Relationships between walk/run performance and cardiorespiratory fitness in adolescents who are overweight. (Research Report).

Obesity during childhood has been identified as a major health problem in the United States. (1-3) Pediatric obesity commonly presages adult obesity (4) and is associated with the development of weight-related comorbid conditions and increased morbidity and mortality. (5-7)

Decreased physical activity and a sedentary lifestyle have been implicated as factors in the development of pediatric obesity. (8-10) Aerobic exercise is widely used in the management of obesity. (11,12) However, there is little information regarding how clinicians should evaluate cardiorespiratory fitness in children and adolescents who are overweight. We believe there is a need for a practical method of measuring fitness, without the requirement of specialized equipment or expertise, in adolescents who are overweight.

Walk/run tests, in which time or distance is the outcome measure, are generally accepted methods of reflecting cardiorespiratory fitness. (13-18) One-mile walk/run test performance is related to peak oxygen uptake (V[O.sub.2]peak) in children and adolescents who are not overweight. (16,17) Time-based walk/run tests of 9 and 12 minutes have been found to yield reliable measurements (with test-retest reliability coefficients of .92-.94) (13,19) and to have concurrent validity with measurements of V[O.sub.2]peak in children and adolescents who are not overweight. (13,14,19) Some investigators have speculated that the 12-minute walk/run test may better measure exercise tolerance among people than tests of shorter duration. (20) Other investigators (21,22) have reported that walk/ run tests of longer duration (eg, 12-15 minutes) versus shorter duration (eg, 5-9 minutes) are more reflective of V[O.sub.2]peak. However, Jackson and Coleman (14) have reported that there are no differences between the 9- and 12-minute tests' relationship with V[O.sub.2]peak. To our knowledge, there have been no systematic investigations of walk/run performance in adolescents who are overweight.

We examined the relationships among walk/run performance, cardiorespiratory fitness, and body composition in adolescents who were obese. We hypothesized that the distance traveled during a 12-minute walk/run test would be highly correlated with oxygen uptake (V[O.sub.2]), peak power (Wpeak), and power at the anaerobic threshold (WAT) measured during cycle ergometry. In addition, we hypothesized that the walk/run distance at 12 minutes would be more closely related to the oxygen uptake, Wpeak, and WAT than the distance at 9 minutes.

Method

Subjects

Ten Caucasian adolescents (5 male, 5 female) and 10 African-American adolescents (5 male, 5 female) who were obese were studied prior to participating in a weight loss study (Tab. 1). All participants were recruited from the greater Washington, DC, metropolitan area by the use of newspaper advertisements and through physician referrals of adolescents who were greatly overweight ([greater than or equal to] 95th percentile for age, sex, and race). Subjects were included only if they were in good general health, with the exception of a predefined list of obesity-related comorbid conditions such as hyperlipidemia and hyperinsulinemia. Subjects were excluded if they had used a suppressant medication within the past 6 months; were pregnant; had major pulmonary, hepatic, or cardiac disorders; or had recently lost weight. Before exercise testing commenced, a medical history and physical examination was done for each subject, and it included a cardiopulmonary examination and a 12-lead electrocardiogram. All subjects were free of musculoskeletal injury as determined by a physician, and American Heart Association guidelines for exercise testing were observed. (23) The subjects' parents provided signed consent statements (and children gave their assent) for all studies.

Each subject was seen on 2 occasions: during the first session, a screening physical and a 12-minute walk/run test were conducted, and, during the second session, a maximal cycle ergometry test and complete body composition studies were performed.

Walk/Run Testing Procedures

A single 12-minute walk/run test was performed. Before exercise, resting blood pressure and heart rate were measured. Each subject was encouraged to give his or her best effort--to walk and/or run to cover as much distance as possible in 12 minutes. A hallway, 47 m in length and 1.8 m in width, was used as the course.

Heart rate was measured and recorded at 3, 6, 9, and 12 minutes into the walk/run test using a heart rate telemeter with a wristwatch receiver (Polar Vantage NV *). The absolute highest heart rate during the walk/run test was defined as the peak walk/run heart rate. The average heart rate for the 12-minute test also was calculated. Total distance (in meters) traveled at 9 minutes ([D.sub.9]) and at 12 minutes ([D.sub.12]) was measured using a measuring wheel (model MM34 ([dagger])). From these 2 measurements, distance traveled between 9 and 12 minutes ([D.sub.9-12]) was determined.

Cycle Ergometry Testing Procedures

Subjects were studied in the afternoon after they ate. Following familiarization with cycle ergometer (Ergoline 800 ([double dagger])) testing procedures, resting heart rate and blood pressure were measured. Subjects were then encouraged to exercise to the limit of their self-determined tolerance and were instructed to maintain pedaling cadence between 60 and 70 revolutions per minute. Exercise began with a 4-minute warm-up period with no additional resistance applied to the pedals, followed by progressively increasing workloads of 15 to 20 W/min, until the subject was unable to continue (signaled by the subject with a "thumbs down" sign) or could no longer maintain the prescribed pedaling cadence. The rate of workload increase for each subject was selected based on predicted maximal power. (24)

Expired gas exchange was measured breath by breath during exercise using a metabolic cart (Sensormedics Vmax ([double dagger])). Peak oxygen uptake, which is the highest V[O.sub.2] achieved during exercise, and respiratory exchange ratio were calculated as average values during the last 20 seconds of exercise. Anaerobic threshold (AT) was determined using the V-slope method. (25) In order to scale V[O.sub.2]peak and AT among subjects, those measures were expressed relative to body weight as mL [O.sub.2]/kg/ min for data analysis. Power at AT (WAT) and at peak exercise (Wpeak) were also determined and recorded in watts per kilogram. Heart rate was measured by a 12-lead electrocardiogram (Sensormedics Max-1 ([double dagger])) during exercise. Blood pressure was measured every 3 minutes during exercise. Peak exercise rating of perceived exertion was measured within the first minute of exercise recovery using the 20-point Borg Rating of Perceived Exertion Scale.

