Higher Self-reported Physical Activity Is Associated With Lower Systolic Blood Pressure: The Dietary Intervention Study in Childhood (DISC)
OBJECTIVE. Children participating in a dietary clinical trial were studied to (1) assess physical activity patterns in boys and girls longitudinally from late childhood through puberty and (2) determine the association of level of physical activity on systolic blood pressure, low-density lipoprotein cholesterol, and BMI.
PATIENTS AND METHODS. In the Dietary Intervention Study in Childhood, a randomized clinical trial of a reduced saturated fat and cholesterol diet in 8- to 10-year-olds with elevated low-density lipoprotein, a questionnaire that determined time spent in 5 intensity levels of physical activity was completed at baseline and at 1 and 3 years. An estimated-metabolic-equivalent score was calculated for weekly activity; hours per week were calculated for intense activities. We hypothesized that weekly self-reported physical activity would be associated with lower systolic blood pressure, low-density lipoprotein, and BMI over 3 years. Longitudinal data analyses were performed for each outcome (systolic blood pressure, low-density lipoprotein, and BMI) by using generalized estimating equations with estimated-metabolic-equivalent score per week as the independent variable adjusted for visit, gender, and Tanner stage (BMI was included in models for systolic blood pressure and low-density lipoprotein).
RESULTS. The initial study cohort comprised 663 youths (362 boys [mean age: 9.7 years] and 301 girls [mean age: 9.0 years), of whom 623 (94%) completed the 3-year visit. For every 100 estimated-metabolic-equivalent hours of physical activity, there was a decrease of 1.15 mmHg of systolic blood pressure. There was a 1.28 mg/dL decline in low-density lipoprotein for a similar energy expenditure. For BMI, an analysis of intense physical activity showed that for every 10 hours of intense activity, there was a trend toward significance with a 0.2 kg/m2 decrease.
CONCLUSIONS. Children with elevated cholesterol levels who lead a more physically active lifestyle have lower systolic blood pressure and a trend toward lower low-density lipoprotein over a 3-year interval. Long-term participation in intense physical activity may reduce BMI as well.
Physical activity is thought to be associated with cardiovascular risk factors at all ages, but there are very few longitudinal data on this relationship in children and adolescents.1 Several cross-sectional, observational, and short-term intervention studies have explored the relationship of physical activity and/or fitness to cardiovascular risk factors; these studies have generally shown small positive results or have been inconclusive.2–5 Two longitudinal studies have shown small but significant relationships of blood pressure and cholesterol level to physical activity participation after sustained intervention over 1 to 2 years. No observational studies have assessed physical activity in relation to cardiovascular risk in boys and girls longitudinally from late childhood through puberty.
The Dietary Intervention Study in Childhood (DISC) was a 3-year randomized trial in children with elevated low-density lipoprotein (LDL) cholesterol levels, 8 to 10 years of age at initial evaluation, that tested the effects of an intervention to promote a reduced saturated fat/reduced cholesterol diet on LDL-cholesterol levels.6 At 3 time points (baseline, year 1, and year 3), an interviewer administered a questionnaire to assess regular physical activity over the course of a full week. The purpose of our analysis was to assess the relationship between regular physical activity and LDL cholesterol level, blood pressure, and BMI over the first 3 years of the DISC. We hypothesized that increased physical activity would be associated with lower blood pressure, LDL cholesterol level, and BMI.
The DISC was a collaborative 6-center randomized, controlled trial of lowering LDL cholesterol level by a dietary educational and behavioral intervention in prepubertal children who had elevated cholesterol levels at initial evaluation. The study was approved by the institutional review boards of all participating centers and a National Heart, Lung, and Blood Institute–appointed data- and safety-monitoring board. Parents provided informed consent.
DISC eligibility criteria, baseline characteristics, study design, and intervention have been described in detail elsewhere.6 In brief, the DISC required baseline LDL-cholesterol levels between the 80th and 98th percentiles for age and gender based on the Lipid Research Clinics data set. Because no evidence for pubertal development could be present, recruitment age was set at 8 years 6 months through 10 years 10 months for boys and 7 years 10 months to 10 years 1 month for girls. Major illnesses and taking lipid-lowering medications were exclusion criteria. A blood pressure level of >125/80 mmHg and weight for height either <5th or >90th percentile on the basis of data from the Bogalusa Heart Study were additional exclusion criteria. For every 100 children screened, 2 were enrolled in the study.
