SEARCH

SEARCH BY CITATION

Keywords:

  • childhood overweight;
  • weight gain;
  • infancy;
  • Special Supplemental Nutrition Program for Women;
  • Infants and Children;
  • breast-feeding

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Research Methods and Procedures
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgement
  9. References

Objective: To determine among a contemporary cohort whether rapid weight gain between birth and 6 months is associated with risk of childhood overweight and if this risk differs by ethnicity and/or breast-feeding history.

Research Methods and Procedures: This was a cross-sectional survey in 1999 to 2000 of parents/guardians of children participating in the Special Supplemental Nutrition Program for Women, Infants, and Children in New York State. Measurements were abstracted by chart review, including weight at birth and 6 months, and height and weight at time of survey and every 6 months subsequently. Overweight at 4 years of age was defined as a BMI ≥ 95th age- and sex-specific percentiles.

Results: The study sample was 32% Hispanic, 19% black, and 49% white; 17% of children were overweight. Rate of infant weight gain (expressed in terms of 100 g/mo) was significantly associated with being overweight at 4 years (odds ratio, 1.4; 95% confidence interval, 1.3 to 1.6 after adjusting for history of breast-feeding, birth weight, and ethnicity). The odds of being overweight at 4 years of age for Hispanic children were twice those of non-Hispanic children (odds ratio, 2.2; 95% confidence interval, 1.5 to 3.3). The population-attributable risk of overweight at 4 years of age was 19% for children in the highest quintile of infant weight gain.

Discussion: Among this contemporary, multi-ethnic cohort, rapid infant weight gain was associated with increased risk of being overweight at 4 years of age, independently of potential confounders. Identification of the risk factors contributing to rapid weight gain during infancy might improve early recognition and guide strategies for optimal nutrition to prevent the development of childhood overweight.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Research Methods and Procedures
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgement
  9. References

Obesity has gained recognition as an important worldwide public health problem and, in the U.S., appears to be overtaking tobacco use as the number one cause of preventable death (1). Prevalence of overweight among U.S. children and adults has increased dramatically, nearly tripling over the past 20 to 30 years (2, 3). Overweight contributes to health problems among children (4), including psychosocial consequences (5) and increased cardiovascular risk factors such as hypertension, hyperlipidemia, or type II diabetes (6, 7). Overweight children are also significantly more likely to become overweight or obese as adults (8, 9). Consequently, identification and prevention of overweight and obesity has become a major public health priority in the U.S. (10).

Because of both a rightward shift in the population distribution of BMI and increased adiposity among the most overweight (11), population-based and risky strategies are needed to combat this epidemic. Studies suggest that there may be critical periods during the life cycle when individuals are at increased risk for the development of obesity (12, 13). Early infancy has been identified as one such critical period. Several studies suggest an association between increased rates of weight gain during the first 4 to 24 months of life and risk of overweight during later childhood or early adulthood (13, 14, 15, 16) These studies of infant weight gain have included limited information about breast-feeding history, which has been associated in several studies with lower rates of childhood overweight or adult obesity (17, 18, 19). Moreover, in the ensuing 40 to 45 years since the cohort studies of Stettler and colleagues (14, 15), there have been many changes in infant feeding practices and in formulas. The prevalence of childhood overweight has also increased nearly 3-fold (3). Furthermore, previous studies have not included Hispanic children, yet the prevalence of overweight is much higher among Hispanics than non-Hispanic blacks or whites (3, 20). The purpose of this study, therefore, was to evaluate whether the rate of infant weight gain among a multi-ethnic sample of contemporary infants predicts overweight status later in childhood and whether this relationship differs by ethnicity or is attenuated by breast-feeding history.

Research Methods and Procedures

  1. Top of page
  2. Abstract
  3. Introduction
  4. Research Methods and Procedures
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgement
  9. References

Study Design and Sample

Children ages 1 up to 5 years, who were participating in the Special Supplemental Nutrition Program for Women, Infants, and Children (WIC)1 at 49 of the 100 local WIC agencies in New York State (NYS), were eligible to participate. Surveys were administered to the child's parent/guardian between July and December 1999 or May and December 2000. The study design was to survey 30 consecutive eligible adults at each WIC agency in each year. Logistical constraints at the WIC clinics, however, prevented accurate documentation of response rates. WIC is a state-administered, federally funded program that serves low-income (<185% of the federal poverty level in NY) women (pregnant, breastfeeding, or postpartum), infants, and children (up to 5 years of age) at medical risk.

