Physical fitness, physical activity, and cardiovascular disease risk factors in adolescents: The Oslo youth study

PREVENTIVE

MEDICINE

17, 12-24 (1988)

Physical Fitness, Physical Activity, and Cardiovascular Disease Risk Factors in Adolescents: The Oslo Youth Study’ GRETHE S.TELL, D.Sc.,M.P.H.*,*

ANDODDD.VELLAR~

*Center for Prevention Research and Biometry, Bowman Gray School of Medicine, Wake Forest University, 300 South Hawthorne Road, Winston-Salem, North Carolina 27103, and tDepartment of Preventive Medicine, University of Oslo, Oslo, Norway

Aerobic fitness, resting pulse rate, and self-reported physical activity were examined along with prevalence of cardiovascular disease risk factors in a population-based study of 413 boys and 372 girls, ages 10 to 14 years. Cardiovascular fitness (VO, max) was predicted from heart rate measured during submaximal bicycle exercise. For both genders, fitness level was significantly and inversely related to body weight, body mass index [weight in kilos/(height in meters)2], triceps skinfold thickness, systolic and diastolic blood pressure, and pulse rate and positively related to high-density lipoprotein/total cholesterol ratio and physical activity. In addition, fitness level was positively related to high-density lipoprotein cholesterol and negatively related to triglycerides in females; it was also negatively related to height, total cholesterol, and hematocrit in males. Analyses of covariance, controlling for sexual maturity ratings, revealed that students in the lowest quartiles of V02 max had significantly higher body mass index and triceps skinfold thickness than students in the higher quartiles. After adjustment for body mass index and sexual maturity ratings, blood pressure and pulse rate in both genders were significantly higher among students in the lower quartiles of VOz max than among the groups who scored higher on the the fitness test. Higher levels of VO, max were also associated with a more favorable lipid profile in females. In gender-specific multiple regression analysis, triceps skinfold thickness was the strongest predictor of VOz max, followed by pulse rate. Our study provides evidence that higher levels of fitness are associated with more favorable risk profiles in adolescents. o t988ACTdemic Press, Inc.

INTRODUCTION

Low levels of physical activity3 have been associated with higher levels of cardiovascular disease (CVD) risk factors in adults (4,9), white adolescent males (7), and black adolescent males and females (8). Studies have also shown that physical activity is a predictor of CVD morbidity and mortality, either directly or via its effect on other risk factors (15). Although the relationship between physical fitness4 and CVD is less clear, cardiorespiratory fitness has been found to be related to lower CVD risk factors in adult men (3) and women (12), as well as in * The Oslo Youth Study was funded by the Norwegian Cancer Society. Additional funding was provided by the Norwegian Confederation of Sports. Preparation of this manuscript was supported by a grant from the National Heart, Lung, and Blood Institute (5-T32-HL-07328). * To whom reprint requests should be addressed. s Physical activity is defined as the physical workload to which a person is usually subjected during his or her work or leisure. 4 Physical fitness, or cardiovascular fitness, is defined as the capacity to perform prolonged heavy work (27), or aerobic capacity. 12 0091-7435/88$3.00 Copyright Q WE8 by Academic Press, Inc. All rights of reproduction in any form reserved.

