Original Article

International Journal of Obesity (2010) 34, 67–74; doi:10.1038/ijo.2009.206; published online 17 November 2009

Gestational weight gain in relation to offspring body mass index and obesity from infancy through adulthood

L Schack-Nielsen1, K F Michaelsen1, M Gamborg2, E L Mortensen3 and T I A Sørensen2

  1. 1Department of Human Nutrition, Centre for Advanced Food Studies, Faculty of Life Science, University of Copenhagen, Frederiksberg C, Denmark
  2. 2Institute of Preventive Medicine, Copenhagen University Hospital, Centre for Health and Society, Copenhagen K, Denmark
  3. 3Institute of Public Health and Center for Healthy Aging, Faculty of Health Science, University of Copenhagen, Copenhagen K, Denmark

Correspondence: Dr TIA Sørensen, Institute of Preventive Medicine, Øster Søgade 18, 1st floor, DK 1357K. E-mail: tias@ipm.regionh.dk

Received 22 February 2009; Revised 28 July 2009; Accepted 10 August 2009; Published online 17 November 2009.





Gestational weight gain (GWG) is associated with childhood obesity. We analyzed whether this effect persists into adulthood and is mediated by effects in childhood.



The design of the study a prospective birth cohort study established in 1959–1961.



The subjects were offspring (n=4234 of whom 2485 had information from the last follow-up) of mothers included in ‘The Copenhagen Perinatal Cohort’ during pregnancy or at birth.



Information on maternal pre-pregnancy body mass index (BMI), GWG and several potential confounders were collected around delivery. Information on offspring BMI was available from various follow-up examinations from 1 to 42 years of age. The association of GWG with offspring BMI was analyzed by regression models including confounders. Using path analysis, the association of GWG with adult BMI was disentangled into an association mediated through childhood BMI and one independent hereof.



GWG was associated with offspring BMI at all ages. At the age of 42 years (n=1540), there was an increasing risk of obesity (odds ratio (OR) 1.08, 95% confidence interval (CI) 1.03–1.14 per kg GWG, P=0.003). Only half of the association of GWG on offspring adult BMI was mediated through birth weight and BMI up to 14 years of age.



Greater GWG is associated with an increased BMI in childhood through adulthood and with an increased risk of obesity in adults. Only part of the association with adult BMI is mediated by childhood BMI, suggesting that excessive GWG induces a persisting susceptibility to obesogenic environments. As GWG is greater in women with small pre-pregnancy body weight, this implies a reinforcement of the obesity epidemic in the next generation. Our findings provide support for avoiding excessive GWG.


gestational weight gain, maternal obesity, birth weight, childhood obesity, programming



The obesity epidemic among adults seems to be preceded by an increased prevalence of obesity among preschool children,1 and the fetal period may be of importance.2, 3 Early growth restriction has received considerably interest as a risk factor for later cardiovascular disease, but in relation to obesity, fetal overnutrition seems a more concerning risk factor. It is well established that a high birth weight is associated with a higher body mass index (BMI) or risk of obesity in later life.4 Furthermore, maternal overweight before pregnancy and maternal diabetes are associated with offspring obesity, possibly due to genetic effects, but perhaps also because of the intrauterine environment.5, 6, 7, 8 This trans-generational association implies that the maternal obesity may, by its effect on the offspring, induce a viscous cycle accelerating the obesity epidemic in successive generations.7, 9, 10 Such acceleration has been observed in the stepwise development of the obesity epidemic in Denmark, where the second phase was much steeper than the first phase.1 Furthermore, a high gestational weight gain (GWG) has recently been associated with an increased risk of overweight in children aged 2–14 years11, 12, 13, 14, 15 and in 21-year olds,16 after adjusting for maternal BMI. Thus, recommendations for GWG may also need to take into account the long-term effects on the offspring.17 We analyzed whether GWG is associated with offspring BMI in childhood through adulthood, and whether the possible effects on adult BMI are direct or indirectly mediated through effects on BMI in earlier life.


