Sex, Obesity, and Blood Pressure Among African American Adolescents: The Jackson Heart KIDS Pilot Study

Abstract

BACKGROUND

This study examined the degree to which sex, weight status, and the presence of hypertension and obesity in parents/grandparents were associated with systolic (SBP) and diastolic (DBP) blood pressure (BP) among African American youth in a pilot offspring study examining obesity-related cardiovascular disease (CVD) risks among adolescents.

METHODS

Data for this study were drawn from 212 adolescents (mean age 15.2 ± 2.2 years, 50.5% female) participating in the Jackson Heart Kids Pilot Study – an offspring cohort study examining cardiovascular disease risks among descendants of Jackson Heart Study participants. SBP (mean 114.4 ± 11.8 mm Hg) and DBP (mean 66.8 ± 9.0 mm Hg) were the primary outcomes. Weight status (46.1% normal weight, 30.1% overweight, 23.6% obese) was the key independent variable. Other covariates were attempts to lose weight, fruit and vegetable consumption, physical activity, and spirituality.

RESULTS

Fully adjusted linear regression models of the total sample produced results indicating that obesity was associated with BP (SBP: β = 7.08, P < 0.01; DBP: β = 8.14, P < 0.001). Sex-stratified analyses indicated that overweight and obesity were associated with SBP (overweight: β = 6.77, P < 0.01; obese: β = 11.65, P < 0.001) and obesity was correlated with DBP (β = 9.86, P < 0.001) among males. For females, overweight was correlated with SBP (β = 4.11, P < 0.05) while obesity was associated with DBP (β = 6.98, P < 0.01). Attempting to lose weight was inversely related to SBP (β = −4.01, P < 0.05) in the full sample and among males (β = −11.94, P < 0.001). Familial presence of hypertension and/or obesity was significantly associated with SBP among adolescent females but not males.

CONCLUSIONS

The relationship between weight status, familial hypertension and obesity status, and BP among adolescents vary by sex. This study underscores the need for additional research investigating the relationship between individual sex, weight status, BP and familial BP, and obesity status on risk among African American adolescents.

Obesity is a major risk factor for hypertension and there has been a concomitant rise in hypertension and obesity over the past 2 decades.¹ Elevated blood pressure (BP) and excess weight have been linked to biologic, behavioral, social, and environmental factors shared across generations^2,3 and research examining the relationship between these cardiovascular disease (CVD) risk factors among young African Americans is critically needed. High BP among youth is growing problem that could have major implications for individuals, families, and communities over the life course.^4,5 Elevated BP can be associated with target organ damage, including left ventricular hypertrophy during childhood and adolescence^5–8 as well as cardiovascular and chronic kidney disease during adulthood.¹ Increasing incidence and prevalence of high BP among children and adolescents is particularly salient for African Americans.⁹ African Americans develop risk factors for hypertension and hypertension-related conditions earlier in the life course than other racial groups.^2,9,10 Few studies have examined factors associated with high BP among African American youth^11,12; however, an even smaller number has focused exclusively on this group.⁴

Studies have established the relationship between elevated BP and excess weight among children from varied racial and ethnic backgrounds^12–16; yet, research is less definitive about gender- or sex-specific patterns in the association between high BP and obesity among younger African Americans.^17,18 The combination of obesity and hypertension can be insidious because of the heightened risk for cardiovascular sequelae^19,20; therefore, research investigating the relationship between obesity and BP among groups with already elevated risks for CVD is important.^11,15,21

The purpose of this study was to examine the degree to which weight status during adolescence can be associated with elevated BP among African American adolescents in the Jackson Heart Study-KIDS Pilot Study (JHS-KIDS). We hypothesized that overweight and obese adolescents would have higher BP levels than normal weight participants in the study. We also tested the hypothesis that the association between weight status and BP varies by sex and is influenced by the presence of hypertension and obesity in parents/grandparents.

METHODS

Study data were drawn from the JHS-KIDS, a pilot study examining the development of obesity, metabolic, and CVD-related risk factors among African American youth in the greater Jackson, Mississippi area. Eligible participants were adolescents between ages 12–19 years and were either children or grandchildren of individuals enrolled in the JHS. After receiving approval from the University of Mississippi Medical Center Institutional Review Board, the research team used multiple methods to recruit participants through contact with JHS participants.^4,22 Interested parents or grandparents called the study phone number and answered screening questions to ensure eligibility. Two hundred twelve adolescents and a parent or grandparent accompanying them participated in the pilot study.

