Regular exercise and a healthy dietary pattern are associated with lower resting blood pressure in non-obese adolescents: a population-based study


The study aims to assess the association of diet and frequency of extracurricular physical activity (PA) on blood pressure (BP) in non-obese adolescents. A total of 7185 non-obese adolescents aged 12–18 years were analysed to elucidate the relationship between BP and exercise/eating habit. Totally, 10.3% of the boys and 4.6% of the girls who responded to the questionnaire reported undertaking regular extracurricular physical exercise 3 times/week and were classified as being physically active. An unhealthy eating habit (UEH) score was constructed by counting the number of ‘yes’ responses to 11 dietary behavioural items considered to be unhealthy. In logistic regression analysis, age, body mass index, exercise frequency and UEH were significantly associated with BP (P<0.001). The odds ratios (ORs) for high BP in physically more active adolescents vs those who were less active was 0.48 (95% confidence interval (CI) 0.30–0.77). The OR for high BP in those with UEH scores in the highest quartile vs those with UEH scores in the lowest quartile was 1.63 (95% CI 1.24–2.15). In conclusion, regular exercise and a healthy diet are positively associated with lower BP even in non-obese adolescents.


Increasing rates of high blood pressure (BP) in children and adolescents have been reported.1, 2, 3, 4 High BP tracks through childhood to adulthood,1, 5, 6, 7 and high BP in adolescents increases cardiovascular disease mortality in adulthood. In 2005, we undertook a cross-sectional prospective study to assess secular trends in height, weight and body mass index in Hong Kong children aged 6–18 years.8 Data from this study were used to establish a BP reference for this population.4 Eight percent of the Hong Kong Chinese adolescents had high BP, as did 28.2% of those who were overweight. Two of the most accepted interventions to manage high BP in adults are the adoption of a healthy diet and the increased participation in physical activities (PAs).9, 10, 11 However, it is less clear whether these interventions are also applicable to adolescents. Diets that contain a high proportion of processed foods and soft drinks have a detrimental impact on the health of adolescents.12, 13, 14, 15 Increased consumption of added sugars by adolescents in the United States has been positively associated with measures known to increase cardiovascular disease risk.15 A cohort study of adolescents in Greece used the Mediterranean diet Quality (KIDMED) Index to measure diet quality in children and adolescents. An inverse relationship was shown between the KIDMED index and the albumin-to-creatinine ratio, suggesting a link between accelerated vascular damage as reflected by albuminuria and poor compliance to a healthy diet. This study also showed that lifestyle factors, including diet and PA, were associated with BP in these children.16, 17 The KIDMED Index has also been shown to have a negative correlation with the augmentation index in the brachial artery independent of obesity.18

Unhealthy eating habits (UEHs) and an inactive lifestyle may have an adverse impact on the health of Hong Kong adolescents. The present study utilises our 2005 data set to study the association of diet and PA on BP in a sub-population of non-obese adolescents.


Details of the 2005/6 Hong Kong Growth Survey have been previously reported.8 The present analysis mainly utilises data for the sub-population of non-obese adolescents.


This study protocol was approved by the Joint The Chinese University of Hong Kong and New Territories East Cluster Clinical Research Ethics Committee and the Ethics Committee of the Department of Health of the Hong Kong Government.


One secondary school from each of the 18 Districts in Hong Kong was randomly selected and invited to participate in the study. From the selected schools, two classes in each grade were selected based on the school timetable and operational needs. Chinese adolescents aged 12–18 years in the selected classes were invited to participate. Parents were informed about the study through a letter distributed by the school and asked to inform the school if they did not wish their child to participate. Parents were told that participating children would be given a record of their body measurements, and that any child with high BP or other abnormal findings would be referred for further assessment. Parents were invited to complete a questionnaire providing demographic information, including gestation and birth weight, and family or personal history related to risk factors for obesity. Verbal consent was obtained from the students who were then asked to complete a self-administered dietary and PA questionnaire.

