Associations between estimated aerobic fitness and cardiovascular risk factors in adults with different levels of abdominal obesity

Abstract

Background We investigated the association between estimated aerobic fitness and cardiovascular risk factors, and how the association is affected by abdominal obesity.

Design Cross-sectional population study.

Methods Participants comprised 3820 adults aged 25 to 64 years from the FIN RISK 2002 Study in Finland. Aerobic fitness was estimated using a non-exercise test. Waist-to-hip ratio (WHR), blood pressure, total cholesterol, high-density lipoprotein cholesterol (HDL-C), triglycerides, HDL-C to total cholesterol ratio, and gamma-glutamyl transferase (GGT) levels were measured by standardized methods.

Results After controlling for age, smoking and alcohol consumption, aerobic fitness was inversely associated with systolic (P=0.027) and diastolic (P<0.001) blood pressure, total cholesterol (P=0.009), triglycerides (P=0.001), and GGT (P<0.001), and directly associated with HDL-C (P<0.001) and HDL-C to total cholesterol ratio (P<0.001) in men. In women, inverse associations were found for diastolic blood pressure (P=0.027) and triglycerides (P<0.001), and direct association for HDL-C (P<0.001) and HDL-C to total cholesterol ratio (P<0.001). Waist-to-hip ratio was independently associated with a better risk factor profile in both sexes. Interactions were found between fitness and WHR in relation to total cholesterol (P=0.001), HDL-C to total cholesterol ratio (P=0.005), triglycerides (P=0.001), and systolic (P=0.009) and diastolic (P<0.001) blood pressure among men only.

Conclusions Our data suggest that good estimated aerobic fitness is associated with a better cardiovascular risk factor profile, regardless of the level of abdominal obesity in Finnish men and women. Men in the highest WHR third seem to have more benefit of aerobic fitness on their cardiovascular risk levels than men with lower WHR. Eur J Cardiovasc Prev Rehabil 12: 126–131 © 2005 The European Society of Cardiology

Keywords

cardiovascular diseases aerobic fitness obesity exercise test epidemiologic studies cross-sectional studies

Introduction

Major risk factors for cardiovascular diseases include smoking, elevated levels of systolic and diastolic blood pressure, and serum total cholesterol, as well as low levels of high-density lipoprotein cholesterol (HDL-C) [1]. In addition, serum triglycerides, alcohol use, and diet have been shown to predict cardiovascular diseases [2, 3]. Obesity, particularly abdominal obesity, and physical activity are considered as both independent and mediating factors in the development of cardiovascular diseases [4, 5]. Gamma-glutamyl transferase (GGT), a liver enzyme, may also be associated with cardiovascular risk [6–8], although less studied in this respect.

Previous studies have shown that cardiovascular disease risk factor profile is better among those with higher levels of cardiorespiratory fitness or physical activity [9–16]. Only few studies have reported the combined effect of cardiorespiratory fitness or physical activity and obesity on risk factor levels, and none of them have used waist-to-hip ratio (WHR) as a measure of obesity [17–20]. Importantly, these studies found that blood pressure or lipids were better among the physically most active, the fittest and the leanest, while good fitness improved the risk factors regardless of the level of obesity. More studies are needed on the associations of cardiorespiratory fitness and abdominal obesity with cardiovascular risk factor levels, particularly in large population samples of both men and women and with a wide age range.

The aim of the study was to investigate the association of estimated aerobic fitness with cardiovascular risk factor profile by using a novel non-exercise fitness test in a large population sample of men and women 25–74 years of age, and further whether the association is affected by the degree of abdominal obesity, when using WHR as a measure of obesity.