Adolescents in the study had their BP, height, and weight and waist circumference measured and were asked to complete self-administered surveys during the study visit. BP readings were obtained by trained staff with a random zero sphygmomanometer.²³ Participants completed a survey containing a number of psychosocial (i.e., religiosity/spirituality) and health behavior (i.e., nutrition, physical activity) measures. The parent/grandparent present at the study visit, regardless of their participation in the JHS, had their BP, height, and weight measured, and completed the Child Feeding Questionnaire if the child participant lived with them.

Study variables

Outcome variables.

BP was the primary outcome variable and was comprised of 2 measurements, systolic BP (SBP) and diastolic BP (DBP).

Independent variables.

Adolescent weight status was a 3-category (normal weight, overweight, obese) variable based on body mass index (BMI) cut points specified by the Childhood Obesity Working Group of the International Obesity Taskforce.²⁴ Participant measured height and weight were used to calculate crude BMI (BMI = weight in kilograms/height in meters²) and were collected using a Shorr Height Measuring Board and a Seca 770 Model scale, respectively. Two measurements were taken to ensure accuracy. Weight status categories were derived by transforming crude BMI to z-scores using the lambda-mu-sigma (LMS) method, allowing for the development of smoothed growth using the curves and the efficient calculation of z-scores simultaneously.^24,25Z-scores were standardized to the reference population for participants’ age and sex using the 2,000 Centers for Disease Control and Prevention Growth Reference in the United States.^26,27 The 3 catagores for 12- to 18-year-old study participants were normal weight (BMI < 85^th percentile), overweight (85^th percentile ≤ BMI < 95^th percentile), and obese (BMI ≥ 95^th percentile). The 3 weight status categories for study participants aged 19 were normal weight (BMI < 25), overweight (25 ≤ BMI < 30), and obese (BMI ≥ 30). Adolescents were asked if they attempted to lose weight and this variable was a derived from the question, “Which of the following are you trying to do about your weight?” The responses options were “lose weight,” “gain weight,” “stay the same,” or “I am not trying to do anything about my weight.” The “lose weight” response was coded “1” and all others were coded “0”.

Parent/grandparent obesity and hypertension status was dichotomous variable derived from anthropometic data collected during a study visit. Parent/grandparent height, weight, and BP were collected using the same procedures for adolescent participants. Obese parents/grandparents had BMIs at or over 30 and hypertensive parents/grandparents had SBP over 120 mm Hg or DBP over 90 mm Hg. Parents/grandparents who were obese and hypertensive were coded “1” and all others were coded “0”.

The covariates were based on self-report data collected during the study visit. All but one instrument were from the Behavioral Risk Factor Surveillance Study.²⁸ The fruit and vegetable consumption variable was a composite of the responses to items about participant intake of fruit and vegetables during the previous week. The physical activity measure was derived from the response to an item asking participants to provide the number of days they had at least 60 minutes of during the past week. The short version of the Daily Spiritual Experiences Scale (DSES) was used to measure spirituality.²⁹ Participants responded to 6 statements (e.g., “I feel God’s presence) responding to the question, “How often do you have the following experiences?” by selecting 1 of 6 possible responses ranging from “never” (coded 0) to “many times a day” (coded 5). A dichotomous variable indicating whether the family member accompanying the adolescent is a grandparent (coded 1) or parent (coded 0). Age (in years) was also included in the analysis.

Analytic strategy

Sample characteristics were described for the total sample, by sex and by attempted weight loss using means and SDs for continuous variables and proportions for categorical variables. T-tests and chi-square tests were used in descriptive analyses to assess how sex- and attempted weight loss-specific groups varied across all variables. Multivariable ordinary least squares regression models were estimated to determine the impact of weight status on BP adjusting for covariates. Models were conducted for the total sample and by sex. P values less than 0.05 were considered significant. All tests were 2-sided. All statistical analyses were conducted with StataSE Version 12.

RESULTS

The description of characteristics of 212 African American adolescents in the JHS-KIDS study is displayed for the total sample and by sex and attempted weight loss in Table 1. The adolescent study population was equally divided between males and females and had an average age slightly over 15 years (15.2 ± 2.2). The average SBP and DBP were 114.4 ± 11.8 mm Hg and 66.8 ± 9.0 mm Hg, respectively. Over half of the cohort (53.7%) were carrying excess weight and nearly half of adolescents (49.5%) reported at least one previous attempt to lose weight. Nearly one-third of the adolescents in the study (30.1%) were overweight and 23.6% were obese. Cohort members reported consuming an average of 6.6 ± 4.7 fruits and vegetables daily and exercising 60 minutes an average of 3.4 ± 2.3 days a week. The mean spirituality score (18.7 ± 6.3) suggested that cohort members had spiritual experiences between 4 and 6 days during a given week. Roughly 30% of the adult participants were grandparents and slightly more than one-fifth of all parent/grandparents in the study (21.7%) were obese and hypertensive. Some measures were found to vary by sex or attempted weight loss. Adolescent males in the study had a higher mean SBP, levels of physical activity, and a smaller proportion of individuals who had tried to lose weight relative to females. Individuals who tried to lose weight had a higher mean DBP and a larger proportion of overweight and obese individuals relative to those did not. There were no differences observed by sex or attempted weight loss for the majority of measures in the analysis.

