The New England Journal of Medicine
e-mail icon  FREE NEJM E-TOC    HOME   |   SUBSCRIBE   |   CURRENT ISSUE   |   PAST ISSUES   |   COLLECTIONS   |    Advanced Search
Sign in | Get NEJM's E-Mail Table of Contents — Free | Subscribe
 
Original Article
PreviousPrevious
Volume 328:538-545 February 25, 1993 Number 8
NextNext

The Association of Changes in Physical-Activity Level and Other Lifestyle Characteristics with Mortality among Men
Ralph S. Paffenbarger, Robert T. Hyde, Alvin L. Wing, I-Min Lee, Dexter L. Jung, and James B. Kampert

 

This Article
-Abstract

Tools and Services
-Add to Personal Archive
-Add to Citation Manager
-Notify a Friend
-E-mail When Cited

More Information
-PubMed Citation
ABSTRACT

Background Recent trends toward increasing physical exercise, stopping cigarette smoking, and avoiding obesity may increase longevity. We analyzed changes in the lifestyles of Harvard College alumni and the associations of these changes with mortality.

Methods Men who were 45 to 84 years of age in 1977 and who had reported no life-threatening disease on questionnaires completed in 1962 or 1966 and again in 1977 were classified according to changes in lifestyle characteristics between the first and second questionnaires. We analyzed changes in their level of physical activity, cigarette smoking, blood pressure, and body weight, and the relation of these factors to mortality between 1977 and 1985.

Results Of the 10,269 men, 476 died during this period (which totaled 90,650 man-years of observation). Beginning moderately vigorous sports activity (at an intensity of 4.5 or more metabolic equivalents) was associated with a 23 percent lower risk of death (95 percent confidence interval, 4 to 42 percent; P = 0.015) than not taking up moderately vigorous sports. Quitting cigarette smoking was associated with a 41 percent lower risk (95 percent confidence interval, 20 to 57 percent; P = 0.001) than continuing smoking, but with a 23 percent higher risk than constant nonsmoking. Men with recently diagnosed hypertension had a lower risk of death than those with long-term hypertension (relative risk, 0.75; 95 percent confidence interval, 0.55 to 1.02; P = 0.057), as did men with consistently normal blood pressure (relative risk, 0.52; 95 percent confidence interval, 0.40 to 0.68; P<0.001). Maintenance of lean body mass was associated with a lower mortality rate than long-term, recent, or previous obesity. The associations between changes in lifestyle and mortality were independent and were largely undiminished by age. Our findings on death from coronary heart disease mirrored those on death from all causes.

Conclusions Beginning moderately vigorous sports activity, quitting cigarette smoking, maintaining normal blood pressure, and avoiding obesity were separately associated with lower rates of death from all causes and from coronary heart disease among middle-aged and older men.


Physical-activity level, physical fitness, and other modifiable lifestyle characteristics may influence the risk of chronic disease and premature death1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18. Changes in lifestyle may therefore promote optimal health and longevity. Harvard College archives from 1916 through 1950 provided physical, social, athletic, and other data on thousands of former students whom we studied to determine whether the men's characteristics at college age predisposed them to chronic diseases, particularly coronary heart disease, later in life19. Follow-up questionnaires mailed to alumni in the 1960s and 1970s reaffirmed the association between adequate higher levels of physical activity and longer survival, and also the associations between sedentary behavior, cigarette smoking, hypertension, and obesity and earlier death4,5. In this study, we examined data on Harvard alumni to determine whether changes in exercise habits and other personal characteristics were associated with lower rates of death from all causes and from coronary heart disease.

Methods

The Study Population

To assess changes in lifestyle, we included only alumni who had responded to mailed questionnaires in 1962 or 1966 and again in 1977. Men who reported on either questionnaire that a physician had given them a diagnosis of coronary heart disease, stroke, chronic obstructive respiratory disease, diabetes, or cancer were excluded in order to avoid bias in the analysis of the influence of physical activity on subsequent mortality. We analyzed associations between the men's personal and lifestyle characteristics and mortality from 1977 through 1985 and the relation of these characteristics to estimated increments in longevity (up to the age of 85). We also examined changes in these characteristics between the first and second questionnaires for any associations such changes had with subsequent mortality (between 1977 and 1985) and length of life.