Patients and methods

The Copenhagen perinatal cohort

The Copenhagen perinatal cohort consists of 9125 individuals born at the Copenhagen University Hospital from 1959 to 1961.18 The hospital preferentially admitted mothers for whom home delivery was ‘inadvisable’ (primarily single mothers) and mothers with current or previous pregnancy or delivery complications. One physician (Dr Aage Willumsen) was responsible for collecting information on the mother's social, general medical and obstetrical history. He interviewed all mothers 5 days after births, and 67% of the mothers at their first visit to antenatal clinic. Of the mothers included in the current study (n=4234), 43% had no complications of the current pregnancy, 29% had anemia, 23% had edema and 25% had at least one of the complications: hypertension, proteinurea and pre-eclampsia, whereas other complications were rare. The parents were invited for an examination of the children when aged 1, 3, and 6 years, carried out at the hospital by Dr Bengt Zachau-Christiansen.18

Follow-up examinations

Information on height and weight (7–14 years) was obtained from mandatory school health examinations for those attending school in the municipality of Copenhagen,1, 4 and for the men from the mandatory draft board examinations.19 At the age of 20–34 years, a subsample participated in a research program on the effect of prenatal exposure to maternal prescribed medication.19 At the age of 41–43 years (referred to as 42 years), 98% of the study participants surviving infancy were still alive and traceable, and they were sent a questionnaire on lifestyle factors and health. This included weight, height and waist circumference measured with an enclosed tape measure, and the response rate was 54%. The follow-up examinations were approved by the local ethical committees, and participants gave written informed consent.


Gestational weight gain was reported as <6, 6–8, 9–10, 11–12, 13–15 or >16kg, and these categories were assigned the values of 5.5, 7.0, 9.5, 11.5, 14.0 and 16.5kg, when GWG was included as a linear variable. Information on how GWG data were obtained was not available, but it was most likely collected from medical records. Potential confounders (based on preliminary analyses) adjusted for were: sex, maternal age (years), pre-pregnancy BMI (kgm−2), parental social class (1–8 point scale, 8 is the highest), breadwinner's education (1–4 point scale, 4 is the highest), single-mother status (yes/no), maternal smoking, edema during pregnancy and preterm birth (yes/no, defined as a gestational age less than or equal to37 weeks due to the original categorization of the variable). Owing to uncertainty about the recorded gestational age, we primarily used prematurity (yes/no) in the analyses, but the findings were similar if gestational age in weeks were used (data not shown). Birth weight was considered to be a potential mediator of the effect of GWG on offspring BMI, and hence analyses were conducted with and without birth weight. On the basis of the preliminary analyses, the following variables were not considered to be potential confounders: parity, duration of breast feeding, age at introduction of complementary foods and, for the female offspring, whether they have had children.

Body mass index from 1 to 42 years, overweight (BMI greater than or equal to25kgm−2) and obesity (BMI greater than or equal to30kgm−2) at the age of 42 years were the outcome variables. The age intervals at the preschool follow-up examinations were broad and only measurements performed within the age intervals of greater than or equal to10–<14 months, greater than or equal to2.5–<3.5 and greater than or equal to5.5–<6.5 years were included and referred to as the 1-, 3- and 6-year examinations, respectively. From the school health records, BMI from age greater than or equal to6.5–<7.5 years was included as a 7-year measurement and so on. All childhood BMI measurements were transformed to BMI Z-scores (BMI-Z), using the British 1990 growth references20 for gender and exact age. For adults (20–34 and 42 years), internal sex-specific BMI-Z was created to compare the results with those obtained in children. Analyses, including BMI at the draft board (16–23 years) and the first adult follow-up examination (20–34 years), were adjusted for current age.

Subjects available for analysis and sample attrition

The original cohort consisted of 9125 infants of whom 996 were stillborn, died during infancy, were from twin or triple births and were excluded. (The high rates were due to the hospitals admission criteria and included infants born down to a birth weight at 250g). Included in the analyses were individuals (n=4234) with information on GWG, birth weight, maternal BMI and at least 1 BMI measurement from 1 year of age or older. The included subsample did not differ from the remaining cohort with respect to BMI from 1 to 20–34 years of age (data not shown), but at the age of 42 years the included women had a slightly lower BMI (24.3±4.7 vs 24.9±5.3kgm−2, P<0.001), whereas no difference was observed for the men (25.7±3.7 vs 25.8±3.8kgm−2), P=0.585). Adult BMI (42 years) was considered to be the most important outcome of the study, and within the male part of the study, the availability of information on this outcome was associated with a higher parental social class, whereas no association was observed for other variables including GWG, birth weight and BMI in earlier life, and for any variables pertaining to the women (Table 1).

