HomeLarge Type Edition
HOME ARCHIVE SEARCH TABLE OF CONTENTS
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Services
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
PubMed
Right arrow PubMed Citation
The Journals of Gerontology Series A: Biological Sciences and Medical Sciences 61:851-858 (2006)
© 2006 The Gerontological Society of America

Longitudinal Changes in Aerobic Capacity: Implications for Concepts of Aging

Milton Hollenberg, Jingrong Yang, Thaddeus J. Haight and Ira B. Tager

1 Department of Veterans Affairs Medical Center, San Francisco, California.
2 Department of Medicine, University of California, San Francisco.
3 Division of Epidemiology, School of Public Health, University of California, Berkeley.

Address correspondence to Milton Hollenberg, MD, Department of Veterans Affairs Medical Center, Cardiology Section–111C3, 4150 Clement Street, San Francisco, CA 94121. E-mail: milton.hollenberg{at}med.va.gov


   Abstract
Top
 Abstract
Methods
 Results
 Discussion
 Appendix A
 References
 
Background. Whereas aerobic capacity declines with age, major factors responsible for such decline have been poorly defined by past studies.

Methods. Participants were relatively healthy older individuals (339 women, 253 men) in whom demographic information and cardiopulmonary physiological measurements were obtained at baseline and biannually for three additional measurements.

Results. The study identified progressive declines in both forced expiratory volume in 1 second (FEV1) and in maximal exercise heart rate as two variables that accounted primarily for the longitudinal decline of aerobic performance in this cohort of relatively fit older persons who achieved high respiratory exchange ratios (RER; mean = 1.08 for women, 1.12 for men). Whereas women achieved a peak oxygen consumption (VO2peak) only 77% that of men, oxygen uptake became similar to men (to 95%) when measured per kilogram of lean body mass rather than per kilogram of total body mass. During the 6 years of the study (four time points) aerobic capacity declined in both sexes, however, less steeply for women than for men (18% vs 24% per decade, respectively). The rate of decline was independent of baseline variables such as body composition (e.g., lean body mass, lean/fat ratio), smoking status, medications, or concomitant health conditions, even though these variables strongly influenced baseline aerobic performance. Inclusion of FEV1 and maximal exercise heart rate into the statistical models, however, accounted for most of the longitudinal decline of aerobic performance. When adjusted for these two variables, aerobic capacity declined 9.7% and 10.4% per decade in women and men, respectively.

Conclusions. Our findings emphasize the primary importance of declining FEV1 and declining maximal exercise heart rate in accounting for the "aging effect" on aerobic capacity. Thus, when comparing longitudinal studies, all estimates of aerobic decline should be interpreted with respect to the specific variables included in the models, which also need to include FEV1 and maximal exercise heart rate.

MAXIMAL aerobic capacity (maximal oxygen uptake, VO2 max) is reported to decline with age, –~10% per decade in sedentary persons >25 years of age (1,2) and ~15% per decade between the ages of 50 and 75 years (3). Such estimates are based mainly on cross-sectional data. The few studies that used longitudinal data have been based on small numbers of selected individuals [e.g., master/endurance athletes (3–6), physical education students (7), men enrolled in a fitness program (8)] or unselected individuals (9). A recent study by Fleg and colleagues (10) studied 810 participants from ages 20 to mid 80s and observed declines in level of oxygen consumption at peak exercise (VO2peak) and oxygen pulse (VO2 ml/heart beat at maximum exercise) that accelerated with advanced decades.

This study describes the longitudinal changes in cardiopulmonary capacity that occur with aging in a large sample (n = 592) of community-dwelling older participants (median age 70 years at enrollment) not selected on the basis of patterns of physical activity or other lifestyle variables. Specifically, we have investigated how the inclusion of a number of physiological variables into such analyses can change the estimates of age-related decline in fitness. In contrast to most other longitudinal studies that have used only two time points to measure change, we took measurements approximately every 2 years up to four time points. The large numbers of observations allowed us to evaluate the effects of several variables on exercise performance, e.g., pulmonary mechanics, smoking, medications, body composition, and physical activity.