PHYSICAL FITNESS VERSUS CVD RISK FACTORS

13

boys (10, 11, 24) and girls (10). Because risk factors and risk-related behavior patterns for CVD have their origins in childhood and adolescence (22), more needs to be learned about the epidemiology of fitness, physical activity, and CVD risk in the younger population. This article presents relationships among physical fitness, physical activity, and CVD risk factors in a population-based sample of Norwegian adolescents, controlling for factors such as relative body weight and pubertal development. METHODS A survey of students in grades 5 to 7 (ages lo- 15 years) in six Oslo schools was conducted from February to June 1979. Of the 1,016 Norwegian students solicited, participation consent was obtained from 431 boys and 397 girls (81%). Measurements included self-reported health behaviors, indirect assessment of maximum oxygen uptake, blood pressure, resting (sitting) pulse rate, height, weight, triceps skinfold thickness, assessment of pubertal development, serum total cholesterol (TC), high-density lipoprotein cholesterol (HDL-C), fasting triglycerides (TG), and hematological parameters. Measurements of physical fitness were obtained from 413 boys and 372 girls, ages lo-14 years. Data from these subjects only are included in the analyses presented in this article. All measurements were taken in the schools. Methods and procedures have been published previously (23). Briefly, resting blood pressure (BP) was measured on a random-zero sphygmomanometer after the student had been seated for 5 min. The onset of first, fourth, and fifth Korotkoff phases was recorded. Two BP readings were performed on the right arm, and the mean obtained from the two readings is reported here. Between the two measurements, the pulse of the right wrist was registered for 30 sec. Anthropometric measurements were performed during the same session. Venipuncture for serum lipids and plasma hematology was performed in a different session, always early in the morning because the students had been fasting overnight. Sexual maturity ratings were made by visual assessment of secondary sex characteristics according to the method moditied by Tanner (19). These ratings are conducted according to ordinal scales which, for males, are based on external genitalia and/or pubic hair development, and for females are based on breast and/or pubic hair development. The scales are divided into five stages based on superficial appearance, stage 1 being preadolescence and stage 5 maturity. Analyses presented in this article are based on breast development in girls and genital development in boys, not on pubic hair growth. Maximum oxygen uptake ( V02 max; milliliters O2 per kilogram body weight per minute) was indirectly assessed by counting the students’ pulses while they performed on bicycle ergometers. This test is based on the principle that from registration of the workload and heartbeat frequency measured in steady state (usually after 4-6 min of pedalling), V02 max can be estimated by extrapolating to the assumed maximum heart rate (2). The tests were performed in the schools’ gymnastic facilities and the students were dressed in training clothes. The height of the saddle on the bicycle ergometer was adjusted for each individual to ensure a slight bending of the knee when the anterior part of the foot was placed on the pedal in its lowest position, Students were instructed to pedal at a steady rate of

14

TELL

AND VELLAR

50 pedal revolutions per minute, guided by the bicycle’s speedometer. Workload was adjusted by a mechanical break mechanism on the wheel. Standard workloads were 50, 100, 150, and 175 W from an assumed end-pulse of 140- 160 beats per minute. Specially trained staff registered the student’s pulse rate by placing a stethoscope on the student’s chest. This test procedure was modified after Astrand and Rodahl’s (2) test method. The test was originally developed for adults with an estimated maximum heart beat frequency of 195 beats/min. In adolescents, who on average have a maximum heart rate above 195 beats/min, the predicted values for maximum oxygen uptake will be underestimated, i.e., lower than those obtained during maximum treadmill running (1, 14). Based on previous studies of Norwegian boys with a mean age of 12.5 years, the direct measurements of V02 max produce values that are on average 1.18 times higher than those from the indirect method (14). Since the mean age of our study population was 12.5 years, and since we are not aware of studies comparing direct and indirect assessments of V02 max in girls, we used a correction factor of 1,18 to calculate maximum oxygen uptake in both girls and boys. Although the validity of this test in predicting maximal oxygen uptake in individual adolescents is imprecise, it has been found to be useful on a group level. Since the main purpose of this article is not to produce reference values for maximum oxygen uptake in adolescents, but rather to investigate the relationships between physical fitness and CVD risk factors, we feel that the bicycle ergometer test provides a simple tool for assessing physical fitness in groups of adolescents. The choice of correction factor will not influence the magnitude of these relationships. Information on the frequency and intensity of physical activity was obtained by a self-report questionnaire of the students. The key question asked was: “How often do you exercise in your leisure time (for at least half an hour) so that you get out of breath and sweat?” The response alternatives were (a) daily, (b) two to three times per week, (c) once per week, (d) two to three times per month, and (e) not that often. Since by this scale, a higher score indicated less physical activity, we recoded the sequence of the answer categories so that in all results presented here, a higher physical activity score indicates more physical activity. All data analyses were performed using BMDP statistical software (5). RESULTS

Some characteristics of the study population are given in Table 1. In each age group, girls had significantly (P s 0.01) larger skinfold thickness values than boys, while the opposite was true for V02 max in the three youngest age groups. Compared with lZyear-old boys, girls of the same age had significantly larger values for pulse rate and systolic @BP) and diastolic (DBP) blood pressure (P < 0.01). Boys had significantly higher values for HDL-C than girls at ages 11 and 12 years (P s 0.03), while girls had higher mean values for fasting TG at ages 12 and 13 years (P s 0.0001). In addition, 17% of males and 8% of females reported exercising daily; 64% of both genders exercised one to three times per week. Among boys, 16% exercised

150.1 157.0 160.8 162.1 157.2

826 139 109 42 372

(10.4) (10.3)

67.1 67.2 65.6 67.4 66.8

a Non-Norwegians surveyed * Includes one lo-year-old.