Methods

Study population and design

This cross-sectional population study was part of the National FINRISK 2002 Study, which was conducted in the spring 2002 to monitor cardiovascular risk factors among Finnish adults aged 25–74 years. A stratified random sample by sex and 10-year age group comprised 13 500 subjects and was drawn from the Population Register in six geographical regions in Finland. Physical activity sub-sample included 8159 men and women aged 25–64 years. Participation rates were 59% (n = 2415) for men and 70% (n = 2873) for women. The Ethics Committee of the National Public Health Institute approved the study protocol and the participants gave their written consent. The study protocol, sampling, anthropometric and blood pressure measurements, as well as laboratory analyses, followed the WHO MONICA Project protocol closely [21] and the later recommendations of the European Health Risk Monitoring Project [22].

Individuals with any self-reported cardiovascular disease or related condition (angina pectoris, myocardial infarction, coronary artery bypass surgery, coronary angioplasty, stroke, heart failure, or medication for hypertension in the last 7 days) were excluded from the further analyses (n = 1192). These conditions and treatments might affect the heart rate variation, which is the essential factor in estimating aerobic fitness by the non-exercise test and thus can cause bias to the fitness estimates. Participants were excluded if any variable used in the analyses was missing. The final sample included 1787 men and 2033 women.

Measurements

Participants were mailed a self-administered questionnaire to be filled in at home, before arriving at the study site. The questionnaire included questions on medical history, smoking and alcohol behaviour, as well as menopausal status and hormone replacement therapy (in women). Smoking behaviour was assessed by a frequency of smoking and dichotomized into daily smokers or non-smokers. Alcohol consumption was calculated as grams of pure alcohol per week using a recall of the preceding week.

At the study site, a trained nurse carried out the following measurements: body weight and height, waist and hip circumference, blood pressure, and fitness test. Blood samples after 4-h fasting were drawn for lipid and GGT analyses. Waist girth was measured at the midpoint between the iliac crest and the lowest rib, and hip girth at the widest part of the pelvis. The WHR was calculated and divided into sex-specific thirds. Cut-off points were 0.93 and 0.98 for men and 0.80 and 0.85 for women. Blood pressure was measured after 5 min rest from the right arm in a sitting position. The mean of the first and the second blood pressure measurements was used in the analyses. From the blood sample, total cholesterol, HDL-C, triglycerides, and GGT levels were analysed, and HDL-C to total cholesterol ratio was calculated. All analyses were carried out at the central laboratory of the National Public Health Institute. Enzymatic methods were used to determine total cholesterol and HDL-C levels (CHOD-PAP; Thermo Elektron Oy, Finland), and triglyceride levels (GPO-PAP; Thermo Elektron Oy). Levels of GGT were measured by a kinetic method (ECCLS; Thermo Elektron Oy) as recommended by the European Committee for Clinical Laboratory Standards.

A non-exercise aerobic fitness test (the Polar Fitness Test; Polar Electro Oy, Kempele, Finland) was carried out. The fitness test estimates maximal aerobic power based on subject's weight, height, age, sex, the score of self-reported physical activity, and resting heart rate measurements. The test was carried out with the Polar S810 heart rate monitor (Polar Electro Oy) at rest, in a sitting position with legs placed at a horizontal level. The different validity tests of the Polar Fitness Test [23] against maximal oxygen consumption measured during a maximal treadmill test have reached correlation coefficients of 0.80 (standard error, SE, ± 5.1 ml/kg per min), 0.95 (SE 4.6 ml/kg per min) and 0.93 (SE 5.1 ml/kg per min), suggesting that the estimated aerobic fitness is a good measure of aerobic fitness. A more detailed description and validation results on Polar Fitness Test and the use of artificial neural networks and heart rate variability in estimating aerobic fitness have been reported elsewhere [23–26]. Based on the estimated aerobic fitness, expressed as ml/kg per min, the participants were first divided into seven sex and age-specific fitness classes: very poor, poor, fair, average, good, very good, and excellent, according to international classification standards published by Shvartz and Reibold [27]. For the analyses, the classes were merged into three major fitness categories: low (very poor, poor, fair), middle (average, good) and high (very good, excellent).