   METHODS
Top
 Abstract
 Methods
Results
 Discussion
 Appendix A
 References
 
Participants
Of 2092 individuals who participated in a population-based longitudinal study of aging in people ≥ 55 years old who lived in the city of Sonoma, California, 998 relatively healthy persons (419 men and 579 women) were eligible for this substudy of aerobic fitness (11). Individuals were excluded if they had: 1) known cardiac disease (myocardial infarction by history or electrocardiogram, history of angina, coronary artery angioplasty, or bypass grafting); 2) cerebrovascular disease (stroke, transient ischemic attacks, or carotid artery surgery); 3) peripheral vascular disease with intermittent claudication; or 4) musculoskeletal impairment (use of a cane and/or walker; lower extremity arthritis that impaired walking) or inability to perform a treadmill exercise test.

Participants were first interviewed and underwent treadmill exercise between June 1993 and March 1995, which then was repeated approximately biannually: 800 (80%; 473 men, 327 women) a second time (T2, September 1995 through April 1997); 598 (60%; 251 men, 347 women) a third time (T3, August 1997 through October 1999); and 490 (43%; 196 men, 234 women) a fourth time (T4, February 2000 through March 2002). Major reasons for participants not to be retested were death (60), intervening illness (263), refusals (238), and moving from the study area (7). The study was approved by the Committee on Human Research, University of California, San Francisco and the Committee for the Protection of Human Subjects, University of California, Berkeley. Written, informed consent was obtained from all participants prior to testing at each time point. We restricted our analyses to 592 participants (253 men and 339 women) who completed at least 3 of the 4 biannual tests.

Treadmill Exercise Testing
Participants were exercised on a computer-driven treadmill (CASE 15; Marquette Electronics, Inc., Milwaukee, WI) with the Cornell modification of the Bruce treadmill exercise protocol (12,13). They were exercised to their self-determined maximal capacity or until the physician stopped the test. Participants could stop the test whenever they felt the need to do so. All participants exercised for ≥4 minutes.

Respiratory Gas Measurements and Spirometry
Continuous breath-to-breath respiratory gas measurements were obtained with a Medical Graphics Cardiopulmonary Exercise (CPX) system (Medical Graphics Corp., St. Paul, MN). Direct measurements of oxygen consumption (VO2), carbon dioxide production (VCO2), minute ventilation (VE), and respiratory rate were obtained. Flow meters and gas analyzers were calibrated daily for accuracy and linearity with precisely analyzed gas mixtures. Gas analyzers were checked by auto-calibration prior to each test. Output from the gas analyses were sampled every 15 seconds and stored for calculation of the Oxygen Uptake Efficiency Slope (OUES). Maximum expiratory maneuvers were performed on a rolling seal Survey Spirometer (Warren E. Collins, Braintree, MA). The mean forced expiratory volume in 1 second (FEV1) was based on two or three trials that met all acceptability criteria for spirometry (14).

We have included FEV1 in our analyses because several long-term follow-up studies (15–18) have demonstrated an association between FEV1 and various causes of mortality that are independent of cigarette smoking, known coronary heart disease, and a variety of risk factors for heart disease. It has been suggested that FEV1 is a marker for general aging processes that affect both the heart and lungs (18) and other organ systems as well. Therefore, in these analyses we consider FEV1 as a surrogate for a number of unmeasured aging processes that could impact aerobic fitness and not as a specific measure of lung function.

Measures of Cardiopulmonary Fitness
VO2peak (ml/kg min–1), the duration of exercise, and the OUES were used to evaluate aerobic fitness. Originally described by Baba and colleagues (19) and verified by us (20), the OUES reflects the relationship between oxygen uptake (VO2 in ml · min–1) and total ventilation (VE in L · min–1), and is described by the following equation:


Formula

where a represents the OUES {(VO2 ml · min–1)/(VE L · min–1)}. OUES can be calculated reliably from a submaximal exercise test of at least 4 minutes duration (20). Its use greatly reduces test variability due to motivational and subjective factors—variables that can influence results when exercise duration or VO2peak are used as test endpoints.