Statistical analyses

At all ages, all individuals with the BMI measurement were included, and the effect of GWG on offspring BMI-Z was explored in models with and without adjustment for covariates. Preliminary analyses indicated GWG to be included as a linear variable at all ages. Logistic regression was used to explore the effect of GWG on the risk of adult (42 years) overweight and obesity. The regression analyses were conducted using SPSS (version 15.0, SPSS Inc., Chicago, IL, US).

Path analysis21 was used to divide the effect of GWG on adult BMI-Z (42 years) into an indirect effect mediated through body size in childhood (birth weight and BMI-Z (1–14 years)) and brought into adulthood by tracking, and a direct effect that cannot be explained by known earlier body size. The path analysis implies simultaneous estimation of regression models for all intermediate measurements (birth weight and BMI-Z from 1 to 14 years) and for the outcome variables (adult BMI-Z). The regression models included the effect of GWG, covariates and all previous growth measurements as independent variables, for example, the 7-year BMI-Z was regressed on GWG, covariates, birth weight and BMI-Z at 1, 3, and 6 years. The final regression was adult BMI-Z on GWG, covariates, birth weight and all measurements of BMI in childhood. The model was reduced to give a good fit by removing all nonsignificant effects, except the effects of covariates and the effect of GWG.



Table 1 shows baseline characteristics. The prevalence of overweight among the mothers was 9%, and among the offspring it was 3–5% during school years (International Obesity Task Force criteria22), and 33 and 53% in adulthood for women and men, respectively.

Associations between GWG, offspring adult BMI and covariates

As shown in Tables 2 and 3, several covariates were associated with both GWG and offspring adult BMI (42 years). High maternal pre-pregnancy BMI was associated with a low GWG (<6kg), and with a high risk of adult obesity in the offspring. High parental social class in early life was associated with an intermediary GWG (11–12kg) and a decreased risk of overweight and obesity in adult offspring.

Regression analyses

In logistic regression analyses adjusting for birth weight and covariates, GWG was positively associated with the risk of offspring adult (42 years) obesity (odds ratio (OR) 1.08, 95% confidence interval (CI 1.03–1.14, P=0.003 per kg GWG, n=1540), and the same trend was observed for overweight (OR 1.03, 95% CI 1.00–1.06, P=0.095, n=1540; Figure 1). This amount to a 2.36 (1.08–5.15)-fold increase in the risk of obesity and a 1.28 (0.89–1.85)-fold increase in the risk of overweight in the highest (greater than or equal to16kg) versus the lowest (<6kg) GWG category. The effect of GWG persisted when restricting the sample to those whose mothers did not have pregnancies complicated by edema, diabetes mellitus type 1, hypertension, pre-eclampsia and proteinurea; the risk of obesity was 1.09 (1.03–1.16, P=0.006, n=1127) and the risk of overweight was 1.04 (1.00–1.08, P=0.058) per kg GWG.

Figure 1.
Figure 1 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

The risk of offspring adult (42 years) overweight (body mass index (BMI) greater than or equal to25kgm−2) or obesity (BMI greater than or equal to30kgm−2) in relation to gestational weight gain. Logistic regression analyses adjusted for sex, maternal age, maternal pre-pregnancy BMI, parental social status, breadwinner's education, single-mother status, prematurity, birth weight, edema and smoking during pregnancy (n=1540).

Full figure and legend (19K)

In linear regression analyses (Figure 2), GWG was positively associated with offspring BMI-Z at all ages, and the associations showed no notable changes by adjustment for confounders, despite sample reduction due to missing values. Further adjustments for birth weight attenuated the effect of GWG, but it remained significant or borderline significant at most ages. Birth weight was significantly positively associated with offspring BMI-Z from 1 to 14 years (data not shown), and the attenuation of the effect of GWG on offspring BMI by inclusion of birth weight to the model was strongest during childhood. In an adjusted model (n=1540), adult BMI increased 0.10 (95% CI 0.04–0.17) kgm−2 per kg gain in GWG.