In the two questionnaires the participants reported detailed information covering most of their lives. In 1962 or 1966, 21,582 (68 percent) of 31,697 men known to be alive who had entered college between 1916 and 1950 returned interpretable questionnaires. Of the 19,359 who were still alive in 1977, 14,800 (76 percent) responded again that year. During follow-up, men who reached 85 years of age were dropped from further assessment, and longevity was measured only up to that arbitrarily chosen age.

The questionnaires asked about physical activity, cigarette smoking, specific diseases diagnosed by physicians, body size, and parental disease and death (an index of familial and hereditary influences). Weekly lists of deaths from the Harvard Alumni Office enabled us to obtain death certificates and identify causes of death. Fewer than 1 percent of the alumni included in the study were lost to follow-up.

Assessment of Physical Activity

On both questionnaires, alumni reported how many city blocks they walked daily, how many flights of stairs they climbed daily, and the type, frequency, and duration of their participation in sports or recreational activities in hours per week. From these data a physical-activity index was computed in kilocalories per week, in which walking 1 mile (1.6 km) was assigned a value of 100 kcal and climbing five flights of stairs (100 stairs), 40 kcal; sports and recreation were classified according to intensity at 5, 7.5, or 10 kcal per minute. The physical-activity index was considered an indicator rather than an absolute measure of energy expenditure.

We further classified sports and other recreational activities according to intensity and duration, defining light sports activity as requiring less than 4.5 metabolic equivalents (METs) and moderately vigorous activity as requiring 4.5 or more METs. One MET is defined as the energy expended per minute while sitting quietly and is equivalent to 3.5 ml of oxygen uptake per kilogram of body weight per minute for an adult weighing 70 kg. Activities were designated light or moderately vigorous on the basis of generally accepted MET values20,21,22. The light activities most commonly reported were golf, walking for pleasure, and a combination of gardening, housework, and carpentry. The most common moderately vigorous sports activities were swimming; tennis, squash, racquetball, or handball; and jogging or running. We also analyzed moderately vigorous sports activity in terms of hours per week.

Other Lifestyle Characteristics

To study the influence of such continuous variables as cigarette smoking and body-mass index, we created three categories that might reveal any gradient in the risk of death. For cigarette smoking in 1977, we grouped men who smoked a pack (20 cigarettes) or more per day, men who smoked less than a pack daily, and nonsmokers. In terms of blood-pressure status in 1977, we divided the sample into men with hypertension diagnosed by a physician, normotensive men whose systolic blood pressure recorded in college was 130 mm Hg or more, and normotensive men whose systolic blood pressure in college was less than 130 mm Hg. The body-mass (Quetelet's) index was computed as the weight in kilograms divided by the square of the height in meters. Alumni were classified as having an index of >= 26, 24 to 25, or <24. The sample was divided with regard to parental death into groups of men both of whose parents had died before 65 years of age by 1977, those with one parent who had died before the age of 65, and those with both parents living.

Statistical Analysis

Data on physical-activity levels and other lifestyle characteristics reported in 1977, and changes in these factors between 1962 or 1966 and 1977, were studied for any association with mortality from all causes and mortality from coronary heart disease during the nine-year period from 1977 through 1985. Mortality rates per 10,000 man-years of observation were adjusted by the indirect method for age (in five-year groups) and other confounding factors among the groups being compared. The rates thus calculated for various subgroups of alumni provided the basis for determining relative risks of death, with the death rate among men whose classification was deemed to entail the highest risk or the lowest risk used as the reference category, depending on the analysis.

Estimates of attributable risk for each category of a variable as of 1977 were computed as potential percentage reductions in population death rates, adjusted for age and each of the other categories considered. The multivariate analyses used for estimating relative and attributable risks of death and added years of life potentially attributable to favorable changes in characteristics were based on proportional-hazards models with Poisson regression methods23.