Lean Body Mass (Bioelectric Impedance) Measurements
Whole-body resistance and reactance were obtained by bioelectric impedance at each visit (Body Composition Analyzer Model B1A-101; RJL Systems, Clinton Twp, MI; 21). Lean and fat mass were estimated from regression equations derived in a validation substudy of 200 study participants who were free of congestive heart failure and kidney or liver disease, who were not taking diuretics, and who underwent bioelectric impedance measurements and dual x-ray absorptiometry (DEXA) (21).

Statistical Analysis
All continuous baseline variables were centered on their median values. Longitudinal effects of interest were estimated with sex-specific repeated measures models (22) that allowed separation of longitudinal effects due to aging from cross-sectional (cohort) effects. Six separate, sex-specific models were tested by repeated measures methods (Appendix A): two cross-sectional models that do not account for time and four longitudinal models to evaluate the effects of time on the estimates of age-related declines.

To account for the effect on the results of various patterns of losses to follow-up, we carried out a series of sensitivity analyses (details of these analyses available from IBT on request). In no instance were the results different from those obtained without adjustment for the censoring.


   RESULTS
Top
 Abstract
 Methods
 Results
Discussion
 Appendix A
 References
 
Participant Characteristics
The characteristics of our original cohort at baseline (579 women and 419 men) have been reported (13,20). For the present analyses, participants who completed at least three of the four biannual cycles of testing (over a mean period of 6.3 years) were included. Tables 1 and 2 show the baseline clinical and exercise characteristics of these 592 participants compared with those of participants who were not able to complete at least three cycles of testing. Of note, of the 14% (n = 37) of women who were taking blood pressure medication, 33 had no other underlying health condition. If we consider these women to be free of underlying health condition, then 85% of women have no self-reported disease. The comparable figure for men was 97%. Participants included in the present study generally had better cardiovascular function than those excluded.


View this table:
[in this window]
[in a new window]
  		Table 1. Comparison of Participants Who Did and Did Not Complete at Least Three Biannual Cycles of Treadmill Exercise.

 

View this table:
[in this window]
[in a new window]
  		Table 2. Statistical Differences Between Participants Who Did and Did Not Complete at Least Three Biannual Cycles of Treadmill Exercise.

 
Tables 3 and 4 show cross-sectional, physical characteristics and exercise responses at each test cycle. Little within-sex variation in body composition occurred as the study progressed. Women have lower body surface area (BSA), body mass index (BMI), total body mass, lean mass, and lean/fat ratio than do men. Such sex differences in body composition contribute to the apparent sex differences in VO2peak. When normalized for lean body (muscle) mass instead of body weight (kg), the sex difference in VO2peak disappeared, and VO2peak uptake in women, expressed as a percentage of that achieved by men, increased from 77% to 95% (13).


View this table:
[in this window]
[in a new window]
  		Table 3. Participant General Characteristics at Each of Four Biannual Cycles of Treadmill Testing.

 

View this table:
[in this window]
[in a new window]
  		Table 4. Treadmill Exercise Outcomes.

 
Decline in Aerobic Capacity with Age
Tables 5 and 6 summarize the results for VO2peak and OUES. Models 1 (baseline data only) and 2 (data as repeated measures) are cross-sectional models (time is not modeled explicitly in Model 2). Models 3-6 are longitudinal models in which time is modeled explicitly. In addition to Time (Model 3), Models 4–6 successively add: age (Model 4), baseline FEV1 and change in FEV1, smoking and medications (Model 5), peak heart rate and change peak heart rate from baseline (Model 6).


View this table:
[in this window]
[in a new window]
  		Table 5. Peak Oxygen Consumption (ml O2/kg/min).

