity) when compared with children never exclusively breastfed (OR 0·73, 95 % CI 0·63, 0·85). In addition, exclusive breastfeeding for 6 months offered slightly more protection than 4 and 5 months combined (OR 0·71, 95 % CI 0·58, 0·85). Findings were consistent between boys and girls. The associations could not be explained by socioeconomic characteristics or maternal overweight.
Since the use of BMI as an appropriate means of assessing childhood weight-related health risks is not clear( Reference Flegal and Ogden 29 ), our group further assessed overweight/obesity with skinfold measures and waist∶height ratio. A logistic regression model and generalised linear regression model respectfully demonstrated waist∶height ratio and per cent body fatness assessed by skinfold measure both agreed with our initial logistic regression findings and support our use of BMI as a binary outcome representing overweight children. Comparing obesity prevalence by BMI categories with waist : height ratio, there was 86·19 % agreement among normal weight children and 67·25 % among overweight/obese. Agreement by skinfold assessment was slightly greater, in the normal weight BMI category there was 86·9 % agreement and in the overweight/obese 75·95 % agreement.
More recently, a novel analysis of variables predictive of adult obesity was carried out by the Foresight Obesity System Map group; a project of the UK government think-tank Foresight( Reference Butland, Jebb and Kopelman 30 ). A two-way framework of past/fixed and present/modifiable factors are presented for a sample (n 11 752)( Reference Potter and Ulijaszek 31 ). Breastfeeding for any amount of time and normal birth weight were included as positive conditioning factors or factors that occurred in the past that cannot be changed( Reference Potter and Ulijaszek 31 ). Individuals identified in the highest risk category at age 11 years were approximately two times as likely to become overweight by age 23 years and obese in their 30s and 40s as opposed to their low-risk peers( Reference Potter and Ulijaszek 31 ). These findings support the protective role of breastfeeding. Further, a systematic literature review that included sixty papers concluded the evidence of a protective dose/duration effect of breastfeeding against overweight and obesity in childhood and adolescence is convincing( Reference Hörnell, Lagström and Lande 32 ). The authors also concluded that there is probable evidence that exclusive breastfeeding for longer than 4 months is associated with slower weight gain from 6 to 12 months and may be the reason behind the reduced risk of later overweight and obesity( Reference Hörnell, Lagström and Lande 32 ). In light of these findings, there is good evidence that breastfeeding is protective against the development of overweight later in life.
Family structure and overweight
In addition to early nutrition, the role of family structure may also have important implications for future childhood overweight and obesity. However, the role of family structure in the development of obesity is not well elucidated. Singleton status was identified as a risk factor for obesity in one Norwegian population study( Reference Júlíusson, Eide and Roelants 33 ), and the research suggests that first-born children receive more ‘quality time’ per day than those born into a household with a sibling( Reference Price 34 ). Furthermore, the influence of social networks is likely an important component of overweight development( Reference Christakis and Fowler 8 , Reference Auld 9 ).
Our group investigated the role of family structure and overweight among the IDEFICS children across eight European countries. In this study, 12 720 children were included based upon completeness of family structure data( Reference Hunsberger, Formisano and Reisch 35 ). All survey centres obtained ethics approval from their respective authority. Parents or legal guardians provided written informed consent on behalf of their children for data collection and specimen collection and children gave oral consent. Detailed information about IDEFICS survey protocol has been previously published( Reference Ahrens, Bammann and Siani 27 ) as have the detailed methods this study( Reference Hunsberger, Formisano and Reisch 35 ).