Statistical methods

We used the Statistical Package for Social Sciences (SPSS, version 11.0. 1; Chicago, Illinois, USA) for all analyses. Each risk factor; systolic and diastolic blood pressure, total cholesterol, HDL-C, HDL-C to total cholesterol ratio, triglycerides, and GGT, were analysed separately using a 3 × 3 analysis of covariance. With this method we were able to investigate the effect of estimated aerobic fitness and WHR separately and independently, and also for their combined effect on each risk factor. This allowed us to interpret the independent effect of aerobic fitness on risk factors at different levels of obesity. In the analyses, we adjusted for the following potential confounders: age, smoking behaviour, and alcohol use, as well as menopausal status and hormone replacement therapy for women. Men and women were analysed separately.

Results

Major characteristics of the study population by sex are described in Table 1. The mean estimated aerobic fitness was 38.2 ml/kg per min for men and 34.9 for women. About 20% of men belonged to the ‘highly fit’ group, whereas in women, 37% were defined as ‘highly fit'.

In men, aerobic fitness had an inverse association with systolic and diastolic blood pressure, and serum total cholesterol, triglyceride and GGT levels (Table 2). A direct association was found between aerobic fitness and HDL-C, and HDL-C to total cholesterol ratio. The WHR had a direct association with systolic and diastolic blood pressure, total cholesterol, triglycerides and GGT, and inverse association with HDL-C and HDL-C to total cholesterol ratio. Statistically significant interactions between fitness and WHR were found in total cholesterol, HDL-C to total cholesterol ratio, triglycerides, and both systolic and diastolic blood pressure. Interactions related to total cholesterol and systolic blood pressure were unsystematic and difficult to interpret, but the association of aerobic fitness with HDL-C to total cholesterol ratio, triglycerides and diastolic blood pressure tended to be stronger among the abdominally obese than the lean men.

Table 1

Characteristics of the participants by sex. Values are means with standard deviations (SD) or percentages (%)

Variable	Men (n = 1787)		Women (n = 2033)
	Mean	SD	Mean	SD
Age (years)	44.0	10.9	43.6	11.2
Body mass index (kg/m²)	26.7	4.0	25.6	4.5
Waist-to-hip ratio	0.96	0.06	0.83	0.06
Estimated aerobic fitness (ml/kg/min)	38.2	7.0	34.9	9.4
Fitness category
High (%)	20.0		37.2
Middle (%)	42.0		36.4
Low (%)	37.9		26.3
Total serum cholesterol (mmol/l)	5.68	1.2	5.37	1.0
High-density lipoprotein (HDL-C) (mmol/l)	1.36	0.4	1.65	0.4
HDL-C to total cholesterol ratio	0.25	0.08	0.32	0.08
Triglycerides (mmol/l)	1.61	1.3	1.14	0.6
Systolic blood pressure (mmHg)	132.3	15.7	125.1	17.4
Diastolic blood pressure (mmHg)	80.6	11.2	75.4	10.4
Gamma glutamyl transferase (units/l)	40.3	36.3	24.4	31.9
Alcohol intake (g/week)	101.6	141.9	39.3	56.0
Non-smoker (%)	63.0		75.3
Pre-menopausal (%)			72.7
Hormone replacement therapy (%)			20.1

In women, aerobic fitness had an inverse association with diastolic blood pressure and triglycerides and a direct association with HDL-C, and HDL-C to total cholesterol ratio (Table 3). Similarly as in men, WHR had a direct association with systolic and diastolic blood pressure, total cholesterol, triglycerides and GGT, and inverse association with HDL-C and HDL-C to total cholesterol ratio. Statistically significant interactions between fitness and WHR were found in triglycerides and GGT. The association of aerobic fitness with serum triglyceride and GGT levels was stronger among obese than lean women.

When the analyses were performed among pre-menopausal women only, the results did not change markedly, except the association between fitness and diastolic blood pressure became weaker. Analyses were not performed separately for post-menopausal women (n = 488), because the number of subjects in certain fitness and obesity groups were very low.

