Adolescent Obesity, Change in Weight Status, and HypertensionNovelty and Significance
We sought to determine whether change in weight status between adolescence and young adulthood was associated with the risk of developing hypertension among adolescents and whether sex and racial/ethnic group differences existed in the National Longitudinal Study of Adolescent Health. The sample was restricted to participants who self-identified as black, Hispanic, or white non-Hispanic (n=8543). Height and weight were measured in adolescence (mean 16 years) and again in adulthood (mean 29 years). We categorized the weight of participants into 4 groups: stayed normal weight; gained weight (normal weight in adolescence and obese in adulthood); lost weight (overweight/obese in adolescence nonobese in adulthood); and chronically overweight/obese. Hypertension was defined as measured systolic blood pressure of at least 140 mm Hg or diastolic blood pressure of at least 90 mm Hg measured in adulthood or use of antihypertensive medications. A higher risk of hypertension was noted for all sex and racial/ethnic groups who became obese in adulthood. Furthermore, those who were chronically overweight/obese were at higher risk of hypertension for all groups, with odds ratios ranging from 2.7 in Hispanic men to 6.5 in Hispanic women. Except for black men, those who lost weight during follow-up had no significant increased risk compared with those who maintained normal weight. Overall, there was an increased risk of hypertension for those who gained weight in adulthood and among those who remained obese from adolescence to young adulthood. These data give further evidence for prevention strategies that begin earlier in life to reduce or delay the onset of chronic disease in young adults.
Childhood obesity is a major public health problem in the United States that disproportionately affects black and Hispanic children, as well as children of lower socioeconomic status.1,2 There are both immediate and long-term impacts of childhood obesity. For example, obese children are more likely to develop diabetes mellitus, hypertension, and elevated cholesterol levels, all cardiovascular risk factors, earlier in life.3 Furthermore, there are long-term consequences of childhood obesity on cardiovascular risk and adult cardiovascular health, which have only been recently examined.
Increasing evidence suggests that overweight/obese children become overweight/obese adults. In the Bogalusa heart Study, black men and women who were overweight in adolescence were 52% and 62%, respectively, more likely to be obese adults.4 Furthermore, evidence suggests there is an immediate impact of childhood obesity on cardiovascular risk. A pooled study of 8 large epidemiological studies noted that children in the upper decile of body mass index (BMI) had odds of hypertension ranging from 2.5 to 3.7 compared with children in the lower decile of BMI, regardless of age, race, and sex.5 In a recent study in Denmark, 50% of obese children referred to a weight loss clinic were hypertensive.6
Despite these findings, the long-term consequences of childhood obesity independent of adult obesity on health are largely unknown. Recent studies suggest that childhood obesity is associated with cardiovascular risk factors in adulthood.7,8 A recent meta-analyses from 4 prospective cohort studies documented an increase in the risk of hypertension among obese adults compared with nonobese adults; the magnitude of the effect was bigger if obese adults had also been overweight/obese adolescents.9 In contrast, those who were overweight/obese during childhood but were nonobese as adults had similar risk of hypertension as those who had a normal BMI during childhood and adulthood, suggesting the effects of childhood obesity are not permanent and there is a benefit of weight loss in adulthood; however, the study was unable to determine whether similar benefits were noted across racial/ethnic groups. Because certain racial/ethnic groups carry a bigger burden of the obesity epidemic, it is important to determine whether a similar effect is noted among various racial/ethnic groups.
There are few data sets that allow for measured height and weight across life stages in the same cohort. Thus, we examine the association between obesity status in adolescence and adulthood and prehypertension and hypertension in adulthood in the National Longitudinal Study of Adolescent Health (Add Health) and explore whether racial/ethnic and sex variations exist in the association.