To assess the associations between changes in lifestyle and mortality, we defined four status categories (remained at high risk, changed from low to high risk, changed from high to low risk, remained at low risk) and examined the mortality rates for each category. The association between each characteristic and mortality was assessed with the relations of age and the other characteristics to mortality held constant. The Mantel extension of the Mantel-Haenszel test was used to determine the significance of mortality trends across categories of selected characteristics24. We computed years of life that may have been gained among men who adopted a more favorable lifestyle, as compared with men whose lifestyle entailed a high risk25. Estimates of significance were derived from two-tailed tests.

Results

The eligible population numbered 10,269 alumni who ranged from 45 to 84 years of age (mean [±SD], 57.5 ±8.8) in 1977. There were 476 deaths during 90,650 man-years of observation in the nine-year follow-up period from 1977 through 1985. There were 208 deaths from cardiovascular disease (130 from coronary heart disease), 156 from cancer, 63 from other natural causes, 45 from trauma, and 4 from unknown causes.

Lifestyle and Mortality from All Causes

Table 1 shows age-adjusted rates and relative risks of death from any cause during follow-up according to lifestyle and personal characteristics as reported on the questionnaires returned in 1977. After more than a decade, these results echoed findings from similar questionnaires returned by the same men in 1962 or 1966 with follow-up through 19784,5. This consistent pattern reaffirms the inverse relation between the level of physical activity and the risk of premature death from any cause. All major trends related to physical-activity level (i.e., the association between lower death rate and increases in walking, stair climbing, moderately vigorous sports activity, and physical-activity index) were statistically significant.

View this table:
[in this window]
[in a new window]
 
Table 1. Age-Adjusted Rates and Relative Risks of Death from All Causes among 10,269 Harvard Alumni from 1977 through 1985, According to Patterns of Physical Activity and Other Characteristics in 1977.

 
Gradients of benefit were consistent throughout, except for light sports activity (fewer than 4.5 METs), which was not associated with a lower mortality from all causes, and the physical-activity index, for which the intermediate levels of energy expenditure (500 to 3499 kcal per week) were associated with similar death rates. Yet the most active men, who expended 3500 kcal per week or more, had half the risk of death of the least active (less than 500 kcal per week). Moreover, when the men were arbitrarily divided according to the physical-activity index at 2000 kcal per week, the death rate associated with a low index was 57.1 per 10,000 man-years and that associated with a high index only 45.1, giving a 21 percent lower relative risk for the more active (95 percent confidence interval, 4 to 35 percent; P = 0.015).

The 1977 questionnaire responses indicated a notably high proportion of men who participated in moderately vigorous sports activity; they accounted for three fourths of the man-years assessed -- a near reversal of the 1962 or 1966 response4,5. From 1977 through 1985, the relative risk of death was halved among the quarter of the alumni who reported three or more hours of moderately vigorous sports activity per week, as compared with the quarter who did not engage in such sports or recreational activity.

Table 1 also shows a direct association between death rates (and relative risks of death) and high-risk characteristics, particularly cigarette smoking, hypertension, and a high body-mass index. A reversed J-shaped curve for mortality risk was seen for the high and low extremes of the body-mass index as compared with intermediate values. Early parental death was largely unrelated to mortality as the alumni neared the age of 65 years themselves.

Table 2 shows the relative and attributable risks of death associated with the presence and absence of each of the five adverse personal characteristics -- sedentary living, cigarette smoking, hypertension, overweight for height (a high body-mass index), and early parental death -- with adjustments for differences in age and in each of the other four characteristics. More detailed adjustments for cigarette-smoking status (never, former, light, and heavy) made little difference in the results for the other characteristics. During the follow-up period, the sedentary alumni (those with a physical-activity index of less than 2000 kcal per week) had a 25 percent higher risk of death than more active men, smokers had an 87 percent higher risk than nonsmokers, men with hypertension had a 69 percent higher risk than men with normal blood pressure, and the more obese men had a 31 percent greater risk than the leaner men, but a history of early death in one or both parents had little relation to longevity. Those with one or more of these adverse lifestyle characteristics, who contributed 83 percent of the man-years of observation, were at 64 percent greater risk of mortality from all causes during the nine-year follow-up period than the men with none of these adverse characteristics.