Discussion

Our results suggest that in the healthy Finnish adult population estimated aerobic fitness is independently associated with a better risk factor profile. Particularly strong and clinically significant associations existed for HDL-C and HDL-C to total cholesterol ratio in both sexes, and for triglycerides, GGT, and blood pressure in men. The WHR had strong associations with all measured risk factors. Those who were both unfit and abdominally obese had the most unfavourable risk factor profile. In general, obese individuals tended to benefit more from aerobic fitness than lean persons in regards their risk factor levels.

Table 2

Mean risk factor levels by estimated aerobic fitness and waist-to-hip ratio (WHR) in men. Adjusted for age, smoking and alcohol use in two-way analysis of covariance

Variable	Estimated aerobic fitness
	Low n = 678	Middle n = 751	High n = 358
Total cholesterol (mmol/l)
1st WHR tertile	5.41	5.54	5.29
2nd WHR tertile	5.78	5.91	5.47
3rd WHR tertile	6.02	5.68	5.73
P-value for trend in estimated aerobic fitness	0.009
P-value for trend in WHR	<0.001
P-value for interaction in fitness and WHR	0.001
High-density lipoprotein cholesterol (HDL-C, mmol/l)
1st WHR tertile	1.46	1.48	1.52
2nd WHR tertile	1.30	1.38	1.41
3rd WHR tertile	1.19	1.25	1.35
P-value for trend in estimated aerobic fitness	<0.001
P-value for trend in WHR	<0.001
P-value for interaction in fitness and WHR	0.39
HDL-C to total cholesterol-ratio
1st WHR tertile	0.28	0.27	0.29
2nd WHR tertile	0.23	0.24	0.27
3rd WHR tertile	0.20	0.23	0.24
P-value for trend in estimated aerobic fitness	<0.001
P-value for trend in WHR	<0.001
P-value for interaction in fitness and WHR	0.005
Triglycerides (mmol/l)
1st WHR tertile	1.16	1.30	1.19
2nd WHR tertile	1.70	1.57	1.28
3rd WHR tertile	2.34	1.86	1.60
P-value for trend in estimated aerobic fitness	0.001
P-value for trend in WHR	<0.001
P-value for interaction in fitness and WHR	0.001
Systolic blood pressure (mmHg)
1st WHR tertile	129.9	130.6	126.6
2nd WHR tertile	131.7	133.9	129.5
3rd WHR tertile	137.6	133.6	134.6
P-value for trend in estimated aerobic fitness	0.027
P-value for trend in WHR	<0.001
P-value for interaction in fitness and WHR	0.009
Diastolic blood pressure (mmHg)
1st WHR tertile	76.9	77.9	75.0
2nd WHR tertile	81.6	81.3	77.8
3rd WHR tertile	86.9	82.8	79.5
P-value for trend in estimated aerobic fitness	<0.001
P-value for trend in WHR	<0.001
P-value for interaction in fitness and WHR	<0.001
Gamma-glutamyl transferase (units/l)
1st WHR tertile	32.9	28.0	27.0
2nd WHR tertile	47.2	37.7	32.0
3rd WHR tertile	58.1	45.3	42.8
P-value for trend in estimated aerobic fitness	<0.001
P-value for trend in WHR	<0.001
P-value for interaction in fitness and WHR	0.14

WHR, waist-to-hip ratio.

Aerobic fitness in most individuals is a result of long-term physical activity behaviours, and perhaps associates more strongly with risk factors than self-reports on physical activity [9, 28, 29]. The Polar Fitness Test is a new method of estimating aerobic fitness. It is simple and safe to carry out, and seems to be useful in population studies. The test uses self-reported leisure time physical activity measurement as one part of the test, which naturally gives a possibility for recall bias. Nevertheless, it should be emphasized that in a study of this magnitude, any other field or laboratory measurement of aerobic fitness is not feasible in the 10-week time frame of our study protocol.