Figure 2.
Figure 2 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Regression analyses with gestational weight gain as dependent variable and offspring body mass index (BMI) Z-scores at different ages as independent variables: (a) Adjusted for sex (n varied from 699 (7 years) to 2577 (1year)). (b) Adjusted for sex, maternal age, maternal pre-pregnancy BMI, parental social status at birth, breadwinner's education, single-mother status, prematurity, edema and smoking during pregnancy (n varied from 434 (7 years) to 1891 (1year)). (c) Adjusted for birth weight in addition to the variables mentioned in (b) (n varied from 434 (7 years) to 1891 (1 year)).

Full figure and legend (38K)

As expected, maternal pre-pregnancy BMI was positively associated with offspring BMI-Z at all ages, but there was no consistent interaction between maternal BMI and GWG in relation to offspring BMI in neither the linear nor logistic regressions. This was observed both with maternal BMI defined as a continuous variable or dichotomized by BMI <21.2 versus greater than or equal to21.2kgm−2, corresponding to the median BMI in the sample, or by BMI <25 versus greater than or equal to25kgm−2. The effect estimates in the linear regression analysis (β (95% CI)) with adult BMI-Z (42 years) as the outcome was for offspring of mothers with a BMI below versus above the median 0.023 (0.003–0.043) and 0.024 (0.002–0.047) BMI-Z per kg GWG, respectively, P-value for interaction 0.389. Furthermore, GWG did not show any interaction with sex or maternal smoking during pregnancy in neither linear nor logistic regression analyses (data not shown).

The regression of offspring adult (42 years) waist circumference on GWG, adjusted for covariates and birth weight, showed a positive association (β (95% CI), 0.29 (0.10–0.46) cm waist circumference per kg GWG (P=0.001)). This effect disappeared by including current BMI in the model (0.05 (−0.07–0.16) cm waist circumference per kg GWG).

Path analysis

Finally, as illustrated in Figure 3, path analysis was used to divide the total effect of GWG on offspring BMI-Z (42 years) into a direct and an indirect effect mediated by birth weight and BMI-Z during childhood (1–14 years). The path analysis was based on the full sample of 4234 individuals, as those with missing information on adult BMI still contribute with information on other associations. The analysis was adjusted for all covariates and resulted in a total effect of GWG on adult BMI at 0.022 (0.011–0.033) BMI-Z per kg GWG, which is very similar to the result obtained by ordinary multiple regression (0.027 (0.013–0.041) as shown in Figure 2. The indirect effect was 0.011 (0.004–0.018), which was computed by adding all products of coefficients found along each path in the path diagram leading from GWG to adult BMI. The direct effect of GWG on adult BMI-Z, that is, the effect that was not mediated by birth weight and BMI-Z up to the age of 14 years was 0.011 (−0.001–0.023).

Figure 3.
Figure 3 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Simplified path analysis diagram of the final model describing the direct effects of the associations between gestational weight gain (GWG) and adult body mass index (BMI)-Z-score. The boxes represent either single measurements (gestational weight gain, birth weight, adult (42 years) BMI-Z) or groups of measurements as for covariates and BMI-Z from 1 to 14 years of age. The arrows in the diagram represent associations, for example, the adult BMI-Z is dependent on birth weight, childhood BMI-Z, GWG and covariates. The estimate pertaining to the path from GWG to adult BMI-Z means that a 1kg increase in GWG is associated with an increase in adult BMI of 0.011 (95% CI −0.01 to 0.023) BMI-Z (the direct effect) in addition to the effect mediated through birth weight and BMI-Z from 1 to 14 years (the indirect effect through the other paths, estimated to be 0.011(0.004–0.018) BMI-Z per kg GWG (this figure is not shown on the figure). From GWG to childhood BMI-Z, the diagram shows a range of the individual regression coefficients from GWG to BMI-Z at 1, 3, 6, 7, 8, 9, 10, 11, 12, 13 and 14 years. These estimates show the effects of GWG not mediated through birth weight. Covariates adjusted for include sex, maternal age, maternal pre-pregnancy BMI, parental social status, breadwinner's education, single-mother status, prematurity, edema and smoking during pregnancy.