View this table:
[in this window]
[in a new window]
 
Table 2. Relative and Attributable Risks of Death from All Causes among 10,269 Harvard Alumni from 1977 through 1985, According to Adverse Lifestyle Characteristics in 1977.

 
Sedentary living was further divided (Table 2) into low levels of walking or stair climbing and lack of moderately vigorous sports activity. The mortality rates associated with each factor have been adjusted for differences in the others as well as for age and the other four adverse characteristics. The subgroup analysis shows that men who walked less than nine miles per week had a 16 percent higher risk of death than those who walked more (P not significant) and those who climbed fewer than 20 flights of stairs per week had a 23 percent higher risk than those who climbed more. Men who did not engage in moderately vigorous sports activity had a 44 percent higher risk of death than those who did.

Estimates of Attributable Risk

The "population-attributable risk" is an estimate of the percentage reduction in the death rate that might have occurred in the total group of 10,269 alumni (some with and some without high-risk characteristics) if all the men with specified adverse characteristics had converted them to healthful levels and if all who already had favorable characteristics had maintained them. Estimates of attributable risk require the assumption of a cause-and-effect relation, persistence of lifestyle patterns through the follow-up period, and equal distribution of any potential confounding factors among the groups being compared. Estimates of attributable risk are given in Table 2. Notably, if all the men had played moderately vigorous sports, the death rates might have been 12 percent lower than was actually the case, and total abstinence from cigarettes might have reduced the overall death rate by 11 percent. If all five of the adverse lifestyle factors listed had been completely absent, the death rate from all causes during the nine years might have been 41 percent lower than that observed. Expressed differently, 195 of the 476 deaths might have been postponed.

Lifestyle Characteristics and Mortality from Coronary Heart Disease

There were 130 deaths from physician-diagnosed coronary heart disease during the follow-up period. When analyzed as in Table 2, sedentary alumni were at 36 percent higher risk of death from coronary heart disease (95 percent confidence interval, -8 to 99 percent) than active men. Men who climbed fewer than 20 flights of stairs per week (one flight consists of 20 steps) and did not engage in moderately vigorous sports activity were at 56 percent (95 percent confidence interval, 7 to 128 percent) and 51 percent (95 percent confidence interval, 2 to 123 percent) higher risk than men who climbed more stairs or engaged in moderately vigorous activity. Cigarette smoking was associated with a doubled risk of death from coronary heart disease (P<0.001), as was hypertension (P<0.001); overweight for height was associated with a 55 percent increase in risk (95 percent confidence interval, 5 to 128 percent); early parental death was associated with a 64 percent increase (95 percent confidence interval, 15 to 134 percent); and having at least one of these adverse characteristics was associated with an added risk of 151 percent (95 percent confidence interval, 23 to 414 percent).

Estimates of the population-attributable risk during follow-up with the elimination of selected characteristics were as follows: not engaging in moderately vigorous sports, 14 percent; cigarette smoking, 13 percent; hypertension, 20 percent; overweight for height, 11 percent; and early parental death, 20 percent. In the absence of all these adverse characteristics, 58 percent of these deaths from coronary heart disease (95 percent confidence interval, 40 to 70 percent) might have been delayed.

Changes in Lifestyle and Mortality from All Causes

Table 3 shows the association of favorable changes in lifestyle between 1962 or 1966 and 1977 with rates of death from all causes from 1977 through 1985. Forty-one percent of the man-years was contributed by men who did not report enough physical activity to reach an index of 2000 kcal or more per week at either assessment. Another 16 percent was contributed by men who dropped below that index level by 1977, 20 percent by men who increased their activity to favorable levels, and 24 percent by men who had been active at the level of 2000 kcal per week at both assessments. Although the difference was not significant, perhaps because of the small numbers, the alumni who increased their level of activity through a combination of walking, climbing stairs, and engaging in sports activity had a 15 percent lower death rate than the men who continued to be sedentary.