The Institutional Review Boards of the Mary Imogene Bassett Hospital and the NYS Department of Health approved this study. At the time of the study, NYS WIC staff had a mandate to collect information annually from participants in the WIC program. Therefore, consent of individual participants was neither necessary nor obtained. Before submission for data entry, WIC staff removed all identifying information from the surveys.

Survey

At each local WIC agency, questionnaires, written in both English and Spanish, were distributed to eligible parents/guardians, who self-administered the survey. The parent/guardian reported: his/her relationship to the child, his/her last grade of school completed, the child's breast-feeding history (yes/no) and duration (in months), and race/ethnicity (Hispanic, non-Hispanic black, non-Hispanic white, Asian, Native American, multicultural, or other). Because of small numbers, Asian, Native American, multicultural, and other race/ethnicities (N = 47) and those with missing race/ethnicity (N = 4) were excluded from analysis, leaving three major racial/ethnic groups: Hispanic, non-Hispanic black, and non-Hispanic white. An Hispanic/non-Hispanic contrast was also constructed by collapsing non-Hispanic black and non-Hispanic white into a “non-Hispanic” category. Educational attainment was defined as less than high school, high school or general equivalency diploma, or more than high school. Children's breast-feeding history was dichotomized into never breast-fed (N = 311) and breast-fed ≥ 1 month (N = 305). Because of concerns that a very limited period of breastfeeding might reflect either child or maternal illness or poor child growth, children who were breast-fed for <1 month were excluded from analysis (N = 48). Records were also excluded from analysis if the respondent was not a parent (N = 44). In the year 2000 survey, respondents self-reported height and weight, and for those who were the child's mother (N = 276), maternal BMI was calculated as weight (kilograms) divided by the square of height (meters).

Anthropometric Measurements

At the time the baseline survey was completed, the child's most recent height or length, weight, and date of measurement were abstracted from the child's WIC chart. Child's weight and height/length were measured using standard protocols (21). Weight was measured using a standard balance beam scale. Standing height, for children over 36 inches tall or over 2 years of age, was measured using a wall-mounted stadiometer. For each study child, beginning in December 2001 through September 2002, WIC staff at each of the 49 local agencies abstracted data from paper charts and/or computer records, which included: the child's birth weight, 6-month weight, and all anthropometric measurements (height, weight, and measurement date) subsequent to the survey date. No measure of gestational age at birth was available; therefore, to exclude very premature children, those with a birth weight <1800 grams were excluded from analysis. These data were linked to the baseline questionnaire data, creating a longitudinal cohort. For each child who did not have a measurement at 48 (±2) months of age but who had at least two follow-up measurements, height and weight at 48 months of age were estimated separately using linear regression analysis.

Measures were converted to metric equivalents, and BMI was computed as weight in kilograms divided by the square of the height in meters. Age- and sex-specific percentiles and z scores were computed from the 2000 Centers for Disease Control and Prevention growth charts for the U.S. (22). Longitudinal data were examined visually for outliers by individually plotting height vs. age, weight vs. age, and BMI vs. age and by identifying values classified as biologically implausible for weight-for-height or BMI-for-age (z score < −4 or >+5) (23). This examination revealed many cases in which the child's growth (height, weight, and BMI) was tracking linearly at high values that would be considered biologically implausible if only the individual measures were assessed. Children with consistently high measures (N = 25), although identified as biologically implausible, were retained for analysis.

The main outcome variables were overweight, which was defined by a BMI at or above the 95th age- and sex-specific percentile, and at-risk-for-overweight, which was defined by a BMI at or above the 85th age- and sex-specific percentile of the Centers for Disease Control and Prevention growth charts (22). Mothers were categorized based on their BMI, as not overweight or obese (BMI < 25 kg/m2), overweight (BMI of 25 to 29.9 kg/m2), or obese (BMI of 30 kg/m2 or greater). The main exposure variable, rate of weight gain during the first 6 months of life, was calculated as the difference (in kilograms) between weight at the 6-month examination (± 1 month) and birth weight divided by the exact age (in months, calculated to the nearest tenth of a month) at the 6-month examination.

Statistical Analysis

Data were analyzed using SAS (version 8.02; SAS Institute Inc., Cary, NC); graphics were produced using S-plus 2000 (Professional Edition for Windows; MathSoft, Inc., Seattle, WA). Proportions or medians were used to describe the characteristics of the study sample. χ2 Statistics or ANOVA were used, as appropriate, to assess differences in characteristics between the study sample and the survey sample.