Measurement of anthropometric parameters

A team of eight trained research staff visited the selected schools on a pre-arranged date to collect the anthropometric data. Standing height without shoes was measured using a Harpenden Stadiometer (Holtain Limited, Crymych, Pembs, UK) to the nearest 0.1 cm. Body weight was measured using a portable Tanita scale (Model BF-522; Tanita Corporation, Tokyo, Japan). BP was measured by Datascope Accutorr Plus (Datascope Corporation, Mahwah, NJ, USA), an oscillometric device previously validated with mercury sphygmomanometer in children.19 Each of the above anthropometric measurements was carried by a single observer. The intra-class (within-observer) correlation coefficients, based on pairs of replicate measurements made by the same observer on 100 subjects on the same day, were 0.998 for weight and height and 0.997 for waist circumference. At each school visit, BP was measured at school in the morning, on the right arm, with the children seated and rested for at least 5 min and with the arm supported at heart level. Appropriate sized cuffs were used (bladder width at least 40% of arm circumference, length 80–100% of arm circumference).20 Two BP measurements were taken with a 1-min interval. The average of the two readings was used for analysis. High BP was identified if systolic and/or diastolic BP was greater or equal to sex, age and height specific 95th percentile.4 Waist circumference was measured midway between the lowest rib and the superior border of the iliac crest with an inelastic measuring tape to the nearest 0.1 cm.

Frequency of PA was assessed by two questions. 1. Do you currently participate in any regular exercise class other than school physical education class? 2. If yes, please tick the frequency of training: 1, 2 or 3 per week. These two questions were designed based on the PA guidelines for adolescents issued from the international consensus conference in 1994.21 An explanation of these two questions with a detailed list of typical exercise training classes commonly available in Hong Kong was provided to all the participants. Hong Kong schools generally allocate only two 45-min lessons per week to physical education. Exercise during the physical education classes was regarded as a constant within the study sample and therefore not taken into account in this analysis.

Students were also asked to complete a ‘1-min’ dietary pattern assessment questionnaire. The questionnaire was validated against 7 individual days of 24-hr dietary recall data from 235 Chinese secondary school students (57% female) to obtain a rapid assessment tool to assess overall diet quality. The percentage of agreement between the questions and the dietary recall ranged from 73% to 95%. Spearman’s correlation coefficient between food intake estimated from the 7 days of 24-hr recalls and the questionnaire was r=0.63 (P=0.004).22

Statistical analysis

Data are presented as percentages for categorical variables and as means (s.d.s.) for continuous variables. An UEH score was constructed by counting the number of ‘yes’ responses to the 11 dietary behavioural items listed in Table 1 to quantify the degree of UEHs. The UEH score was categorised into three levels (namely, low, moderate and high) of healthy/UEHs using the lower and upper quartiles of the scores of all the study subjects as the cutoffs, with a high score being considered ‘unhealthy’ and low score being considered ‘healthy’. The Pearson’s Chi-square test and t-test, as appropriate, were used to compare the studied variables between adolescents with and those without high BP. Those variables with P-values <0.25 in univariate analysis were chosen as candidate independent variables for multivariable stepwise logistic regression analysis to delineate factors independently associated with high BP among non-obese Chinese adolescents aged 12–18 years. The results of the final multivariable logistic regression model were presented by the odds ratios (OR) and their associated 95% confidence intervals (CIs) of the significant factors identified. All the statistical analyses were performed using SPSS 17.0 (SPSS, Chicago, IL, USA). All statistical tests were two-sided and a P-value <0.05 was considered statistically significant.

Table 1 Characteristics of the 7185 non-obese adolescents aged 12–18 years


Eighteen secondary schools, one from each of the 18 Hong Kong districts, were selected. All adolescents in this study were ethnic Chinese. The sample consisted of 8315 students aged 12–18 years (boys=4078 and girls=4237). Ten percent of eligible students did not participate in this study, mainly as a result of being absent from school. In all, 1130 adolescents (690 boys and 440 girls) were considered overweight or obese based on International Obesity Task Force definitions.23 In total, 7185 non-obese adolescents were included in the analysis.