View this table:
[in this window]
[in a new window]
 
Table 3. Rates and Relative Risks of Death from All Causes among 10,269 Harvard Alumni from 1977 through 1985, According to Changes in Patterns of Physical Activity and Other Characteristics between 1962 or 1966 and 1977.

 
The findings for moderately vigorous sports activity resemble those for the physical-activity index, even though the former was defined by intensity ( >= 4.5 METs) and the latter in terms of kilocalories per week. Men who discontinued moderately vigorous sports activity had a 15 percent higher risk of mortality than men who had never reported such activity (P not significant). Thirty-eight percent of the men, however, took up moderately vigorous sports activity, and they had a 23 percent lower risk of mortality, close to the 29 percent lower risk for men who habitually engaged in moderately vigorous sports activity.

Hardly anyone took up cigarette smoking during the period between the questionnaires. Sixteen percent of the men continued to smoke; 18 percent abandoned smoking, joining the 64 percent who were nonsmokers, for a total of 82 percent nonsmokers in 1977. Men who quit smoking had a 41 percent lower death rate, but even this substantial lowering of mortality did not reach the level for nonsmokers, which was half that for persistent smokers.

Alumni in whom hypertension developed during the period between questionnaires amassed more man-years (13 percent) than the men in whom hypertension developed earlier (8 percent), but they had a lower relative risk of mortality from all causes (0.75). Risk for the men who continued to have normal blood pressure was half that for those with long-term hypertension.

The risk of death from all causes was higher for men who were persistently overweight for height, for those whose body-mass index increased to 26 or more, and for those whose index dropped below 26, as compared with those whose body-mass index remained below that level. Normal-weight alumni had a 23 percent lower risk of mortality than the persistently overweight.

The associations of increases in the physical-activity index, taking up moderately vigorous sports activity, quitting cigarette smoking, and maintaining normal blood pressure with age-specific mortality from all causes are shown in Figure 1. Although many of the specific comparisons were not statistically significant, the trends of the associations are apparent. The difference in the risk of death associated with an increase in the physical-activity index changed from a 55 percent lower risk to a 10 percent higher risk as age increased, but taking up moderately vigorous sports activity conferred a steady advantage of 28, 31, 26, and 21 percent from the youngest to the oldest age group. Both quitting cigarette smoking and maintaining normal blood pressure were more strongly associated with lower mortality from all causes in the older age groups than in the youngest.


View larger version (58K):
[in this window]
[in a new window]
 
Figure 1. Rates and Relative Risks of Death from All Causes among 10,269 Harvard Alumni, According to Age in 1977 and Changes in Patterns of Physical Activity and Other Characteristics between 1962 or 1966 and 1977.

Relative risks are expressed in relation to men who did not make the change, as indicated in each panel. The men ranged in age from 45 to 84 years in 1977. The follow-up period, from 1977 through 1985, totaled 90,650 man-years of observation. Relative risks and 95 percent confidence intervals are shown above the bars.

 
Changes in Lifestyle and Mortality from Coronary Heart Disease

The risk of death from coronary heart disease was examined according to changes in lifestyle, in analyses that paralleled those summarized in Table 3. Again, the small numbers of deaths from coronary heart disease limited the statistical power of the observations. Alumni who increased their physical-activity index to 2000 kcal or more per week had a 17 percent lower risk of death from coronary heart disease than those who remained more sedentary (P = 0.507). But men who took up moderately vigorous activity had a 41 percent lower risk than those who continued not to engage in such activity (P = 0.044). Men who quit cigarette smoking were at 44 percent lower risk than continuing smokers (P = 0.052). Men who maintained normal blood pressure were at 49 percent lower risk than those with hypertension (P<0.001). And the men who continued to be lean had a 41 percent lower risk than those whose body-mass index increased to at least 26 (P = 0.085).

