OBJECTIVE: To assess longitudinally the relationship between measures of adiposity in children over the first 8 y of life with that of their parents and to explore the role of parental adiposity in the development of childhood adiposity.
DESIGN: Longitudinal study of measures of adiposity in children.
SUBJECTS: A community sample from three health service systems including 114 children followed annually from infancy to age 8 and their 228 biological parents.
METHODS: Measurements were assessed at baseline for parents (6 months post-partum for mothers) and at regular intervals for children beginning at age 2 months. Measurements included weight, height, triceps skinfold, subscapular skinfold, midarm circumference, waist and hip.
RESULTS: The major findings were: (1) significant correlations between parental body mass index (BMI), both maternal and paternal, and their biological offspring first emerged at age 7; (2) children with two overweight parents had consistently elevated BMI compared to children with either no overweight parents or one overweight parent. These differences became significant beginning at age 7.
CONCLUSIONS: This study supports the hypothesis that familial factors (biological and/or environmental) affecting the development of adiposity emerge at specific ages and are related to the adiposity of both parents.
Increasing percentages of Americans, adults and children, are now overweight or obese.1,2 Because childhood obesity increases the risk of obesity in adulthood2,3,4 and adult obesity has well-documented associations with disorders such as hypertension, coronary artery disease and stroke,5 the prevention of obesity in childhood is an important public health issue.
While studies of adoptive families and twins (reared together and apart) support the role of hereditary factors in childhood obesity, the extent of that role remains the subject of active debate.2,6,7,8 For example, Stunkard and colleagues9 recently challenged commonly held assumptions regarding associations between parental adiposity with that of their offspring by reporting the absence of a relationship between parental body mass index (BMI) and infant body weight over the first 2 y of life. The implication of their findings, these authors argue, is that the ‘genetic influences on the body weights of infants are independent of genetic influences on the body weights of adults.’
One way of understanding this finding is to propose, as do researchers in the field of developmental genetics such as Cardon,6 that the genes influencing adiposity have differential effects at distinct childhood ages. To test this hypothesis, prospective studies of community samples of children, along with data from their parents, are needed to accurately track the development of obesity over the life-span. While there are a number of studies tracking obesity from childhood to adulthood10,11 and tracking relationships between parental and offspring fatness over time (see Parsons et al12 for a recent review), the majority of these use cross-sectional data or are retrospective. Unfortunately the number of studies investigating the longitudinal relationship between parents and offspring fatness are few.12 The Stunkard and colleagues study, for example, only followed children for 2 y and did not involve a normally distributed, naturalistic sample of children (half were born to obese mothers and half to lean mothers). Other studies, such as one by Klesges and colleagues,2 only used a 3 y period over which to investigate the role of parental BMI in predicting weight changes in children, and one study that did cover a much longer period of time, by Kaplowitz and colleagues,13 used an outcome measure based on skinfold measures, not BMI.
The Colorado Adoption Project,6 which examined a cohort of both adopted and non-adopted children, is the only published longitudinal study to our knowledge that followed its participants over 9 y (measurements were performed yearly from birth to age 7 and again at age 9) and included parental data. In that study, path analyses were used to estimate heritability. For children between the ages of 0 and 2, heritabilities between parents and biological offspring were very low (0.01–0.09), but increased after age 3. However, maternal and paternal BMI were not analyzed separately, but were averaged to obtain one ‘parent’ BMI score. This may have been due to the fact that only 20% of biological fathers were directly assessed.
The purpose of the current investigation is to examine the correlations between measures of adiposity in parents and their biological offspring over nine time periods—birth through age 8—using a cohort that includes both maternal and paternal data for all subjects. In addition, the yearly pattern of children's BMI changes from birth to age 8 in relation to parental adiposity status was explored.
The sample of 114 children and their 228 biological parents was a subset of the Stanford Infant Growth study, an ongoing longitudinal study. Descriptions of the methods of recruitment, determination of weaning status, and socio-demographic information on the original cohort of 216 healthy full-term infants are provided in detail by Hammer and colleagues.14 The study was approved by the Stanford University Committee for the Protection of Human Subjects.
