OBJECTIVE. Childhood overweight and obesity may result in premature onset of cardiovascular risk factors such as hypertension. Rural populations in North America may be at increased risk for overweight. We evaluated whether overweight and obesity were associated with prehypertension and hypertension in a well-characterized population of children in rural Canada.
METHODS. The study population for this cross-sectional study was composed of children (aged 4–17 years) who were participants of the Walkerton Health Study (Canada) in 2004. Prehypertension and hypertension were defined on the basis of percentiles from the average of 3 blood pressure measures taken on a single occasion. Percentiles for BMI and blood pressure were calculated by using the 2000 Centers for Disease Control and Prevention growth charts. Multinomial logistic regression was used to evaluate the odds for prehypertension and hypertension resulting from overweight and obesity.
RESULTS. Of 675 children (98.7% white), 122 (18.1%) were overweight and 77 (11.4%) were obese. Prehypertension and hypertension were detected in 51 (7.6%) and 50 (7.4%), respectively. After adjustment for family history of hypertension and kidney disease, obesity was associated with both prehypertension and hypertension. Overweight was associated with hypertension but not prehypertension. These associations were observed across the genders and children aged <13 and ≥13 years, except that overweight was not associated with hypertension among girls.
CONCLUSIONS. In this population of children who lived in a rural community in Canada, overweight and obesity were strongly associated with elevated blood pressure. Whether blood pressure normalizes with improvements in diet, physical activity, and environment is an area for additional study.
The prevalence of childhood obesity is increasing at an alarming rate in developed countries. Between the years 1981 and 1996, the prevalence of obesity among Canadian children aged 7 to 13 years has more than doubled, from 5% to 14% among boys and to 11% among girls.1 Increasing evidence suggests that this epidemic of childhood obesity is causing premature onset of hypertension, insulin resistance, and dyslipidemia,2 resulting in increased risk for adult coronary heart disease.3,4 Some have predicted that the current generation of children may be the first to have poorer health outcomes and a shorter life span than their parents.5 This has become a significant public health concern, and in October 2005, the Ontario Medical Association published a position statement calling for evidence-based analysis of the scope of the obesity problem and its risk factors, prevention strategies, and changes required at the level of public policy.6
Rural populations in North America may be at increased risk for overweight as a result of a clustering of risk factors that occur more frequently in rural than urban populations. These include lower socioeconomic status, poor dietary habits, and limited recreational facilities and opportunity for physical activity.7–10 Whereas studies suggest that blood pressure (BP) in North American youth is increasing in parallel with weight,11–13 hypertension may be underdiagnosed given that BP varies with age, gender, and height. Accurate diagnosis requires the use and interpretation of standardized growth charts.13 Underdiagnosis may be of greater concern in rural communities, where access to physician care is limited.14 In this study of children who lived in the rural community of Walkerton, Ontario, we evaluated whether prehypertension and hypertension were associated with overweight and obesity while controlling for other factors.
Design and Participants
The data for this cross-sectional study came from the Walkerton Health Study (WHS), a prospective cohort study that evaluated the long-term health sequelae from exposure to Escherichia coli O157 and Campylobacter in May 2000 in Walkerton, Ontario (Fig 1). 15 Methodologic details of the WHS are described elsewhere.15–17 In brief, all individuals who either were residents of the Walkerton area or had consumed municipal water at the time of the outbreak were invited to participate in the WHS irrespective of whether they had developed an acute illness. Recruitment began in 2002, and new participants were allowed to join in subsequent years. Participants of this study were those who attended the WHS clinic in 2004 (n = 3161), 687 of whom were younger than 18 years and eligible for this study. The sample has previously been shown to be representative of the target population at the time of the outbreak: the proportion of children who were younger than 20 years in the WHS (31%) was similar to Canadian Census figures for the rural communities of Walkerton (27%) and Brockton (28%).18 Written consent was obtained from all participants. Ethics approval for the WHS was obtained from the University of Western Ontario's Research Ethics Board for Health Sciences.
Measures and Definitions
Participants attended an annual clinic at the local hospital or high school and completed a 20- to 45-minute face-to-face, computer-assisted interview that contained questions on previous and current health conditions, family history, and smoking status. The presence of health conditions that predated the outbreak was corroborated by a medical chart review. Trained study personnel measured each participant's height, weight, and BP by using standardized protocols. Standing height with shoes removed was measured with a measuring tape to the nearest millimeter. Body weight was measured to the nearest 0.1 kg on calibrated digital scales. One manual and 2 automated (Dinamap, Tampa, FL) BP readings, separated by 1 minute, were taken of seated participants with arm resting and palm facing upward and by using a cuff size that was appropriate for the measured arm circumference.19 The mean of the 3 BP measures was used for this analysis. Crohbach's α for the 3 systolic and diastolic pressures was .86 and .73, respectively. As part of the study protocol, within-subject variability of >20 mmHg (for systolic BP [SBP]) or >13 mmHg (for DBP) was flagged by using an electronic algorithm.
