Journal Logo


Associations between physical activity and risk factors for coronary heart disease: The Cardiovascular Risk in Young Finns Study


Author Information
Medicine & Science in Sports & Exercise: August 1997 - Volume 29 - Issue 8 - p 1055-1061
  • Free


Coronary heart disease (CHD) is recognized as a pediatric problem even though the clinical symptoms of CHD do not appear until much later in life (17). CHD risk factors- elevated serum lipids, obesity, hypertension, smoking, diabetes mellitus, and physical inactivity-can already be identified in childhood. Several studies indicate that many children who are obese or have high blood pressure or dyslipidemia are likely to retain these CHD risk factors as adults(24,36). In adults the beneficial effect of physical activity on CHD risk factors, such as obesity, serum lipids and lipoproteins (32,38), diabetes(20), and to some extent on hypertension(14,21) is widely approved. In addition, smoking and overeating are less frequent in physically active subjects(8). The studies on the effects of physical activity on CHD risk factors have been, however, more limited in children and young adults(9,26,27,35,39,40). The aim of this study was to study the relationships between physical activity and CHD risk factors, i.e., body composition, blood pressure, serum lipids, apolipoproteins, and insulin levels in children and young adults.


Study population. Cardiovascular Risk in Young Finns Study is a multicenter study on atherosclerosis precursors in Finnish children and young adults. The initial cohort included 3,596 children and young adults aged 3, 6, 9, 12, 15, and 18 years who were chosen from the national population register. A baseline cross-sectional study was done in 1980. The initial study population consisted of 4,320 children and adolescents aged 3, 6, 9, 12, 15, and 18 years. Of those invited, 3,596 subjects (83.2%) participated. Details of the baseline study in 1980 have been published earlier(3). Follow-up studies were carried out in 1983 and 1986 with the same protocol (2). The loss of subjects was approximately 20% and 30% after 3 and 6 years, respectively. The present report includes results of 1,114 males and 1,244 females who participated in the study in 1986 (Table 1). The study year of 1986 was selected because the widest selection of biochemical measurements was then available. The study protocol was approved by Ethics Committees of each of the five participating universities (medical schools of Helsinki, Kuopio, Oulu, Tampere, and Turku).

Anthropometric and lifestyle variables. Height was measured by Seca anthropometer and weight by Seca weighing scale. Subscapular skinfolds were measured were measured from the nondominating arm using a Harpenden skinfolds caliper. BMI was calculated from the formula: BMI = weight(kg)/(height (m))2.

Participation in leisure-time physical activity was assessed by using a questionnaire. Participants were asked the frequency of participation in physical activity and its intensity outside school hours. There were seven multiple choice answers for frequency: 1) never, 2) “less than once a month,” 3) “once a month,” 4) “2-3 times a month,” 5) “once a week,” 6) “2-6 times a week,” and 7) “once a day.” The multiple choice answers for intensity included the following three alternatives: 1) “never sweating and becoming breathless,” 2) “some sweating and becoming breathless,” and 3) “heavy sweating and becoming breathless.”

An index for leisure-time physical activity was calculated from the product of intensity, estimated duration, and monthly frequency(5,37). For intensity we used coefficient values 4(“never sweating and becoming breathless”) corresponding to light aerobic activity, 6 (“some sweating and becoming breathless”) corresponding to moderate aerobic activity, and 10 (“heavy sweating and becoming breathless”) corresponding to intense aerobic activity. The coefficient values 4, 6 and 10 for different levels of intensity were chosen in order to estimate the average metabolic cost of each intensity level. The values were estimated from existing tables(1,5,37). An average value of 30 min(coefficient 0.5) for duration was used in case of other physical activity than supervised exercise. In case of supervised exercise (participation to sports club session), an average value of 45 min for duration (coefficient 0.75) was used. Coefficient value 0 for duration was used in case of no leisure-time physical activity. Coefficient values for monthly frequency of activity were 0.5 (less than once a month), 1 (once a month), 2.5 (two to three times a month), 4.3 (once a week), 17 (two to six times a week) and 30(once a day). The range of the index was 0 to 225.