The Add Health study is a nationally representative school-based, longitudinal study of the health-related behaviors of adolescents and their outcomes in young adulthood. An in-school questionnaire was administered to a nationally representative sample of students in grades 7 through 12, plus selected oversampled minority groups, stratified by age and sex, during the 1994–1995 school year in 132 schools. Four waves of in-home interviews (wave 1, 1994–1995; wave 2, 1996; wave 3, 2001–2002; wave 4, 2007–2008) were subsequently conducted. The study design has been described in detail elsewhere.10 Briefly, 80 high schools representative of US schools were selected with respect to region of country, urbanicity, size, type, and ethnicity. Eligible schools included an 11th grade and enrolled >30 students. The first wave of follow-up in-home interviews was conducted 1 year after the initial survey between 1994 and 1995 (wave 1), and ≈20 745 adolescents, between the ages of 12 and 18 years, completed the in-home questionnaire at that time. Wave 2 follow-up (1996, mean age 16 years), conducted 1 year after wave 1, included adolescents who would still be enrolled in high school during 1996, including dropouts. Older youths who were high school graduates in wave 1 were not followed in wave 2. Wave 2 serves as the baseline for this analysis because it is the first wave at which height and weight were objectively measured. Wave 4 (mean age 29 years) included 11 863 of wave 2 respondents. In these analyses, our sample consists of participants who self-identified as black, white non-Hispanic, or Hispanic and who had sampling weight information (n=10 065). Participants who were pregnant during wave 4 (n=351) or who were missing BMI data (n=407) or blood pressure (n=233), as well as those missing information on smoking status or sociodemographics were excluded, leaving 8543 as the final sample size for these analyses. The Add Health study was approved by the institutional review board of the University of North Carolina, Chapel Hill. These analyses were approved by the institutional review board of Columbia University, New York.
During the second wave of follow-up, 1 year after the baseline examination, home visits were conducted to assess the adolescent’s height and weight. BMI was calculated from measured height and weight (kg/m2). The Centers for Disease Control and Prevention BMI growth reference11 was used to determine age- and sex-specific BMI percentiles. Overweight during adolescence was defined as BMI ≥85th percentile at the wave 2 assessment. During the fourth wave of follow-up home visits were again conducted, which included height and weight measurement of participants. Obesity in adulthood was defined as BMI ≥30 at the wave 4 assessment. Four categories to define obesity status across adolescence and adulthood were created: (1) participants who were normal weight in adolescence and were not obese in adulthood were classified as having normal weight; (2) those who were not overweight/obese in adolescence but were obese in adulthood were classified as gaining weight; (3) participants who were overweight/obese in adolescence but were not obese in adulthood were classified as losing weight; and (4) those who were overweight/obese in adolescence and obese in adulthood were classified as chronically overweight/obese.
During the fourth wave of follow-up (2007–2008, mean age 29 years), an in-home assessment was conducted and 3 blood pressure measurements were obtained with a Microlife automatic blood pressure monitor after the participant was seated for 5 minutes. The average of the last 2 measurements was used to calculate the average systolic blood pressure (SBP) and diastolic blood pressure. Participants were asked to provide to the interviewer all medications they were currently taking or had taken in the previous 6 months. Antihypertensive medications (β-adrenergic blockers, calcium channel blockers, angiotensin-converting enzymes) were identified from all medications provided. Prehypertension was defined as having SBP between 121 and 140 mm Hg or a diastolic blood pressure between 81 and 89 mm Hg. Hypertension was defined as having an elevated SBP ≥140 mm Hg or diastolic blood pressure ≥90 mm Hg or reported using antihypertensive medications (n=256).
Questionnaires ascertained information on sociodemographic factors, including age, sex, race/ethnicity, parental highest education level achieved (wave 2), highest education level achieved (wave 4), individual-level household income (wave 4), and current daily smoking (wave 4). Highest education level was categorized as follows: less than high school, high school graduate, some college, college graduate, or graduate education. Parental education was classified as follows: less than high school, high school graduate, some college, college graduate, or graduate education. Household income in adulthood was categorized as <$20 000 or higher. Current daily smoking in adulthood was defined as smoking at least 1 cigarette per day in the past 30 days.
We examined the prevalence of overweight/obesity status across adolescence (wave 2) and adulthood (wave 4), as well as the prevalence of hypertension by race/ethnicity and sex. Multinomial logistic regression analyses were then conducted to estimate the association between obesity status and prehypertension and hypertension, adjusting for sociodemographic factors (age, education, household income, and smoking status), stratified by race/ethnicity and sex. An additional analysis was conducted to estimate the association between weight status and SBP adjusted for sociodemographic factors and stratified by race and sex. All analyses were weighted, accounted for clustering of the sample design, and conducted in SAS version 9.0 (SAS Institute, Cary, NC).
In the final sample of 8543 participants, 52% were men, 15% were black, and 12% were Hispanic (Table 1). The mean age of participants at adolescence was 16 years, and the mean age in the adult follow-up was 29.0 years. In adolescence, 30% of participants were overweight or obese; at follow-up the prevalence of obesity rose to 36%. Relative weight status across both time points differed by sex and race/ethnicity (Table 2). Black women and Hispanic women were more likely to be chronically overweight/obese (35% and 27%, respectively) compared with white women (19%; P<0.05). Among men, Hispanic men were more likely to be chronically overweight/obese (24%), although not statistically different from white men (22%). As expected, differences in the prevalence of hypertension by sex were noted; 28% of men were hypertensive compared with 13% of women. Racial/ethnic differences in the prevalence of hypertension were also noted. Among women, a higher prevalence of hypertension was noted among black women compared with white women (16% versus 12%; P<0.05). Among men, Hispanic men had the highest prevalence of hypertension (29%), although not statistically different from white men (25%).