BMI was categorized by using the 2000 Centers for Disease Control and Prevention growth charts for boys and girls aged 2 to 20 (underweight: <5th percentile; normal weight: 5th–85th percentiles; overweight: 85th–95th percentiles; obese: ≥95th percentile).20 The definitions of prehypertension and hypertension are shown in Table 1. Participants were also defined as having hypertension when they were taking antihypertensive medications or had previously received a diagnosis of hypertension.
BMI for gender and age percentiles were calculated from height and weight by using the Nutstat component of Epi Info 3.4, which uses the 2000 CDC growth charts for boys and girls (aged 2–20 years) as a reference standard.21 SBP and DBP for age, height, and gender percentiles were calculated by using the 2000 CDC growth charts20 and BP tables from the Fourth Report on the Diagnosis, Evaluation, and Treatment of High Blood Pressure in Children and Adolescents.19
Bivariable analyses were performed by using the χ2 test for association and Fisher's exact test. Multinomial logistic regression was used to evaluate the odds of prehypertension and hypertension resulting from overweight and obesity while controlling for family history of hypertension or kidney disease. Normal BMI was the reference category, and participants with an underweight BMI expected for age and gender were excluded from the multivariable analysis because of the small sample size (n = 22). Analyses were conducted by using SPSS 16.0 (SPSS, Inc, Chicago, IL).
Of 687 participants who were younger than 18 years and completed the WHS clinic in 2004, complete data on height, weight, and BP were available for 675 (98.3%). The average age of participants was 11 years (range: 4–17 years), and the majority were white (98.7%). A family history of hypertension or kidney disease was reported by 15.1% and <1% of participants, respectively. The number of children who were overweight or obese was 122 (18.1%) and 77 (11.4%), respectively. Prehypertension and hypertension were detected in 51 (7.6%) and 50 (7.4%) children, respectively. Eighteen of those with hypertension were defined as such on the basis of antihypertensive medication use (n = 1) or previous medical diagnosis (n = 17; 10 who had normotension plus 7 who had prehypertension on the basis of BP data). Characteristics of the study sample across BP status are shown in Table 2. Girls were more likely to have normotension than boys (89.0% vs 81.1%). Children aged 4 to 12 years were more likely to have normotension than children aged 13 to 17 years (87.7% vs 80.6%). The presence of hypertension was markedly higher among obese children, 19.5%, compared with children with a BMI in the normal range (4.0%). Similarly, the prevalence of prehypertension was 18.2% among the obese versus 5.7% in those with a normal BMI.
After adjustment for family history of hypertension or kidney disease, obesity was associated with both prehypertension (odds ratio [OR: 4.5 [95% confidence interval (95% CI): 2.2–9.2]) and hypertension (OR: 7.0 [95% CI: 3.3–14.9]; Fig 2). Overweight was associated with hypertension (OR: 3.7 [95% CI: 1.8–7.6]) but not prehypertension (OR: 1.6 [95% CI: 0.8–3.5]). These associations were observed across both age groups (Fig 3) and genders (Fig 4), except that overweight was not associated with hypertension among girls. The associations between obesity and both prehypertension and hypertension were stronger among children aged 4 to 12 years compared with those aged 13 to 17.
The results of the multivariable analysis did not differ from a sensitivity analysis that combined underweight and normal BMI into 1 category. In a second sensitivity analysis that defined hypertension on the basis of BP data alone (eg, information on previous diagnosis or medication use was ignored), prehypertension and hypertension were significantly associated with both overweight and obesity, which differed from the main analysis in which prehypertension was associated with obesity but not overweight.
In this well-characterized population of children who lived in a rural Canadian community, we observed a large association between obesity and elevated BP across age groups and genders. This association was independent of family history of hypertension and kidney disease and suggests that a substantial component of pediatric hypertension in rural populations is explained by overweight and obesity.
