IJE Advance Access originally published online on September 16, 2005
International Journal of Epidemiology 2005 34(6):1387-1394; doi:10.1093/ije/dyi193
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Article |
Cardiorespiratory fitness, all-cause mortality, and risk of cardiovascular disease in Trinidadian men—the St James survey
1 Medical Research Council Cardiovascular Group, Wolfson Institute of Preventive Medicine, Barts and The London Queen Mary's School of Medicine and Dentistry, Charterhouse Square, London, UK
2 Centre for Cardiovascular Genetics, Royal Free and University College London Medical School, London, UK
3 Division of Diabetes Translation, National Center for Chronic Disease Prevention and Health Promotion, Centers for Disease Control and Prevention, Atlanta, GA 30341, USA
* Corresponding author. Medical Research Council Cardiovascular Group, Department of Environmental and Preventive Medicine, Wolfson Institute of Preventive Medicine, Charterhouse Square, London, EC1M 6BQ, UK. E-mail: g.miller{at}qmul.ac.uk
 |
Abstract |
|---|
 Top Abstract MethodsResultsDiscussionReferences |
|---|
Background This study examined whether cardiorespiratory fitness is a risk factor for cardiovascular disease, myocardial infarction, and all-cause mortality in a low- to middle-income Trinidadian community of African, South Asian Indian, and European origin. Those of Indian descent have a distinctively high rate of myocardial infarction.
Methods The St James Study is a prospective total community survey located in Port-of-Spain, Trinidad, West Indies. A random sample of 626 men aged 35–69 years, without angina of effort, previous myocardial infarction, partial or complete atrio-ventricular conduction defect, complete heart block, or exercise-induced asthma, was used for the assessment of cardiorespiratory fitness by cycle ergometry. Surveillance for morbidity and mortality was maintained for an average of 7.3 years.
Results When the subjects were grouped into those with an age- and fat-free mass-adjusted peak oxygen uptake above and below the mean of 60.4 mmol/min (1.34 l/min), the hazard ratios (below/above) (95% confidence interval) for all-cause mortality, cardiovascular disease incidence, and incidence of myocardial infarction, after allowance for conventional cardiovascular risk factors, were 2.08 (1.23–3.52), 2.13 (1.22–3.69), and 2.36 (0.84–6.67), respectively. For those unable to achieve a level of work requiring an oxygen uptake of 67 mmol/min (1.5 l/min) during progressive exercise, the respective hazard ratios were 3.49 (1.57–7.76), 2.29 (1.21–4.33), and 5.45 (1.22–24.34). Indian ethnicity remained a predictor of myocardial infarction after allowance for cardiorespiratory performance.
Conclusion Low cardiorespiratory fitness is a risk factor for cardiovascular disease morbidity and mortality in the low- to middle-income developing community of Trinidad.
Keywords Cardiorespiratory fitness, cardiovascular disease, myocardial infarction, mortality, cohort study, men, developing community
Accepted 22 August 2005
Studies in high-income industrialized societies have shown that a low level of cardiorespiratory fitness increases the risk of mortality from myocardial infarction1,2 and cardiovascular diseases (CVDs).3 Low cardiorespiratory fitness is also associated with type II diabetes mellitus4–7 and hypertension,8,9 both being major cardiovascular risk factors. These diseases are increasingly prevalent in developing low- and middle-income countries,10 among which the Caribbean territory of Trinidad and Tobago is a prime example. In this community, people whose ancestry stems from the Indian subcontinent (hereafter referred to as ‘Indians’) have high rates of diabetes and coronary heart disease (CHD), and hypertension is especially prevalent in adults of African descent.11–13 To date, however, low cardiorespiratory fitness as a risk factor for CVD has received far less attention in low- to middle-income populations than in high-income industrialized communities.10 On this account a prospective survey was conducted in Port-of-Spain, Trinidad, to explore the relations of cardiorespiratory fitness with ethnicity, blood pressure, diabetes, all-cause mortality, and the incidence rates of CVD and myocardial infarction.