 

View this table:
[in this window]
[in a new window]
  		Table 6. Oxygen Uptake Efficiency Slope.

 
The cross-section parameter estimate for age (Model 1) largely reflects differences in the varying cohorts that make up the study participants. Model 2 mixes longitudinal and cohort effects together by treating data as repeated measures. The rate of decline of VO2peak with age is similar for men and women but, for OUES, is approximately 2-fold greater in men.

Model 3 estimates unadjusted longitudinal change (Time) in VO2peak and OUES over the average 6.3 years of the follow-up. For both measures, the decline for men is greater than for women, particularly for OUES. In Model 4, the cross-sectional (cohort) effects of age are added into the model (e.g., 0.23 ml O2 min–1 · kg–1 needs to be subtracted for each year of a participant's age over 65 years). There is virtually no change in the estimates of the magnitude of the longitudinal declines (Time), which appear to be larger than the cross-sectional effect (Age) (–0.39 vs –0.23 ml O2 · kg–1 · min–1 in women, –0.69 vs –0.34 ml O2 · kg–1 · min–1 in men). Model 5 adds a number of baseline covariates that result in little change in the longitudinal point estimate of decline (–0.39 to –0.36 ml O2 · kg–1 · min–1 in women; –0.69 to –0.58 ml O2 · kg–1 · min–1 in men) (–5.8 ml O2 · kg–1 · min–1 per decade for any given stratum implied in the model). Model 6 includes the change in peak heart rate at the end of the most recent exercise test from the value achieved at the end of the baseline test. With this inclusion, the estimate for longitudinal change (Time) in VO2peak in women is decreased by 39% (–0.36 to –0.22 ml O2 · kg–1 · min–1) and is similar for men (46%, Table 5, Model 6). Thus, the estimated rate of longitudinal decline in women is –0.22 ml O2 · kg–1 · min–1 · y–1 or 9.7% per decade and in men is –0.31 ml O2 · kg–1 · min–1 · y–1 or 10.4% per decade. Although fewer than 10% of participants were taking beta adrenergic blocking drugs for hypertension, the above analyses were repeated after censoring these participants. There was little effect on the decline in heart rate (Table 4) and little change in the regression results, most likely due to the small number of participants receiving these medications.

With regard to the cross-sectional parameter point estimates of VO2peak (Age), these also decrease successively as one progresses from Model 4 to Model 6 (from –0.23 to –0.16 to –0.08 ml O2 · kg–1 · min–1 in women; from –0.34 to –0.27 to –0.15 ml O2 · kg–1 · min–1 in men).

Similar results to the above for the variables of Time and Age are noted when OUES is used as the outcome variable (Table 6, Model 6), although the percentage changes in the age and time parameters are not as great as for VO2peak.

Table 7 provides sex-specific estimates of the percentage declines in VO2peak and OUES per decade. Each estimate represents the percentage decline for participants age 65 years at baseline not taking cardiovascular medications and whose FEV1, peak heart rate, as well as their change during the study were set to the sex-specific medians for each test round. Based on Models 3–5, the conclusion would be that longitudinal estimates of decline per decade are larger than cross-sectional estimates (Model 1). The inclusion of change in peak heart rate achieved during exercise results in approximately a 40% and 50% decline in the estimated per decade decline in VO2peak for women and men, respectively, compared to Model 5 where change in heart rate is not included. The effect of peak heart rate change is much smaller for OUES (Table 7).


View this table:
[in this window]
[in a new window]
  		Table 7. Per Decade Percent Declines in Peak Oxygen Consumption and Oxygen Utilization Efficiency Slope.

 
Inclusion of lean body mass, the ratio of lean/fat mass, and change in these variables during the course of the study resulted only in small changes in aerobic performance at baseline (1.8% and 1.3% per kg difference in lean body mass in women and men, respectively; 1.8% and 0.9% per 0.1 unit difference in the lean/fat ratio in women and men, respectively). Changes in lean body mass and lean/fat ratio during the 6.3 years of the study were small and did not affect the age-related decline in exercise performance. Inclusion of self-reported physical activity over the 4 years did not modify the observed age-related decline in exercise performance. Exclusion of the small number of participants with underlying health conditions (e.g., diabetes, lung conditions, cancer) had little effect on the results.