Measures of association between covariates were assessed with Pearson correlation coefficients or χ2 statistics, as appropriate. Student's t tests or ANOVA analyses were used, as appropriate, to assess whether rate of infant weight gain differed by child sex, the three primary racial/ethnic groups, Hispanic ethnicity, birth weight, maternal BMI, or breast-feeding history. The effect of rate of infant weight gain on risk of overweight at 4 years, adjusted for race/ethnicity, was estimated by a single logistic regression equation. Predicted probabilities of overweight at age 4 years were also estimated separately for non-Hispanic white, non-Hispanic black, and Hispanic children for the main exposure variable, rate of infant weight gain, using individual logistic regression equations.

The population-attributable risk for children in the highest quintile for infant weight gain was calculated as the difference in the prevalence of overweight among the study population and the prevalence of overweight among the exposed, divided by prevalence of overweight among the study population, and multiplied by 100%.

Univariate logistic regression assessed the associations among participant characteristics, including the main exposure variable, 6-month rate of infant weight gain, and child weight status at 4 years of age. Multiple logistic regression models (Model I) with overweight or at-risk-for-overweight as outcome variables included Hispanic ethnicity, birth weight, child sex, breast-feeding history, and rate of infant weight gain. Continuous variables were checked for linearity before inclusion in the models. To test for systematic differences between each WIC site, a variable for site was included in the initial multiple logistic regression models and found to be non-significant; thus, it was removed from the final models. All first order interactions were also included in the initial analyses, including interactions between ethnicity and rate of infant weight gain and between birth weight and rate of infant weight gain. Interaction terms were found to be non-significant and were removed from subsequent models.

A second set of multiple logistic regression equations (Model II) included maternal BMI (which was available for just under one-half the study population). Model II equations also assessed separately for breast-fed and non-breast-fed children. Odds ratios (ORs) and 95% confidence intervals (CIs) are presented to show the association between exposure and outcome variables. Rate of weight gain during the first 6 months of life was expressed in terms of 100 g/mo. Therefore, an OR of 1.4 would indicate that the odds of being overweight at 4 years of age would be 1.4 times greater for each 100 g/mo increase in infant weight gain during the first 6 months. All statistical tests were two-tailed, with a value of p < 0.05 considered statistically significant. Adequacy of multiple logistic regression models was assessed using Hosmer-Lemeshow statistics (24).

Birth weight and weight gain between birth and 6 months were entered into logistic regression equations with either overweight or at-risk-for-overweight as the response variable. Using the predicted probabilities, contour plots were constructed to visually show the relationship between the probability of being overweight or at-risk-for-overweight at 4 years of age by both birth weight and 6-month weight gain.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Research Methods and Procedures
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgement
  9. References

Surveys were completed by parents or guardians of 2801 children, ages 1 through 4 years. Birth weight and 6-month weight were provided for 1377 (49%) of the survey sample. Children in the final study sample (N = 616) had at least two follow-up measurements or a measured height and weight at 4 years, birth weight of at least 1800 grams, 6-month weight, were either never breast-fed or breast-fed for at least 1 month, and had a parent as the survey respondent (Table 1). Because younger children were more likely to have two follow-up visits, it was expected that the study sample would be younger than the total survey sample. Local WIC agencies located outside the metro New York City area (N = 16) were better able to access paper records than those located within New York City (N = 33), which resulted in higher follow-up rates. Thus, the follow-up rates for clients outside metro New York City (who were predominately non-Hispanic whites) were higher. The study sample also had a slightly greater mean rate of infant weight gain, which most likely contributed to a higher mean 6-month weight than the survey sample.

Table 1. . Characteristics of the study subsample compared with remainder of survey sample*
  Study sample (N = 616)Remainder of survey sample (N = 2185)
VariablesNMean(SD)%NMean(SD)%P
  • *

    χ2 Tests used to compare sample proportions, Student's t tests used to compare continuous variables.

  • BMI calculated as weight in kilograms divided by square of height in meters.

  • Overweight was defined as a BMI at or above the 95th sex-specific BMI percentile for age for children at least 2 years old, a weight-for- height at or above the 95th percentile for children less than 2 years old.

  • §

    Metro NYC includes metropolitan NYC and surrounding counties of Nassau, Rockland, Suffolk, Sullivan, and Westchester.

