Interpreting the continued decline in the average age at menarche: results from two nationally representative surveys of U.S. girls studied 10 years apart

OBJECTIVES: To determine whether average age at menarche declined in the United States during the past decade, and whether associations between menarcheal timing, weight status, and race/ethnicity changed. STUDY DESIGN: Relative weight, race/ethnicity, and menarcheal status of girls (n = 1577) in the National Health and Nutrition Examination Survey (NHANES) III (1988-1994) were compared with those of girls (n = 1720) in NHANES 1999-2002. Probit analysis estimated average age at menarche overall and also by race/ethnicity. Logistic regression assessed associations of relative weight and race/ethnicity with menarcheal status. RESULTS: In the United States, average age at menarche declined from 12.53 years (95% confidence interval [CI] = 12.43 to 12.63 years) in 1988-1994 to 12.34 years (95% CI = 12.24 to 12.45 years) in 1999-2002. By race/ethnicity, average age at menarche estimates were as follows: non-Hispanic whites, 12.57 years (95% CI = 12.45 to 12.69 years) and 12.52 years (95% CI = 12.38 to 12.67 years); non-Hispanic blacks, 12.09 years (95% CI = 11.82 to 12.36 years) and 12.06 years (95% CI = 11.81 to 12.32 years); and Mexican Americans, 12.24 years (95% CI = 11.88 to 12.59 years) and 12.09 years (95% CI = 11.81 to 12.37 years). Higher relative weight was consistently associated with increased likelihood of having reached menarche. CONCLUSIONS: Average age at menarche in the United States declined by 2.3 months between 1988-1994 and 1999-2002; by race/ethnicity, declines were considerably smaller. Changes in the population distribution of race/ethnicity and relative weight should be considered when interpreting trends in age at menarche.     

Sex differences in the association between sleep and body mass index in adolescents

OBJECTIVE: To determine whether an association between short sleep duration and increased body mass index (BMI) exists in a sample of U.S. adolescents. STUDY DESIGN: Public-use dataset of the National Longitudinal Study of Adolescent Health. Final sample included 4486 adolescents (51% female). Main outcome was BMI transformed into z-scores for age and sex using reference values from the Centers for Disease Control/National Center for Health Statistics. Overweight was defined as > or =95th percentile. Linear and logistic regression models were calculated. Sleep duration was self-reported in hours. A quadratic term for sleep was added to test curvilinear association. Covariates included age, race, parental education, activity and inactivity scores. RESULTS: Among males, linear regression indicated that sleep duration significantly predicted BMI z-score (Beta = -0.08, 95% CI: -0.12, -0.03). Logistic regression indicated that sleep duration predicted risk of overweight among males (OR = 0.90, 95% CI: 0.82, 1.00). Sleep duration was not a significant predictor among females in either regression model. Quadratic term for sleep was not significant for either sex. CONCLUSIONS: Longer sleep duration was weakly associated with lower BMI and risk of overweight among male adolescents only. This sex-related difference may be due to differences in the physiology of puberty or in sleep characteristics.     

Television viewing in early childhood predicts adult body mass index

OBJECTIVES: To examine the effects of duration, timing and type of television (TV) viewing at age 5 years on body mass index (BMI) in adult life. STUDY DESIGN AND METHODS: 1970 British Birth Cohort, followed up at 5 (N=13,135), 10 (N=14,875), and 30 years (N=11,261). OUTCOME MEASURES: Weekday and weekend TV viewing at 5 years, type of programs, and maternal attitudes toward TV at age 5 years. BMI z-score at 10 and 30 years. RESULTS: Mean daily hours of TV viewed at weekends predicted higher BMI z-score at 30 years (coefficient=0.03, 95% CI: 0.01, 0.05, P=.01) when adjusted for TV viewing and activity level at 10 years, sex, socioeconomic status, parental BMIs, and birth weight. Each additional hour of TV watched on weekends at 5 years increased risk of adult obesity (BMI > or =30 kg/m2) by 7% (OR=1.07, 95% CI 1.01, 1.13, P=.02). Weekday viewing, type of program and maternal attitudes to TV at 5 years were not independently associated with adult BMI z-score. CONCLUSIONS: Weekend TV viewing in early childhood continues to influence BMI in adulthood. Interventions to influence obesity by reducing sedentary behaviors must begin in early childhood. Interventions focusing on weekend TV viewing may be particularly effective.     