Table 3

Mean risk factor levels by estimated aerobic fitness and waist-to-hip ratio in women. Adjusted for age, smoking, alcohol use, menopausal status, and hormone replacement therapy in two-way analysis of covariance

Variable	Estimated aerobic fitness
	Low n = 535	Middle n = 741	High n = 757
Total cholesterol (mmol/l)
1st WHR tertile	5.19	5.33	5.26
2nd WHR tertile	5.25	5.34	5.31
3rd WHR tertile	5.49	5.49	5.61
P-value for trend in estimated aerobic fitness	0.26
P-value for trend in WHR	<0.001
P-value for interaction in fitness and WHR	0.58
High-density lipoprotein cholesterol (HDL-C, mmol/l)
1st WHR tertile	1.70	1.80	1.86
2nd WHR tertile	1.59	1.69	1.70
3rd WHR tertile	1.42	1.54	1.52
P-value for trend in estimated aerobic fitness	<0.001
P-value for trend in WHR	<0.001
P-value for interaction in fitness and WHR	0.59
HDL-C to total cholesterol- ratio
1st WHR tertile	0.33	0.35	0.36
2nd WHR tertile	0.31	0.32	0.33
3rd WHR tertile	0.27	0.29	0.28
P-value for trend in estimated aerobic fitness	<0.001
P-value for trend in WHR	<0.001
P-value for interaction in fitness and WHR	0.32
Triglycerides (mmol/l)
1st WHR tertile	0.99	0.95	0.94
2nd WHR tertile	1.15	1.08	1.03
3rd WHR tertile	1.50	1.32	1.36
P-value for trend in estimated aerobic fitness	<0.001
P-value for trend in WHR	<0.001
P-value for interaction in fitness and WHR	0.37
Systolic blood pressure (mmHg)
1st WHR tertile	122.7	122.2 1	22.9
2nd WHR tertile	125.7	125.3 1	23.3
3rd WHR tertile	129.8	127.1 1	27.4
P-value for trend in estimated aerobic fitness	0.23
P-value for trend in WHR	<0.001
P-value for interaction in fitness and WHR	0.45
Diastolic blood pressure (mmHg)
1st WHR tertile	74.1	73.1	73.6
2nd WHR tertile	75.5	74.8	74.0
3rd WHR tertile	79.3	76.7	78.1
P-value for trend in estimated aerobic fitness	0.027
P-value for trend in WHR	<0.001
P-value for interaction in fitness and WHR	0.35
Gamma-glutamyl transferase (units/l)
1st WHR tertile	19.6	20.8	21.9
2nd WHR tertile	23.4	24.5	19.7
3rd WHR tertile	33.6	25.0	30.1
P-value for trend in estimated aerobic fitness	0.48
P-value for trend in WHR	<0.001
P-value for interaction in fitness and WHR	0.035

WHR, waist-to-hip ratio.

Distribution into fitness categories was different in men and women. While 20% of men belonged to the ‘high fitness’ group, a much larger proportion, 37%, of women were classified as highly fit. Most of these highly fit women were older, from the age group 55–64 years of age. Potential explanations for this high fitness group could be that the healthy older women in this population sample were actually truly fit, or that the Polar Fitness Test overestimated the aerobic fitness of these older women [24].

Some of the associations between estimated aerobic fitness and risk factors were stronger in men than women, and no interactions were found among women. This could result from different distributions of fitness categories in men and women, from menopausal status, or from hormone replacement therapy in women. However, the results remained unchanged in pre-menopausal women suggesting that hormonal status is an unlikely explanation. It is also noteworthy that WHR might function differently as a measure of obesity for men and women, mainly due to sex-specific propensities to gain extra fat at different locations of the body.