Changes in Lifestyle and Longevity

Table 4 gives estimates of the years of life added for alumni who reported adopting low-risk characteristics between 1962 or 1966 and 1977, based on their rates of mortality from all causes from 1977 through 1985. The data represent estimates of the extension of life associated with specific favorable lifestyle characteristics, adjusted for differences in each of the other characteristics listed and for early parental mortality. The gains shown are according to 10-year age groups and for the composite total, with added years estimated only up to the age of 85.

View this table:
[in this window]
[in a new window]
 
Table 4. Additional Years of Life up to the Age of 85 Associated with Adoption or Maintenance of a Favorable Physical-Activity Level and Other Characteristics between 1962 or 1966 and 1977, as Estimated from Mortality Rates among 10,269 Harvard Alumni from 1977 through 1985.

 
Age adjustments made for the entire group of men from 45 to 84 years of age showed that added years of life were associated with specific changes in lifestyle. Noteworthy were an additional 0.72 year associated with taking up moderately vigorous sports activity, 1.46 associated with quitting cigarette smoking, and an impressive 2.49 years associated with both changes.

Discussion

The questionnaire responses of the Harvard alumni whom we studied reflected recent trends in the United States toward increased leisure-time exercise, abandonment of cigarette smoking, and control of body weight22. The changes in lifestyle made by these men during the period between the questionnaires and their corresponding rates of mortality from all causes and from coronary heart disease from 1977 through 1985 fit the hypothesis that these trends have a favorable effect on mortality. In particular, the hundreds of aging alumni who took up moderately vigorous sports activity had a substantial reduction in mortality from all causes (23 percent) and from coronary heart disease (41 percent) as compared with their less vigorous classmates. However, the data do not prove a cause-and-effect relation between the adoption of a more active lifestyle and a lower death rate.

Our observations are reminiscent of earlier findings4,5,26,27 that former varsity athletes who discontinued their sports activities had higher rates of disease and death thereafter than their teammates who continued energetic exercise. Moreover, in an early example of favorable change, alumni who had avoided athletics as college students but subsequently took up a more active lifestyle had the same low risk as classmates who had been vigorously active all along. The testimony of these patterns -- derived from more than 300,000 man-years of observation (since college entrance) -- supports the thesis that a reduced risk of death from coronary heart disease and from all causes, leading to longer life, results from an adequate exercise program.

Although in 1977 the 10,269 alumni were free of diagnosed chronic disease that might limit their physical activity, including sports activity, some men may have had subclinical or undiagnosed disease that, in turn, resulted in both sedentary habits and premature death. In observational studies, such selective influences cannot be eliminated entirely, as might be possible in appropriately designed clinical trials. But the impracticality of controlled trials that would assess the health effects of a physically active way of life makes such an undertaking virtually impossible. Further limitations of our study result from incomplete questionnaire returns (response rates of 68 to 76 percent) and possible confounding due to any differences in dietary intake between active and inactive men. Nonrespondents to the 1977 questionnaire had a two thirds higher death rate than respondents through 1985 (206.8 vs. 123.3 per 10,000 man-years). Dietary data collected in the questionnaires were insufficient for meaningful analysis.

Data are unavailable on intermediate variables -- such as dyslipoproteinemia, hyperinsulinemia, hyperglycemia, cardiac dysfunction, and actual blood pressure levels -- that might provide a causal pathway between physical inactivity and higher death rates from coronary heart disease and from all causes28. In addition, the relative importance of the intensity as compared with the quantity of exercise for optimal health benefit is not easily determined, since men who engaged in moderately vigorous sports activity were also those who had a higher physical-activity index. Yet in earlier analyses,26,29,30 the lower risk of death associated with a physically active lifestyle was even lower among the men who engaged in moderately vigorous sports activity each week. Although the Multiple Risk Factor Intervention Trial showed little or no added benefit of exercising beyond a moderate level,8 British civil servants had lower rates of coronary heart disease and death only when they engaged in moderately vigorous recreational activity1,13,31. What kinds of physical activity should be prescribed, how much, how intense, and for whom if optimal health and longevity are to be achieved remain unanswered questions that require further clarification.