Table 1.

Distribution of characteristics of African American adolescents in the Jackson Heart Kids Pilot Study for the total sample, by sex, and by attempted weight loss

Variable		Sex			Attempted weight loss
	Total sample, N = 212	Females, N = 107	Males, N = 105	P value	No attempts, N = 107	Prior attempts, N = 105	P value
Systolic blood pressure (mm Hg) [mean ± SD]	114.4 ± 11.8	111.6 ± 9.9	117.4 ± 12.9	0.001	115.4 ± 13.0	113.4 ± 10.4	0.200
Diastolic blood pressure (mm Hg) [mean ± SD]	66.8 ± 9.0	67.8 ± 8.7	65.7 ± 9.2	0.090	65.5 ± 7.8	68.1 ± 9.9	0.030
Sex [%]							0.002
Female	50.5	–	–		40.1	61.0
Male	49.5	–	–		59.8	39.1
Age (years) [mean ± SD]	15.2 ± 2.2	15.2 ± 2.1	15.1 ± 2.2	0.930	15.1 ± 2.2	15.2 ± 2.2	0.868
Grandparent [%]	29.7	29.9	29.5	0.951	26.1	33.3	0.254
Weight status [%]				0.310			0.001
Normal weight	46.2	41.1	51.4		73.8	18.1
Overweight	30.1	33.6	26.7		22.4	38.1
Obese	23.6	25.2	21.9		3.7	43.8
Obese and hypertensive parent/ grandparent [%]	21.7	22.4	20.9	0.794	20.6	22.9	0.685
Ever attempted to lose weight [%]	49.5	59.8	39.1	0.002	–	–	–
Fruit and vegetable consumption score [mean ± SD]	6.6 ± 4.7	6.1 ± 4.5	7.1 ± 4.9	0.150	6.1 ± 4.1	7.2 ± 5.2	0.067
60+ mins physical activity (days) [mean ± SD]	3.4 ± 2.3	2.9 ± 2.1	3.9 ± 2.3	0.001	3.6 ± 2.4	3.2 ± 2.1	0.240
Spirituality score [mean ± SD]	18.7 ± 6.3	19.2 ± 5.9	18.3 ± 6.7	0.330	18.0 ± 6.7	19.5 ± 5.9	0.073

Bold indicates where P value is significant at the P < 0.05 level.

SBP and weight status

The relationship between SBP and weight status for the total sample and by sex is shown in Table 2. The results in the fully adjusted model for the total sample indicate that overweight and obese adolescents had SBPs that were on average 5.33 mm Hg (P < 0.01) and 7.08 mm Hg (P < 0.01) higher than normal weight study participants, respectively. Controlling for potential confounders, overweight, and obese adolescent males in the study had SBPs that were on average 6.77 mm Hg (P < 0.01) and 11.65 mm Hg (P < 0.001) higher than normal weight adolescent males, respectively. By contrast, overweight adolescent females had SBPs that were on average 4.11 mm Hg (P < 0.05) higher than normal weight adolescent females while the differences in the average SBPs for obese compared to normal weight adolescent females did not reach statistical significance.

Table 2.

Relationship between systolic blood pressure and weight status among 212 African American adolescents in the Jackson Heart Kids Pilot Study for the total sample and by sex

Variable	Total sample β (SE)	Females β (SE)	Males β (SE)
Male	5.19 (1.6)***	–	–
Age (years)	1.21 (0.35)***	0.77 (0.48)	1.28 (0.53)*
Grandparent	−0.49 (1.68)	1.42 (2.04)	−1.47 (2.60)
Weight status
Overweight	5.33 (1.95)**	4.11 (2.28)*	6.77 (3.13)**
Obese	7.08 (2.42)**	4.34 (2.80)	11.65 (3.97)**
Obese and hypertensive parent/grandparent	−0.71 (1.88)	4.48 (2.25)*	−4.90 (2.94)
Ever attempted to lose weight	−4.01 (1.97)*	1.17 (2.21)	−11.94 (3.35)***
Fruit and vegetable consumption score	−0.19 (0.17)	−0.28 (0.22)	0.03 (0.26)
60+ minutes physical activity (days)	0.45 (0.38)	0.13 (0.55)	0.17 (0.57)
Spirituality score	−0.17 (0.12)	−0.36 (0.16)*	0.10 (0.19)

*P < 0.05, **P < 0.01, ***P < 0.001.