The combined prevalence of overweight and obesity in this study (30%) was similar to that of another study of youth who lived in rural Ontario (29%)22 and slightly higher than estimates from children who attended schools in urban centers, which range from 15% to 28%.23–25 Urban–rural differences in pediatric obesity are more pronounced in the United States, where the risk for obesity is reported to be 38% to 82% higher among children who reside in rural versus urban areas.8,10,26 Some researchers suggested that rural communities are among those that support “obesogenic” environments, which discourage energy expenditure and encourage energy intake.10,26–30 Rural communities have limited access to recreational facilities: opportunities for jogging, walking, and biking may be restricted because of a lack of sidewalks, paved roads, and trails, and transportation issues may make it difficult for children to stay after school to participate in sports activities.10,29–31 A recent survey indicated that only 10% of Canadian schools provide late bussing for students who participate in physical activity programs after school.32 Decreasing physical activity and increasing obesity among rural populations may be partially attributable to the incorporation of modern, labor-saving technology in agricultural practices.33,34 In studies of contemporary, nontechnologic farming communities (eg, Old Order Mennonites, Amish), thought to be representative of mainstream Canadian communities 150 years ago, children were fitter and substantially more physically active despite a lack of organized sports or institutional physical education. In 1 study, the prevalence of obesity was only 1.4% as compared with 12% in this study.33,34
Our findings are similar to other studies that showed a positive association between BMI and BP. In the Bogalusa Heart study, for instance, overweight children were 4.5 times more likely to have elevated SBP.35 A strong relationship between BMI and BP is consistently observed among both preadolescent and adolescent age groups,36–38 although fewer studies have evaluated this relationship in younger age groups. When we stratified according to age, the associations between obesity and both prehypertension and hypertension were stronger among children aged 4 to 12 years compared with those aged 13 to 17. This alarming finding has been documented in previous reports and suggests that the harmful effects of obesity are present even among prepubescent children.35,39
The prevalence of hypertension in the literature shows great variation. The prevalence of hypertension in this study (7.4%) was higher than that reported in a review of 8 large studies of American youth (4%l 1978–1991)36; however, the combined prevalence of prehypertension and hypertension in this study (15.0%) was lower than that observed in a Canadian study of youth aged 9 to 16 from Quebec, where the prevalence of high-normal or elevated SBP was estimated to be between 12% and 23%.12 In our study, hypertension was based on the average of 3 BP measurements that were taken on a single occasion. This differs from the definition recommended by the National Institutes of Health, which bases hypertension on elevated BP that is measured on 3 separate occasions. As a result, the prevalence of prehypertension and hypertension in this study is likely overestimated relative to those that follow the National Institutes of Health definition exactly. In addition, estimates of hypertension may vary depending on whether BP was measured by using an auscultatory versus oscillometric device. Whereas auscultation is subject to observer error, oscillometric devices measure mean arterial pressure and then calculate systolic and diastolic values, which can vary with company algorithms.19 Although our study observed strong reliability among 3 BP measurements by using both techniques, this may not always be the case and may explain some of the variation in prevalence of hypertension observed in the literature.
Other limitations of this study include the use of BMI as a measure of overweight and obesity and the use of cross-sectional data. The limitations of BMI are widely known,40 particularly among children for whom adiposity is influenced by age, gender, race, pubertal stages, and waist-to-hip ratio41; however, BMI is currently the most acceptable measure for quantifying childhood obesity for research purposes and is improved through the adjustment of age and gender, as was done here.41 Although longitudinal studies of children have demonstrated that increased BMI is prospectively related to increased BP,2,42 our data are cross-sectional and therefore no conclusions regarding causality can be drawn. Finally, we were not able to conduct a comprehensive analysis for confounding, and, as with any observational study, the possibility for unmeasured confounding exists. For instance, obesity may not be a direct cause of hypertension per se; rather, children who eat a lot may consume greater amounts of salt and experience elevated BP as a result. This possible confounder should be evaluated in future research.
Because this study took place within the context of an environmental disaster during which a substantial number of participants were exposed to E coli O157 and Campylobacter, we considered the possibility that this exposure may have affected the results; however, no association was observed between exposure status (defined by the presence of acute gastrointestinal symptoms at the time of the outbreak16) and hypertension or BMI in this study (data not shown). This result is in agreement with a previous study of WHS participants who showed no long-term renal sequelae or BP differences for the children who were exposed to E coli O757:H1 gastroenteritis.16 Thus, generalization of these findings to children outside the Walkerton area is reasonable.
Risk factors for overweight/obesity are similar in Canada and the United States and include lower socioeconomic status,23,43 sedentary behavior, few family meals, a diet high in fast foods, early childhood obesity, and elevated parental BMI.41 Pharmacologic and surgical management of pediatric obesity is controversial, and most interventions have focused on diet and physical activity, which seem to be successful for some but not all patients.2 In general, interventions related to increasing physical activity among children are most effective when they (1) are delivered in the school setting, (2) include both physical education and activity breaks, and (3) include family strategies.31 Between 2001 and 2006, the proportion of Canadian schools that had policies regarding daily physical education classes increased significantly, yet only 22% of students actually receive daily physical education according to parental reports.32 Hopefully, reports such as this will help end the discord between policy and practice. The results of this study should also encourage the development of policies that specifically target rural populations.