We defined the three-level classification of physical activity at 1986 as follows: the subjects with a physical activity index level higher than or equal to 90 for males (N = 277) or 80 for females (N = 284) were considered as physically active. The subjects with a physical activity index level less than or equal to 15 (males, N = 373; females, N = 530) were considered as physically inactive. The other subjects with a physical activity index between these two cut-points (males,N = 486; females N = 466) were classified as moderately active. The upper cut-point limit was different between genders to select approximately the highest quartile in both genders into the highest activity class.

The pubertal stages were estimated according to Tanner and Whitehouse(30). The five Tanner stages were grouped into three classes. The first class included subjects with puberty not started(N = 373), i.e., all of the Tanner parameters (public hair, genitals/breast) stage 1. The second class included subjects with puberty ongoing (N = 922), i.e., any of the parameters at least stage 2 but not stage 5. The third class included subjects with puberty passed(N = 1045), i.e., at least one of the parameters stage 5.

Blood sampling. Blood was drawn from the right antecubital vein with the subject lying recumbent after an overnight fast. Fasting was confirmed by an interview. Serum was separated and stored at -20°C until assayed.

Determination of serum lipids, apolipoproteins, insulin, and LCAT. All lipid determinations were done in duplicate and in the same laboratory (Research and Development Unit, Social Insurance Institution, Turku, Finland) using standard enzymatic methods for serum total cholesterol(TC) (Boehringer CHOD-PAP) and triglycerides (TG) (Boehringer). Serum high-density cholesterol (HDL-C) was measured from the serum supernatant after precipitation of very-low-density- and low-density lipoproteins with dextran sulphate (DS-500 000) and MgCl2(10). The precipitation was done at the blood collection site. Low-density lipoprotein cholesterol (LDL-C) concentrations were calculated by the Friedewald formula(11). HDL2-C and HDL3-C subfractions were determined by the dual precipitation method (13). Details of the methods have been presented elsewhere(34). Apolipoproteins A-I and B were determined by immunoturbidometry (25). Serum insulin was measured using a modification of the immunoassay method of Herbert et al.(16). Serum lecithin:cholesterol acyltransferase (LCAT) activity was determined using exogenous substrate (4).

Statistical methods. Analyses were performed separately for males and females. Least-squares regression was used to produce age-adjusted averages of measurements taken in the beginning of the study in individuals who participated in all of the studies and in those who were lost to the follow-up. The P-values for differences in these averages for each gender were calculated in a supplementary analysis. Differences in physical activity level between the genders were tested with a nonparametric Mann Whitney U-test. Differences in CHD risk factors between groups of physical activity were analyzed by two-way ANOVA (pubertal stage as the second dimension). Linear trends in CHD risk factors with physical activity were analyzed by linear regression model adjusted for pubertal stage. Analyses were performed using version 6.04 of the Statistical Analysis System (SAS) scientific software for microcomputers.


The representativeness of the current study cohort (Table 1) was tested by comparing it with the subjects who were lost to follow-up or excluded because of missing data on exercise or serum lipid variables(Table 2). No significant difference was found in the proportion of males lost (36.4%) and females lost (33.6%) (P = 0.079). The study cohort was 3.3 years younger on the average than the lost cohort (Table 2). The lost cohort was on the average more obese and had lower levels of HDL-C and higher levels of TG than the study cohort (Table 2). Furthermore, the study cohort was physically more active than the lost cohort (Table 2).

On average males were physically more active than females. The mean physical activity index value was 59 (median 51) in males, and 45 (median 51) in females (P < 0.001). The distribution of the physical activity index was skewed to the right in both genders, i.e., there were less frequently high levels than low levels of the index (data not shown).Figure 1 shows the distribution of physical activity level by age and gender. The lower end of the floating bar indicates the age- and gender-specific 25th percentile of physical activity index, and the upper end indicates the 75th percentile.

Body composition. Among males, higher physical activity levels were associated with lower BMI levels and thinner skinfolds(Table 3). Among females, skinfolds were thinner in physically active than among inactive (Table 3). Significant dose-related effects were also observed in these associations. In females, no statistically significant association was seen between physical activity and BMI.

Blood pressure. No relation between physical activity and systolic or diastolic blood pressure was found in either sex(Table 4).

TC, LDL-C, and apoB. Among males, TC and LDL-C tended to be lower with high levels of physical activity although statistical significance was not reached (Table 5). ApoB (Apolipoprotein B) levels were significantly lower among physically active males and a significant dose-related effect was found (Table 5). No significant differences nor trends as regards TC, LDL-C or ApoB were found among females.