   DISCUSSION
Top
 Abstract
 Methods
 Results
 Discussion
Appendix A
 References
 
Few longitudinal studies have been performed to define age-related decline in aerobic fitness. Dehn and Bruce (23) suggested that the decline (25% per decade) in individuals studied serially was approximately double the rate of decline reported in cross-sectional analyses. In the study by Jackson and colleagues (8) of persons ages 25–70 years, cross-section estimates likely are confounded by cohort effects due to enrollment over an unspecified period of time; their estimates of longitudinal change are imprecise due to relatively little change over the study interval (4.1 years). Although Fleg and colleagues included individuals who were much younger than ours, the per decade estimates derived from our mixed Model 5 for VO2peak (Table 7) are very similar to their results for persons older than 50 years (see Figure 3 of reference 10). In both studies, percentage declines were greater for men than for women. Moreover, our study and the Fleg study show that estimates based on longitudinal analyses are larger than those based on cross-sectional data. Neither the study of Fleg and colleagues nor any of the other studies, however, evaluated the extent to which the estimated rates of decline were sensitive to the variables included in their regression models.

We have demonstrated that estimated rates of decline in VO2peak and OUES associated with aging are sensitive to variables included in the models. Estimated declines in VO2peak, based on longitudinal analysis, were 18% per decade in women and 24% per decade in men when only the longitudinal effects observed over the years of the study were considered (Model 3, Table 5). When FEV1 and peak exercise heart rate and their respective changes were included, however (Model 6, Table 5), aerobic capacity declined only 9.7% per decade in women and 10.4% per decade in men. Similar results were observed for OUES (Tables 5 and 6). Thus, the observed decline in aerobic capacity can be accounted for primarily by a decline in these two physiological variables. Of note, neither habitual physical exercise nor high aerobic capacity is known to modify the normal deterioration in resting FEV1 or lung function that occurs with age (24). The study by Fleg and colleagues also showed that high intensity physical activity did not affect patterns of decline in VO2peak, although levels at any age were ordered by degree of participation in high level activities (10).

On the basis of our data, we suggest that aging with respect to cardiopulmonary function, and, as with many other age-related processes [e.g., physical functioning and disability (25)], is a relative concept the interpretation of which depends on the types of variables included in models as well as the model that is chosen to characterize the change with age (cross-section vs longitudinal).

Our analysis provides an explanation for the differences between estimates of decline in aerobic capacity based on cross-sectional and longitudinal data. Estimates based on cross-sectional studies mix aging and cohort effects together, whereas those from longitudinal studies can separate the effects. In cross-sectional analyses, the survivor bias will tend to decrease the average estimate of decline compared to what would have been observed longitudinally. Therefore, if one is interested in an estimate of longitudinal change, then the Time variable in our models is that estimate. If one is interested in the overall effect of aging on aerobic decline, then the sum of the Age (cross-sectional) and Time variables provides that estimate.

Age-related decline in aerobic capacity, when not adjusted for other variables, was less in women than in men (e.g., 18% vs 24% per decade for VO2peak and 12% vs 20% per decade for OUES, Model 3, Table 7). This difference has been found in other studies (2,9,24,26). Adjustment for differences in body composition had little effect on these gender differences in the rates of decline. The decline in peak exercise heart rate with age appears to be the major cardiac determinant of the age-related decline in VO2max. Moreover, the lack of effect of physical activity in unselected older age groups in this process of decline would suggest that activity patterns early in life that maximize aerobic fitness may be particularly important in the preservation of adequate aerobic reserve in the oldest old.