41.1 46.1 49.5 50.5 46.5

40.0 42.7 47.9 52.0 45.1

X

(9.5) (9.5) (9.3)

(8.6)

(7.6)

(9.1) (8.7)

(8.2)

(7.0)

(6.9)

(SD)

59.4 59.8 60.8 61.9 60.3

56.4 55.6

58.2 54.3 54.7

TI

(10.3) (10.3) (10.4) (12.0) (10.5)

(11.0) (10.2)

(10.2) (10.4) (9.4)

(SD)

Phase 5

18.2 18.6 19.0 19.1 18.7

17.7 18.1 18.5 19.3 18.3

X

13.3 13.1 13.2 13.9 13.3

10.8 10.4 10.5 10.8 10.6

4.73 4.79 4.65 4.92 4.75

4.50 4.64

4.77 4.53 4.75

51

Y

(0.68) (0.70) (0.61) (0.81) (0.69)

(0.83) (0.80)

(0.84) (0.78) (0.76)

(SD)

45.1 46.0 47.2 48.7 46.5

52.1 52.3 52.3 51.3 52.1

X

1.34 1.31 1.29 1.33 1.31

1.29 1.36

1.44 1.29 1.42

TI

/

(0.29) (0.28) (0.25) (0.27) (0.27)

(0.28) (0.30)

(0.32) (0.30) (0.28)

(SD)

HDL cholesterol (mmoVliter)

(5.0)

(5.8)

(5.4) (5.1) (4.3)

(4.6)

(6.1)

(3.8) (3.9) (4.9)

(SD)

(10.6) (9.8) (9.5) (11.8) (10.1)

(11.8) (10.3) (10.3) (12.0) (10.9)

(SD)

VO, max (ml O,/kg x min)

OSLO ADOLESCENTS

Triceps skinfold thickness (mm)

Total cholesterol (mmohhter)

(2.6) WI (2.8) (2.6)

(2.5)

(2.0) (2.1) (2.6) (2.1)

(1.9)

(SD)

BMI (kg/mz)

(N = 24) are not included.

P-8)

(10.2)

(8.3)

(10.0) (9.5)

(8.4) (10.0) (9.3)

(SD)

65.4 63.5

64.1 63.4 62.3

Boys 11 13 12

14 Total Girls 11 12 13 14 Total

51

6.3) 03.1)

(7.5)

(6.6) (6.8)

(9.4)

(8.8) (8.8)

(7.0) (7.0)

(SD)

Weight (kg)

blood pressure (mm Hg)

Phase 4

Diastolic

149.9 153.2 160.4 163.8 156.4

X

98* 115 144 56 413

Na

Age (years)

Boys 11 12 13 14 Total Girls 11 12 13 14 Total

Age (years)

Height (cm)

CHARACTERISTICSOFTHEEXAMINED

0.59 0.74 0.75 0.78 0.72

0.71 0.57

0.52 0.60 0.50

51

(13.5) (15.6) (15.3) (15.3) (15.3)

(11.4) (11.9) (13.9) (12.5) (12.6)

(SD)

(0.32) (0.44) (0.37) (0.45) (0.40)

(04 (0.30)

(0.25) (0.30) (0.21)

(SD)

Triglycerides (fasting) (mmolihter)

85.6 89.9 84.1 79.7 86.1

82.7 81.4 83.0 80.0 82.1

X

Sitting pulse (beats/mm)

39.1 40.3 41.3 41.5 40.5

42.5 40.7

39.2 41.4 39.9

51

Hematocrit

110.2 111.6 109.3

(3.1) (2.4) (2.7)

Cm

(2.7)

(2.8)

(2.4)

(2.2) (2.9)

(2.7)

(SD)

(%)

(11.7) (12.1) (12.5)

(12.2)

(13.5)

105.1 110.3

(10.4) (12.3) (12.2) (11.6)

(9.6)

(SD)

104.2 106.8 111.3 113.1 108.6

x

Systolic blood pressure (mm Hg)