|
Methods |
|---|
|
TopAbstractMethodsResultsDiscussionReferences |
|---|
Subjects
Census of the St James urban community, described previously,11–13 identified 1390 men aged 35–69 years eligible for recruitment, and out of them 1246 (89.6%) responded. Recruitment took place between 1977 and 1981, respondents giving consent after receiving information from the Ministry of Health, Trinidad and Tobago, which granted the ethical approval. A random sample of 839 (60%) of the census population was identified for test exercise, and 721 (86%) accepted. Out of them 77 men (10.7%) were not exercised owing to definite angina of effort, a history of myocardial infarction, complete or partial atrio-ventricular block, complete heart block, or exercise-induced asthma. In 18 subjects (2.5%) the exercise test was technically inadequate, leaving 626 eligible respondents with satisfactory tests.
Baseline characteristics
A questionnaire was administered for medical history, including angina pectoris and pain of possible myocardial infarction,14 alcohol consumption in the previous week (units), and smoking habit (current, ex-, or non-smoker). Breathlessness on exertion was graded using a standardized questionnaire.15 Ethnicity was determined from grandparental origins. Those with fewer than three grandparents of common descent (mainly Afro-European) were grouped with very small numbers of Lebanese and Syrian as ‘Other Trinidadian’.
Fasting blood glucose level was determined by the Nelson–Somogyi method. Subjects with a concentration of 6.17 to 7.72 mmol/l (111–139 mg/dl) were given a 50 g oral glucose tolerance test and diabetes defined as closely as possible13 to World Health Organization (WHO) criteria, with allowance for the 50 g load and the analysis of whole venous blood.16 Measurement of fasting serum lipids and high density lipoprotein (HDL) cholesterol concentration is described previously.17 Blood pressure was recorded twice on each of the two occasions after the subject was seated for 5 min, using a random-zero sphygmomanometer (Gelman-Hawksley). Average systolic pressure (SBP) and average diastolic V pressure were used for the statistical analysis. Cardiac frequency was recorded at rest. Body mass index was calculated as weight (kg)/height (m)2. Triceps, biceps, subscapula, and suprailiac skinfolds were measured with Harpenden callipers and summed to estimate the fat-free mass.18 A resting ECG was recorded with chest leads limited to V5 unless CHD was suspected or a QS complex found in V5, in which case a full standard 12-lead ECG was recorded. Records were coded and left ventricular hypertrophy diagnosed according to the Minnesota criteria.19
Assessment of cardiorespiratory fitness
Exercise tests in the 644 eligible respondents were medically supervised with resuscitation equipment to hand. Illness within the previous 10 days led to deferral. Progressive exercise was performed on a cycle ergometer (Lode's Instrumentan BV, Groningen, The Netherlands) in a laboratory maintained at 23–24°C. The power output was raised by 15 W/min and the subjects exercised to a cardiac frequency equal to 85% of the predicted maximum20 unless stopped beforehand owing to dyspnoea or fatigue (22%), angina pectoris (0%), SBP 
250 mmHg (12%), or an abnormal exercise ECG (4%) (ST junction depression 1 mm with a horizontal or downward sloping ST segment, or ectopic beats 10% of recorded cycles).19
A Cambridge model 3030 monitor recorded cardiac frequency and electrocardiographic lead V5 throughout the exercise and recovery, while SBP was measured manually at 2 min intervals. Respiratory gas exchange was measured by using the mixing-chamber method, in which mixed expired air was fed into CO2 and O2 analysers calibrated with special gas mixtures. Inspired minute ventilation was displayed on a ventilation meter (PK Morgan Ltd, Chatham, UK). Cardiac frequency (fC), SBP, and ventilation (VE, l/min, body temperature and pressure, saturated) were plotted against oxygen consumption, inspected visually by a single observer, and the regression analysis was used to derive values at an oxygen consumption of 45 mmol/min (1.0 l/min) by interpolation (SBP45, fC45, VE45) as indices of standardized sub-maximal exercise performance, in accordance with the International Labour Organization recommendations.21,22 A subject weighing 71 kg requires an oxygen consumption of 
45 mmol/min when walking at 4.7 km (3 miles)/hr on a level ground.23
Peak oxygen uptake (mmol/min), recorded at end-exercise, was associated inversely with age and directly with body muscle expressed as fat-free mass (see Results section). On this account, age- and fat-free mass-adjusted peak oxygen uptake was used as an index of cardiorespiratory performance. The utility of a simplified index of cardiorespiratory performance, not necessarily requiring exercise to peak performance nor measurement of oxygen consumption, was also explored, based on the knowledge that a subject weighing 71 kg requires an oxygen uptake of 67 mmol/min (1.5 l/min) when walking at 4.7 km/hr up a 5% slope.23 Those failing to achieve this level of oxygen uptake during progressive cycling exercise (i.e. it exceeded their peak oxygen uptake) were classified as ‘Unfit’, whereas those achieving or exceeding this level were classified as ‘Fit’.