Full figure and legend (29K)



This study suggests a consistent positive effect of GWG on offspring BMI during childhood through adulthood. Apparently, half of the effect of GWG on offspring adult BMI (42 years) was mediated through birth weight and BMI from 1 to 14 years of age. At the age of 42 years, there was a significant, but small effect, on the average BMI (0.10kgm−2 per kg GWG), and a significantly increased risk of obesity with increasing GWG (8% per kg GWG). This corresponds to the general observation that the obesity epidemic primarily reflects an increase in the proportion of obese subjects rather than an increase in the overall level of BMI in the population.1, 23 These findings were obtained in a cohort reflecting a marked increase in the prevalence of overweight; from 9% among the mothers to 53 and 33% for the male and female adult offspring, respectively, which corresponds to the current prevalence in Denmark.23 No Danish data for trend in GWG are available to quantify the potential contribution of GWG to the obesity epidemic, but a Finnish study reported an approximately 1kg increase in mean GWG from the 1960s to the mid-1980s.24 If this is associated with an 8% increase in obesity prevalence, as suggested by this study, we find this of great public health relevance.

Our findings agree with other studies11, 12, 13, 14, 15 examining children aged 2–14 years and young adults aged 21 years.16 In three of these studies,11, 12, 13 the prevalence of maternal and childhood overweight was higher than in our study. This consistency is particularly important because our cohort had an overrepresentation of mothers with previous or current pregnancy complications as well as single mothers, whereas the study populations of the other studies were characterized as socially advantaged,11, 15 nationally representative12 or multiethnic.14 In contrast, studies with other primary focuses have reported no effect of GWG on offspring BMI in childhood25 or young adulthood.26

GWG is positively associated with birth weight even when stratifying on factors such as maternal BMI, age, parity and educational level.27 In accordance with others,11, 12, 13, 14, 15, 16 our data suggest that birth weight is not a predominant mediator of the effect of GWG on offspring BMI. Instead, these intrauterine effects could be related to permanent modulation of appetite control, neuro-endocrine pathways influencing adipose tissue development and function or energy metabolism causing development of obesity in later life.5, 28 It is also possible that birth weight is not sensitive enough to measure GWG influences on fetal growth, which is suggested by an observed effect of GWG on body fat percentage at birth independent of pre-pregnancy BMI.29

To explore further whether the effect of GWG on adult BMI (42 years) was mediated through BMI during childhood, we used path analysis.21 The direct effect constituted theoretically half of the total effect and was independent of the effect of GWG mediated through birth weight and BMI up to the age of 14 years. We find it likely that the same pattern applies to adult obesity (42 years), that is, a considerable part of the effect of GWG is not expressed until after childhood. In another study, normal-weight children were compared with children with early (<6 years) and late onset of overweight (6–10 years), and GWG was found to predict only early-onset overweight.12 It is possible that the effect of GWG is to make the offspring more susceptible to an obesogenic environment, which would imply that the expression of the effect is dependent on environmental exposures more than on the age of the individual.

Our finding of a consistent positive effect of maternal pre-pregnancy BMI on offspring BMI at all ages is supported by numerous studies12, 25, 26 and is likely to primarily reflect the well-known genetic effects on obesity.30 In addition, in agreement with other studies,31, 32 women who had the highest pre-pregnancy BMI had the lowest GWG. The largest of the studies14 on GWG and offspring obesity revealed an interaction such that the effect of GWG on the offspring risk of overweight was strongest in underweight mothers. In our study with a majority of normal-weight mothers and in a US study with a higher prevalence of overweight mothers,11 there were, however, no indication of an interaction between GWG and maternal pre-pregnancy BMI, suggesting that a high GWG may not be more harmful in obese women than in the normal-weight group with respect to offspring BMI.

The major strength of our study is the use of information on GWG and potential confounders collected in early life, and the long-term multiple follow-up examinations through childhood and adulthood. A limitation is that the precision of the association between GWG and offspring BMI would have been higher, if information on exact dates for GWG measurements had been available, and if GWG had been recorded in absolute values instead of categories. The cohort consisted of a high proportion of both unmarried mothers and mothers with previous or current pregnancy complications, but the findings were robust in that they did not change by adjustment for pregnancy complications as edema, hypertension and pre-eclampsia. If the high-risk status of the cohort should have affected the results and the conclusions, it would imply that the high-risk features in some way operate as effect modifiers on the association between GWG and offspring obesity, which we find an unlikely hypothesis given the outcome of our analysis. Furthermore, the consistency between our findings and other childhood studies and between childhood and adult data in the present study supports the generalizability of the findings. Information on adult weight, height and waist circumference was subject to participation bias and potential misreporting, but an important bias seem unlikely as the findings were in accordance with the analyses based on mandatory examinations in childhood and at draft board. We find the limitations of this study to most likely contribute to a conservative estimate of the effect size.