Child         
 Sex (male)616  50.52185  51.40.71
 Birth weight (grams)6163340(547) 17433282(605) 0.03
 Rate of weight gain (100 g/mo)6167.4(1.7) 11527.2(1.8) 0.01
 Weight at 6 months (grams)6167827(1129) 11527624(1267) <0.001
 Breastfed (>1 month) %616  49.52183  47.80.45
  Duration (months)3058.4(5.6) 10428.7(6.3) 0.44
 Age (months) (baseline)61628.5(11.6) 218532.5(13.8) <0.001
 BMI (kg/m2) (baseline)37116.8(1.9) 146516.8(2.2) 0.67
 Overweight (baseline) %585  17.11972  16.70.82
Location of WIC site616   2185   0.01
 Metro NYC§   58.9   65.0 
 NY excluding Metro NYC   41.1   35.0 
Parent         
 Age (years)61028.4(8.1) 215229.0(9.7) 0.17
 Race/ethnicity616   1951   0.001
  Hispanic   31.5   38.2 
  Non-Hispanic, white   49.5   41.2 
  Non-Hispanic, black   19.0   20.7 
 Education604   2128   0.39
 <High school   32.1   29.8 
 High school/GED   42.1   41.9 
 >High school   25.8   28.3 
Maternal BMI (kg/m2)27627.4(6.2) 87027.5(6.3) 0.67

Unadjusted Analyses

At baseline, 17% of the total child sample was overweight. The prevalence of childhood overweight differed by ethnicity, with non-Hispanic blacks and non-Hispanic whites having significantly lower rates than Hispanics (p < 0.05), whereas there were no differences between blacks and whites. Six-month rate of infant weight gain did not differ by category of breast feeding duration, birth weight (1800 to 2500 grams vs. ≥ 2500 grams), parental education attainment, or mother's BMI group (all p > 0.05). Mean [standard deviation (SD)] 6-month rate of infant weight gain, however, was greater for boys than for girls [7.8, 100 g/mo (1.8) vs. 7.0, 100 g/mo (1.6), respectively (p < 0.0001)]. Birth weight and rate of 6-month weight gain were not significantly correlated (p > 0.05).

Controlling for the rate of infant weight gain, Hispanic children had twice the odds of being overweight at age 4 years compared with non-Hispanic white children (OR, 2.0; 95% CI, 1.3 to 3.2), whereas the odds of being overweight for non-Hispanic black children were comparable with those for non-Hispanic white children (OR, 0.8; 95% CI, 0.4 to 1.5). Because the odds of childhood overweight based on infant weight gain were similar for non-Hispanic blacks and whites, the two groups were combined in subsequent analyses (Figure 1).

image

Figure 1. Probability of childhood overweight (BMI ≥ 95th sex-specific percentile-for-age) at 4 years of age vs. rate of infant weight gain (between birth and 6 months of age) by race/ethnicity.

Download figure to PowerPoint

Breast-feeding history also differed by ethnicity. Hispanic mothers were twice as likely as non-Hispanic mothers to report having breast-fed for at least 1 month (OR, 2.0; 95% CI, 1.4 to 2.8; p < 0.0001), and they reported breast-feeding for a longer time period than non-Hispanic mothers [mean (SD), 5.2 (6.0) vs. 3.7 (5.6) months, respectively (p < 0.01)]. Maternal BMI [mean (SD), 27.7 kg/m2 (5.7) vs. 27.1 kg/m2 (6.2), p > 0.05] and prevalence of maternal overweight (35.3% vs. 29.5%, p > 0.05) or obesity (29.4% vs. 28.1%, p > 0.05), however, did not differ between Hispanic and non-Hispanic mothers, respectively. Infant birth weight [mean (SD), 3.3 kg (0.5) vs. 3.3 kg (0.6), respectively] and the rate of infant weight gain [mean (SD), 7.5, 100 g/mo (1.7) vs. 7.3, 100 g/mo (1.7), respectively] were similar between Hispanic and non-Hispanic mothers, respectively (both p > 0.05). The population-attributable risk of overweight at 4 years of age was 18.6% for children in the highest quintile for infant weight gain.

Univariate Logistic Regressions

The rate of infant weight gain, weight at 6 months of age, maternal BMI, and Hispanic ethnicity were associated with significantly increased odds of the child being overweight or being at-risk-for-overweight at 4 years of age (Table 2). The odds of the child being overweight or at-risk-for-overweight were ∼40% higher for each additional 100 g/mo weight gain between birth and 6 months of age.