Ascertainment of mortality and incident non-fatal events
Follow-up was maintained up to 31st December 1986. All subjects were contacted at least once each year, and respondents were encouraged to contact the staff for their medical problems. Those who emigrated were mailed a questionnaire annually and examined on return visits when possible. Subjects with symptoms suggestive of CVD (defined below) attended for diagnostic evaluation and (if needed) referral for care. Morbidity and mortality records were compiled in this manner together with regular inspection of hospital records, registers of death, and obituaries (cancers were also documented in the same manner). Non-fatal events and underlying cause of death were arrived at after the review of all documentation (including hospital notes, necropsy records, laboratory reports, and history from the subject and family) by two physicians, with a third as arbiter when required. Myocardial infarction was diagnosed on WHO criteria24 and terminating events classified according to the International Classification of Diseases, 9th Revision (CVD, codes 390–448.9; myocardial infarction, code 410).25
Statistical analysis
Statistical analysis was performed with the STATA Version 8.2 software (STATA Corporation, TX). Variables were log-transformed wherever appropriate. Group differences were explored using 2-sample t-test or Mann–Whitney U-test for continuous variables, and using chi-squared or Fisher's exact test for categorical data. All tests of statistical significance were two-sided. Differences were considered statistically significant when the probability of occurrence by chance alone was 
5%.
To adjust the peak oxygen consumption for body size, separate regression analyses were performed on weight and fat-free mass. The association was stronger with fat-free mass and therefore only fat-free mass-adjusted values are reported. Age was also a determinant of peak oxygen uptake (R = –0.33, P < 0.0001). The multiple regression was
 |
To assess risk in men free of CHD at the baseline, 48 (6.7%) of those with technically satisfactory exercise tests were excluded from the further statistical analysis because of a major electrocardiographic Q/QS complex and/or a history of chest pain considered non-cardiac at recruitment (subjects with an indefinite history of angina of effort are reported to have an increased cardiovascular mortality26). Hazard ratios [95% confidence intervals (95% CIs)] for all-cause mortality, CVD, and myocardial infarction in the remaining 578 subjects were obtained from Cox proportional hazards models with adjustments made by inclusion of the following factors as covariates: peak oxygen consumption (age- and fat-free mass-adjusted) and/or Unfit/Fit, age, ethnicity, diabetes, SBP, total cholesterol, HDL cholesterol, smoking habit, body mass index, fat-free mass, and an abnormal exercise ECG. Collinearity diagnostics confirmed that in the regression analyses none of the independent variables was strongly correlated with any other. The lowest tolerance (1 – R2) observed was 0.4 and the highest variance inflation factor was 2.4 (values of <0.1 and >10, respectively, would indicate significant problems with collinearity in the models). The final models were obtained by stepwise elimination of variables for which the 95% CI about the associated hazard ratio spanned unity. Forward and backward elimination procedures were performed to confirm the consistency.