Our results are in concordance with many earlier studies on aerobic fitness and cardiovascular risk factors [9, 19, 30–32]. Most population studies use self-reported physical activity instead of aerobic fitness and thus their results are not fully comparable to ours. Nevertheless, our findings agree with many larger studies that self-reported physical activity is associated with a favourable blood lipid profile and lower blood pressure [14, 15, 20, 33]. One Finnish prospective study found a similar association between physical activity and hypertension [34]. Unexpectedly, we found no association between estimated aerobic fitness and systolic blood pressure in women. Concordant findings have been reported also from other studies [35, 36]. The plausible causes for these unexpected results could be related to the validity of estimated fitness level and their cut-off points, as the mean systolic blood pressure levels declined by fitness categories in women, but the decline was not dose dependent. Also, WHR is more strongly associated with blood pressure than aerobic fitness in women.

We found significant associations of aerobic fitness and WHR with GGT in men, but not in women. Similar findings have been reported previously with occupational physical activity and GGT [16]. Population studies on the association of fitness or physical activity with GGT are sparse, have included mainly men, and have not studied the inter-relations of physical activity and obesity [7, 37]. Gamma-glutamyl transferase is used both as a marker of obesity and long-term alcohol use. We additionally found that the leaner and fitter the men were, the lower was their self-reported alcohol consumption. In women, the more obese reported heavier alcohol use, but no difference in the alcohol use was found across the aerobic fitness categories.

Inter-relation between aerobic fitness and obesity has not been studied much in relation to cardiovascular risk profile. Our results agree with previous data, showing that overweight persons with high level of aerobic fitness also have a better blood lipid profile [18] and lower hypertension prevalence [17, 19]. A recent large cross-sectional study [20] suggested that obese individuals improved their risk profile more with physical activity than leaner persons. Similar results on the interaction were also found for coronary heart disease [38], all-cause and cardiovascular mortality [39, 40]. Contradictory, one Finnish study found no interaction between fitness and obesity with respect to all-cause and cardiovascular mortality [28].

These results provide new information on the associations of estimated aerobic fitness, obesity, and cardiovascular risk factors among a population-based random sample of relatively healthy individuals. The cross-sectional design of our study, however, limits the interpretation of the associations of the aerobic fitness with the risk factors to some extent. It is possible that healthier people with a better risk factor profile are more physically active, but because we excluded 1192 individuals with cardiovascular disease or related condition and hypertension medication, the effect of selection bias is most probably minor.

In conclusion, two important public health messages can be conveyed from this study. First, having both a good aerobic fitness and a low WHR seems to be beneficial to cardiovascular risk profile. Second, aerobic fitness may be more beneficial to abdominally obese than normal weight persons. More pronounced research on the association between aerobic fitness, obesity, and risk factor profile among women should be implemented in the future.

Footnotes

Acknowledgements

Some parts of the manuscript were presented at the Annual Meeting of the European College of Sport Science, July 2004, Clermont-Ferrand, France. This study was supported by a grant from the Finnish Ministry of Education and from the Polar Electro Oy. In addition, the corresponding author, K. B. received personal salary from the Doctoral Programs in Public Health, University of Helsinki. R. L. works as a Director, Exercise Science, at Polar Electro Oy.

References

1 Conroy

Pyorala

Fitzgerald

Sans

Menotti

De Backer

, et al. Estimation of ten-year risk of fatal cardiovascular disease in Europe: the SCORE project. Eur Heart J 2003; 24: 987–1003.

2 LaRosa

Hunninghake

Bush

Criqui

Getz

Gotto

Jr. , et al. The cholesterol facts. A summary of the evidence relating dietary fats, serum cholesterol, and coronary heart disease. A joint statement by the American Heart Association and the National Heart, Lung, and Blood Institute. The Task Force on Cholesterol Issues, American Heart Association. Circulation 1990; 81: 1721–1733.

3 Kannel

D'Agostino

. Update of old coronary risk factors. Cardiovasc Risk Factors 1996; 6: 244–253.

4 Hubert

Feinleib

McNamara

Castelli

. Obesity as an independent risk factor for cardiovascular disease: a 26-year follow-up of participants in the Framingham Heart Study. Circulation 1983; 67: 968–977.