Body mass index charts: useful yet underused

OBJECTIVES: To determine how frequently pediatricians use body mass index (BMI) and whether pediatricians are more likely to regard a child as "too fat" and manifest greater concern about health sequelae when presented with BMI versus height and weight charting. STUDY DESIGN: North Carolina Pediatrics Society members completed a self-administered, quasi-experimental mail survey, using two different case vignette versions. They were given a clinical vignette but systematically received either the same hypothetical overweight child's height and weight data, percentile, and charts (Ht and Wt Group) or her BMI, percentile, and chart (BMI Group). They rated levels of fatness and concern by using Likert scales and the frequency of use of methods to determine overweight. RESULTS: Adjusted response rate was 71% (N = 356). The BMI Group rated the hypothetical child with a higher mean on a scale of fatness (P < .0001) and reported higher levels of concern about all consequences (all P values < or = .01) than those in the Ht and Wt Group. Only 11% of respondents reported "always," and 31% reported "never," using BMI. CONCLUSIONS: BMI charting prompted greater recognition of a weight problem than height and weight charting, yet BMI is inconsistently used. Interventions to help pediatricians adopt this tool may be warranted.     

Prevalence of overweight in children with developmental disorders in the continuous national health and nutrition examination survey (NHANES) 1999-2002

OBJECTIVE: To estimate the prevalence of overweight in children identified with developmental disorders on the basis of nationally representative survey data. STUDY DESIGN: We estimated the prevalence of overweight in children with developmental disorders on the basis of a recent large nationally representative survey. The continuous National Health and Nutrition Examination Survey (NHANES) 1999-2002 included 4 questions to identify children with developmental disorders. Height and weight were used to calculate body mass index (BMI). BMI percentiles were estimated relative to the age- and sex-specific Centers for Disease Control and Prevention growth reference. The 85th percentile BMI defined at-risk-for-overweight and the 95th percentile BMI defined overweight. RESULTS: We found a higher prevalence of at-risk-for overweight and overweight among children with limitations in physical activity and a higher prevalence of overweight in girls with learning disabilities, compared with children without these conditions, after adjustment for age and race-ethnicity. CONCLUSION: To the extent that children with developmental disorders are included in large representative surveys, the data suggest that children with developmental disorders have a risk for overweight that is at least as great as that of typically developing children.     

Neighborhood disadvantage as a risk factor for pediatric obstructive sleep apnea

OBJECTIVE: Low socioeconomic status (SES) is associated with a range of health outcomes. Our objective was to study the relationship between residence in a neighborhood of severe socioeconomic disadvantage and childhood obstructive sleep apnea (OSA). STUDY DESIGN: Cross-sectional analysis of 843 (49% female, 36% African-American) children 8 to 11 years of age from a community-based cohort. Data on neighborhood conditions were obtained from the 2000 US Census. The main outcome measure was OSA, defined as an obstructive apnea hypopnea index >5 events per hour or an obstructive apnea index >1 event per hour. RESULTS: Residence in a neighborhood of severe socioeconomic disadvantage was significantly associated with OSA after adjusting for effects of previously established risk factors: premature birth, obesity, and African-American ethnicity (OR = 3.44, 95% CI = 1.53-7.75). Secondary analyses showed that neighborhood disadvantage remained significantly associated with OSA: (1) in the African-American subgroup, after controlling for effects of prematurity and obesity; and (2) after controlling for indicators of household-level SES or other health characteristics. CONCLUSIONS: Childhood OSA is associated with low SES as measured by an index describing severe neighborhood disadvantage, emphasizing the potential importance of environmental factors, particularly those associated with neighborhood distress, as risk factors for OSA.