One potential limitation on the interpretation of our results could be the occurrence of selection bias that is suggested by the loss of participants between the 3rd and 4th round of testing. When we dropped the round four data and included participants who completed all of the first three rounds, however, the results were unchanged. In addition, a more formal analysis to reweight the sample for various censoring patterns also gave nearly identical results. Therefore, we believe that the results for the full sample are valid. This view is reinforced by the quantitative and qualitative similarity of our results to those of Fleg and colleagues (10) who also used a population-based sample. Unfortunately, the participants in our study were predominantly white and middle class (as were those in the study of Fleg and colleagues) which does limit the findings of this study to these groups.

Summary
This study demonstrates that estimates of effects of aging on aerobic capacity are sensitive to the inclusion of physiological factors that also decline with age. In the present study, FEV1 and maximal exercise heart rate (together with their changes during the course of the study) accounted for most of the longitudinal decline of aerobic capacity with age. Our findings also corroborate, in a much larger cohort, the findings of others that the rate of decline is greater in men than in women—an observation that remains largely unexplained. In a more general context, quantitative estimates of aging of any physiological process, as measured in clinical and epidemiologic studies, represent the residual variance not explained by factors included in the study and/or any mismeasurement of those factors. The apparently larger estimates of decline observed in some longitudinal analyses likely represent incomplete selection of variables in the models used. Thus, attempts to reconcile differences and/or compare results between studies of the effects of aging on aerobic fitness need to recognize this heretofore neglected concept.


   APPENDIX A
Top
 Abstract
 Methods
 Results
 Discussion
 Appendix A
References
 
Statistical Details.-- Continuous baseline variables were centered on their median values. Longitudinal effects of interest were estimated with sex-specific repeated measures models (5)


Formula

where i indexes participant and j indexes trial number. xi1 is the vector of baseline covariates (age, BSA, median centered mean reproducible FEV1, and smoking status). yij represents a vector of the variables that captures the change at time j of FEV1 and peak heart rate from their respective baseline values. wij is the vector of time-dependent variables (time [years from baseline visit], use of cardiac medications known to affect exercise performance). ß0 is the estimated mean VO2peak/kg or OUES adjusted to the median values of baseline age, BSA, and FEV1 for nonsmokers who are not on any cardiac medications. {delta}B represents the cross sectional difference in average VO2peak/kg or OUES for groups that differ by 1 unit in xi1 at baseline, conditional on other covariates. {delta}L represents the expected longitudinal difference in VO2peak/kg or OUES per unit change in yij, conditional on other covariates. ß2 is the difference in average VO2peak/kg or OUES for groups that differ by 1 unit in wij over time, conditional on other covariates. The part of ß2 that is the coefficient for time captures changes in VO2peak/kg or OUES due to any other unmeasured variability specific to the settings of different rounds, conditional on all other covariates being held constant. The conditional longitudinal effect of aging on VO2peak/kg or OUES can be estimated through the parameter estimates for Age and Time as follows: {delta}ageAgei,baseline + ßtimeTimei. The coefficients for the cardiac medications represent the estimated difference in VO2peak/kg or OUES when switching between being on a drug not thought to influence VO2peak/kg or OUES (or not being on a drug at all) to being on beta blockers or cardiac drugs that influence exercise performance, conditional on all other covariates being held constant.

A baseline visit is defined as the first time that the participant performed treadmill exercise. The OUES regression slope for each participant was calculated by regression of VO2 on the log10 (VE) for each time j. The inverse of the variance of the slope estimates (1/Var(OUES)ij) was used as the weight in the repeated measures model to account for the differences in data available to fit each participant's OUES (a minimum of eight data points was required). Although longitudinal data were obtained for BSA, only baseline BSA was included in the model, as there was very little variability in the measure over time. A similar, repeated measures model was fit for VO2peak/kg/min. BSA was excluded in this latter model, because the measure is already normalized for body weight.

Descriptive and regression analyses were performed using SAS System statistical software (version 8.2; SAS Institute Inc., Cary, NC). Models were fit using the MIXED procedure with the repeated statement.