The 114 children in the present study were chosen on the basis of their having had yearly measurements of at least body weight and length from birth through age 8. They represent 53% of the original sample, a sample which included primarily well-educated, non-Hispanic white middle class parents. The children included in the study were not statistically different from children missing at least one yearly measurement with respect to gender, birth BMI, parental age, parental ethnicity, or maternal BMI. However, the children in the study compared to those with missing data had lower mean paternal BMI (25.2 vs 26.7, t (195)=−3.07, P=0.002) and higher maternal education scores (6.2 vs 6.0, t (198)=2.27, P=0.024).
Measurements (weight, height, triceps skinfold, subscapular skinfold, midarm circumference, waist, hip) were assessed at baseline for both parents (6 months post-partum for mothers) and at regular intervals for children, beginning at 2 months of age. Weights and heights were obtained using standard laboratory methods.15 Skinfolds were assessed at two sites (triceps and subscapular) using Lange calipers. Waist circumference was measured at the umbilicus and hip circumference at the widest part of the hip region. Research assistants were trained before the study so that a satisfactory level of agreement was obtained. Ongoing assessment of inter-observer agreement took place on a randomly chosen 20% of all observations.
Spearman correlations were calculated between the child's BMI, skinfold measurements (subscapular, triceps and midarm circumference) and waist–hip ratios and those of their mother and father.
To assess the effects of parental adiposity on childhood BMI, a repeated measures analysis across nine time periods (birth through age 8) was performed, using parental adiposity status as the independent variable. Overweight was defined as a BMI greater than 27 and parental adiposity status was divided into four groups (1) having an overweight mother but non-overweight father; (2) having a non-overweight mother but overweight father; (3) having both parents overweight; and (4) having neither parent overweight. Eleven children (9.6%) had only an overweight mother, 16 (14%) had only an overweight father, 11 (9.6%) had two overweight parents, and 76 (67%) had two non-overweight parents.
Table 1 displays BMI correlations between parents (mothers and fathers) and their children over the first 8 y of life. Maternal–offspring correlations include those between mothers and children (both genders inclusive), between mothers and daughters, and between mothers and sons. Likewise, paternal–offspring correlations include those between fathers and children (both genders inclusive), fathers and daughters, and fathers and sons. Consecutive significant correlations were found between children's BMI at ages 7 and 8 and parental BMI (both maternal and paternal).
There were few significant correlations between parents and their offspring on the other anthropometric measures of triceps skinfold, subscapular skinfold, midarm circumference, or waist–hip ratio. The only consecutive significant correlations were those between paternal midarm circumference and offspring midarm from ages 6 to 8. These seem primarily accounted for by the significant father–son midarm correlations at these ages, as no significant father–daughter correlations at these ages were found. None of the other skinfold measurements nor the waist–hip measurements showed any consistent correlations over more than one year.
A repeated measures ANOVA found parental obesity status to be a significant factor (F(24,880)=2.57, P<0.001) affecting the pattern of BMI changes across the nine yearly time periods (birth to 8 y). Children with two overweight parents had consistently higher BMI compared to children with no overweight parents or one overweight and one non-overweight parent. Post-hoc univariate tests using the Tukey-HSD test found these differences to be non-significant between birth and age 6 but to become so at age 7 (F(3,110)=2.9, P=0.04) and more so at age 8 (F(3,110)=6.3, P<0.001). The graph in Figure 1 displays the children's patterns of BMI changes over time according to parental adiposity status (eg father-only overweight, mother-only overweight, both parents overweight, or neither parent overweight).
Our findings extend those found by Stunkard and colleagues,9 who examined a cohort of children during their first 2 y of life and found no statistical relationship between the body weights of parents and offspring. Their results challenged what seems to be a widely accepted belief that genetic influences on obesity are exerted throughout the life cycle.