In this population of children who lived in a rural community in Canada, overweight and obesity were strongly associated with elevated BP. This association was independent of family history of hypertension and was present in both preadolescent and adolescent children. Whether BP normalizes with improvements in diet, physical activity, and environment is an area for additional study.
The WHS is funded by the Ontario Ministry of Health.
We thank Ms Arlene Richards and the WHS staff for efforts in ensuring accurate and timely data collection and the people of Walkerton for participation and patience.
- Accepted June 23, 2008.
- Address correspondence to Marina Salvadori, MD, 800 Commissioner's Rd E, Westminster Tower E6-315, London, Ontario, Canada N6A 5W9. E-mail:
The authors have indicated they have no financial relationships relevant to this article to disclose.
What's Known on This Subject
Childhood overweight and obesity may result in premature onset of cardiovascular risk factors such as hypertension. Children who live in rural communities may be at particular risk.
What This Study Adds
Overweight and obesity are strongly associated with hypertension among children who live in a rural Canadian community. This association is independent of a family history of hypertension or kidney disease.
- ↵Tremblay MS, Willms JD. Secular trends in the body mass index of Canadian children. CMAJ.2000;163 (11):1429– 1433
- ↵Sorof J, Daniels S. Obesity hypertension in children: a problem of epidemic proportions. Hypertension.2002;40 (4):441– 447
- ↵House of Commons Canada. Healthy weights for healthy kids: report of the Standing Committee on Health; 2007. Available at: http://cmte.parl.gc.ca/Content/HOC/committee/391/hesa/reports/rp2795145/hesarp07/hesarp07-e.pdf. Accessed June 20, 2008
- ↵Ontario Medical Association Child Health Committee. An Ounce of Prevention or a Ton of Trouble? Is There an Epidemic of Obesity in Children? Toronto, Ontario, Canada: Ontario Medical Association; 2005
- ↵Paradis G, Lambert M, O'Loughlin J, et al. Blood pressure and adiposity in children and adolescents. Circulation.2004;110 (13):1832– 1838
- ↵Garg AX, Moist L, Matsell D, et al. Risk of hypertension and reduced kidney function after acute gastroenteritis from bacteria-contaminated drinking water. CMAJ.2005;173 (3):261– 268
- ↵National High Blood Pressure Education Program Working Group on High Blood Pressure in Children and Adolescents. The fourth report on the diagnosis, evaluation, and treatment of high blood pressure in children and adolescents. Pediatrics.2004;114 (2 suppl 4th report):555– 576
- ↵Centers for Disease Control and Prevention, National Center for Health Statistics. 2000 CDC growth charts: United States. Available at: www.cdc.gov/growthcharts. Accessed June 20, 2008
- ↵Centers for Disease Control and Prevention. Epi Info 3.4. Available at: www.cdc.gov/epiinfo. Accessed June 20, 2008
- ↵Parks SE, Housemann RA, Brownson RC. Differential correlates of physical activity in urban and rural adults of various socioeconomic backgrounds in the United States. J Epidemiol Community Health.2003;57 (1):29– 35
- ↵Salmon J, Booth ML, Phongsavan P, Murphy N, Timperio A. Promoting physical activity participation among children and adolescents. Epidemiol Rev.2007;29 :144– 159
- ↵Cameron C, Wolfe R, Craig C. Opportunities for physical activity in Canadian schools: trends from 2001–2006; 2007. Available at: www.cflri.ca/eng/statistics/surveys/documents/2006capacity.pdf. Accessed June 20, 2008
- ↵Freedman DS, Dietz WH, Srinivasan SR, Berenson GS. The relation of overweight to cardiovascular risk factors among children and adolescents: the Bogalusa Heart Study. Pediatrics.1999;103 (6 pt 1):1175– 1182
- ↵Rosner B, Prineas R, Daniels SR, Loggie J. Blood pressure differences between blacks and whites in relation to body size among US children and adolescents. Am J Epidemiol.2000;151 (10):1007– 1019
- ↵Schneider MB, Brill SR. Obesity in children and adolescents. Pediatr Rev.2005;26 (5):155– 162
- ↵Sinaiko AR, Donahue RP, Jacobs DR Jr, Prineas RJ. Relation of weight and rate of increase in weight during childhood and adolescence to body size, blood pressure, fasting insulin, and lipids in young adults: the Minneapolis Children's Blood Pressure Study. Circulation.1999;99 (11):1471– 1476
- ↵Veugelers PJ, Fitzgerald AL. Prevalence of and risk factors for childhood overweight and obesity. CMAJ.2005;173 (6):607– 613
- Copyright © 2008 by the American Academy of Pediatrics