DISC participants were randomly assigned to treatment (n = 334) or usual care (n = 329). The intervention consisted of individual, group, and family education to achieve dietary goals of ≤28% of calories from total fat, <8% of calories from saturated fat, and <150 mg/day of cholesterol.7 Although exercise was encouraged in the intervention group and in the usual-care group, exercise was not a formal component of the intervention.
In this report we present an analysis of all DISC participants who had data on physical activity, blood pressure, LDL cholesterol level, and BMI on at least 1 of 3 time points: baseline, 1 year, and 3 years. The DISC data set was collected between 1987 and 1991.
An interviewer-administered recall questionnaire was designed to capture physical activity for a full 24 hours each day for a period of 1 week.8 The questionnaire was administered twice to collect data for weekdays and weekends. Amount of time spent in physical activities was collected for each of 5 activity intensity levels. An estimated-metabolic-equivalent (MET) score was calculated by multiplying the number of hours spent at each level of activity intensity by a MET (multiple of resting energy expenditure) for that intensity level. The levels and MET scores used were: sleeping (1 MET), sedentary activity (1.5 METs), light activity (4 METs), moderate activity (6 METs), and intense activity (10 METs). Specific examples of activities were provided for each intensity level to guide assignment. The weekday score was multiplied by 5, and the weekend score was multiplied by 2; the sum represented a weekly MET score, which can be interpreted as MET-hours per week. For example, over 1 week, a participant who ran for 4 hours (intense: 10 METs), walked briskly for 14 hours (moderate: 6 METs), spent 64 hours in standing activities or light household chores (light: 4 METs), slept 8 hours per night (56 hours at 1 MET), and spent the remainder of the time sitting at school or watching television (1.5 METs) would have a MET score of 481.
Blood pressure was measured at rest twice by using a Hawksley random-zero sphygmomanometer, and the average measure, after correction for the random zero, was used. Korotkoff phase IV was used for diastolic blood pressure (DBP). Height was measured with shoes off in centimeters with stadiometers constructed by the Medical Instruments Unit of the University of Iowa (Iowa City, IO). Weight in light clothing was measured in kilograms by using scales that are calibrated weekly. Strict quality-assurance measures were used to guarantee measurement accuracy. Each measurement was made twice and averaged; if agreement was poor, a third measurement was made. Lipids and lipoproteins were measured in a Centers for Disease Control and Prevention–standardized laboratory (Johns Hopkins University, Baltimore, MD) using enzymatic procedures. LDL cholesterol was calculated by using the Friedewald equation.
Descriptive statistics were calculated for all study variables and are presented according to gender. Histograms for change in MET score over time were generated. Initially, analyses included an interaction term for treatment-group assignment. Because this term was nonsignificant, data from the 2 treatment groups were merged, and those results are reported.
Generalized estimating equations (GEEs) were used to identify the relationship between risk factors as continuous outcomes (blood pressure, LDL-cholesterol level, and BMI) and self-reported physical activity. The analyses were implemented by using proc genmod in SAS 9.1 (2002–2003, SAS Institute, Inc, Cary, NC), taking into account clustering resulting from multiple measurements per participant with appropriate distribution and link functions to model the continuous outcomes and the continuous predictor variable self-reported physical activity (MET score or hours in intense activity). Type III Wald χ2 tests were used to test the significance of predictors in the GEE models. Additional analyses were performed to determine if specific components of the score (hours in moderate or intense activity) were independently associated with an effect. Models for DBP, triglycerides, and high-density lipoprotein cholesterol were created, but no significant relationships were obtained (data not shown).
The characteristics of the cohort for critical study variables are presented in Tables 1 (boys) and 2 (girls). LDL-cholesterol values are at the high end of the normal distribution per the study protocol. Approximately half the cohort entered puberty by year 1, and very few were prepubertal at year 3.
MET scores for boys and girls are presented in Tables 3 and 4, respectively. MET scores were similar for both genders, with boys having slightly higher scores at baseline and scores increasing slightly from examination to examination. The gender difference may be explained by the fact that boys participated in slightly more intense activity (Fig 1); participation in moderate activity was similar in both genders (Fig 2).
Results of the GEE analyses showing associations between physical activity and cardiovascular risk factors are presented in Table 5. The most important finding is the relationship between systolic blood pressure (SBP) and activity level. A 100-U higher MET score was associated with a 1.15-mmHg decline in SBP. There was a trend toward a 1.28 mg/dL lower LDL-cholesterol level with a similar change in activity. No relationship between MET score and BMI was obtained, but a trend toward a lower BMI was observed in relation to time spent in intense activity.