|
Results |
|---|
|
TopAbstractMethodsResultsDiscussionReferences |
|---|
The 721 men volunteering for test exercise and the remaining 525 respondents in the parent study were of similar mean age (52.9 and 52.1 years, respectively; P = 0.13) and ethnicity (African: 37.1 and 39.2%; Indians: 29.0 and 27.8%; Europeans: 10.5 and 6.7%, respectively). Height, weight, smoking habit, alcohol consumption, blood lipids, blood glucose, and prevalence of diabetes (11.8 and 11.0%, respectively; P = 0.62) were also similar. The exercise group had a slightly greater sum of four skinfolds [geometric mean (SD), 39.2 (16.6) and 37.3 (15.7) mm, respectively; P = 0.05], lower mean (SD) SBP [134.7 (21.0) and 140.1 (21.8) mmHg, respectively; P < 0.0001], and diastolic V blood pressure [83.6 (12.0) and 86.7 (13.5) mmHg, respectively; P < 0.0001] than the remainder.
Cardiorespiratory fitness and associated characteristics
No medical emergency or chest pain occurred in the 644 men tested. The unadjusted mean (SD) peak oxygen uptake, available for 626 of those tested, was 60.5 (18.2) mmol/min (1.34 l/min). Subjects were grouped into those with an age- and fat-free mass-adjusted peak oxygen uptake above and below the adjusted mean of 60.4 mmol/min [average values 74.1 mmol/min (1.65 l/min) and 49.2 mmol/min (1.09 l/min), respectively]. Table 1 shows that the two groups differed in mean age by only 1.5 years, the below-average group being the younger. Those of Afro-European or minority descent were over-represented, and Indians under-represented, in the above-average group. Breathlessness on walking with others of similar age on a level ground was infrequent, but was reported more so by those of below-average performance. Smoking habit, drinking habit, body size, and adiposity did not differ significantly between the groups. Those of above-average performance had a significantly lower fasting blood glucose, but the prevalence of diabetes, mean levels of blood lipoprotein lipids, and resting blood pressure did not differ significantly between the two groups. ‘Fitness’ was almost three times more common in the above-average than the below-average group.
|
Only 1% of both the groups were taking a beta-adrenoceptor blocking drug. Table 2 presents the resting electrocardiographic findings and indices of sub-maximal exercise performance. Minor ST segment depression and negative T waves were significantly more common in the below-average group, but major Q/QS complexes, left ventricular hypertrophy, and minor conduction defects were of similar frequency in both the groups. Median resting cardiac frequency was 4/min lower in the above-average group (P = 0.06). Mean fC45 was lower by 17.1/min in the above-average group (P < 0.0001). The mean peak fC achieved during exercise was the same in both groups. The mean VE45 was 3.7 l/min lower and mean SBP45 13.5 mmHg lower in the above-average group (P < 0.0001 in both cases).
|
Sixteen (5.1%) of the below-average group and 9 (2.9%) of the above-average group were stopped during exercise owing to the development of electrocardiographic abnormality (P = 0.15).
Morbidity and mortality
Average (range) follow-up was 7.2 (0.4–9.2) and 7.4 (0.7–8.7) years in those with an adjusted peak oxygen uptake above and below the population mean, respectively. There were 83 deaths, 77 CVD events and 23 first myocardial infarctions during surveillance in the 644 men exercised, giving rates (per 1000 person-years) of 18.5 for all-cause mortality, 17.4 for CVD, and 5.1 for myocardial infarction. Table 3 presents the age-adjusted incidence rates in the two groups, showing uniformly poorer outcomes in those of below-average performance, with highly statistically significant hazard ratios of 2.42 for all-cause mortality and fatal plus non-fatal CVD, and 2.69, of borderline significance, for fatal plus non-fatal myocardial infarction. Similar hazard ratios were found when fatal and non-fatal events were considered separately, but because of smaller numbers in the groups only that for non-fatal CVD was statistically significant.