Totally, 10.3% of the boys and 4.6% of the girls were physically active, exercising outside of school at least 3 times a week (Table 1). The UEH score was categorised to represent three levels (namely, low: 25th percentile, moderate: >25th–75th percentile and high: >75th percentile; Table 1). Seven percent of the adolescents had high BP. Age, body mass index, reduced frequency of PA and UEH score were greater among the adolescents with high BP (Table 2). Girls were less active than boys (4.6% vs 10.3% having extra PA at least 3 times a week) but there was no gender difference between subjects with and those without high BP. The multivariable logistic regression analysis, which adjusted for age, gender, parental hypertension and body mass index, showed that high BP was significantly associated with the frequency of PA and the UEH score among non-obese adolescents (Table 3). A higher frequency of PA was associated with a decreased OR of high BP in the non-obese adolescents but not in the overweight and obese adolescents; the adjusted OR for those who had once, twice and three times or more extra sessions of extracurricular exercise training per week vs those who had none were 0.8 (95% CI: 0.6–1.0), P=0.068, 0.6 (95% CI: 0.4–0.9), P=0.016 and 0.4 (95% CI: 0.3–0.7), P=0.001, respectively. The non-obese adolescents with high UEH scores were more likely to have a high BP; the adjusted ORs for those who had moderate and high UEH scores vs those who had low UEH scores were 1.3 (95% CI: 1.0–1.7), P=0.026 and 1.5 (95% CI: 1.1–2.0), P=0.004, respectively. The overweight and obese adolescents with high UEH scores were also more likely to have a high BP; the adjusted ORs for those who had moderate and high UEH scores vs those who had low UEH scores were 1.9 (95% CI: 1.1–3.3), P=0.027 and 2.1 (95% CI: 1.2–3.8), P=0.015, respectively.

Table 2 Characteristics of 7185 non-obese adolescents with and without high BP
Table 3 Estimated effect of associated factors on high BP

Figure 1 shows the prevalence of high BP among the non-obese adolescents by the frequency of extra exercise training and the UEH score. It indicates that there may be an additive effect on BP by training frequency and UEH; the prevalence of high BP increases with UEH score and the prevalence among those who have low training frequency are higher than those who have high training frequency. We further assessed any multiplicative interaction effect of training frequency and UEH by multivariable logistic regression but found no such interaction effect, that is, our results showed an additive but not synergistic effect between these two variables. We also demonstrated an inter-relationship between UEH score and the frequency of extra exercise training, with those having a higher frequency of PA being less likely to have a high UEH score, P=0.003 (Kruskal–Wallis test).

Figure 1

Prevalence of high BP among 7185 non-obese adolescents by training frequency and UEH score.


We excluded from the main analysis, the overweight and obese adolescents who are typically considered to be at high risk of hypertension. Few studies have evaluated the association of healthy diet and PA with resting BP in non-obese adolescents. We show that the risk of high BP in an inactive non-obese adolescent with a high UEH is greater than that in his/her more active counterparts. However, there was no association between PA and high BP in the overweight and obese group. We and others have previously reported the association between higher rates of PA and lower rates of hypertension.13, 24, 25, 26 Although diet has been recognised as an important factor that can influence BP in adults, there is limited evidence that diet can affect the BP in children or adolescents.27 The CYKIDS Study16, 17 enrolled 622 Cypriot school children (11.7±0.83 years), and using the KIDMED index, it was shown that good adherence to a Mediterranean diet was associated with lower systolic and diastolic BP levels. Although the CYKIDS study showed a positive association between watching TV and BP levels, no association between PA and BP levels was found in contrast to our study.

A recent US study has reported that a high consumption of added sugars is associated with measures known to increase cardiovascular disease risk in adults but no significant association between BP and increased intake of added sugars was found.28 Our UEH score was associated with high BP. This score was derived from ‘1-min’ dietary pattern assessment questionnaire to reflect consumption of a range of unhealthy foods that are more likely to be highly processed or contain added sugars.

There are a number of limitations of our cross-sectional study. First, it is not possible to determine any causality between the associations identified. Second, we only measured BP on one occasion, whereas the Forth Report on the diagnosis of pediatric high BP recommends that measurements should be taken on three separate occasions.20 A single measurement is likely to overestimate the prevalence of elevated BP due to the white-coat effect.29 High BP was found in 8.0% of our school children, which is similar to reported rates from Shanghai, China (6.9%),30 Delaware, USA (7.2%),31 and Sousse, Tunisia (9.6%).32 Although white-coat hypertension was believed to be harmless, it is now viewed as a potential risk factor for cardiovascular events.3 Thus, overestimating the rate of hypertension in our study is unlikely to alter the implications of our findings. A third limitation of our study is that our assessment of UEHs using our UEH score is arbitrary. However, there is no definition or universally accepted measure to assess healthy and UEHs. Given the perception that unhealthy eating reflects the increased consumption of highly processed foods that are rich in fat and/or sugar, we believe that our UEH categorisation provides a measure of such eating behaviours. A fourth limitation of our study is that the self-reported dietary questionnaire could have introduced biases as it does not explore the amount of daily food intake. An important consideration is whether there could be independent effects of different nutrients on the BP level or whether the effects of different nutrients in combination could be synergistic.33 Detailed dietary assessments would be required to tease out such relationships. A fifth limitation of our study is that the two simple questions used to assess PA only give limited insight into the nature and the intensity of overall activity. The heterogeneous nature of the various activities undertaken by children made it impossible for us to characterise or quantify these activities. However, the answers to these two questions likely reflect an intention to participate in additional extra-curricular PA. Finally, as detailed data on socioeconomic status and sedentary lifestyle behaviours such as TV watching were not available, we cannot evaluate the effect of these factors on BP.