In adjusted analyses examining prehypertension, significant associations were noted among black, Hispanic, and white women who became obese in adulthood or who were chronically overweight/obese in adolescence and remained obese in adulthood compared with women of normal weight at both time points (Table 3). Hispanic men and white men who became obese in adulthood, as well as white men who were chronically overweight/obese in adolescence and remained obese in adulthood, were more likely to be prehypertensive compared with men of normal weight at both time points.
In examining hypertension, significant associations were noted across all racial/ethnic and sex groups among those who were overweight/obese in adolescence and remained obese in adulthood or who were of normal weight during adolescence but became obese in adulthood (Table 3). Those who were formerly overweight but were not obese in adulthood had no increased risk of prehypertension and hypertension compared with those who were of normal weight during both time periods; this finding was noted across all groups except for black men.
We furthermore examined the relationship between change in weight status between both time points and SBP in adulthood (Figures 1 and 2). Black, Hispanic, and white men who were of normal weight in adolescence but became obese in adulthood had significantly higher SBP (4.4, 4.0, and 5.3 mm Hg, respectively) compared with men with normal weight across both time periods. However, higher effects were noted among black, Hispanic, and white men who were chronically overweight/obese (11.0, 5.3, and 5.9 mm Hg, respectively). Significant associations were also noted among black (8.3 mm Hg), Hispanic (7.9 mm Hg), and white women (6.5 mm Hg) who became obese in adulthood or who were chronically overweight/obese (6.9, 7.9, and 9.3 mm Hg, respectively) compared with women with normal weight across both time periods.
In this longitudinal study of diverse adolescents, we examined the association between change in obesity status, from adolescence to adulthood, and hypertension and explored whether racial/ethnic and sex differences exist. Our results are similar to the few studies that have noted a benefit of losing weight in adulthood among those who were overweight/obese in adolescence. In a recent pooled analyses of 4 US and European cohorts, men and women who were overweight/obese in childhood but nonobese as adults had similar risk of hypertension and other cardiovascular indices as those who remained of normal weight from childhood to adulthood.9 We extend this work by exploring racial/ethnic variations and note that losing weight in adulthood was associated with the same risk of hypertension as those who remained of normal weight from adolescence to adulthood for all racial/ethnic groups except for black men. It is plausible that other underlying conditions, which we did not account for, are driving this association.
We also noted a higher risk of hypertension among those who were chronically overweight/obese compared with those of normal weight across both time points. These effects were consistent across both sexes and for most racial/ethnic groups. The magnitude of the effect was higher among those chronically overweight/obese compared with those who became obese in adulthood, supporting the notion that the length of time one spends in an obese state has a detrimental effect on blood pressure. This is consistent with other studies,8,12 which have noted associations between childhood obesity and cardiovascular risk factors in adulthood. In the CARDIA (Coronary Artery Risk Development In Young Adults) study, weight gain (defined as gaining >5 pounds in a 15-year follow-up) was associated with adverse changes in blood pressure, with a differential effect noted by baseline weight status.13 Those who were overweight at baseline had a higher increase in SBP than those who were of normal weight at baseline. In contrast, Wright et al14 in a cohort of children from Newcastle (Newcastle Thousand Families Study) noted that childhood BMI was not associated with blood pressure in adulthood once adult BMI was accounted for in the analyses. However, they failed to consider that adult BMI is not independent of childhood BMI.
We also note racial/ethnic differences in the magnitude of effects. For all racial/ethnic groups, being chronically overweight seemed to infuse a higher risk of hypertension than gaining weight. For black women and Hispanic men, the magnitude of the effect was not very different for those who gained weight versus those who were chronically overweight. It is plausible that within these 2 groups, being overweight/obese in childhood does not confer an added risk of hypertension. Future studies should explore these racial/ethnic variations in other diverse cohorts and what factors are contributing to the differences across groups. Furthermore, we note sex differences in the relationship between weight status and blood pressure. Women across all racial/ethnic groups, who were chronically overweight/obese, were at an increased risk of being prehypertensive. In contrast, this relationship was only noted among white men. Furthermore, a greater change in SBP was noted among chronically obese white and Hispanic women compared with normal weight women than effects noted among white and Hispanic men. Sex differences in the association between obesity and hypertension have been previously noted15; however, results have not been consistent, with some studies reporting a stronger association among men, whereas others report stronger associations among women. The discrepancy in results could be a result of the different age groups and ethnicities represented in these studies.