High-density lipoproteins and LCAT. HDL-C was high in physically active males with a significant dose-related effect (Table 6). High HDL2-C levels were associated with high levels of physical activity (Table 6), while no such association was found in HDL3-C. No significant associations or trends between physical activity and HDL were found among females. Also, HDL-C to TC ratio and ApoAI(Apolipoprotein A-I) to ApoB ratio were higher in physically active males in a dose-related manner (Table 7). No significant association was found between physical activity and LCAT activity (Table 6).

TG and insulin. Among males, a low serum TG level was associated in a dose-related manner with a high level of physical activity(Table 8). Serum insulin levels were significantly lower among physically active males compared with the physically less active(Table 8). Among females, low serum TG levels were associated in a dose-related manner with high levels of physical activity(Table 8).


We found that among young males there are many beneficial relationships between physical activity and many CHD risk factors, such as obesity, ApoB, HDL-C, HDL2-C, TG, and insulin. For example, physically active males had on average 4% higher serum HDL-C levels compared with physically inactive males. These findings were consistent with each stage of puberty, and trend analyses indicated significant dose-related effects between physical activity and these CHD risk factors. The statistical models were adjusted for pubertal stage instead of age since among young people pubertal status may be more relevant confounder than years of age per se. Our findings on the relation between physical activity and CHD risk factors in males are therefore in agreement with many previous studies(9,19,26,27,33,35,40).

Among females, the benefits of activity were seen in obesity variables and serum TG levels only. Thus, our findings suggest that there may be a gender difference in the lipid response to exercise. Large epidemiologic studies have shown that both men and women who are physically active have higher concentrations of HDL-C than sedentary individuals(12,15). However, intervention studies suggest that exercise training programs alone do not cause HDL-C levels to increase notably in middleaged women and that high amounts of exercise are needed to raise HDL-C in young women (31). In the present study, females were physically somewhat less active than males. Thus, the absolute level of physical activity may have been too low to induce differences in other lipids than TG in females. This may offer an explanation for the observed gender difference.

There were certain differences in the main study variables between the study cohort and the subjects lost to follow up. For example, subjects lost were older, more obese, and physically less active compared with the study cohort at the baseline. Therefore, it may be that the study cohort is not ideally representative of the original baseline cohort. This means that we have to be careful when generalizing these results directly to the general population. However, it is unlikely that the strength of associations between physical activity level and other risk factors would be greatly affected by the selection caused by lost to follow-up.

No universally accepted simple questionnaire methods exist to evaluate the intensity of exercise in population studies at present(7,18). Usually people tend to overestimate the amount of their exercise (18) which declines the validity of questionnaire methods. Assessing physical activity levels in adults using inventories is a difficult task. It is even more difficult to do this in children. Estimates of the physical activity level in children and adolescents based on questionnaire data may be inadequate since children have difficulties of recalling their activities accurately(28). Therefore, the youngest participants in the present study (9 and 12 years) were encouraged to fill the questionnaire with the aid of their parents.

Macroscopic and microscopic studies provide evidence that fatty streaks and intimal lipid accumulations occur in coronary arteries in children(29). Furthermore, serum lipid levels have been linked to early arterial lesions in the aorta and the coronary arteries(22,23). Also Klag et al.(17) showed recently that TC measurements in young adulthood are powerful predictors of CHD in middle-age. Thus, cardiovascular mortality may be reduced by correcting the CHD risk factor levels at a time when irreversible degenerative changes have not yet occurred in the cardiovascular system. Strategies for intervention include both a population approach and an individualized approach (6). The population approach includes, e.g., nutrition recommendations to lower the average population levels of serum cholesterol. The results of the present study show that physical activity induces favorable changes in obesity and serum lipoproteins in children and young adults. Physical activity may therefore be recommended as an additional preventive measure for population approach to lower the average population levels of CHD risk factors.

Figure 1-Physical activity levels by age and gender. The lower end of the floating bar indicates the age- and gender-specific 25th percentile of physical activity index, and the upper end indicates the upper 75th percentile of the index.
Figure 1-Physical activity levels by age and gender. The lower end of the floating bar indicates the age- and gender-specific 25th percentile of physical activity index, and the upper end indicates the upper 75th percentile of the index.