E

g

R g 9

-

E z

3 g c

y

F

g 2 ij

16

TELL AND VELLAR

less than two to three times per month; the corresponding figure for girls was

22%. In Tables 2 and 3 Pearson’s simple linear correlations between maximum oxygen uptake, pulse rate, and some measured variables are shown. In boys, higher aerobic capacity was significantly associated with lower resting pulse rate, height, weight, body mass index (BMI), triceps skinfold thickness, SBP, diastolic phase 4 (DBP4), and diastolic phase 5 (DBPS) blood pressure, TC, and hematocrit (HCR) and with more frequent physical activity and greater HDL-C/TC ratio. Pulse rate was positively correlated with triceps skinfold thickness, SBP, DBP4, and DBPS, and negatively related to physical activity. In girls (Table 3), maximum oxygen uptake was negatively related to resting pulse rate, weight, BMI, triceps skinfold thickness, SBP, DBP4, DBPS, fasting TG, and sexual maturity and positively related to HDL cholesterol, HDL-C/TC ratio, chronologic age, and physical activity. Pulse rate was positively related to SBP, DBP4, and TG and negatively related to HDL-C, HDL-C/TC ratio, and age in years. In addition to the significant correlation with V02 max and pulse rate, self-reported physical activity of boys was significantly correlated with triceps skinfold thickness (r = -0.10, P < 0.05), but not to other risk factors. For girls, in addition to V02 max, physical activity was related to DBP4 (r = -0.10, P < 0.05), TG (r = -0.13, P < O.Ol), and self-reported cigarette smoking frequency (r = -0.17, P < 0.0001). To further explore the relationship between risk factors and physical fitness, we analyzed several variables by quartiles of maximum oxygen uptake, adjusting for TABLE 2 PEARSON’SCORRELATIONS,BOYS

Sitting pulse rate Physical activity’ Height Weight Body mass index Triceps skinfold thickness Systolic blood pressure (BP) Diastolic (IV) BP Diastolic (V) BP Total cholesterol High-density lipoprotein (HDL-C) cholesterol HDL/total cholesterol l’kiglycerides Hematocrit Sexual maturity Chronologic age

VO, max

Sitting pulse

-0.36t 0.11* -0.11* -0.32t -0.41t -0.45-I -0.14** - 0.20t -0.24t -0.10* 0.05 0.12* -0.09 -0.12* 0.04 - 0.02

-0.13** -0.01 0.04 0.07 0.11* 0.17t 0.16t 0.18t 0.04 -0.01 -0.05 -0.01 0.04 - 0.02 -0.04

Note. VO, max, milliliters 0, per kilogram body weight per minute. 0 A higher score indicates more physical activity. * P < 0.05. ** P < 0.01. *** P < 0.001. i P < 0.0001.

17

PHYSICAL FITNESS VERSUS CVD RISK FACTORS TABLE 3 F%ARSON'SCORRELATIONS,GUUS

Sitting pulse Physical activity” Height Weight Body mass index ‘Biceps skinfold thickness Systolic blood pressure (BP) Diastolic (IV) BP Diastolic (V) BP Total cholesterol High-density lipoprotein (HDL-C) cholesterol HDLfIC Triglycerides Hematocrit Sexual maturity Chronological age

VOz max

Sitting pulse

- 0.29t 0.12* -0.06 - 0.28t - 0.36t -0.38t -0.22t - 0.28t -0.18t - 0.07 0.14** 0.18t - 0.25t - 0.02 -0.12* 0.10*

-0.04 -0.05 -0.06 -0.04 -0.05 0.25t 0.11* 0.04 0.01 -0.12* -0.11* 0.13** -0.03 - 0.06 -0.13**

Note. VO, max, milliliters 0, per kilogram body weight per minute. LIA higher score indicates more physical activity. * P < 0.05. ** P < 0.01. *** P < 0.001. t P < 0.0001.

BMI and sexual maturity ratings using analyses of covariance. The results of these analyses are shown in Table 4. In both genders, BMI and triceps skinfold thickness were successively lower with increasing VOz max quartiles, indicating that the more fit students were leaner than the less fit students. Those who were in the lowest quartile of VO, max (and in the poorest physical condition) had higher SBP, DBP4, and DBPS compared with students who scored higher on the bicycle test. Pulse rate was significantly different in the four V02 max quartiles, with students that were in better physical condition having lower values. Among boys, no significant differences were seen between the four quartiles of VO, max and serum lipids. Girls in the highest VO, max quartile had a higher HDL and HDL-C/TC ratio, and those who were more fit had lower TG. Similar analyses were performed with self-reported physical activity as the dependent variable. As seen in Table 5, the five answer alternatives of the question were grouped into three categories; alternatives 1 and 2 were collapsed as “at least 2-3 times per week,” alternatives 3 and 4 as “weekly or 2-3 times per month,” and category 5 as “not that often. ” When we looked at trends with level of activity frequency, significant differences in the means over the three categories were seen in mean VO, max (adjusted for BMI and sexual maturity) for both boys and girls; students who exercised more frequently had higher levels of VOz max. Male students who exercised more frequently also had lower resting pulse rates and were leaner than those who exercised less frequently or never. No such differences were observed among the female participants; however, girls who ex-