|
The total person-years of follow-up in the 578 men entered into the analysis of morbidity and mortality was 4008. Table 4 summarizes the models from the Cox proportional hazards analysis with age-and fat-free mass-adjusted peak oxygen uptake as the measure of cardiorespiratory fitness. Adjusted peak oxygen uptake, age, diabetes, and an abnormal exercise ECG necessitating premature termination of exercise were highly significant independent predictors of all-cause mortality. Independent predictors of CVD were adjusted peak oxygen uptake, SBP, and HDL cholesterol. The association of adjusted peak oxygen uptake with the incidence of myocardial infarction was not statistically significant (P = 0.10) when the independent effects of Indian ethnicity and fat-free mass were taken into account.
|
The association of fat-free mass in the predictive model for myocardial infarction, given in Table 4, had not arisen because of the prior adjustment of peak oxygen uptake for fat-free mass. Fat-free mass was not a significant predictor of myocardial infarction when included alone in the model, the hazard ratio being 1.42 (0.89–2.25). It emerged as a significant predictor only when ethnicity was also taken into account [hazard ratio 1.99 (1.18–3.35); P = 0.01]. The hazard ratio increased still further when unadjusted peak oxygen uptake was also added to the model [2.08 (1.25–3.44); P = 0.005].
Age was not a predictor of myocardial infarction in this study. When used alone in the regression model the hazard ratio for a 5-year increase in age was 0.98 (95% CI, 0.77–1.26). In contrast, age was a predictor of CVD, the corresponding hazard ratio being 1.19 (1.02–1.38); P = 0.02. However, age was not associated with CVD after adjustment for SBP and HDL cholesterol [hazard ratio 1.14 (0.97–1.33)], whereas both SBP and HDL cholesterol were significant predictors of CVD after adjustment for age (P < 0.01 in both cases). With further adjustment by inclusion of unadjusted peak oxygen uptake in the model, the hazard ratio for age was 1.11 (0.94–1.30). Thus, the employment of age- and fat-free mass-adjusted peak oxygen uptake in the model presented in Table 4 was not a reason for the lack of predictive power of age for myocardial infarction and CVD.
When both adjusted peak oxygen uptake and ‘Unfitness/Fitness’ were included in the Cox proportional hazards analysis, the former was no longer a significant independent predictor of any end point, while ‘Unfitness’ was a significant independent predictor of all-cause mortality (P = 0.02) and an independent predictor of borderline significance (P = 0.06) of CVD and myocardial infarction. The analysis was therefore repeated with ‘Unfitness/Fitness’ as the sole index of cardiorespiratory performance (Table 5). ‘Unfitness’ was found to be a significant independent predictor of all the three end points. The independent associations of other covariates entered into the model were essentially unchanged from those shown in Table 4.
|
Cancer incidence was not associated with peak oxygen uptake (data not shown).
|
Discussion |
|---|
|
TopAbstractMethodsResultsDiscussionReferences |
|---|
The response rate was very satisfactory at 86%. However, 77 of the 721 respondents were ineligible for exercise testing on predefined criteria, and the test was technically unsatisfactory in 18, leaving 626 with estimates of exercise performance. A further 48 had to be excluded from the analysis of cardiorespiratory fitness as a predictor of morbidity and mortality, owing to evidence for possible CHD at the baseline.
The 721 respondents for exercise testing were similar to the remaining respondents in the parent survey, apart from a lower blood pressure and a slightly larger skinfold thickness. These small differences will not have affected the reliability of the estimate of age- and fat-free mass-adjusted peak oxygen uptake in the population, since none of the characteristics was related to this index of cardiorespiratory performance. Only 35% of the survey population appeared capable of sustained activities demanding an oxygen uptake of 67 mmol/min, such as walking at 4.7 km/hr up a 5% slope.23 This result, together with the low mean peak oxygen uptake and high fC45 in comparison with some other populations,2,21 indicated that middle-aged men in this community were generally sedentary.