Table 2. . Associations between participants’ characteristics and overweight status (BMI ≥ 95th percentile or BMI ≥ 85th percentile) at age 4 years
 BMI ≥ 85th percentileBMI ≥ 95th percentile
CovariatesNOR95% CIOR95% CI
  • *

    Results significant at p < 0.05.

  • Results significant at p < 0.01.

  • Results significant at p < 0.001.

Rate of infant weight gain (100 g/mo)6161.371.23 to 1.531.411.25 to 1.60*
Birth weight (100 g)6161.051.02 to 1.09*1.030.99 to 1.07
Parent education604    
 Less than high school vs. high school 1.370.92 to 2.021.210.76 to 1.94
 More than high school vs. high school 1.460.96 to 2.220.900.53 to 1.53
Breastfed (yes vs. no)6160.920.66 to 1.280.880.59 to 1.32
Ethnicity616    
 Hispanic vs. non-Hispanic 1.781.26 to 2.532.201.46 to 3.32
Sex (girls vs. boys)6160.870.63 to 1.210.850.56 to 1.27
Maternal BMI (kg/m2)2761.051.01 to 1.09*1.071.02 to 1.12
Weight at 6 months (100 g)6161.061.04 to 1.081.061.04 to 1.08

Multiple Logistic Regressions

In multivariate logistic regression, controlling for Hispanic ethnicity, birth weight, child gender, and breast-feeding history (Model I), increased infant weight gain (100 g/mo) continued to be significantly associated with increased odds that the child would be overweight or at-risk-for-overweight at age 4 years (Table 3). Children of Hispanic (vs. non-Hispanic) mothers continued to be twice as likely to be overweight or at-risk-for-overweight after controlling for potentially confounding variables. All first order interactions were entered into each model with no significant effects. Hosmer-Lemeshow goodness of fit statistics supported the adequacy of each model (all p > 0.05) for predicting overweight and at-risk-for-overweight.

Table 3. . Multiple logistic regression for association between rate of weight gain (birth to 6 months) and BMI at age 4 years controlling for selected covariates
 BMI ≥ 85th percentileBMI ≥ 95th percentile
CovariatesROR95% CIROR95% CI
  • *

    Results significant at p < 0.05.

  • Results significant at p < 0.01.

  • Results significant at p < 0.001.

Model I (N = 616)0.14  0.13  
 Rate of weight gain (100 g/mo) 1.431.27 to 1.60 1.441.26 to 1.65
 Ethnicity      
  Hispanic vs. non-Hispanic 1.821.26 to 2.64* 2.281.47 to 3.53
  Birth weight (100 grams) 1.071.04 to 1.11* 1.041.00 to 1.09*
 Sex (girls vs. boys) 1.270.88 to 1.83 1.200.77 to 1.88
 Breast-fed (≥1 vs. 0 months) 0.780.55 to 1.11 0.730.48 to 1.13
Model II (n = 276)0.14  0.24  
 Rate of weight gain (100 g/mo) 1.411.18 to 1.68 1.731.35 to 2.21
 Ethnicity      
  Hispanic vs. non-Hispanic 1.460.84 to 2.56 2.311.13 to 4.72*
 Birth weight (100 grams) 1.071.02 to 1.13* 1.070.99 to 1.15
 Sex (girls vs. boys) 1.530.88 to 2.65 2.151.02 to 4.54*
 Maternal BMI (kg/m2) 1.051.01 to 1.10* 1.091.03 to 1.15*

A second set of models (Model II) also included maternal BMI in the regression equation. Again, all first order interactions were entered into the equation and found to be non-significant and were, therefore, removed from the model. Breast-feeding history continued to have no significant relationship to either response variable and was removed from the model. The second set of models (Model II) confirmed the significant association between increased rate of infant weight gain and increased odds of childhood overweight and at-risk-for-overweight at 4 years of age (Table 3). Furthermore, the increased odds of childhood overweight associated with Hispanic ethnicity persisted after adjusting for maternal BMI and other potential confounders. In additional analyses, stratified by infant breast-feeding history (>1 month vs. never), the ORs associated with the rate of infant weight gain and other covariates were similar to the non-stratified analysis (data not shown).

