PEDIATRIC ORIGINAL ARTICLE Weight and weight gain during early infancy predict childhood obesity: a case-cohort study LG Andersen 1 , C Holst 1 , KF Michaelsen 2 , JL Baker 1 and TIA Sørensen 1 BACKGROUND: Infant weight and weight gain are positively associated with later obesity, but whether there is a particular critical time during infancy remains uncertain. OBJECTIVE: The aim was to investigate when and how weight and weight gain during infancy become associated with childhood obesity. METHODS: In a cohort representing 28 340 children born from 1959–67 and measured in Copenhagen schools, 962 obese children (2007 World Health Organization criteria), were compared with a 5% randomly selected sub-cohort of 1417 children. Information on weight at birth, 2 weeks, 1, 2, 3, 4, 6 and 9 months was retrieved from health visitors’ records. Odds ratios and 95% confidence intervals (CI) for childhood obesity by tertiles of weight at each age and by change in tertiles of weight between two consecutive measurements were estimated using multivariate logistic regression with adjustment for indicators of socioeconomic status, preterm birth, and breastfeeding. RESULTS: Compared with children in the middle weight-tertile, children with a weight in the upper tertile had a 1.36-fold (CI, 1.10–1.69) to 1.72-fold (CI, 1.36–2.18) higher risk of childhood obesity from birth through 9 months, whereas children in the lower weight-tertile had almost half the risk of obesity from 2 through 9 months. The risk of childhood obesity associated with change in weight-tertile in each interval was stable at B1.5-fold per weight-tertile increase throughout infancy. CONCLUSIONS: Infant weight and weight gain are associated with obesity in childhood already during the first months of life. Determinants of weight gain shortly after birth may be a suitable target for prevention of obesity. International Journal of Obesity (2012) 36, 1306–1311; doi:10.1038/ijo.2012.134; published online 21 August 2012 Keywords: birth weight; breastfeeding; infant; weight gain; childhood obesity INTRODUCTION An obesity epidemic is rapidly evolving all over the world 1 bringing with it short- and long-term adverse consequences of premature morbidity and mortality. 2 Early life factors may contribute to the risk of developing obesity, and increasing attention has been given to potential prevention strategies that begin early in life. 3,4 Rapid weight gain in the first year of life has consistently been associated with an increased risk of obesity in childhood and later life. 5–9 A recent individual-level meta-analysis found a twofold higher risk of childhood obesity per increase in weight s.d. score from birth to 1 year, 6 but which age periods during infancy are most important remains uncertain. 5–9 A study of exclusively formula-fed infants suggested that weight gain already during the first postnatal week is associated with increased risk of adulthood overweight. 10 However, the majority of studies have not investigated monthly weight measurements in the infancy period, and various methodological issues in the studies with repeated measurements make the interpretation difficult. 5,7,8,10 Identification of the most sensitive age window is needed to elucidate the mechanisms 11 and to target possible prevention strategies. 12 We investigated when and how weight and weight gain during infancy become associated with childhood obesity in a large general population of schoolchildren. SUBJECTS AND METHODS Study population The source population for this study was The Copenhagen School Health Records Register (CSHRR), which contains information on 372 636 children, born during the years 1930–1989, who underwent mandatory health examinations in schools in the Copenhagen municipality. 13 The records include measurements of height and weight routinely performed by school health doctors and nurses, and birth weight reported by the mother at school entry. 13 Information about infancy was collected from records made by infant health visitors who examined the children and provided guidance on nutrition and baby care through free home visits in the Copenhagen municipality. 14 The original records from these visits have been kept for B80 000 of infants born from 1959–67. The present study was approved by the Danish Data Protection Agency. The study employed a case-cohort sampling design with cases defined as obese children in the CSHRR. For most children born 1959–67, annual measurements of height and weight at the school health examinations were available. Cases were children who from 7 through 13 years of age had at least one body mass index (BMI; kg m À 2 ) value exceeding two s.d. scores of the 2007 World Health Organization (WHO) Growth Reference. 15 Of 42 049 children, 1410 children (3.3%) were identified as being obese at least once at any of these ages, of whom 365 were obese at 7 years and 307 at 13 years. Infant health visitor records were retrieved for 962 (68%) of these children. For comparison with the case group, a sub-cohort corresponding to 5% of the children in the CSHRR (n ¼ 2081) was randomly selected, among whom infant health visitors’ records were retrieved for 1417 children (68%). Records were not retrieved for all children mainly 1 Institute of Preventive Medicine, Copenhagen University Hospital, Copenhagen, Denmark and 2 Department of Nutrition, Exercise and Sports, Faculty of Science, University of Copenhagen, Copenhagen, Denmark. Correspondence: LG Andersen, Institute of Preventive Medicine, Frederiksberg Hospital, Hovedvejen 5, Nordre Fasanvej 57, 2000 Frederiksberg, Denmark. E-mail: lg@ipm.regionh.dk Received 30 March 2012; revised 27 June 2012; accepted 11 July 2012; published online 21 August 2012 International Journal of Obesity (2012) 36, 1306–1311 & 2012 Macmillan Publishers Limited All rights reserved 0307-0565/12 www.nature.com/ijo