We found the emergence of statistically significant correlations between the BMI of children beginning at the age of 7 and their parents. This finding indicates that familial factors, rather than influencing children and adults independently, find expression in later childhood.
We also found that children with two overweight parents had a significantly different BMI growth pattern from birth to age 8 compared to children with no or only one overweight parent. While a number of studies show that having an obese parent increases the risk of obesity in adulthood,7,16,17,18 less is known about the emergence of this relationship in early childhood.
In addition to examining correlations between the BMI of parents and children, another way to explore the influence of familial factors on BMI is by examining longitudinal patterns of BMI changes in children over time. Klesges and colleagues2 found parental adiposity status to be a predictor for accelerated rates of growth in children aged 3–5. Our findings extend those of Klesges and colleagues by including children from birth through age 8.
A review of the literature reporting familial relationships (mother–child, father–child, parent–child) for the BMI shows correlation coefficients that consistently range between 0.20–0.30.8 These values, obtained largely through cross-sectional studies3,19 in which data on children of various ages were included, are in general agreement with those reported in this study for children at least 7 or 8. Given the consistency of a familial BMI correlation across studies, the question is raised as to why our prospective study demonstrated these coefficients only beginning with age 7. In other words, since there are familial effects on a child's relative growth rate, why are they not evidenced earlier?
One way of understanding the generally low parent–child BMI correlations before children reach the age of 7 is to take into account the phenomenon of ‘adiposity rebound.’ As described by Rolland-Cachera and colleagues,20 BMI increases over the first year of life and decreases thereafter until between the ages of 5½ and 7, when it begins to climb again. Figure 1 reveals that the BMI of the children with two overweight parents experience this rebound earlier (eg begin to increase at age 3–4), whereas the mean BMI of the children with no overweight parents does not exhibit adiposity rebound until at least age 6. With the BMI of some children increasing and some, over the same span of years decreasing, a ‘washout’ effect could obscure the detection of significant correlations.
Another possible reason for the low correlations between parents and their children over the first 6 y of life is that there may be a certain time span necessary before the effects of the environment, such as eating and activity habits, are observed in individuals carrying a particular genetic make-up. The emergence of diabetes type 2 may be an example of this pattern of genetic expression mediated via interactions with the environment.21
Finally, the low correlations between parents and their children over the first 6 y of life may be because the BMI is not an accurate enough measure of fatness; other indexes of adiposity might correlate more highly. However, despite some controversy about the utility of the BMI vs other measures of adiposity, the consensus (including conclusions from the International Obesity Task Force22) is that the BMI is a reasonable measure with which to assess fatness in children and adults.22,23,24,25 In this study, the correlations for parents and children using other anthropometric measures of adiposity were almost always lower than BMI correlations and showed no consistent pattern over more than one time period. Correlations for midarm circumference between fathers and sons beginning at age 5 provided the only exception to this. By contrast there was an absence of significant correlations between father–daughter midarm circumference at these same ages.
The finding of no overall significant parent–child correlations for the adiposity measures of triceps skinfold and subscapular skinfold is puzzling. Other researchers have demonstrated moderately high correlations between the BMI and both triceps and subscapular skinfolds.23,26 However, especially with young children, skinfolds can be difficult to measure.26,27 Also, the great deal of variation in fat distribution in both children and adults further limits the likelihood of any particular adiposity measure accurately reflecting the totality of an individual's adiposity status.
One unexpected finding was that children with only an overweight mother had higher mean BMI over time and exhibited a growth pattern more similar to that of two overweight parents than the group with fathers only overweight, although these differences did not reach statistical significance. Theoretically, there should be no difference between the effects of genes for overweight from one's mother and from one's father, as each parent contributes half of his/her genes to the child. While the effects of heredity and the environment cannot be distinguished in this study, this greater similarity may be due to children's food environments usually being shaped more by mothers than fathers. Perhaps mothers are more often involved in food preparation and share more meals with their children than fathers do. Interestingly, in the study of adopted children aged 7–13 y by Sorenson and colleagues,28 the BMI of the adoptees showed a weak correlation with their adoptive mothers but no correlation at all with their adoptive fathers.