Contour Plots

The relationship between infant weight gain and birth weight to the probability of childhood overweight at 4 years of age is visually shown in the contour plot (Figure 2). It should be noted that the probability of being overweight, for a given birth weight, is higher for children who gain more weight during the first 6 months of life than for children who gain less weight. Due to the small sample size, however, the probabilities associated with the contour lines should be cautiously interpreted.

image

Figure 2. Contour plot, probability of childhood overweight (BMI ≥ 95th sex-specific percentile-for-age) at 4 years of age by birth weight and infant weight gain.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Research Methods and Procedures
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgement
  9. References

In this study, the rate of infant weight gain during the first 6 months of life among a contemporary, multiethnic sample of children was associated with a significantly increased risk of childhood overweight at 4 years of age. This study confirms the associations noted previously among black and white infants and extends the finding to Hispanics. Hispanic ethnicity is also demonstrated to be an additional independent risk factor for childhood overweight, with Hispanics being more than twice as likely as non-Hispanics to be overweight at 4 years of age, after adjusting for infant weight gain, breast-feeding history, and other potential confounders.

The association noted between infant weight gain and risk of childhood overweight in this study (OR, 1.4; 95% CI, 1.2 to 1.6) among a contemporary cohort (born during 1995 to 1999) is remarkably similar to that found among a large cohort of U.S. children born some 40 to 45 years earlier (OR, 1.4; 95% CI, 1.3 to 1.4) (14). Additionally, the population-attributable risk of being overweight at 4 years of age associated with being in the highest quintile of 6-month weight gain (19%) is very similar to that previously reported by Stettler (21%) (14). These similarities are despite a 3-fold higher prevalence of childhood overweight currently, compared with the earlier study (19.0% vs. 5.4%, respectively) and a longer observed period of infant weight gain (6 vs. 4 months, respectively). Although the earlier study included limited data about breast-feeding (14), a strength of this study is that information regarding initiation and duration of breast-feeding was available for all children. Consistent with a recent study, the relationship between infant weight gain and risk of childhood overweight persisted after adjusting for breast-feeding history (13). The protective effect of breast-feeding on the risk of childhood overweight, however, was not found to be statistically significant in this study, possibly reflecting insufficient power (due to limited sample size). Nonetheless, the observed OR (OR, 0.73; CI, 0.48 to 1.13) was very similar in magnitude to that reported in a recent meta-analysis (OR, 0.78; CI, 0.71 to 0.85) (19).

A limitation of this study was that no growth data were available between birth and 6 months of life, and no information was collected about the exclusivity of breast-feeding or the age at which solids/foods other than milk or formula were introduced. In this study, breast-feeding was defined as at least 1 month in duration, whereas other studies have used a duration of 3, 4, or 6 months (25, 27). The rate of infant weight gain appears to vary during the first 4 to 6 months of life by the type of feeding. Studies suggest that breast-fed infants gain more weight during the first 1 to 2 months, whereas formula-fed infants gain more weight after the 3rd month of life (26). At 4 months of age, exclusively breast-fed infants consumed 20% fewer calories per day than did partially breast-fed infants (i.e., who were also fed formula or cow's milk) (26). These differences might be partially due to parents of formula-fed infants feeding more formula and/or introducing solids at an earlier age (27) or because breast-fed infants regulate their energy intakes at lower levels than formula-fed infants (28). Limited definitions of breast-feeding would not capture these variations, and a 6-month window would reflect the sum of these differences in growth.

A second limitation of this study is that the study population is not representative of all children, only children living in low-income families who participate in the WIC program in NYS. This, however, might be a strength because it reduces confounding due to income level. It is also possible that there was some unmeasured selection bias due to possibly non-random selection of clients who filled out the surveys. It should be noted, however, that the prevalence of childhood overweight in the study sample (17%) is very similar to that reported for the total NYS WIC population (16%) (30).

Finally, because gestational age at birth was not available, children with a birth weight <1800 grams were excluded from analysis. Thus, the effects of low birth weight due to small-for-gestational-age-size could not be evaluated separately in these analyses.