In summary, the current study, with its large population-based sample, provides evidence that regular exercise and avoidance of unhealthy eating behaviours could be important non-pharmacological approaches in the prevention of high BP in non-obese adolescents. However, experimental studies are needed to elucidate the causality and possible mechanisms of these associations.


  1. 1

    Luma GB, Spiotta RT . Hypertension in children and adolescents. Am Fam Physician 2006; 73: 1558–1568.

  2. 2

    Papa A, Dal CA, Capuano A, Conte G, D’Avanzo L, D’Anna F et al. Epidemiology of hypertension in children. Proc Eur Dial Transplant Assoc 1983; 20: 551–556.

  3. 3

    Collins RT, Alpert BS . Pre-hypertension and hypertension in pediatrics: don’t let the statistics hide the pathology. J Pediatr 2009; 155: 165–169.

  4. 4

    Sung RY, Choi KC, So HK, Nelson EA, Li AM, Kwok CW et al. Oscillometrically measured blood pressure in Hong Kong Chinese children and associations with anthropometric parameters. J Hypertens 2008; 26: 678–684.

  5. 5

    Bouziotas C, Koutedakis Y, Nevill A, Ageli E, Tsigilis N, Nikolaou A et al. Greek adolescents, fitness, fatness, fat intake, activity, and coronary heart disease risk. Arch Dis Child 2004; 89: 41–44.

  6. 6

    Lauer RM, Clarke WR, Lauer RM, Clarke WR . Childhood risk factors for high adult blood pressure: the Muscatine Study. Pediatrics 1989; 84: 633–641.

  7. 7

    MacMahon S, Peto R, Cutler J, Collins R, Sorlie P, Neaton J et al. Blood pressure, stroke, and coronary heart disease. Part 1. Prolonged differences in blood pressure: prospective observational studies corrected for the regression dilution bias. Lancet 1990; 335: 765–774.

  8. 8

    So HK, Nelson EAS, Li AM, Wong EMC, Lau JTF, Guldan GS et al. Secular changes in height, weight and body mass index in Hong Kong children. BMC Public Health 2008; 8: 320.

  9. 9

    Bacon SL, Sherwood A, Hinderliter A, Blumenthal JA . Effects of exercise, diet and weight loss on high blood pressure. Sport Med 2004; 34: 307–316.

  10. 10

    Erdine S, Aran SN . Current status of hypertension control around the world. Clin Exp Hypertens 2004; 26: 731–738.

  11. 11

    Woo J, Leung SS, Ho SC, Lam TH, Janus ED . Dietary intake and practices in the Hong Kong Chinese population. J Epidemiol Community Health 1998; 52: 631–637.

  12. 12

    Brendan O . Is this what you call junk food?, BBC NEWS 20060 November 30. URL:

  13. 13

    Fasting MH, Nilsen TI, Holmen TL, Vik T . Life style related to blood pressure and body weight in adolescence: cross sectional data from the Young-HUNT study, Norway. BMC Public Health 2008; 8: 111.

  14. 14

    Kourlaba G, Panagiotakos DB, Mihas K, Alevizos A, Marayiannis K, Mariolis A et al. Dietary patterns in relation to socio-economic and lifestyle characteristics among Greek adolescents: a multivariate analysis. Public Health Nutr 2009; 12: 1366–1372.

  15. 15

    Wiles NJ, Northstone K, Emmett P, Lewis G . ‘Junk food’ diet and childhood behavioural problems: results from the ALSPAC cohort. Eur J Clin Nutr 2009; 63: 491–498.