This study has several limitations. First, while we have demonstrated the relationship between parental and childrens' BMI, we cannot draw any conclusions regarding how much of the relationship is due to genetic or environmental factors, or an interaction between them. In other words, parental adiposity represents a cumulative influence of not only genetic make-up but also lifestyle and behavior. It is beyond the scope of this current study to include measures of lifestyle and behavior in these analyses.
An additional limitation stems from the fact that data are only available for children through age 8. Also, because the sample was predominantly well-educated, white and middle class, generalizability to other socioeconomic and ethnic classes is unclear. Finally, the sample size may have been too small to detect some significant parent-child correlations before age 7. In particular, there were not enough children with two overweight parents to explore whether parent–child BMI correlations in this group would be higher than when based on mean BMI for the entire sample.
In summary, this study extends the findings of Stunkard and colleagues9 by prospectively measuring parent–offspring BMI correlations over nine yearly time periods and showing that parental influences on the child's BMI first emerge at age 7. Our findings also suggest that adiposity status of both parents significantly affects their child's BMI growth patterns. Interestingly, this two-parent influence also begins at age 7.
These findings support the hypothesis that familial factors affecting the development of adiposity emerge at specific ages. Longitudinal studies designed to measure environmental influences on adiposity are needed to help design programs to prevent or limit the expression of these familial factors.
Yager J . Weighty perspectives contemporary challenges in obesity and eating disorders Am J Psychiat 2000 157: 851–853.
Klesges RC, Klesges LM, Eck LH, Shelton ML . A longitudinal analysis of accelerated weight gain in preschool children Pediatrics 1995 95: 126–130.
Price RA, Stunkard AJ, Ness, R, Wadden T, Heshka S, Kanders B, Cormillot A . Childhood onset (age<10) obesity has high familial risk Int J Obes 1990 14: 185–195.
Clarke WR, Woolson RF, Lauer RM . Changes in ponderosity and blood pressure in childhood: the Muscatine Study Am J Epidemiol 1986 124: 195–206.
Pi-Sunyer FX . NHLBI Obesity Education Initiative Expert Panel on the Identification, Evaluation, and Treatment of Overweight and Obesity in Adults—the evidence report Obes Res 1998 6: (Suppl 2): 51S–209S.
Cardon L . Genetic influences on body mass index in early childhood. In: Turner L, Cardon L, Hewitt J (eds). Behavior genetic approaches to behavioral medicine. Plenum Press: New York 1995 133–143.
Whitaker RC, Wright JA, Pepe MS, Seidel KD, Dietz WH . Predicting obesity in young adulthood from childhood and parental obesity New Engl J Med 1997 337: 869–873.
Meyer JM, Stunkard AJ . Genetics and human obesity. In: Stunkard AJ, Wadden TA (eds). Obesity theory and therapy, 2nd edn. Raven Press: New York 1993 137–149.
Stunkard AJ, Berkowitz RI, Stallings VA, Cater JR . Weights of parents and infants is there a relationship? Int J Obes Relat Metab Disord 1999 23: 159–162.
Power C, Lake JK, Cole TJ . Measurement and long-term health risks of child and adolescent fatness Int J Obes Relat Metab Disord 1997 21: 507–526.
Serdula MK, Ivery D, Coates RJ, Freedman DS, Williamson DF, Byers T . Do obese children become obese adults? A review of the literature Prev Med 1993 22: 167–177.
Parsons TJ, Power C, Logan S, Summerbell CD . Childhood predictors of adult obesity: a systematic review Int J Obes Relat Metab Disord 1999 8: (suppl): S1–S107.