Conclusions

  1. Top of page
  2. Abstract
  3. Introduction
  4. Research Methods and Procedures
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgement
  9. References

In conclusion, the consistency and robustness of the relationship between rapid infant weight gain and the increased odds of childhood overweight supports infancy as a critical period of increased susceptibility for becoming overweight. A better understanding of the factors contributing to increased infant weight gain, however, is needed. Nonetheless, recommendations for exclusive breast-feeding during the first 4 to 6 months of life are to be supported (28, 29). More prescriptive feeding advice (e.g., amount and frequency of formula for those not breast-feeding, and when, how much, and what complementary solid foods to start) might be warranted. The causes of rapid infant weight gain are not known. Overfeeding is one of the most likely causes, although decreased physical activity by increased use of confining devices such as infant seats, swings, or strollers could also play a role. Improved monitoring of infant growth by using growth charts and better assessment of dietary intake might lead to earlier identification of infants with rapid growth and/or excessive intake. To this end, the NYS WIC program now recommends routine measurement of length and weight for all infants at 6 months of age. Pediatric care providers, who routinely see infants at more frequent intervals, are even better positioned to identify and evaluate infants who are gaining excessive weight.

There is, however, a real need for more research to understand the parent/child mechanisms, biological/genetic, cultural/ethnic determinants, and environmental factors that convey increased risk for rapid infant weight gain and the development of childhood overweight and adult obesity. Only then can targeted strategies be developed, applied, and evaluated to decrease the risk, to prevent childhood overweight, and to stem the tide of the global obesity epidemic.

Acknowledgement

  1. Top of page
  2. Abstract
  3. Introduction
  4. Research Methods and Procedures
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgement
  9. References

B.A.D., H.H.S., and R.M.P. were supported, in part, by a grant from the Economic Research Service, Food Assistance and Nutrition Research Program, U.S. Department of Agriculture (Grant No. 43-3AEM-1-80075). We acknowledge the contributions of Liyi Cen in compiling this database, conducting statistical analyses, and preparing Figure 2. The views expressed reflect those of the authors and not necessarily those of the U.S. Department of Agriculture or the New York State Department of Health.