1. Ainsworth, B. E., W. L. Haskell, A. S. Leon, et al. Compendium of physical activities: classification of energy costs of human physical activities. Med. Sci. Sports Exerc. 25:71-80, 1993.
2. Åkerblom, H. K., M. Uhari, E. Pesonen, et al.Cardiovascular Risk in Young Finns. Ann. Med. 23:35-39, 1991.
3. Åkerbom, H. K., J. Viikari, M. Uhari, et al. Atherosclerosis precursors in Finnish children and adolescents. I. General description of the cross-sectional study of 1980, and an account of the children's and families' state of health. Acta Paediatr. Scand. Suppl. 318:49-63, 1985.
4. Alcindor, L., A. Dusser, M. C. Piot, R. Infante, and J. Polonowski. A rapid method for lecithin:cholesterol acyltransferase estimation in human serum. Scand. J. Lab. Invest. 38(Suppl. 150):12-14, 1978.
5. American College Of Sports Medicine. Position Stand: The recommended quantity and quality of exercise for developing and maintaining cardiorespiratory and muscular fitness in healthy adults. Med. Sci. Sport. Exerc. 22:265-274, 1990.
6. National Cholesterol Education Program. Report of expert panel on blood cholesterol levels in children and adolescents.Pediatrics 89(Suppl. 1):525-577, 1992.
7. Arroll, B. and R. Beaglehole. Potential misclassification in studies of physical activity. Med. Sci. Sport. Exerc. 23:1176-1178, 1991.
8. Blair, S. N., D. R. Jacobs, and K. E. Powell. Relationships between exercise or physical activity and other health behaviors. Public Health Rep. 100:172-180, 1985.
9. Durant, R. H., T. Baranowski, T. Rhodes, et al. Association among serum lipid and lipoprotein concentrations and physical activity, physical fitness, and body composition in young children. J. Pediatr. 123:185-192, 1993.
10. Finley, P. R., R. B. Schifman, J. Williams, and D. A. Lichti. Cholesterol in high density lipoprotein: use of Mg2+/dextran sulphate in its enzymatic measurement. Clin. Chem. 24:931-937, 1978.
11. Friedewald, W. T., R. Levy, and D. S. Fredrickson. Estimation of the concentration of low-density lipoprotein cholesterol in plasma without use of the preparative ultracentrifuge. Clin. Chem. 18:499-502, 1972.
12. Fulton-Kehoe, D. L., R. H. Eckel, S. M. Shetterly, and R. F. Hamman. Determinants of total high density lipoprotein cholesterol and high density lipoprotein subfraction levels among Hispanic and non-Hispanic white persons with normal glucose tolerance: The San Luis Valley Diabetes Study. J. Clin. Epidemiol. 45:1191-1200, 1992.
13. Gidez, L. I., G. J. Miller, M. Burstein, S. Slagle, and H. A. Eder. Separation and quantitation of subclasses of human plasma high density lipoproteins by a simple precipitation procedure. J. Lipid Res. 23:1206-1210, 1982.
14. Gilders, R. M. and G. A. Dudley. Endurance exercise training and treatment of hypertension: the controversy. Sports. Med. 13:71-77, 1992.
15. Helmert, U., B. Herman, and S. Shea. Moderate and vigorous leisure-time physical activity and cardiovascular disease risk factors in West Germany, 1984-1991. Int. J. Epidemiol. 23:285-292, 1994.
16. Herbert, V., K. S. Lau, C. W. Gottlieb, and S. J. Bleicher. Coated charcoal immunoassay of insulin. J. Clin. Endocrinol. Metab. 25:1375-1384, 1965.
17. Klag, M. J., D. E. Ford, L. A. Mead, et al. Serum cholesterol in young men and subsequent cardiovascular disease. N. Engl. J. Med. 328:313-318, 1993.
18. Klesges, R. C., L. H. Eck, M. W. Mellon, W. Fulliton, G. W. Somes, and C. L. Hanson. The accuracy of self-reports of physical activity. Med. Sci. Sport. Exerc. 22:690-697, 1990.
19. Linder, C. W. and R. H. Durant. Exercise, serum lipids, and cardiovascular disease-risk factors in children. Pediatr. Clin. N. Am. 29:1341-1354, 1982.
20. Manson, J. E., E. B. Rimm, M. J. Stampfer, et al. Physical activity and incidence of non-insulin-dependent diabetes mellitus in women. Lancet 338:774-778, 1991.