Kaplowitz HJ, Wild KA, Mueller WH, Decker M, Tanner JM . Serial and parent–child changes in components of body fat distribution and fatness in children from the London Longitudinal Growth Study, ages two to eighteen years Hum Biol 1988 60: 739–758.
Hammer LD, Bryson S, Agras WS . Development of feeding practices during the first 5 y of life Arch Pediatr Adolesc Med 1999 36: 569–572.
Gordon CC, Chumlea WC, Roche A . Stature, recumbent length, and weight. In: Lohman TG, Roche AF, Martorell R (eds). Anthropometric standardization reference manual. Human Kinetics: Champaign, IL 1988
Garn SM, Clark DC, Guire KE . Growth, body composition and development of lean and obese children. In: Winick M (ed). Childhood obesity. John Wiley: New York 1975
Laskarzewski PM, Khoury P, Morrison JA, Kelly K, Mellies MJ, Glueck CJ . Familial obesity and leanness Int J Obes 1983 7: 505–527.
Eck LH, Klesges RC, Hanson RC, Slawson D . Children at familial risk for obesity: an examination of dietary intake, physical activity and weight status Int J Obes Relat Metab Disord 1992 16: 71–78.
Khoury P, Morrison JA, Laskarzewski PM, Gluek CJ . Parent–offspring and sibling body mass index associations during and after sharing of common household environments: the Princeton School District Family Study Metabolism 1983 32: 82–89.
Rolland-Cachera MF, Deheeger M, Bellisle F, Sempe M, Guilloud-Bataille M, Patois E . Adiposity rebound in children: a simple indicator for predicting obesity Am J Clin Nutr 1984 39: 129–135.
Groop LC . Pathogenesis of insulin resistance in type 2 diabetes: a collision between thrifty genes and an affluent environment Drugs 1999 58: (Suppl 1): 11–12.
Dietz WH, Bellizzi MC . Introduction: the use of body mass index to assess obesity in children Am J Clin Nutr 1999 70: (Suppl): 123S–125S.
Frisancho AR, Flegel PN . Relative merits of old and new indices of body mass with reference to skinfold thickness Am J Clin Nutr 1982 36: 697–699.
Hammer LD, Kraemer HC, Wilson DM, Ritter PL, Dornbusch SM . Standardized percentile curves of body-mass index for children and adolescents Am J Dis Child 1991 145: 259–263.
Kraemer HC, Berkowitz RI, Hammer LD . Methodological difficulties in studies of obesity, 1: measurement issues Ann Behav Med 1990 23: 112–118.
Rolland-Cachera MF, Sempe M, Guilloud-Bataille M, Patois E, Pequignot-Guggenbuhl F, Fautrad V . Adiposity indices in children Am J Clin Nutr 1982 36: 178–184.
Longini IM, Higgins MW, Hinton PC, Moll PP, Keller JB . Genetic and environmental sources of familial aggregation of body mass in Tecumseh, Michigan Hum Biol 1984 56: 733–757.
Sorenson TI, Stunkard AJ . Childhood body mass index—genetic and familial environmental influences assessed in a longitudinal adoption study Int J Obes Relat Metab Disord 1992 16: 705–714.
This research was partially supported by a National Institute of Mental Health (NIMH) post-doctoral fellowship (5-T32MH19938) awarded to Dr Safer.
About this article
Cite this article
Safer, D., Agras, W., Bryson, S. et al. Early body mass index and other anthropometric relationships between parents and children. Int J Obes 25, 1532–1536 (2001). https://doi.org/10.1038/sj.ijo.0801786
- body mass index
- parent–child relationship
Maternal and Child Health Journal (2014)
Dietary pattern and leisure time activity of overweight and normal weight children in Germany: sex-specific differences
Nutrition Journal (2013)
Associations between severity of obesity in childhood and adolescence, obesity onset and parental BMI: a longitudinal cohort study
International Journal of Obesity (2011)
European Journal of Pediatrics (2011)
Parent Overweight Predicts Daughters’ Increase in BMI and Disinhibited Overeating from 5 to 13 Years*