DBP and weight status

The relationship between DBP and weight status for the total sample and by sex is shown in Table 3. The results from fully adjusted model for the total sample indicate that obese adolescents had DBP levels 8.14 mm Hg (P < 0.001) higher than normal weight study participants. The sex-specific models yielded similar results. Controlling for potential confounders, obese adolescent females, and males in the study had DBPs that were on average 6.98 mm Hg (P < 0.01) and 9.86 mm Hg (P < 0.001) higher than normal weight adolescent females and males, respectively.

Table 3.

Relationship between diastolic blood pressure and weight status among 212 African American adolescents in the Jackson Heart Kids Pilot Study for the total sample and by sex

Variable	Total sample β (SE)	Females β (SE)	Males β (SE)
Male	−1.7 (1.15)	–	–
Age (years)	1.32 (0.26)***	1.27 (0.39)***	1.14 (0.36)**
Grandparent	−0.76 (1.21)	2.92 (1.64)	−3.40 (1.75)
Weight status
Overweight	1.98 (1.41)	0.65 (1.83)	3.30 (2.11)
Obese	8.14 (1.75)***	6.98 (2.25)**	9.86 (2.67)***
Obesity and hypertensive parent/grandparent	−1.93 (1.36)	2.67 (1.81)	−5.40 (1.99)
Ever attempted to lose weight	−1.09 (1.42)	0.32 (1.79)	−3.8 (2.26)
Fruit and vegetable consumption score	−0.13 (0.13)	−0.12 (0.18)	−0.80 (0.18)
60+ minutes physical activity (days)	−0.05 (0.28)	−0.28 (0.44)	−0.20 (0.38)
Spirituality score	0.003 (0.09)	−0.17 (0.13)	0.16 (0.13)

*P < 0.05, **P < 0.01, ***P < 0.001.

Familial hypertension and/or obesity and adolescent BP

Familial presence of hypertension and/or obesity was significantly associated with SBP among adolescent females in the study. Adolescent females with parents/grandparents who were obese and hypertensive had SBPs that were 4.48 mm Hg higher than their corresponding female peers. By contrast, there was no significant association with adolescent males.

Additional covariates and BP

Others factors were also found to be associated with BP. The results in the fully adjusted model for the total sample indicated that attempting to lose weight is inversely associated with SBP. Adolescents in the sample who reported earlier attempts to lose weight has SBPs that were on average 4.01 mm Hg (P < 0.05) lower than those never attempting to lose weight. Results in the fully adjusted stratified SBP models indicated that individuals attempting to lose weight were similar to those not attempting to lose weight on all of the variables except DBP, sex, and weight status and attempting to lose weight was most robust for males. After controlling for potential confounders, the SBP for males who attempted to lose weight were 11.94 mm Hg (P < 0.001) lower on average than males who never attempted to lose weight. A final finding of note was the relationship between spirituality and SBP among female sample members. A 1-point increase in the spirituality score was associated with a 0.33 mm Hg (P < 0.05) reduction in SBP among females in the fully adjusted model.