66.6 67.3 63.8 2.99 0.032

109.3 108.2 107.3 3.04 0.029

13.7 12.3 11.1 20.00 0.000

0.000

P value

0.000

P value

o Adjusted for sexual maturity rating. b Adjusted for sexual maturity rating and body mass index.

19.2 18.1 17.8 17.68

39.8-46.0 46.1-52.6 rS2.7 F (ANCOVA)

19.7

62.2 61.8 2.67 0.048 69.4

106.9 108.4 4.80 0.003 112.5

8.4 9.3 34.78 0.000 16.1

17.3 17.9 29.99

959.6 52.9-59.5 F (ANCOVA)

Girls c39.7

66.0 64.2

112.1 107.2

13.8 10.8

19.7 18.5

Boys c45.3 45.3-52.8

(mr%Ig)

Phase 4 (mm Hg)

62.2 59.8 61.2 58.3 2.38 0.069

59.1 56.9 53.3 53.5 6.75 0.000

Phase 5 (mm Hg)

Diastolic BP

BMP’ Wmz)

Systolic

vo, max (ml 04 (kg x tin))

Triceps skinfold thickness” (mm)

TABLE 4

90.7 90.5 85.3 78.2 13.66 0.000

88.4 82.9 81.2 76.2 14.68 0.000

Pulse* (beatdmin)

4.85 4.70 4.72 4.74 0.82 0.484

4.78 4.58 4.57 4.65 1so 0.215

Total cholesterop (mmof/liter)

MEAN RISKFACTORLEVE@BY VO, &~AXQUARTILES

1.33 1.27 1.27 1.38 3.47 0.016

1.36 1.37 1.37 1.34 0.26 0.853

HDL cholesterop (mmoyliter)

-

0.28 0.27 0.27 0.29 2.43 0.065

0.29 0.31 0.30 0.29 1.28 0.282

HDL/total cholesterol*

0.80 0.75 0.71 0.60 4.05 0.008

0.58 0.55 0.58 0.56 0.31 0.818

Triglycerides* (mmoUliter)

$

is F

5

8 F

PHYSICAL FITNESS VERSUS CVD RISK FACTORS

19

ercised more frequently had lower fasting TG levels compared with girls who did not get as much physical activity. No significant relationships were seen between exercise frequency and other serum lipids or any of the BP variables, and so these are not included in the table. Multiple regression analysis was performed to identify variables predictive of estimated maximum oxygen uptake. A stepwise regression technique was used to allow the independent variables to enter in the order of their explanatory ability. Gender-specific equations were generated. Independent variables entered in the analyses were BP, pulse rate, self-reported physical activity (entered as a j-point categorical variable), height, triceps skinfold thickness, HCR, sexual maturity ratings, TC and HDL-C, HDL-C/K ratio, and fasting TG. Because triceps skinfold thickness was more strongly correlated with VOz max than were weight and BMI in both genders, this variable was included in the list of independent variables. For boys, triceps skinfold thickness and pulse rate were the two strongest predictors, contributing respectively 22.4 and 14.4% to R2. Including additional variables that contributed to at least 1% of the R2, the linear regression equation for boys was V02 max (ml O,/kg x min) = - 1.04 (triceps) - 0.37 (pulse)

- 12.00 (height) + 112.06. The R2 for this equation was 0.38. For girls, triceps skinfold thickness and pulse rate were the two strongest predictors, contributing respectively 12.7 and 11.O%to R*. The strongest regression equation was (including variables contributing to at least 1% of the R2) V02 max (ml O,/kg x min) = - 0.69 (triceps) - 0.18 (pulse)