While breastfeeding is protective of overweight, being an only child is not. IDEFICS children without siblings were more likely (OR 1·52, 95 % CI 1·34, 1·72) to be overweight than their peers with siblings when controlling for factors related to childhood overweight (parental education, parental weight, maternal age, child's age, birth weight and gender) and for survey country. In fact, the longer a child remains an only child, the stronger the association with overweight. As previously reported by our group, among older singletons the association with overweight (OR 1·70, 95 % CI 1·44, 2·01) was greater than for younger singletons (OR 1·32, 95 % CI 1·10, 1·60)( Reference Hörnell, Lagström and Lande 32 ). The potential behavioural mediators examined (playtime outdoors, screen time/d, propensity to consume sugar or fat, parental attitudes towards food rewards and television in the child's bedroom) did not attenuate the relationship between only children and overweight with associated OR ranging from 1·51 to 1·58 when each mediator was explored individually and in combination. In a sensitivity analysis, having a sibling was protective regardless of birth order (twins or those with older siblings v. only younger sibling) when only children were compared with those with siblings. Additional analyses suggested that children who have always had a sibling were slightly more protected from overweight than those that were previously a singleton until the arrival of a younger sibling but this finding was not statistically significant.
New findings from our group further demonstrate only children are at increased risk for overweight. Among the IDEFICS sample, children with the greatest number of siblings had the lowest BMI z-scores in cross-sectional analyses and over a 2-year follow-up period the risk of incidence of overweight/obesity was significantly lowered by the greater the number of siblings living in the household( Reference Formisano, Hunsberger and Bammann 36 ).
Limited research has examined the role of siblings in the home. Other studies have focused on family structure but have placed emphasis on the adults in the home rather than siblings. For example, a recent study in South Korea examined the relationship between child weight status by household structure but did not report on sibling status( Reference Eidsdóttir, Kristjánsson and Sigfúsdóttir 37 ). In contrast to the limited studies that have found a relationship between family structure and childhood weight status, an Icelandic research group reported no association between family structure and the prevalence of overweight and obesity among 16- to 20-year-old adolescents( Reference Lee, Duffey and Kim 38 ).
The present review highlights two factors that may be important to achieving a healthy weight both early in life and in the future. While overweight is multi-factorial in nature, exclusive breastfeeding and family structure are two factors that are likely to have an impact. Exclusive breastfeeding exposure was protective of overweight in our study, in part because we made a distinction between exclusive breastfeeding and partial or any breastfeeding. Our findings support the recommendation to exclusively breastfeed children for 6 months as do the findings of a recent systematic review( Reference Hörnell, Lagström and Lande 32 ). Breastfeeding policies that support women in the endeavour to exclusively breastfeed for 4–6 months, with 6 months being an optimal goal, may improve public health as many nations struggle to reduce childhood overweight prevalence. For a thorough review of the topic, see the systematic analysis conducted by the WHO( Reference Horta, Bahl and Martines 22 ) and two systematic reviews of the literature, the first includes both published and unpublished results and the latter only published results( Reference Owen, Martin and Whincup 23 , Reference Hörnell, Lagström and Lande 32 ).
Further, a limited body of research appears to demonstrate that only children are at increased risk for overweight. In our study, we found that only children were at risk for overweight and this risk was not attenuated by a number of potential behavioural mediators. The longer a child remains an only child in the household the stronger the association with overweight. This may have important parenting implications such that parents of only children may wish to limit screen time and encourage physical play time in addition to other lifestyle factors, such as feeding practices, that contribute to overweight. More research is needed to fully understand the associations between family structure and risk for overweight.
Our group will continue to explore many aspects of childhood overweight development and prevention in the continuation study I.Family. To learn more about IDEFICS or I.Family visit http://www.idefics.eu/Idefics/ and http://www.ifamilystudy.eu, respectively. I.Family is EC funded FP7 Project no. 266044.
The author thanks IDEFICS collaborators: Lauren Lissner, Annarita Formisano, Alfonso Siani, Gabriele Eiben, Anna Reeske, Anne Lanfer, Charalampos Hadjigeorgiou, Dénes Molnar, Karin Bammann, Lucia A. Reisch, Luis A. Moreno, Michael Tornaritis, Paola Russo, Stefaan De Henauw and Toomas Veidebaum.
The research presented in this work has been supported by the European community within the 6th RTD Framework Programme Contract No. 016181 (FOOD).
Conflicts of Interest
This proceedings paper is based upon work previously published by the author and the author's collaborators who are acknowledged. This work was completed by the presenting author on behalf of the IDEFICS Consortium.