  16. 16

    Lazarou C, Panagiotakos DB, Matalas AL . Foods E-KINDEX: a dietary index associated with reduced blood pressure levels among young children: the CYKIDS study. J Am Diet Assoc 2009; 109: 1070–1075.

  17. 17

    Lazarou C, Panagiotakos DB, Matalas AL . Lifestyle factors are determinants of children’s blood pressure levels: the CYKIDS study. J Hum Hypertens 2009; 23: 456–463.

  18. 18

    Lydakis C, Stefanaki E, Stefanaki S, Thalassinos E, Kavousanaki M, Lydaki D . Correlation of blood pressure, obesity, and adherence to the Mediterranean diet with indices of arterial stiffness in children. Eur J Pediatr 2012; 171 (9): 1373–1382.

  19. 19

    Wong SN, Tz Sung RY, Leung LC . Validation of three oscillometric blood pressure devices against auscultatory mercury sphygmomanometer in children. Blood Press Monit 2006; 11: 281–291.

  20. 20

    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: 555–576.

  21. 21

    Sallis JF, Patrick K . Physical activity guidelines for adolescents: consensus statement. Pediatr Exerc Sci 1994; 6: 302–314.

  22. 22

    Guldan GS, Wong S, Chan J . Validation of the one-minute dietary assessment to assess the dietary behaviors of Hong Kong adolescents. Fourth Annual Conference of the International Society of Behavioral Nutrition and Physical Activity (ISBNPA) [3 - 50544] 2005.

  23. 23

    Cole TJ, Bellizzi MC, Flegal KM, Dietz WH . Establishing a standard definition for child overweight and obesity worldwide: international survey. BMJ 2000; 320: 1240–1243.

  24. 24

    Boreham C, Twisk J, van MW, Savage M, Strain J, Cran G . Relationships between the development of biological risk factors for coronary heart disease and lifestyle parameters during adolescence: The Northern Ireland Young Hearts Project. Public Health 1999; 113: 7–12.

  25. 25

    Marti B, Vartiainen E . Relation between leisure time exercise and cardiovascular risk factors among 15-year-olds in eastern Finland. J Epidemiol Community Health 1989; 43: 228–233.

  26. 26

    So HK, Sung RY, Li AM, Choi KC, Nelson EA, Yin J et al. Higher exercise frequency associated with lower blood pressure in Hong Kong adolescents: a population-based study. J Hum Hypertens 2010; 24: 646–651.

  27. 27

    Couch SC, Daniels SR . Diet and blood pressure in children. Curr Opin Pediatr 2005; 17: 642–647.

  28. 28

    Welsh JA, Sharma A, Cunningham SA, Vos MB . Consumption of added sugars and indicators of cardiovascular disease risk among US adolescents. Circulation 2011; 123: 249–257.

  29. 29

    Stabouli S, Kotsis V, Toumanidis S, Papamichael C, Constantopoulos A, Zakopoulos N . White-coat and masked hypertension in children: association with target-organ damage. Pediatr Nephrol 2005; 20: 1151–1155.

  30. 30

    Wan YP, Zhang XM, Lu LP, XU RY, Cao D, Luo KL . Distribution tendency of hypertension in 7326 primary and middle school students in Shanghai and analysis of the associated factors. J Clin Pediatr 2007; 25: 1005–1008.

  31. 31

    Falkner B, Gidding SS, Ramirez-Garnica G, Wiltrout SA, West D, Rappaport EB . The relationship of body mass index and blood pressure in primary care pediatric patients. J Pediatr 2006; 148: 195–200.

  32. 32

    Bao W, Threefoot SA, Srinivasan SR, Berenson GS . Essential hypertension predicted by tracking of elevated blood pressure from childhood to adulthood: the Bogalusa Heart Study. Am J Hypertens 1995; 8: 657–665.

  33. 33

    Couch SC, Daniels SR . Diet and blood pressure elevation in children and adolescents. Arch Pediatr Adolesc Med 2004; 158: 418–419.

Download references


We thank the school principals, teachers, parents and students who participated in the study. This research project received financial support from Departmental funds and the Hong Kong Pediatric Society. Parental consent was obtained for conducting the study.

Author information

Correspondence to E A S Nelson.

Ethics declarations

Competing interests

The authors declare no conflict of interest.

Rights and permissions

Reprints and Permissions

About this article


  • blood pressure
  • physical activity
  • diet pattern
  • adolescent

Further reading