- 0.16 (DBP4) - 0.11 (SBP) + 94.94. The R2 for this equation was 0.29. When BMI was substituted for triceps skinfold thickness in the regression analyses, almost identical results were obtained. For boys, pulse rate and BMI contributed 19.4 and 19.1% of R2, respectively, and the regression equation was vG2 max (ml &/kg X min) = - 0.37 (pulse) - 2.21 (BMI) + 2.29 (sexual maturity) - 17.09 (height) + 144.60. The R2 for this equation was 0.37. Similarly, for girls, BMI and pulse rate were the two strongest predictors in the new equation, contributing respectively 13.3 and 9.8% to the R2. V02 max (ml 02/kg x min) = - 1.36 (BMI) - 0.20 (pulse)

- 0.21 (DBP4) - 1.64 (TC) + 0.42 (HCR) + 94.48. The R2 for this equation was 0.29. Separate regression analyses were performed with physical activity frequency (PA) as the dependent variable and with the physiological variables as independent variables. For boys, pulse rate and V02 max were the only significant vari-

261

85

67

Boys At least 2-3 times per week

Weekly or 2-3 times per month

Not that often

103

82

Weekly or 2-3 times per month

Not that often

I X

11.36 0.000 *t = 4.69 P= o.ooo **t = 2.29 P = 0.022

P= P=

**t =

*t =

P = 0.002

-

x

0.55 0 58 --

81.2

86.9

85.4

3,89 0.021 2.32 0.021 2.04 0.042

84.0

84.2 I

80.6

J

*t P **t P

= = = =

-0.55

0.59

13.3

13.7

13.0

4.98 o.007 2.68 0.008 2.22 0.027

11.5

11.6

10.1

Triceps skinfold thickness” (mm)

xx 1

Sitting pulseb (beats/mm)

*t = 3.08

4.80 0.009

51.8

49.5

53.2

VO, maxb (ml O,/kg x min)

a Adjusted for sexual maturity rating. b Adjusted for sexual maturity and body mass index.

F (ANCOVA) P value

187

Girls At least 2-3 times per week

P value

F (ANCOVA)

N

Physical activity frequency

5

MEAN RISK FACTORLEVELS BY PHYSICAL ACTIVITY FREQUENCY

TABLE

X

0.70 50 L0

18.5

18.9

18.7

5.09 o.007 *t = 3.01 P = 0.003

18.6

18.91

18.1

BMI” Wm*)

F: $

8

I 3.27 0.029 **r = 2.55 P= 0.011

0.81

0.71

0.68

0.30 0.74

xx

: - 8F zis

0.57

0.54

0.57

Triglyceridesb (mmohliter)

PHYSICAL FITNESS VERSUS CVD RISK FACTORS

21

ables, contributing respectively 4.4 and 1.7% to the R2, explaining a total of 6.1% of the variation in PA: PA = 0.02 (pulse) - 0.02 (V02 max) + 2.14. A different picture was seen for girls, where DBP4, fasting TG, and HCR entered significantly into the equation, contributing respectively 2.0, 1.8, and 1.4% to the variation in R2 and explaining a total of 5.2% of the variation in PA: PA = 0.02 (DBP4) + 0.53 (TG) - 0.06 (HCR) + 3.63. DISCUSSION Results of this study indicate that there are important relationships between risk factors for CVD and physical fitness, resting pulse, and physical activity. In general, higher levels of fitness were associated with more favorable risk profiles. The same tendency was found by Fripp et al. (11) among 10th grade male students; subjects with higher levels of fitness had more favorable risk profiles with lower BMI, SBP, DBP, and plasma TG levels and higher plasma HDL-C levels. Because CVD risk factors have been found to be related more strongly to sexual maturity than to chronological age in this population (20, 21), and because different sexual maturity levels may mask or enhance true relationships between physical fitness and the variables studied here, we adjusted for sexual maturity ratings in the analyses of covariance presented. Thus the observed relationships between physical fitness and CVD risk factors are not due to confounding by pubertal development. As seen in Table 3, V02 max is negatively related to sexual maturity but positively related to age in years. Since girls who mature earlier are often heavier than girls who mature later (e.g., triceps skinfold thickness is significantly correlated with sexual maturity [r = 0.191, but not with age [r = 0.03]), the relationship between fitness and sexual maturity may be confounded by body fat level. In fact, while the linear correlation coefficient between VO, max and sexual maturity is - 0.12, the partial correlation coefficient between sexual maturity and fitness, after the linear effect of triceps skinfold thickness is removed, is -0.05. Cardiovascular fitness was negatively related to BMI and triceps skinfold thickness in our study. Similar relationships have been observed in other studies of adolescents (11, 16), and there have been some suggestions that the relationship between better physical fitness and lower BP may be due to lower body fat level. However, in this study we found that physical fitness was related to BP independent of BMI. This is in agreement with results of Fraser et ~1. (IO), who found that after adjustment for skinfold thickness and an index of lean arm mass, the SBP of male and female adolescents who were above average in fitness was lower than the SBP of those with below average fitness. In our study, the most important predictors of physical fitness were triceps skinfold thickness in boys and BMI in girls. Perhaps part of the explanation of the commonly observed rela-