Supported by grants from the National Heart, Lung, and Blood Institute (R01 HL 34174) and the National Cancer Institute (R01 CA 44854).


Source Information

From the Division of Epidemiology, Stanford University School of Medicine, Stanford, Calif. (R.S.P., R.T.H., D.L.J., J.B.K.), and the Department of Epidemiology, Harvard School of Public Health, Boston (R.S.P., A.L.W., I-M.L.).

Address reprint requests to Dr. Paffenbarger at the Department of Health Research and Policy, HRP Bldg., Rm. 113, Stanford University School of Medicine, Stanford, CA 94305-5092.

References

  1. Chave SPW, Morris JN, Moss S, Semmence AM. Vigorous exercise in leisure time and the death rate: a study of male civil servants. J Epidemiol Community Health 1978;32:239-243. [Free Full Text]
  2. Paffenbarger RS Jr, Brand RJ, Sholtz RI, Jung DL. Energy expenditure, cigarette smoking, and blood pressure level as related to death from specific diseases. Am J Epidemiol 1978;108:12-18. [Free Full Text]
  3. Salonen JT, Puska P, Tuomilehto J. Physical activity and risk of myocardial infarction, cerebral stroke and death: a longitudinal study in Eastern Finland. Am J Epidemiol 1982;115:526-537. [Free Full Text]
  4. Paffenbarger RS Jr, Hyde RT, Wing AL, Hsieh C-c. Physical activity, all-cause mortality, and longevity of college alumni. N Engl J Med 1986;314:605-613. [Abstract]
  5. Physical activity and longevity of college alumni. N Engl J Med 1986;315:399-401. [Medline]
  6. Pekkanen J, Marti B, Nissinen A, Tuomilehto J, Punsar S, Karvonen MJ. Reduction of premature mortality by high physical activity: a 20-year follow-up of middle-aged Finnish men. Lancet 1987;1:1473-1477. [Medline]
  7. Kaplan GA, Seeman TE, Cohen RD, Knudsen LP, Guralnik J. Mortality among the elderly in the Alameda County Study: behavioral and demographic risk factors. Am J Public Health 1987;77:307-312. [Erratum, J Public Health 1987;77:818.] [Free Full Text]
  8. Leon AS, Connett J, Jacobs DR Jr, Rauramaa R. Leisure-time physical activity levels and risk of coronary heart disease and death: the Multiple Risk Factor Intervention Trial. JAMA 1987;258:2388-2395. [Free Full Text]
  9. Karvonen MJ. Determinants of cardiovascular diseases in the elderly. Ann Med 1989;21:3-12. [Medline]
  10. Benfante RJ, Reed DM, MacLean CJ, Yano K. Risk factors in middle age that predict early and late onset of coronary heart disease. J Clin Epidemiol 1989;42:95-104. [CrossRef][Medline]
  11. Slattery ML, Jacobs DR Jr, Nichaman MZ. Leisure time physical activity and coronary heart disease death: the US Railroad Study. Circulation 1989;79:304-311. [Free Full Text]
  12. Blair SN, Kohl HW III, Paffenbarger RS Jr, Clark DG, Cooper KH, Gibbons LW. Physical fitness and all-cause mortality: a prospective study of healthy men and women. JAMA 1989;262:2395-2401. [Free Full Text]
  13. Morris JN, Clayton DG, Everitt MG, Semmence AM, Burgess EH. Exercise in leisure-time: coronary attack and death rates. Br Heart J 1990;63:325-334. [Free Full Text]
  14. Sytkowski PA, Kannel WB, D'Agostino RB. Changes in risk factors and the decline in mortality from cardiovascular disease: the Framingham Heart Study. N Engl J Med 1990;322:1635-1641. [Abstract]
  15. Lindsted KD, Tonstad S, Kuzma JW. Self-report of physical activity and patterns of mortality in Seventh-Day Adventist men. J Clin Epidemiol 1991;44:355-364. [CrossRef][Medline]