This study suggests that greater GWG may contribute to the obesity epidemic among children and adults, which is of particular importance because the greatest GWG occurs among non-obese women. Thus, this effect seems to recruit subjects to the obese segment of the population from the non-obese segment in the preceding generation. GWG seems to have a long-term direct effect on offspring adult BMI, which is independent of the effects of maternal pre-pregnancy BMI and partly independent of the effects of birth weight and BMI in childhood. This suggests that GWG may permanently change the susceptibility of the offspring to obesogenic environments, and for this reason excessive GWG should be avoided.


Conflict of interest

The authors declare no conflict of interest.



  1. Olsen LW, Baker JL, Holst C, Sørensen TIA. Birth cohort effect on the obesity epidemic in Denmark. Epidemiology 2006; 17: 292–295. | Article | PubMed
  2. Oken E, Gillman MW. Fetal origins of obesity. Obes Res 2003; 11: 496–506. | Article | PubMed | ISI
  3. Lawlor DA, Chaturvedi N. Treatment and prevention of obesity--are there critical periods for intervention? Int J Epidemiol 2006; 35: 3–9. | Article | PubMed
  4. Rugholm S, Baker JL, Olsen LW, Schack-Nielsen L, Bua J, Sørensen TIA. Stability of the association between birth weight and childhood overweight during the development of the obesity epidemic. Obes Res 2005; 13: 2187–2194. | Article | PubMed
  5. Lawlor DA, Smith GD, O'Callaghan M, Alati R, Mamun AA, Williams GM et al. Epidemiologic evidence for the fetal overnutrition hypothesis: findings from the mater-university study of pregnancy and its outcomes. Am J Epidemiol 2007; 165: 418–424. | Article | PubMed
  6. Dabelea D, Hanson RL, Lindsay RS, Pettitt DJ, Imperatore G, Gabir MM et al. Intrauterine exposure to diabetes conveys risks for type 2 diabetes and obesity: a study of discordant sibships. Diabetes 2000; 49: 2208–2211. | Article | PubMed | ChemPort |
  7. Fetita LS, Sobngwi E, Serradas P, Calvo F, Gautier JF. Consequences of fetal exposure to maternal diabetes in offspring. J Clin Endocrinol Metab 2006; 91: 3718–3724. | Article | PubMed | ChemPort |
  8. Levin BE, Govek E. Gestational obesity accentuates obesity in obesity-prone progeny 2. Am J Physiol 1998; 275 (Pt 2): R1374–R1379. | PubMed | ChemPort |
  9. Catalano PM. Obesity and pregnancy--the propagation of a viscous cycle? J Clin Endocrinol Metab 2003; 88: 3505–3506. | Article | PubMed | ChemPort |
  10. Ebbeling CB, Pawlak DB, Ludwig DS. Childhood obesity: public-health crisis, common sense cure. Lancet 2002; 360: 473–482. | Article | PubMed | ISI
  11. Oken E, Taveras EM, Kleinman KP, Rich-Edwards JW, Gillman MW. Gestational weight gain and child adiposity at age 3 years. Am J Obstet Gynecol 2007; 196: 322–328. | Article | PubMed
  12. Li C, Goran MI, Kaur H, Nollen N, Ahluwalia JS. Developmental trajectories of overweight during childhood: role of early life factors. Obesity (Silver Spring) 2007; 15: 760–771. | Article | PubMed
  13. Moreira P, Padez C, Mourao-Carvalhal I, Rosado V. Maternal weight gain during pregnancy and overweight in Portuguese children. Int J Obes (London) 2007; 31: 608–614. | Article | ChemPort |
  14. Wrotniak BH, Shults J, Butts S, Stettler N. Gestational weight gain and risk of overweight in the offspring at age 7 y in a multicenter, multiethnic cohort study. Am J Clin Nutr 2008; 87: 1818–1824. | PubMed | ChemPort |
  15. Oken E, Rifas-Shiman SL, Field AE, Frazier AL, Gillman MW. Maternal gestational weight gain and offspring weight in adolescence. Obstet Gynecol 2008; 112: 999–1006. | PubMed |