Motivation, body muscle, age, and diseases at the baseline represent potential drawbacks associated with the use of peak oxygen uptake to explore relations of cardiorespiratory fitness with morbidity and mortality. Motivation did not appear to differ between those of above-average and below-average exercise performance, because the peak cardiac frequency achieved was the same in both the groups. Some highly motivated subjects in both groups might have achieved a higher peak oxygen uptake had they not been stopped on reaching their target cardiac frequency. Mean peak oxygen uptake may therefore have been slightly underestimated. However, this possibility was of little practical importance, because work at >85% predicted maximum cardiac frequency is sustainable only for a brief period, even in the highly motivated.
The higher resting cardiac frequency in those with a below-average exercise performance agreed with previous studies.27 The significantly higher fC45 and VE45 in those with a below-average age- and fat-free mass-adjusted peak oxygen uptake confirmed their relatively low level of cardiorespiratory fitness. When standardized for body muscle and energy expenditure, cardiac frequency during submaximal exercise (which was unrelated to age) provides an index of cardiorespiratory fitness and an indirect estimate of habitual activity.21 With respect to ventilation, the level of activity at which anaerobic metabolism occurs is related to the degree of physical training. Those who are unfit accumulate more lactic acid in the blood and hence experience a stronger drive to ventilation from acidaemia during exercise.
Little evidence was found for a higher rate of cardiac disorders at the baseline in those with a below-average peak oxygen uptake that might have accounted for their poorer exercise performance and relatively high cardiovascular morbidity and mortality. The only significant differences on the ECG were increases in minor ST segment depression and negative T waves at rest in those of below-average exercise performance, together with a slightly greater proportion in this group who developed electrocardiographic abnormalities necessitating termination of exercise (not statistically significant). The latter finding was a significant predictor of all-cause mortality, but not CVD or myocardial infarction, and the association of a low peak oxygen uptake with increased mortality persisted when the exercise ECG was taken into account. Breathlessness on exertion occurred with similar frequency in both groups. Age-adjusted resting blood pressure did not differ between groups, though SBP45 was higher in those of below-average cardiorespiratory performance, possibly owing to work requiring an oxygen uptake of 45 mmol/min being closer to the maximum tolerable in this group. Angina and myocardial infarction were contra-indications to the test exercise, and no subject experienced chest pain during the procedure. Men with a history of chest pain, even when considered non-cardiac in origin, were excluded from the estimates of multiple risk factor-adjusted hazard ratios because they are reported to be at increased risk of cardiovascular death.26
In previous analyses, Indian ethnicity was associated with an increased risk of CHD in this community, whereas men of mixed ethnicity (mainly Afro-European) were at significantly reduced risk.12 These associations were replicated in the sample of men exercised (data not shown), but unlike Indian ethnicity, mixed ethnicity was not an independent predictor of myocardial infarction when cardiorespiratory fitness was taken into account. Thus, while cardiorespiratory fitness may have contributed to the reduced risk in Afro-European men, it did not explain the high risk in Indian men.
The positive association of fat-free mass with risk of myocardial infarction, after controlling for other cardiovascular risk factors, including cardiorespiratory performance, was unexpected and unexplained.
The findings in Trinidad agreed with other studies in which cardiorespiratory fitness was predictive of all-cause and cardiovascular mortality.28 A novel finding was the superiority of a simplified index of cardiorespiratory fitness to predict cardiovascular morbidity and all-cause mortality, as compared with the more conventional index peak oxygen uptake. This index did not necessarily demand maximum performance from the subject, and the ability to achieve an oxygen uptake of 67 mmol/min could be judged indirectly from the ability to walk up a 5% gradient at 4.7 km/hr. The predictive power associated with the ability to achieve an oxygen uptake of 67 mmol/min during progressive exercise persisted when age and fat-free mass were included in a Cox proportional hazard analysis, so it appeared to reflect the importance of cardiorespiratory fitness for health. On this account, adults in Trinidad and similar largely sedentary populations are likely to benefit from an increase in physical activity sufficient to improve cardiorespiratory performance. Even relatively low levels of exercise are thought to have a training effect in unfit subjects.29 This study has important public health implications for developing communities that, like Trinidad, are experiencing an increasing burden of CVDs.30 A simple measure of cardiorespiratory performance, such as the Unfitness index used here, may have some utility in clinical and public health settings where exercise laboratory facilities are not available.