   Acknowledgments
Top
 Abstract
 Methods
 Results
 Discussion
 Appendix A
 References
 
The work was supported by grant R01 AG09389.

We acknowledge Donna Turner and Paula DiCamillo, who performed the treadmill exercise testing.


   Footnotes
Top
 Abstract
 Methods
 Results
 Discussion
 Appendix A
 References
 
Decision Editor: Luigi Ferrucci, MD, PhD

Received May 16, 2005

Accepted February 24, 2006


   References
Top
 Abstract
 Methods
 Results
 Discussion
 Appendix A
 References
 

  1. Heath GW, Hagberg JM, Ehsani AA, Holloszy JO. A physiological comparison of young and older endurance athletes. J Appl Physiol. 1981;51:634-640.[Abstract/Free Full Text]
  2. Inbar O, Oren A, Scheinowitz M, Rotstein A, Dlin R, Casaburi R. Normal cardiopulmonary responses during incremental exercise in 20- to 70-yr-old men. Med Sci Sports Exerc. 1994;26:538-546.[Medline]
  3. Rogers MA, Hagberg JM, Martin WH, 3rd, Ehsani AA, Holloszy JO. Decline in VO2max with aging in master athletes and sedentary men. J Appl Physiol. 1990;68:2195-2199.[Abstract/Free Full Text]
  4. Eskurza I, Donato AJ, Moreau KL, Seals DR, Tanaka H. Changes in maximal aerobic capacity with age in endurance-trained women: 7-yr follow-up. J Appl Physiol. 2002;92:2303-2308.[Abstract/Free Full Text]
  5. Katzel LI, Sorkin JD, Fleg JL. A comparison of longitudinal changes in aerobic fitness in older endurance athletes and sedentary men. J Am Geriatr Soc. 2001;49:1657-1664.[Medline]
  6. Pollock ML, Mengelkoch LJ, Graves JE, et al. Twenty-year follow-up of aerobic power and body composition of older track athletes. J Appl Physiol. 1997;82:1508-1516.[Abstract/Free Full Text]
  7. Astrand P, Bergh U, Kilbom A. A 33-yr follow-up of peak oxygen uptake and related variables of former physical education students. J Appl Physiol. 1997;82:1844-1852.[Abstract/Free Full Text]
  8. Jackson AS, Beard EF, Wier LT, Ross RM, Stuteville JE, Blair SN. Changes in aerobic power of men, ages 25–70 yr. Med Sci Sports Exerc. 1995;27:113-120.[Medline]
  9. Stathokostos L, Jacob-Johnson S, Petrella RJ, Paterson DH. Longitudinal changes in aerobic power in older men and women. J Appl Physiol. 2004;97:781-789.[Abstract/Free Full Text]
  10. Fleg JL, Morell CH, Bos AG, et al. Accelerated longitudinal decline of aerobic capacity in healthy older adults. Circulation. 2005;112:674-682.[Abstract/Free Full Text]
  11. Tager IB, Haight TJ, Hollenberg M, Satariano WA. Physical functioning and mortality in older women: an assessment of energy costs and level of difficulty. J Clin Epidemiol. 2003;56:807-813.[Medline]
  12. Okin PM, Ameisen O, Kligfield PA. A modified exercise protocol for computer-assisted analysis of the ST segment/heart rate slope: methods and reproducibility. J Electrocardiol. 1986;19:311-318.[Medline]
  13. Hollenberg M, Ngo LH, Turner D, Tager IB. Treadmill exercise testing in an epidemiologic study of aging and physical activity: application of the Cornell Protocol. J Gerontol Biol Sci. 1998;53A:B259-B267.[Abstract]
  14. American Thoracic Society. Standardization of spirometry, 1994 update. Am J Respir Crit Care Med. 1995;152:1107-1136.[Medline]
  15. Sorlie PD, Kannel WB, O'Connor G. Mortality associated with respiratory function and symptoms in advanced age. Am Rev Respir Dis. 1989;140:379-384.[Medline]