5 Lakka

Venalainen

Rauramaa

Salonen

Tuomilehto

Salonen

. Relation of leisure-time physical activity and cardiorespiratory fitness to the risk of acute myocardial infarction. N Engl J Med 1994; 330: 1549–1554.

6 Jousilahti

Rastenyte

Tuomilehto

. Serum gamma-glutamyl transferase, self-reported alcohol drinking, and the risk of stroke. Stroke 2000; 31: 1851–1855.

7 Wannamethee

Ebrahim

Shaper

. Gamma-glutamyl transferase: determinants and association with mortality from ischemic heart disease and all causes. Am J Epidemiol 1995; 142: 699–708.

8 Jousilahti

Vartiainen

Alho

Poikolainen

Sillanaukee

. Opposite associations of carbohydrate-deficient transferrin and gamma-glutamyl transferase with prevalent coronary heart disease. Arch Intern Med 2002; 162: 817–821.

9 McMurray

Ainsworth

Harrell

Griggs

Williams

. Is physical activity or aerobic power more influential on reducing cardiovascular disease risk factors? Med Sci Sports Exerc 1998; 30: 1521–1529.

10.

10 Fagard

. Exercise characteristics and the blood pressure response to dynamic physical training. Med Sci Sports Exerc 2001; 33: S484–S492.

11.

11 Leon

Sanchez

. Response of blood lipids to exercise training alone or combined with dietary intervention. Med Sci Sports Exerc 2001; 33: S502–S515.

12.

12 Kokkinos

Holland

Narayan

Colleran

Dotson

Papademetriou

. Miles run per week and high-density lipoprotein cholesterol levels in healthy, middle-aged men. A dose-response relationship. Arch Intern Med, 1995; 155: 415–420.

13.

13 Williams

. Relationships of heart disease risk factors to exercise quantity and intensity. Arch Intern Med 1998; 158: 237–245.

14.

14 Jakes

Day

Khaw

Luben

Oakes

Welch

, et al. Television viewing and low participation in vigorous recreation are independently associated with obesity and markers of cardiovascular disease risk: EPIC-Norfolk population-based study. Eur J Clin Nutr 2003; 57: 1089–1096.

15.

15 Bassett

Jr Fitzhugh

Crespo

King

McLaughlin

. Physical activity and ethnic differences in hypertension prevalence in the United States. Prev Med 2002; 34: 179–186.

16.

16 Nilssen

Forde

. Seven-year longitudinal population study of change in gamma-glutamyl transferase: the Tromso Study. Am J Epidemiol 1994; 139: 787–792.

17.

17 Fagard

. Physical activity in the prevention and treatment of hypertension in the obese. Med Sci Sports Exerc 1999; 31: S624–S630.

18.

18 Wei

Macera

Hornung

Blair

. Changes in lipids associated with change in regular exercise in free-living men. J Clin Epidemiol 1997; 50: 1137–1142.

19.

19 Carnethon

Gidding

Nehgme

Sidney

Jacobs

Jr Liu

. Cardiorespiratory fitness in young adulthood and the development of cardiovascular disease risk factors. JAMA 2003; 290: 3092–3100.

20.

20 Fransson

Alfredsson

de Faire

Knutsson

Westerholm

. Leisure time, occupational and household physical activity, and risk factors for cardiovascular disease in working men and women: the WOLF study. Scand J Public Health 2003; 31: 324–333.

21.

21 The World Health Organization MONICA Project

(monitoring trends and determinants in cardiovascular disease): a major international collaboration. WHO MONICA Project Principal Investigators.

J Clin Epidemiol 1988; 41: 105–114.

22.

22 Tolonen

Kuulasmaa

Laatikainen

Wolf

. Project EHRM, European Health Risk Monitoring Project. Recommendation for indicators, international collaboration, protocol and manual of operations for chronic disease risk factor surveys. www-publications from the EHRM Project. www-publications from the EHRM Project, URL:http://www.ktl.fi/ehrm/index.html. 2002, Helsinki, Finland: National Public Health Institute.

23.