DISCUSSION

Hypertension is a growing problem among children and adolescents and research examining its risk factors is critically needed. This is the first study to our knowledge to examine the association between weight status and BP by sex in adolescent African Americans and its relationship to the presence of hypertension and obesity in parents/grandparents. A number of studies have examined the relationship between excess weight and BP in children,^{11–13,15,20} but only a few report data from African American youth.^11,21,30 There have been mixed findings on the impact of weight loss on BP in youth,^31–33 but our study was novel in assessing and finding that intent to lose weight was associated with lower BP, although the reasons for this finding remain elusive. Rosner and colleagues³⁰ combined data from 8 large epidemiological sources to examine BP differences in relation to body size among African American and White children and adolescents and found at lower levels of BMI African American children had high levels of BP, but at higher levels of BMI White children had higher BP levels. Brady and colleagues²¹ reported higher BP levels in African American youth compared to White children in a cohort of 184 children with a mean BMI percentile >85%. Finally, Freedman and colleagues¹¹ reported minimal changes in BP over time despite a nearly tripling in obesity in over 11,000 African American and White adolescents and no significant racial differences. Our study extends these lines of research with an exclusive focus on African Americans and provides important data about the relationship between excess weight and elevated pressure among a population with elevated risks for obesity and hypertension. Our findings from a cohort of adolescent JHS descendants provides a more refined understanding of the relationship between weight and BP among African Americans as they highlight the manner in which sex and weight status can have implications for hypertension risk. The descriptive results suggest that gender might be an important factor associated with BP among African American youth as males have higher mean SBP than females in the study. The difference in SBP between males and females is considerable; however, this finding is supported by studies using national-level data.³⁴ Over half of the sample is overweight or obese with no significant difference in the proportions of males and females carrying excess weight. These notable findings indicate that the levels of overweight and obesity among males in our cohort are equivalent to those for females. Recent national-level data indicate that adolescent females have higher a prevalence of overweight and obesity than adolescent males¹⁷; however, additional studies are needed to confirm our findings. Nonetheless, results from both analyses suggest that future pediatric obesity prevention and treatment efforts should seek to address both males and females.⁴ Gender parity has been difficult to achieve in intervention studies of African Americans to increase healthy eating and physical activity.³⁵

Our findings suggest that gender can have implications for the relationship between weight status and BP among African American adolescents. Sex remains a significant factor for SBP in the overall model even after controlling for known confounders. This finding is consistent with a recent study using national-level data³⁶ and indicates a need for further exploration. Results from our sex-stratified regression models provide evidence suggesting the relationship between weight status and BP varies by sex. SBP among males was significantly influenced by being overweight and obese while only overweight status was found to be associated with SBP among adolescent females in the study. The nuanced relationship between weight status and BP among adolescent females is consistent with findings from earlier studies. Using national-level adult data, Crossrow and Falkner³⁷ present findings indicating that disproportionately high prevalence of obesity did not translate into equally high rates of obesity-related comorbidties such as hypertension. We also found sex differences in relationship between familial obesity and hypertension and BP. SBP among female adolescents was significantly influenced by having a parent or grandparent with obesity-related comorbidities. The findings for female adolescents highlight the need for additional research identifying other social, behavioral, and biological factors affecting BP in sex-specific ways.

Our results also suggest that weight management can have positive implications for BP as adolescents attempting to lose weight had lower SBP than individuals who never tried to lose weight. Weight loss is an established hypertension prevention strategy for adolescents carrying excess weight⁵ and our study suggests that even an attempt to lose weight can have cardiovascular benefit, especially for males. Other weight-related behaviors are not significant in the overall models; however, the weight management variable may be confounding the effects of fruit and vegetable consumption or physical activity. Larger studies are needed to parse out these effects.

This study is significant because the results associated with overweight and BP demonstrate how subgroup analyses can highlight patterns masked in pooled models. This research contributes to our understanding of the relationship between sex, weight status, and high BP among African American adolescents; however, there are notable limitations. The analytic models are estimated using data drawn from a sample of African American adolescents who reside in the South and are descendants of individuals enrolled in a longitudinal study. Members of our study population are also heavier and have higher SBPs and DBPs than a those in a recent study using national-level data.³⁶ Therefore, the results may not be generalizable to the larger population of African American adolescents. However, it is important to understand determinants and patterns of obesity in the South given the high rates of childhood obesity in Mississippi and other Southern states.³⁸ The models in this study were estimated using cross-sectional data, which does not allow for the specification of temporal events or determination of causal inferences. The sample size limits the number of independent variables included in regression analysis, thereby limiting the number of factors considered. Limitations associated with self-report data such as recall bias and social desirability also apply to this study. However, the self-reported data uses items from well-established national surveys and BP is measured using a random zero sphygmomanometer, the highest standard in BP measurement.

CONCLUSION

While we found that obesity was associated with increased BP among African American adolescents, this study provides insight into the complex relationship between sex, weight status, and high BP status among African American youth. Additional studies are needed to examine which social, psychological, behavioral, biologic, and familial factors have implications for the relationship between not only weight status and high BP, but more robust markers of early cardiovascular risk. Research is also needed to specify the manner through which early weight gain and what constellation of associated risk factors accelerate the development and progression of premature CVD-related diseases among male and female children and adolescents over the life course, thereby laying the foundation for tailored interventions that can reduce risks for premature morbidity, disability, and mortality among African Americans.

DISCLOSURE

Drs Bruce, Beech, Norris, Griffith, Sims, and Thorpe have no financial relationships with the Department of Health and Human Services’ Office of Minority Health or the National Institutes of Health and other authors declared no conflict of interest.