  16. Knuiman MW, James AL, Divitini ML, Ryan G, Bartholomew HC, Musk AW. Lung function, respiratory symptoms, and mortality: results from the Busselton Health Study. Ann Epidemiol. 1999;9:297-306.[Medline]
  17. Schunemann HJ, Dorn J, Grant BJB, Winkelstein W, Trevisan M. Pulmonary function is a long term-predictor of mortality in the general population: 29-year follow-up of the Buffalo Health Study. Chest. 2000;118:656-664.[Abstract/Free Full Text]
  18. Zureik M, Benetos A, Neukirch C, et al. Reduced pulmonary function is associated with central arterial stiffness in men. Am J Respir Crit Care Med. 2001;164:2181-2185.[Abstract/Free Full Text]
  19. Baba R, Nagashima M, Goto M, et al. Oxygen uptake efficiency slope: a new index of cardiorespiratory functional reserve derived from the relation between oxygen uptake and minute ventilation during incremental exercise. J Am Coll Cardiol. 1996;28:1567-1572.[Abstract]
  20. Hollenberg M, Tager IB. Oxygen uptake efficiency slope: an index of exercise performance and cardiopulmonary reserve requiring only submaximal exercise. J Am Coll Cardiol. 2000;36:194-201.[Abstract/Free Full Text]
  21. Sternfeld B, Ngo L, Satariano WA, Tager IB. Associations of body composition with physical performance and self-reported functional limitation in elderly men and women. Am J Epidemiol. 2002;156:110-121.[Abstract/Free Full Text]
  22. Diggle P, Liang KY, Zeger S. Analysis of longitudinal data. New York: Oxford University Press; 1994.
  23. Dehn MM, Bruce RA. Longitudinal variations in maximal oxygen intake with age and activity. J Appl Physiol. 1972;33:805-807.[Free Full Text]
  24. McClaran SR, Babcock MA, Pegelow DF, Reddan WG, Dempsey JA. Longitudinal effects of aging on lung function at rest and exercise in healthy active fit elderly adults. J Appl Physiol. 1995;78:1957-1968.[Abstract/Free Full Text]
  25. Tager IB, Haight T, Sternfeld B, Yu Z, van Der Laan M. Effects of physical activity and body composition on functional limitation in the elderly: application of the marginal structural model. Epidemiology. 2004;15:479-493.[Medline]
  26. Fleg JL, Lakatta EG. Role of muscle loss in the age-associated reduction in VO2 max. J Appl Physiol. 1988;65:1147-1151.[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 J. Appl. Physiol.Home page
K. F. Herspring, L. F. Ferreira, S. W. Copp, B. S. Snyder, D. C. Poole, and T. I. Musch
Effects of antioxidants on contracting spinotrapezius muscle microvascular oxygenation and blood flow in aged rats
J Appl Physiol, December 1, 2008; 105(6): 1889 - 1896.
[Abstract] [Full Text] [PDF]

Home page
 CirculationHome page
R. Arena, J. Myers, M. A. Williams, M. Gulati, P. Kligfield, G. J. Balady, E. Collins, and G. Fletcher
Assessment of Functional Capacity in Clinical and Research Settings: A Scientific Statement From the American Heart Association Committee on Exercise, Rehabilitation, and Prevention of the Council on Clinical Cardiology and the Council on Cardiovascular Nursing
Circulation, July 17, 2007; 116(3): 329 - 343.
[Full Text] [PDF]

Home page
 Journals of Gerontology Series A: Biological Sciences and Medical SciencesHome page
L. Ferrucci and E. M. Simonsick
A Little Exercise
J. Gerontol. A Biol. Sci. Med. Sci., November 1, 2006; 61(11): 1154 - 1156.
[Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Services
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
PubMed
Right arrow PubMed Citation

HOME ARCHIVE SEARCH TABLE OF CONTENTS