Article Text

other Versions

Cohort study
Excess weight gain in the first 18 months of life is associated with later childhood overweight, obesity and greater arterial wall thickness
  1. Nicolas Stettler1,
  2. Madeleine Baumann2
  1. 1Exponent, Washington, DC, USA
  2. 2International Affairs, Haute Ecole de la Santé La Source, Lausanne, Switzerland
  1. Correspondence to: Dr Nicolas Stettler, Exponent, 1150 Connecticut Avenue NW, Suite 1100, Washington, DC 20036, USA; nstettler{at}

Statistics from

Commentary on: OpenUrlAbstract/FREE Full Text

Implications for practice and research

  • The association between early postnatal weight gain and later childhood obesity suggests that programmes providing culturally tailored, universal, affordable or free support to families of newborns and infants are essential.

  • Nursing care should emphasise the importance of optimal, rather than maximal, infant growth.

  • Nursing research, in particular qualitative methods, could strengthen understanding of infant feeding practices and theory-based behaviour modification research could explore non-judgemental and culturally tailored interventions.


Ischaemic heart disease and stroke are the leading causes of death globally with a rapidly increasing incidence in low and intermediate income countries.1 Intrauterine growth retardation (IUGR) is associated with an increased risk of metabolic and cardiovascular disease (CVD) in adulthood. As IUGR prevention is difficult to achieve, the causality of this association is difficult to establish in humans and the practical implications for prevention are limited. Rapid infancy weight gain has been recognised as a possible modifiable risk factor for the development of obesity,2 but the impact on CVD has been less extensively studied. The aim of the study undertaken by Skilton and colleagues was to assess a possible association between rapid infancy weight gain and early markers of CVD among Australian children.


The findings reported were based on a secondary analysis of data from an observational birth cohort study of 395 participants, with infants followed up to 8 years of age. The weight and length of the infants at birth were extracted from birth records; height and weight were measured at 18 months and 8 years of age with standardised methods; blood pressure and carotid ultrasounds for intima-media thickness (IMT) were measured at 8 years of age. A fasting blood sample was also taken from a large subsample of participants to measure total cholesterol, high-density lipoprotein cholesterol and high-sensitivity C reactive protein (hsCRP). Data were analysed using multivariate linear regression to determine the association between early infant weight gain and parent life style factors. Data collected at 8 years of age were analysed using Pearson's correlation, multivariate linear regression and logistic regression to determine the association between early infant weight gain and cardiovascular risk.


Male gender, early birth, birth length and formula feeding were positively associated with infancy weight gain in analyses unadjusted and adjusted for length. Length-adjusted weight gain was positively associated with overweight status as were high blood pressure, elevated hsCRP and larger carotid IMT at age 8 years before and after adjustment for confounding factors. The associations identified are clinically significant; a 2 kg difference in length-adjusted infancy weight gain was associated with an increase in IMT comparable to that associated with type 1 diabetes. The associations were in part explained by weight status. Lipid profile was not associated with infancy weight gain.


This is a well-conducted study using appropriate methods. A few limitations are identified by the authors. First, participants were children at increased risk for asthma because this was an inclusion criterion for the original study. Second, a 36% loss of participants to follow-up was observed between birth and 8 years of age. This could have led to a biased estimate of the association if the associations differed between completers and children lost to follow-up. Third, there was potential for residual confounding factors, such as family income, family eating or physical activity patterns. Finally, birth weight and length were extracted from hospital records; yet the clinical measurements of birth length are notoriously imprecise. However, these limitations are unlikely to provide a complementary explanation to the findings so generalisability to other populations is likely.

An important distinction, not extensively discussed by the authors, is related to the difference between unadjusted infancy weight gain and length adjusted infancy weight gain. The first is a reflection of energy balance, regardless of whether the growth occurs in length or in adiposity. The second reflects weight gain mathematically maintaining the length constant, thus reflecting mostly adipose tissue deposit. Physiologically, the two are markers of different hormonal mechanisms, regulations and possibly long-term consequences; therefore, they should be clearly differentiated in these types of cohort studies.

Since postnatal growth is likely to be more modifiable than prenatal growth, the findings of this study strengthen the prospect that CVD could be prevented over the life course, rather than only in the years preceding these events. If randomised intervention trials confirm these observational findings, it could be that short interventions during critical periods have important long-term benefits for the major global public health problem caused by CVD.

View Abstract


  • Competing interests None.

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.