22

TELL

AND VELLAR

tionship between BP and indices of obesity may involve a lower overall physical fitness in obese persons. DuRant et al. (6, 7) found that the levels of habitual physical activity in white boys ages 11-17 years, and in black adolescent males and females, were associated with higher levels of HDL-C relative to TC and low-density lipoprotein cholesterol. Thorland and Gilliam (24) reported that habitually active 8- to llyear-old boys had significantly higher HDL-C/TC ratios and lower TG levels than less active subjects. In our study, HDL-CiTC ratios showed significant associations with VO, max in both genders and with pulse rate in girls. Also, more fit girls had significantly higher HDL-C and lower TG levels. Because maintaining a high level of HDL-C relative to TC appears to favorably influence the developmental process of atherosclerosis (13), it would seem advantageous to identify factors that have a positive influence on HDL-C and other blood lipids in children and adolescents. Thus, if the level of physical activity and physical fitness favorably affects the lipid profile in children and adolescents, this may have important implications for the prevention of CVD. Resting heart rate is a determinant of mortality in adults (8). It is also an indicator of physical fitness (2), and in our study it was the second best predictor of VO, max for both genders. A higher pulse rate has been associated with a higher level of blood pressure in adults (18) and adolescents (25) and with the future development of hypertension in college students (17). In the Oslo Youth Study we found significant positive correlations between resting pulse rate and BP in both genders. Since resting heart rate is an easily measureable parameter of cardiovascular system status, we suggest that it should be routinely assessed in epidemiological studies of CVD risk factors in children and adolescents, as well as in adults. Although physical activity is a mandatory subject in all grades in Norway, the exercise intensity level in these classes varies greatly. The majority of students who are active in sports do so outside of school. More than half of our students surveyed indicated participation in organized sports clubs, while 17% reported not engaging in sports outside of school at least once a week. Since physical activity with a frequency of three times per week for 20-30 min at 80% maximal heart rate is thought to be necessary for cardiovascular fitness, we used a survey question that would indicate both frequency and level of intensity of physical activity. Although this question was significantly correlated (although moderately so) with VO, max for both genders in our population sample, it may not be equally suitable for students of other populations, especially for those where most sports played are part of the school curriculum. In summary, results of the Oslo Youth Study indicate that there is an inverse relationship between physical fitness and CVD risk factors in this sample of Norwegian adolescents. In particular, higher levels of physical fitness were associated with lower levels of resting heart rate, BP, triceps skinfold thickness, and BMI, and a more favorable lipid profile in girls. Further, high levels of physical activity were associated with less smoking in girls. Results of the Oslo Youth Study indicate that frequent participation in vigorous physical activity enhances

PHYSICAL FITNESS VERSUS CVD RISK FACTORS

23

cardiovascular fitness and may conceivably be of benefit in the reduction of future CVD risk. REFERENCES 1. Astrand, I. Aerobic work capacity in men and women with special reference to age. Actu Physiol. Stand. 49 (Suppl. 169), 45-60 (1960). 2. Astrand, P. O., and Rodahl, K. “Textbook of Work Physiology: Physiological Bases of Exercise,” 2nd ed. McGraw-Hill, New York, 1977. 3. Cooper, K. H., Pollock, M. L., Martin, R. P., White, S. R., Linnerud, A. C., and Jackson, A. Physical fitness levels vs selected coronary risk factors: A cross-sectional study. JAMA 236, 166-169(1976). 4. Dawber, T. R. Risk factors in young adults: The lessons from epidemiologic studies of cardiovas-

cular disease: Framingham, Tecumseh, and Evans County. J. Amer. Co/l. Health Assoc. 22, 84-95 (1973). 5. Dixon, W. J. (Ed.) “BMDP Statistical Software.” Univ. of California Press, Berkeley, CA, 1983.