The study has several strengths; we use a nationally representative sample of US high school adolescents with objectively measured BMI and blood pressure. The large and diverse sample allows for the examination of sex and racial/ethnic differences. Furthermore, we are able to adjust for sociodemographic factors and smoking status, addressing a limitation of one of the major studies in this area.9 There are also several limitations to the study. Our hypertension definition was based on a single time point measurement in adulthood, thus not capturing the potential variability of blood pressure. Although we use standard BMI cut points to define obesity, it is possible that these cut points are not capturing weight gain or loss that could also have an impact on hypertension, regardless of whether it changes weight status. Future studies should consider an additional measure of obesity such as waist circumference, which may aid in the identification of cardiovascular risk. We furthermore do not have information on obesity earlier in childhood. However, our goal was to characterize weight status as adolescents transition to adulthood, which has been shown to be a period of high risk for the development and maintenance of obesity.16 Future studies should explore trajectories of weight status from an earlier time point.
Overall, these data provide further support to the prevailing paradigm of prevention, which emphasizes prevention of risk factors early in life to reduce the impact of chronic diseases, such as cardiovascular disease and diabetes mellitus, in mid and later life. A salient message from this observational research and demonstrated in clinical trials17–19 is that losing weight is beneficial for clinical parameters such as hemoglobin A1C and blood pressure and even prevalence of disease. Long-term follow-up of those trials shows sustained weight loss results in sustained health benefits. Our study adds to this message and extends it to include primary prevention of weight gain early in life. Future studies should confirm these results in other diverse cohorts and further determine critical periods in the life course to target for prevention intervention efforts.
Sources of Funding
Dr Suglia was supported by grant number K01HL103199. Dr Gary-Webb was supported by K01-HL084700. Dr Clark was supported by grant numbers 1R03HD068045-01A1, UL1 RR033183, and 1KL2RR033182-0. This research uses data from National Longitudinal Study of Adolescent Health (Add Health), a program project directed by Kathleen Mullan Harris and designed by J. Richard Udry, Peter S. Bearman, and Kathleen Mullan Harris at the University of North Carolina at Chapel Hill and funded by grant P01-HD31921 from the Eunice Kennedy Shriver National Institute of Child Health and Human Development, with cooperative funding from 23 other federal agencies and foundations. Special acknowledgment is due to Ronald R. Rindfuss and Barbara Entwisle for assistance in the original design. Information on how to obtain the Add Health data files is available on the Add Health website (http://www.cpc.unc.edu/addhealth). No direct support was received from grant P01-HD31921 for this analysis.
- Received September 20, 2012.
- Revision received November 8, 2012.
- Accepted November 15, 2012.
- © 2013 American Heart Association, Inc.
- Bethell C,
- Simpson L,
- Stumbo S,
- Carle AC,
- Gombojav N
- Rosner B,
- Prineas R,
- Daniels SR,
- Loggie J
- Moriarty-Kelsey M,
- Daniels S
- Popkin BM,
- Udry JR
- 11.↵Statistics NCfH. 2000 CDC growth charts: United states. 2004.
- Wright CM,
- Parker L,
- Lamont D,
- Craft AW
- Gordon-Larsen P,
- Adair LS,
- Nelson MC,
- Popkin BM
- Pi-Sunyer X,
- Blackburn G,
- Brancati FL,
- et al
- Knowler WC,
- Fowler SE,
- Hamman RF,
- Christophi CA,
- Hoffman HJ,
- Brenneman AT,
- Brown-Friday JO,
- Goldberg R,
- Venditti E,
- Nathan DM
Novelty and Significance
What Is New?
We explore sex and racial/ethnic differences in the association between change in weight status from adolescence to adulthood and risk of developing hypertension.
What Is Relevant?
The epidemic of childhood obesity has been well characterized, yet the consequences of this epidemic, particularly over the long-term, have received less attention. Although it is known that adult obesity is a risk factor for the development of hypertension in adulthood, it is unknown whether obesity in adolescence constitutes a sensitive time period that is more relevant for the development of hypertension.
Our findings provide evidence that adolescent obesity has long-term implications for developing hypertension among both men and women, yet the effects are reversible if normal weight is achieved in adulthood for some racial/ethnic groups.