| KEY MESSAGES What do we know?
|
|
Acknowledgments |
|---|
The survey was conducted at the Caribbean Epidemiology Centre, Trinidad. The assistance of the then director Dr Patrick Hamilton (deceased) and his staff is gratefully acknowledged. The authors also thank Dr John E Cotes for advice and support. Funding was provided by the Government of Trinidad and Tobago, the UK Medical Research Council, and the Pan-American Health Organization.
|
References |
|---|
|
TopAbstractMethodsResultsDiscussionReferences |
|---|
1 Peters RK, Cady LD, Bischoff DP, Bernstein L, Pike MC. Physical fitness and subsequent myocardial infarction in healthy workers. JAMA 1983;249:3052–56.
2 Lakka TA, Venalainen JM, Rauramaa R, Salonen R, Tuomilehto J, Salonen JT. Relation of leisure-time physical activity and cardiorespiratory fitness to the risk of acute myocardial infarction in men. N Engl J Med 1994;330:1549–54.
3 Farrell SW, Kampert JB, Kohl HW et al. Influences of cardiorespiratory fitness levels and other predictors on cardiovascular disease mortality in men. Med Sci Sports Exerc 1999;30:899–905.
4 Perry IJ, Wannamethee SG, Walker MK, Thomson AG, Whincup PH, Shaper AG. Prospective study of risk factors for development of non-insulin dependent diabetes in middle aged British men. BMJ 1995;310:560–64.
5 Wei M, Gibbons LW, Mitchell TL, Kampert JB, Lee CD, Blair SN. The association between cardiorespiratory fitness and impaired fasting glucose and type 2 diabetes mellitus in men. Ann Intern Med 1999;130:89–96.
6 Goodyear LJ, Kahn BB. Exercise, glucose transport, and insulin sensitivity. Annu Rev Med 1998;49:235–61.[CrossRef][Web of Science][Medline]
7 Van Dam RM, Schuit AJ, Feskens, EJM, Seidell JC, Kromhout D. Physical activity and glucose tolerance in elderly men: the Zutphen Elderly Study. Med Sci Sports Exerc 2002;34:1132–36.[Web of Science][Medline]
8 American Heart Association. Heart and Stroke Statistical Update. Dallas: American Heart Association, 1999.
9 Whelton SP, Chin A, Xin X, He J. Effect of aerobic exercise on blood pressure: a meta-analysis of randomised, controlled trials. Ann Intern Med 2002;136:493–503.
10 Yusuf S, Reddy S, Ounpuu S, Anand S. Global burden of cardiovascular diseases. Circulation 2001;104:2746–53, 2855–64.
11 Miller GJ, Beckles GLA, Maude GH et al. Ethnicity and other characteristics predictive of coronary heart disease in a developing community: principal results of the St James Survey, Trinidad. Int J Epidemiol 1989;18:808–17.
12 Beckles GLA, Miller GJ, Kirkwood BR, Alexis SD, Carson DC, Byam NTA. High total and cardiovascular disease mortality in adults of Indian descent in Trinidad, unexplained by major risk factors. Lancet 1986;1:1298–301.[CrossRef][Web of Science][Medline]
13 Miller GJ, Kirkwood BR, Beckles GLA, Alexis SD, Carson DC, Byam NTA. Adult all-cause, cardiovascular and cerebrovascular mortality in relation to ethnic group, systolic blood pressure and blood glucose concentration in Trinidad, West Indies. Int J Epidemiol 1988;17:62–69.