  16. Arraiz GA, Wigle DT, Mao Y. Risk assessment of physical activity and physical fitness in the Canada Health Survey mortality follow-up study. J Clin Epidemiol 1992;45:419-428. [CrossRef][Medline]
  17. Shaper AG, Wannamethee G. Physical activity and ischaemic heart disease in middle-aged British men. Br Heart J 1991;66:384-394. [Free Full Text]
  18. Wannamethee G, Shaper AG. Physical activity and stroke in British middle aged men. BMJ 1992;304:597-601.
  19. Paffenbarger RS Jr, Notkin J, Krueger DE, et al. Chronic disease in former college students. II. Methods of study and observations on mortality from coronary heart disease. Am J Public Health Nations Health 1966;56:962-971. [Medline]
  20. Caspersen CJ, Bloemberg BPM, Saris WHM, Merritt RK, Kromhout D. The prevalence of selected physical activities and their relation with coronary heart disease risk factors in elderly men: the Zutphen Study, 1985. Am J Epidemiol 1991;133:1078-1092. [Free Full Text]
  21. Ainsworth BE, Haskell WL, Leon AS, et al. Compendium of physical activities: classification of energy costs of human physical activities. Med Sci Sports Exerc 1993;25:71-80. [Medline]
  22. Lee I-M, Paffenbarger RS Jr, Hsieh C-c. Time trends in physical activity among college alumni, 1962-1988. Am J Epidemiol 1992;135:915-925. [Free Full Text]
  23. Frome EL. The analysis of rates using Poisson regression models. Biometrics 1983;39:665-674. [CrossRef][Medline]
  24. Mantel N. Chi-square tests with one degree of freedom: extensions of the Mantel-Haenszel procedure. J Am Stat Assoc 1963;58:690-700.
  25. Hill AB. Principles of medical statistics. 9th ed. New York: Oxford University Press, 1971.
  26. Paffenbarger RS Jr, Wing AL, Hyde RT. Physical activity as an index of heart attack risk in college alumni. Am J Epidemiol 1978;108:161-175. [Free Full Text]
  27. Paffenbarger RS Jr, Hyde RT, Wing AL, Steinmetz CH. A natural history of athleticism and cardiovascular health. JAMA 1984;252:491-495. [Free Full Text]
  28. Manson JE, Tosteson H, Ridker PM, et al. The primary prevention of myocardial infarction. N Engl J Med 1992;326:1406-1416. [Medline]
  29. Paffenbarger RS Jr, Hyde RT, Wing AL. Physical activity and physical fitness as determinants of health and longevity. In: Bouchard C, Shephard RJ, Stephens T, Sutton JR, McPherson BD, eds. Exercise, fitness, and health: a consensus of current knowledge. Champaign, Ill.: Human Kinetics Books, 1990:33-48.
  30. Paffenbarger RS Jr, Hyde RT, Wing AL, Lee I-M, Kampert JB. Some interrelationships of physical activity, physiological fitness, health, and longevity. In: Bouchard C, Shephard RJ, Stephens T, McPherson BD, eds. Physical activity, fitness, and health. Champaign, Ill.: Human Kinetics Books (in press).
  31. Morris JN. Physical activity, epidemiology, and the public's health: the role of exercise in the prevention of coronary heart disease. Med Sci Sports Exerc (in press).

 

This Article
-Abstract

Tools and Services
-Add to Personal Archive
-Add to Citation Manager
-Notify a Friend
-E-mail When Cited

More Information
-PubMed Citation

This article has been cited by other articles:



HOME  |  SUBSCRIBE  |  SEARCH  |  CURRENT ISSUE  |  PAST ISSUES  |  COLLECTIONS  |  PRIVACY  |  TERMS OF USE  |  HELP  |  beta.nejm.org

Comments and questions? Please contact us.

The New England Journal of Medicine is owned, published, and copyrighted © 2009 Massachusetts Medical Society. All rights reserved.