6. DuRant, R. H., Linder, C. W., Harkess, J. W., and Gray, R. G. The relationship between physical activity and serum lipids and lipoproteins in black children and adolescents. J. Adolesc. Health Care 4, 55-60 (1981). 7. DuRant, R. H., Linder, C. W., and Mahoney, 0. M. Relationship between habitual physical activity and serum lipoprotein levels in white male adolescents. J. Adolesc. Health Care 4, 235-240(1983). 8. Dyer, A. R., Persky, V., Stamler, J., Paul, O., Shekelle, R. B., Berkson, D. M., Lepper, M.,

Schoenberger, J. A., and Lindberg, H. A. Heart rate as a prognostic factor for coronary heart disease and mortality: Findings in three Chicago epidemiologic studies. Amer. J. Epidemiol. 112, 736-749(1980). 9. Fox, S. M., Naughton, J. P., and Haskell, W. L. Physical activity and the prevention of coronary

heart disease. Ann. Clin. Res. 3, 404-432 (1971). 10. Fraser, G. E., Phillips, R. L., and Harris, R. Physical fitness and blood pressure in school children. Circulation 67, 405-412 (1983). 11. Fripp, R. R., Hodgson, J. L., Kwiterovich, P O., Werner, J. C., Schuler, H. G., and Whitman, V. Aerobic capacity, obesity, and atherosclerotic risk factors in male adolescents. Pediatrics 75, 813-818 (1985). 12. Gibbons, L. W., Blair, S. N., Cooper, K. H., and Smith, M. Association between coronary heart disease risk factors and physical fitness in healthy adult women. Circulntion 67, 977-983 (1983). 13. Gordon, T., Castelli, W. P., Hjortland, M. C., Kannel, W. B., and Dawber, T. R. High density lipoprotein as a protective factor against coronary heart disease. Amer. J. Med. 62, 707-714 (1977).

14. Hermansen, L., and Oseid, S. Direct and indirect estimation of maximal oxygen uptake in prepubertal boys. Acta Puediat. Stand. 217 (Suppl.), 18-23 (1971). 15. LaPorte, R. E., Adams, L. L., Savage, D. D., Brenes, G., Dearwater, S., and Cook, T. The spectrum of physical activity, cardiovascular disease and health: An epidemiologic perspective. Amer. J. Epidemiol. 120, 507-517 (1984). 16. Mocellin, R., and Rutenfranz, J. Investigations of the physical working capacity of obese children. Acta Paediat. Stand. 217 (Suppl.), 77-79 (1971). 17. Paffenbarger, R. S., Thome, M. C., and Wing, A. L. Chronic disease in former college students. VIII. Characteristics in youth predisposing to hypertension in later years. Amer. J. Epidemiol. 88, 25-32 (1968).

18. Stamler, J., Rhomberg, P., Schoenberger, J. A., Shekelle, R. B., Dyer, A., Shekelle, S., Stamler, R., and Wannamaker, J. Multivariate analysis of the relationship of seven variables to blood pressure: Findings of the Chicago Heart Association Detection Project in Industry, 1967-1972. J. Chronic Dis. 28, 527-548. (1975). 19. Tanner, J. M. “Growth at Adolescence,” 2nd ed. Blackwell Scientific, Oxford, 1962.

24

TELL AND VELLAR

20. Tell, G. S. Cardiovascular disease risk factors related to sexual maturation: The Oslo Youth Study. .I. Chronic Dis. 38, 633-642 (1985). 21. Tell, G. S., Mittelmark, M. B., and Vellar, 0. D. Cholesterol, high density lipoprotein cholesterol and triglycerides during puberty: The Oslo Youth Study. Amer. J. Epidemiol. 122, 750-761 (1985). 22. Tell, G. S., Btomilehto, J., Epstein, F. H., and Strasser, T. Studies of atherosclerosis determinants and precursors during childhood and adolescence. WHO Bull. 64, 595405 (1986). 23. Tell, G. S., Vellar, 0. D., and Monrad-Hansen, H. P. Risk factors for chronic diseases in Norwegian school children: Preliminary findings of a baseline survey as part of an intervention study (The Oslo Youth Study). Prev. Med. 10, 211-225 (1981). 24. Thorland, W. G., and Gilliam, T. B. Comparison of serum lipids between habitually high and low active pre-adolescent males. Med. Sci. Sports Exercise 13, 316-321 (1981). 25. Voors, A. W., Webber, L. S., and Berenson, G. S. Resting heart rate and pressure-rate product of children in a total biracial community. Amer. J. Epidemiol. 116, 276-286 (1982).