The mean change in MET score according to gender is shown in Fig 3. Although the median change in score is slightly >0 U, there was a wide variation in MET score from visit to visit. The difference between the upper and lower quartile of the distribution is 75 U, and the difference from the 95th or 5th percentile to the median is ∼100 U.
This study shows a significant relationship in the direction of improved cardiovascular risk factors between self-reported level of physical activity and SBP over a 3-year interval in children going through puberty. Trends were observed toward lower LDL-cholesterol level, and, for participation in intense activity, BMI. These data are supportive of current recommendations for increased participation in physical activity to prevent development of cardiovascular risk factors.9 Because children's blood pressures normally increase 1 to 2 mmHg/year, the benefit from prevention over a 3-year interval of a 1-mmHg rise in SBP, as observed in this study, has potential public health significance.10
The DISC adds to the literature on longitudinal assessment of the relationship of physical activity to cardiovascular risk factors in children. A comparable study is the Northern Ireland Young Hearts Project, which assessed cardiovascular risk in a random cohort of schoolchildren at 12 and 15 years of age.2 That study showed a significant relationship between increased self-reported physical activity and lower blood pressure over the 3-year interval. In the Muscatine Study, a small subset of the cohort (mean age: 10 years at the onset of the study) underwent rigorous measurement of physical fitness over a 5-year interval.4 Measures of fitness and strength showed relationships to lipid levels, adiposity, and blood pressure over the follow-up interval. In a study of 5-year-old inner-city children followed for an average of 19 months, Shea et al3 showed that the age-related rise in blood pressure was attenuated by increased aerobic fitness or decreased BMI.
There have been several randomized, controlled trials that examined the effects of controlled physical activity interventions on blood pressure in children, and these were recently reviewed.5 Interventions were generally conducted over several months; the longest study conducted lasted for 36 weeks. Sample sizes varied widely, with most studies having between 16 and 188 participants; however, 2 studies had sample sizes of ≥500. Collectively, these studies showed a 1% to 3% reduction in blood pressure of marginal statistical significance; the types of interventions varied widely, from weight training to regular aerobic activity programs to participation in activity-related games and other activities. The magnitude of effect is similar to that shown in our study. Although the magnitude of change in blood pressure with increased activity is not sufficient to treat established hypertension, intervention studies and longitudinal studies such as this one all suggest that the age-related rise in blood pressure may be blunted by frequent or increased physical activity participation.10 The change in activity associated with a significant change in SBP, although large, did occur spontaneously among DISC participants over the course of the study.
Our study did not show a significant relationship between self-reported total physical activity and BMI. However, because significant obesity was an exclusion criteria for study participation, and very few of the children in this study became overweight, it would be a misinterpretation of the data to suggest that physical activity participation is not useful in either obesity treatment or prevention. The trend toward lower BMI with participation in intense activity does suggest that regular participation in vigorous activities may play a role in obesity prevention; in adolescent girls physical activity participation is directly related to acquisition of obesity.11 In this study BMI was an average of 3 U lower in girls classified as “active” compared with those “inactive” over a 10-year follow-up period.11 A study of successful weight reduction in adolescents showed a significant reduction in blood pressure (∼10 mmHg) when BMI was reduced ∼10% using both a diet and exercise intervention.12
Some intervention studies that have related measured change in physical fitness to risk factors have shown risk-factor reduction, others have reported no change, and none have shown adverse changes.13–17 Recently, exercise intervention studies in overweight children have shown favorable effects of regular exercise on markers of inflammation, insulin sensitivity, endothelial function, and vascular reactivity.18–22 Population-based studies of children in schools have shown associations of higher BMI with sedentary activity and prevention of the acquisition of overweight with regular increases in scheduled physical education classes.23,24 Cross-sectional studies in children have generally shown that participation in physical activity improves cardiovascular risk factors. Results are often inconsistent, however, with some studies showing positive effects on blood pressure, whereas others show benefit with respect to dyslipidemia. No studies have shown statistically significant adverse effects of increased physical activity on cardiovascular risk.25–28
Although our study showed that physical activity has a statistically significant benefit on cardiovascular risk factors, particularly SBP, the change in physical activity required to produce a measurable change is large. These data support the development of clinical trials to test whether public health efforts to increase regular physical activity lead to long-term attenuation of the rise of SBP over time. It would be interesting to compare results from newer methods of physical activity assessment, such as accelerometers, to results from questionnaires to determine if these newer methods show stronger correlations with risk factors.29
An important component of the DISC was data collection with rigorous quality-control procedures.6 This high-quality data set allowed for careful analysis of subtle effects of physical activity on risk. A second advantage is the relatively large sample size (>600 children). A third advantage is the relatively narrow age range of the subjects, with all children being prepubertal at baseline.