14 Rose GA. The diagnosis of ischaemic heart pain and intermittent claudication in field surveys. Bull World Health Organ 1962;27:645–58.[Web of Science][Medline]
15 Medical Research Council Committee on the Aetiology of Chronic Bronchitis. Standardized questionnaires on respiratory symptoms. BMJ 1960;2:1665.
16 World Health Organization Expert Committee on Diabetes Mellitus. Second Report. Geneva: WHO, 1980.
17 Miller GJ, Beckles GLA, Byam NTA et al. Serum lipoprotein concentrations in relation to ethnic composition and urbanization in men and women of Trinidad, West Indies. Int J Epidemiol 1984;13:413–21.
18 Durnin JVGA, Womersley J. Body fat assessed from total body density and its estimation from skinfold thickness: assessments in 481 men and women aged from 16 to 72 years. Br J Nutr 1974;32:77–97.[CrossRef][Web of Science][Medline]
19 Prineas RJ, Crow RS, Blackburn H. The Minnesota Code Manual of Electrocardiographic Findings. Boston: John Wright, 1982.
20 Astrand PO, Rodahl K. Textbook of Work Physiology. 3rd edn. London: McGraw-Hill, 1986, pp. 354–90.
21 Cotes JE. Lung Function. Assessment and Application in Medicine. 5th edn. London: Blackwell Scientific Publications, 1993, pp. 429–33.
22 International Labour Organization. Respiratory function tests in Pneumoconiosis. Occupational Safety and Health Series. No. 6. Geneva: International Labour Office, 1966, p. 10.
23 Allied Dunbar National Fitness Survey. Main Findings. Report. London: Sports Council and Health Education Authority, 1994, pp. 11–19.
24 World Health Organization. Myocardial Infarction Registers. Public Health in Europe. Vol. 5. Copenhagen: WHO, 1976.
25 World Health Organization. Manual of the International Statistical Classification of Diseases, Injuries and Causes of Death. 9th Revision. Geneva: WHO, 1977.
26 Bodegard J, Erikssen G, Bjornholt JV, Thelle D, Erikssen J. Possible angina detected by the WHO angina questionnaire in apparently healthy men with a normal exercise ECG: coronary heart disease or not? A 26 year follow up study. Heart 2004;90:627–32.
27 Van Ganse W, Versee L, Eylenbosch W, Vuylsteek K. The electrocardiogram of athletes: comparison with untrained subjects. Br Heart J 1970;32:160–64.
28 Sandvik, L, Erikssen J, Thaulow E, Erikssen G, Mundal R, Rodahl K. Physical fitness as a predictor of mortality among healthy, middle-aged Norwegian men. N Engl J Med 1993;328:533–37.
29 Swain DP, Franklin BA. VO2 reserve and the minimal intensity for improving cardiorespiratory fitness. Med Sci Sports Exerc 2002;34:152–57.[Web of Science][Medline]
30 Lambert EV, Lambert MI, Hudson K, Steyn K, Levitt NS, Charlton K. Role of physical activity for health in communities undergoing epidemiological transition. World Rev Nutr Diet 2001;90:110–26.[Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  S. Kodama, K. Saito, S. Tanaka, M. Maki, Y. Yachi, M. Asumi, A. Sugawara, K. Totsuka, H. Shimano, Y. Ohashi, et al. Cardiorespiratory Fitness as a Quantitative Predictor of All-Cause Mortality and Cardiovascular Events in Healthy Men and Women: A Meta-analysis JAMA, May 20, 2009; 301(19): 2024 - 2035. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 X. Sui, M. J. LaMonte, and S. N. Blair Cardiorespiratory Fitness as a Predictor of Nonfatal Cardiovascular Events in Asymptomatic Women and Men Am. J. Epidemiol., June 15, 2007; 165(12): 1413 - 1423. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||