Footnotes
  • 1

    Nonstandard abbreviations: WIC, Special Supplemental Nutrition Program for Women, Infants, and Children; NYS, New York State; OR, odds ratio; CI, confidence interval.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Research Methods and Procedures
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgement
  9. References
  • 1
    Mokdad, A. H., Marks, J. S., Stroup, D. F., Gerberding, JL. (2004) Actual causes of death in the United States, 2000. JAMA 291: 12381245.
  • 2
    Troiano, R. P., Flegal, K. M., Kuczmarski, R. J., Campbell, S. M., Johnson, Cl. (1995) Overweight prevalence and trends for children and adolescents: The National Health and Nutrition Examination Surveys, 1963 to 1991. Arch Pediatr Adolesc Med. 149: 10851091.
  • 3
    Hedley, A. A., Ogden, C. L., Johnson, C. L., Carroll, M. D., Curtin, L. R., Flegal, KM. (2004) Prevalence of overweight and obesity among US children, adolescents, and adults, 1999–2002. JAMA 291: 28472850.
  • 4
    Dietz, WH. (1998) Health consequences of obesity in youth: childhood predictors of adult disease. Pediatrics 101: 518525.
  • 5
    Strauss, RS. (2000) Childhood obesity and self-esteem. Pediatrics 105: e15
  • 6
    Freedman, D. S., Dietz, W. H., Srinivasan, S. R., Berenson, GS. (1999) The relation of overweight to cardiovascular risk factors among children and adolescents: the Bogalusa Heart Study. Pediatrics 103: 11751182.
  • 7
    Fagot-Campagna, A., Pettitt, D. J., Engelgau, M. M., et al (2000) Type 2 diabetes among North American children and adolescents: an epidemiologic review and a public health perspective. J Pediatr. 136: 664672.
  • 8
    Whitaker, R. C., Wright, J. A., Pepe, M. S., Seidel, K. D., Dietz, WH. (1997) Predicting obesity in young adulthood from childhood and parental obesity. N Engl J Med. 337: 926927.
  • 9
    Guo, S. S., Wu, W., Chumlea, W. C., Roche, AF. (2002) Predicting overweight and obesity in adulthood from body mass index values in childhood and adolescence. Am J Clin Nutr. 76: 653658.
  • 10
    Healthy People 2010, Nutrition and Overweight http:www.healthpeopledocumenthtmlvolume219nutritionhtm (Accessed March 30, 2001).
  • 11
    Flegal, K. M., Troiano, RP. (2000) Changes in the distribution of body mass index of adults and children in the US population. Int J Obes. 24: 807818.
  • 12
    Rolland-Cachera, M. F., Deheeger, M., Bellisle, F., Sempe, M., Guilloud-Bataille, M., Patois, E. (1984) Adiposity rebound in children: a simple indicator for predicting obesity. Am J Clin Nutr. 39: 129135.
  • 13
    Ong, K. K., Ahmed, M. L., Emmett, P. M., Preece, M. A., Dunger, DB. (2000) Association between postnatal catch-up growth and obesity in childhood: prospective cohort study. Br Med J. 320: 967971.
  • 14
    Stettler, N., Zemel, B. S., Kumanyika, S., Stallings, VA. (2002) Infant weight gain and childhood overweight status in a multicenter, cohort study. Pediatrics 109: 194199.
  • 15
    Stettler, N., Kumanyika, S. K., Katz, S. H., Zemel, B. S., Stallings, VA. (2003) Rapid weight gain during infancy and obesity in young adulthood in a cohort of African Americans. Am J Clin Nutr. 77: 13741378.
  • 16
    Toschke, A. M., Grote, V., Koletzko, B., von Kries, R. (2004) Identifying children at high risk for overweight at school entry by weight gain during the first 2 years. Arch Pediatr Adolesc Med. 158: 449452.
  • 17
    Grummer-Strawn, L., Mei, Z. (2004) Does breastfeeding protect against pediatric overweight? Analysis of longitudinal data from the Centers for Disease Control and Prevention Pediatric Nutrition Surveillance System. Pediatrics 113: e81e86.
  • 18
    Dewey, KG. (2003) Is breastfeeding protective against child obesity? J Hum Lact. 19: 918.
  • 19
    Arenz, S., Ruckerl, R., Kolezko, B., von Kries, R. (2004) Breast-feeding and childhood obesity: a systematic review. Int J Obesity 28: 12471256.
  • 20
    Mei, Z., Scanlon, K. S., Grummer-Strawn, L. M., Freedman, D. S., Yip, R., Trowbridge, FL. (1998) Increasing prevalence of overweight among U.S. low-income preschool children: the Centers for Disease Control and Prevention Pediatric Nutrition Surveillance, 1983–1995. Pediatrics 101: E12
  • 21
    New York State Department of Health (September 1996) Women, Infants and Children (WIC) Program Manual, section #1183-C2 –Anthropometrics for infants and children 141.
  • 22
    Kuczmarski, R. J., Ogden, C. L., Guo, S. S., et al (2002) 2000 CDC growth charts for the United States: methods and development. Vital Health Stat 11 246: 1190.
  • 23
    World Health Organization (1995) Physical Status: The Use and Interpretation of Anthropometry: WHO Technical Report Series 854 World Health Organization Geneva, Switzerland.
  • 24
    Stokes, M. E., Davis, C. S., Koch, GG. (2000) Categorical Data Analysis Using the SAS System 2nd ed. Cary NC, SAS Institute, Inc..
  • 25
    Ong, K. K., Preece, M. A., Emmett, P. M., Ahmed, M. L., Dunger, DB. (2002) Size at birth and early childhood growth in relation to maternal smoking, parity and infant breast-feeding: longitudinal birth cohort study and analysis. Pediatr Res. 52: 863867.
  • 26
    Butte, N. F., Wong, W. W., Hopkinson, J. M., et al (2000) Infant feeding mode affects early growth and body composition. Pediatrics 106: 13551366.
  • 27
    Neumann, C. G., Alpaugh, M. (1976) Birthweight doubling time: a fresh look. Pediatrics 57: 469473.
  • 28
    Dewey, KG. (2001) Nutrition, growth, and complementary feeding of the breastfed infant. Pediatr Clin North Am. 48: 87104.
  • 29
    Haisma, H., Coward, W. A., Albernaz, E. (2003) Breast milk and energy intake in exclusively, predominately, and partially breast-fed infants. Eur J Clin Nutr. 57: 16331642.
  • 30
    Edmunds, L. S., Wolfel, M. L., Dennison, B. A., Stratton, H. L., Pruzek, R. (2006) Overweight trends among children enrolled in the New York State Special Supplemental Nutrition Program for Women, Infants, and Children. J Am Diet Assoc. 106: 113117.
  • 31
    American Academy of Pediatrics, Committee on Nutrition (2004) Breastfeeding. Pediatric Nutrition Handbook 5578.5th ed, American Academy of Pediatrics Elk Grove Village, IL.
  • 32
    World Health Organization The Optimal Duration of Exclusive Breastfeeding: Report of an Expert Consultation, Geneva, Switzerland, 28–30 March 2001 (Document WHO/NHD/01.09) http:www.who.intnutinf.htm.