21. Marti, B. Health effects of recreational running in women. Some epidemiological and preventive aspects. Sports. Med. 11:20-51, 1991.
22. Newman, W. P., D. S. Freedman, A. W. Voors, et al. Relation of serum lipoprotein levels and systolic blood pressure to early atherosclerosis. N. Engl. J. Med. 314:138-144, 1986.
23. PDAY Research Group. Relationship of atherosclerosis in young men to serum lipoprotein cholesterol concentrations and smoking: a preliminary report from the Pathobiological Determinants of Atherosclerosis in Youth (PDAY) Research Group. JAMA 264:3018-3024, 1990.
24. Porkka, K. V. K., J. S. A. Viikari, S. Taimela, M. Dahl, and H. K. Åkerblom. Tracking and predictiveness of serum lipid and lipoprotein measurements in childhood: a 12-year follow-up. The Cardiovascular Risk in Young Finns Study. Am. J. Epidemiol. 140:1096-1110, 1994.
25. Riepponen, P., J. Marniemi, and T. Rautaoja. Immunoturbidimetric determination of apolipoproteins A1 and B in serum.Scand. J. Clin. Lab. Invest. 47:739-747, 1987.
26. Rönnemaa, T., A. Lehtonen, H. Järveläinen, and J. Viikari. The effect of exercise on plasma lipids and lipoproteins of young men and the effect of their sera on cultured human aortic smooth muscle cells. Scand. J. Sports. Sci. 2:33-38, 1980.
27. Saris, W. H. M. Aerobic power and daily physical activity in children with special reference to methods and cardiovascular risk indicators. Academic Dissertation. University of Nijimegen, Wageningen, Netherlands, 1982.
28. Saris, W. H. M. The assesment and evaluation of daily physical activity in children. A review. Acta Paediatr. Scand. Suppl. 318:37-48, 1985.
29. Stary, H. C. Evolution and progression of atherosclerotic lesions in coronary arteries of children and young adults.Arteriosclerosis 9(Suppl.):L19-L32, 1989.
30. Tanner, J. M. and R. H. Whitehouse. Clinical longitudinal standards for height, weight, height velocity, and stages of puberty. Arch. Dis. Child. 51:170-179, 1976.
31. Taylor, P. A. and A. Ward. Women, high-density lipoprotein cholesterol and exercise. Arch. Intern. Med. 153:1178-1184, 1993.
32. Thompson, P. D., B. Lazarus, E. Cullinane, et al. Exercise, diet or physical characteristics as determinants of HDL-levels in endurance athletes. Atherosclerosis 46:333-339, 1983.
33. Välimäki, I., M. L. Hursti, L. Pihlakoski, and J. Viikari. Exercise performance and serum lipids in relation to physical activity in school children. Int. J. Sports. Med. 1:132-136, 1980.
34. Viikari, J., T. Rönnemaa, A. Seppänen, et al. Serum lipids and lipoproteins in children, adolescents and young adults in 1980-1986. Ann. Med. 23:53-59, 1991.
35. Wanne, O., J. Viikari, R. Telama, et al. Physical activity and serum lipids in 8-year-old Finnish schoolboys. Scand. J. Sports. Sci. 5:10-14, 1983.
36. Webber, L. S., S. R. Srinivasan, W. A. Wattigney, and G. S. Berenson. Tracking of serum lipids and lipoproteins from childhood to adulthood: The Bogalusa Heart Study. Am. J. Epidemiol. 133:884-899, 1991.
37. Wilson, P. W. F., R. S. Paffenbarger, J. N. Morris, and R. J. Havlik. Assessment methods for physical activity and physical fitness in population studies: Report of a NHLBI workshop. Am. Heart J. 111:1177-1192, 1986.
38. Wood, P. D., W. L. Haskell, S. N. Blair, et al. Increased excercise level and plasma lipoprotein concentrations: a one-year randomized study in sedentary middle-aged men. Metabolism 32:31-39, 1983.
39. Ylitalo, V. Treatment of obese schoolchildren.Acta Paediatr. Scand. Suppl. 290:1-108, 1981.
40. Zonderland, M. L. Lipid and apolipoprotein profiles in premenarcheal athletes. The relation with training, nutrition and biological maturation. Academic Dissertation. University of Utrecht, Utrecht, Netherlands, 1985.


©1997The American College of Sports Medicine