  16. Mamun AA, O'Callaghan M, Callaway L, Williams G, Najman J, Lawlor DA. Associations of gestational weight gain with offspring body mass index and blood pressure at 21 years of age: evidence from a birth cohort study. Circulation 2009; 119: 1720–1727. | Article | PubMed
  17. Committee on the Impact of Pregnancy Weight on Maternal and Child Health. Board on Children, Youth and Families Division of Behavioral and Social Sciences and Education and Food and Nutrition Board Institute of Medicine. Influence of Pregnancy Weight on Maternal and Child Health: Workshop Report. National Research Council and Institute of Medicine. National Academies Press: Washington, DC, 2007.
  18. Zachau-Christiansen B, Ross EM. Babies: Human Development During the First Year. JW Arrowsmith Ltd: Bristol, 1975.
  19. Reinisch JM, Mortensen EL, Sanders SA. The prenatal development project. Acta Psychiatr Scand Suppl 1993; 370: 54–61. | Article | PubMed | ChemPort |
  20. Cole TJ, Freeman JV, Preece MA. Body mass index reference curves for the UK, 1990. Arch Dis Child 1995; 73: 25–29. | Article | PubMed | ISI | ChemPort |
  21. Gamborg M, Andersen PK, Baker JL, Budtz-Jørgense E, Jørgensen T, Jensen G et al. Life course path analysis of birth weight, childhood growth and adult systolic pressure. A J Epidemiol 2009; 169: 1167–1178. | Article
  22. Cole TJ, Bellizzi MC, Flegal KM, Dietz WH. Establishing a standard definition for child overweight and obesity worldwide: international survey. Br Med J 2000; 320: 1240–1243. | Article | ISI | ChemPort |
  23. Due P, Heitmann BL, Sørensen TIA. Prevalence of obesity in Denmark. Obes Rev 2007; 8: 187–189. | Article | PubMed | ChemPort |
  24. Kinnunen TI, Luoto R, Gissler M, Hemminki E. Pregnancy weight gain from 1960s to 2000 in Finland. Int J Obes Relat Metab Disord 2003; 27: 1572–1577. | Article | PubMed | ChemPort |
  25. Whitaker RC. Predicting preschooler obesity at birth: the role of maternal obesity in early pregnancy. Pediatrics 2004; 114: e29–e36. | Article | PubMed | ISI
  26. Koupil I, Toivanen P. Social and early-life determinants of overweight and obesity in 18-year-old Swedish men. Int J Obes (London) 2008; 32: 73–81. | Article | ChemPort |
  27. Seidman DS, Ever-Hadani P, Gale R. The effect of maternal weight gain in pregnancy on birth weight. Obstet Gynecol 1989; 74: 240–246. | PubMed | ChemPort |
  28. Levin BE. Metabolic imprinting: critical impact of the perinatal environment on the regulation of energy homeostasis. Phil Trans R Soc B 2006; 361: 1107–1121. | Article | PubMed | ChemPort |
  29. Catalano PM, Thomas A, Huston-Presley L, Amini SB. Phenotype of infants of mothers with gestational diabetes. Diabetes Care 2007; 30 (Suppl 2): S156–S160. | Article | PubMed
  30. Clemént K, Sørensen TIA. Obesity: Genomics and Postgenomics. Informa Healthcare USA, Inc.: New York, 2007.
  31. Cedergren M. Effects of gestational weight gain and body mass index on obstetric outcome in Sweden. Int J Gynaecol Obstet 2006; 93: 269–274. | Article | PubMed | ChemPort |
  32. Baker JL, Michaelsen KF, Sørensen TIA, Rasmussen KM. High prepregnant body mass index is associated with early termination of full and any breastfeeding in Danish women. Am J Clin Nutr 2007; 86: 404–411. | PubMed | ChemPort |


We are grateful to the late Drs Aage Willumsen and Bengt Zachau-Christiansen for the collection of the primary data, and we thank Prof June M Reinisch and colleagues for contributing height and weight data on the subjects examined at draft and at age 20–34 years.

Extra navigation