The major limitation of the DISC cohort is the highly selective enrollment criteria. The DISC cohort represents a generally middle- to upper-middle-class group of families, with an interest in participating in a 3-year rigorous dietary intervention, rather than a population-based sample. Elevated LDL-cholesterol levels were required. Extremes of the BMI distribution were excluded. The finding in this study that physical activity increased over time (including overall, moderate, and vigorous intensity) are not consistent with other studies. For example, in the population-based National Growth and Health Study, conducted simultaneously, physical activity declined substantially over time, and this decline was associated with an increased prevalence of obesity.29,30 This difference in pattern over time may suggest that this highly selected cohort was not representative, which may have led to a narrower variation in daily physical activity, perhaps contributing to an underestimation of the effect.
This study showed beneficial long-term effects of increased physical activity on cardiovascular risk factors in developing children. Of particular importance is the potential of higher levels of participation in physical activity to counteract age-related increases in SBP and, potentially, dyslipidemia and BMI. Increased physical activity in youth is a critical component of primordial prevention of heart disease (ie, the prevention of risk-factor acquisition).19
This work was supported by cooperative agreements U01-HL37947, U01-HL37948, U01-HL37954, U01-HL37962, U01-HL37966, U01-HL37975, and U01-HL38110 from the National Heart, Lung, and Blood Institute, National Institutes of Health. Dr Gidding acknowledges support of grant 1 P20 RR020-173-01 from the National Center for Research Resources, a component of the National Institutes of Health.
- Accepted July 20, 2006.
- Address correspondence to Samuel S. Gidding, MD, Nemours Cardiac Center, 1600 Rockland Rd, Wilmington, DE 19899. E-mail:
- Address reprint requests to Bruce A. Barton, PhD, Maryland Medical Research Institute, 600 Wyndhurst Ave, Baltimore, MD 21210. E-mail:
The authors have indicated they have no financial relationships relevant to this article to disclose.
- ↵Williams CL, Hayman LL, Daniels SR, et al. Cardiovascular health in childhood: a statement for health professionals from the Committee on Atherosclerosis, Hypertension, and Obesity in the Young (AHOY) of the Council on Cardiovascular Disease in the Young, American Heart Association [published correction appears in Circulation. 2002;106:1178]. Circulation.2002;106 :143– 160
- ↵Boreham C, Twisk J, van Mechelen W, 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
- ↵Shea S, Basch CE, Gutin B, et al. The rate of increase in blood pressure in children 5 years of age is related to changes in aerobic fitness and body mass index. Pediatrics.1994;94 :465– 470
- ↵Janz KF, Dawson JD, Mahoney LT. Increases in physical fitness during childhood improve cardiovascular health during adolescence: the Muscatine Study. Int J Sports Med.2002;23(suppl 1) :S15– S21
- ↵Sallis JF, Haskell WL, Wood PD, et al. Physical activity assessment methodology in the Five-City Project. Am J Epidemiol.1985;121 :91– 106
- ↵US Department of Health and Human Services. Physical Activity and Health: A Report of the Surgeon General. Atlanta, GA: Centers for Disease Control and Prevention, President's Council on Physical Fitness and Sports; 1996
- ↵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
- ↵Becque MD, Katch VL, Rocchini AP, Marks CR, Moorehead C. Coronary risk incidence of obese adolescents: reduction by exercise plus diet intervention. Pediatrics.1988;81 :605– 612
- ↵Hansen HS, Froberg K, Hyldebrandt N, Nielsen JR. A controlled study of eight months of physical training and reduction of blood pressure in children: the Odense Schoolchild Study. BMJ.1991;303 :682– 685
- Woo KS, Chook P, Yu CW, et al. Effects of diet and exercise on obesity-related vascular dysfunction in children. Circulation.2004;109 :1981– 1986
- Boreham CA, Twisk J, Savage MJ, Cran GW, Strain JJ. Physical activity, sports participation, and risk factors in adolescents. Med Sci Sports Exerc.1997;29 :788– 793
- Craig SB, Bandini LG, Lichtenstein AH, Schaefer EJ, Dietz WH. The impact of physical activity on lipids, lipoproteins, and blood pressure in preadolescent girls. Pediatrics.1996;98 :389– 395
- Copyright © 2006 by the American Academy of Pediatrics