23 Kinnunen

Vainamo

Hautala

Makikallio

Tulppo

Nissila

. Artificial neural network in predicting maximal aerobic power. Med Sci Sports Exerc 2000; 32: S308, abstract 1535.

24.

24 Borodulin

Lakka

Laatikainen

Laukkanen

Kinnunen

Jousilahti

. Associations of self-rated fitness and different types of leisure time physical activity with predicted aerobic fitness in 5979 Finnish adults. JPAH 2004; 1: 142–153.

25.

25 Vainamo

Nissila

Makikallio

Tulppo

Roning

. Artificial neural networks for aerobic fitness approximation. In ICNN, International Neural Network. 1996, Washington, DC, USA.

26.

26 Laukkanen

. Non-exercise test for aerobic fitness assessment. In The 6th Annual Congress of the ECSS. 2003, Salzburg, Austria.

27.

27 Shvartz

Reibold

, Aerobic fitness norms for males and females aged 6 to 75 years: a review. Aviat Space Environ Med 1990; 61: 3–11.

28.

28 Haapanen-Niemi

Miilunpalo

Pasanen

Vuori

Oja

Malmberg

. Body mass index, physical inactivity and low level of physical fitness as determinants of all-cause and cardiovascular disease mortality—16 y follow-up of middle-aged and elderly men and women. Int J Obes Relat Metab Disord 2000; 24: 1465–1474.

29.

29 Blair

Cheng

Holder

. Is physical activity or physical fitness more important in defining health benefits? Med Sci Sports Exerc 2001; 33: S379–S399.

30.

30 Blair

Goodyear

Gibbons

Cooper

. Physical fitness and incidence of hypertension in healthy normotensive men and women. JAMA 1984; 252: 487–490.

31.

31 Haddock

Hopp

Mason

Blix

Blair

. Cardiorespiratory fitness and cardiovascular disease risk factors in postmenopausal women. Med Sci Sports Exerc 1998; 30: 893–898.

32.

32 Murphy

Nevill

Neville

Biddle

Hardman

. Accumulating brisk walking for fitness, cardiovascular risk, and psychological health. Med Sci Sports Exerc 2002; 34: 1468–1474.

33.

33 Thune

Njolstad

Lochen

Forde

. Physical activity improves the metabolic risk profiles in men and women: the Tromso Study. Arch Intern Med 1998; 158: 1633–1640.

34.

34 Hu

Barengo

Tuomilehto

Lakka

Nissinen

Jousilahti

. Relationship of physical activity and body mass index to the risk of hypertension: a prospective study in Finland. Hypertension 2004; 43: 25–30.

35.

35 Greendale

Bodin-Dunn

Ingles

Haile

Barrett-Connor

. Leisure, home, and occupational physical activity and cardiovascular risk factors in postmenopausal women. The Postmenopausal Estrogens/Progestins Intervention (PEPI) Study. Arch Intern Med 1996; 156: 418–424.

36.

36 Wilmore

Stanforth

Gagnon

Rice

Mandel

Leon

, et al. Heart rate and blood pressure changes with endurance training: the HERITAGE Family Study. Med Sci Sports Exerc 2001; 33: 107–116.

37.

37 Robinson

Whitehead

. Effect of body mass and other factors on serum liver enzyme levels in men attending for well population screening. Ann Clin Biochem 1989; 26: 393–400.

38.

38 Lakka

Lakka

Tuomilehto

Salonen

. Abdominal obesity is associated with increased risk of acute coronary events in men. Eur Heart J 2002; 23: 706–713.

39.

39 Lee

Blair

Jackson

. Cardiorespiratory fitness, body composition, and all-cause and cardiovascular disease mortality in men. Am J Clin Nutr 1999; 69: 373–380.

40.

40 Wei

Kampert

Barlow

Nichaman

Gibbons

Paffenbarger

Jr , et al. Relationship between low cardiorespiratory fitness and mortality in normal-weight, overweight, and obese men. JAMA 1999; 282: 1547–1553.