Body Composition Assessment by Air Displacement Plethysmography

For determination of body composition, subjects were studied in the morning, after an overnight fast. They were instructed to void before measurements were obtained. Body density was assessed using an air displacement plethysmography body composition system (Badpod ([section])) according to the manufacturer's directions using the procedures previously described. (26)

Measurements were taken with all subjects wearing minimal clothing (either underwear or a tight-fitting bathing suit) and wearing a swim cap. Thoracic gas volume was measured during tidal breathing and during exhalation against a mechanical obstruction. The procedure has been described in detail elsewhere. (26) Percentage of body fat was determined from body density using the standard 2-compartment model. (27) Measurements of percentage of body fat were converted into kilograms of body fat before analysis by multiplying percentage of body fat and total body weight together. We did not examine the reliability of these measurements.

Statistical Analysis

Data were analyzed on a Macintosh PowerPC using Superanova 1.11 and StatView 4.5 software. ([parallel]) Simple linear regression was used to determine the relationships of [D.sub.9], [D.sub.12], and [D.sub.9-12] with each other, with body composition, with heart rate during the walk/run test, and with measurements derived from the cycle ergometer test (Tab. 2). To determine whether [D.sub.12] was superior as a predictor of cycle ergometry performance compared with [D.sub.9], 2 prediction models for V[O.sub.2]peak, V[O.sub.2]peak/kg lean body mass, AT, Wpeak, WAT, and peak respiratory exchange ratio were constructed--one containing Do and the other containing both [D.sub.9] and [D.sub.9-12]. For each variable of interest, we compared the amount of variance accounted for by these 2 models using the F test for analysis of concomitant variables. (28) This test determines whether inclusion of a factor, even one that may be correlated with other variables in the model, adds information and improves the ability of the model to predict the dependent variable.

To examine the importance of variables used in predicting fitness, 2 equations were used to calculate predicted values for V[O.sub.2]peak and AT. The V[O.sub.2]peak and AT values were predicted first from a previously published equation that included height and age (29) and second from a previously published equation that included weight and age. (30)

Results

All 20 subjects completed the 12-minute walk/run test. Eighteen subjects completed the maximal cycle ergometer test. One subject was unable to complete the maximal cycle ergometer test because of an inability to maintain a consistent pedaling cadence. Another subject stopped exercise soon after beginning because of "general fatigue." Normal exercise electrocardiographic and blood pressure responses were observed in all subjects during exercise and recovery.

Results from the 18 subjects who completed both the walk/run test and the cycle 'ergometer test are given in Table 2. Average ([+ or -]SD) 9- and 12-minute walk/run distances were 883.3 [+ or -] 162.2 m ([D.sub.9]) and 1,174.4 [+ or -] 208.1 m ([D.sub.12]), respectively. Mean heart rate rose between the initial pre-exercise level and 3 minutes of the walk/run test (from 83 bpm [SD=9.2, range=68-102] to 165 bpm [SD=18.1, range=118-200], P<.001), and remained stable thereafter (Fig. 1). Although there was some change in mean heart rate between 9 and 12 minutes of the walk/run test (from 165.0 bpm [SD=18.0, range=127-199] to 171 bpm [SD=13.0, range=140-192]), this increase was not statistically significant. The [D.sub.9], [D.sub.12], and [D.sub.9-12] measurements were not different between the male and female subjects and were not correlated with age. Both [D.sub.12] (P<.004) and [D.sub.9-12] (P <.008), but not [D.sub.9], were significantly correlated with height. The [D.sub.9], [D.sub.12], and [D.sub.9-12] measurements were each negatively correlated with total body weight, BMI, and body fat mass (Tab. 2, all r values between -.51 and -.82, P<.005). The [D.sub.9], [D.sub.12], and [D.sub.9-12] measurements were positively correlated with each other (Tab. 2, all r values between .64 and .97, P <.005). Average walk/run heart rate was related to [D.sub.9], and [D.sub.12], and peak walk/run heart rate was related to [D.sub.12] (Tab. 2, P<.05).

[FIGURE 1 OMITTED]

During the cycle ergometer test, mean peak respiratory exchange ratio was 1.21 [+ or -] 0.11, mean peak heart rate was 180 [+ or -] 9 bpm, and mean peak rating of perceived exertion (6-20 Borg scale) was 17.8 [+ or -] 1.5 (Tab. 2). Based on values predicted from age and body weight, mean V[O.sub.2]peak, and AT were 53% [+ or -] 21% and 60% [+ or -] 25% of predicted values, respectively (both P<.001). In contrast, there were no differences between measured and predicted V[O.sub.2]peak and AT when predicted values were determined using height and age.

Walk/run and cycle ergometer test results were found to be related to one another (Tab. 2). Measurements of V[O.sub.2]peak, V[O.sub.2]peak/kilograms of lean body mass, AT, Wpeak, and WAT were all positively related to [D.sub.9], [D.sub.12], and [D.sub.9-12] (Tab. 2). The strongest relationships were found between walk/run distance and either Wpeak (Tab. 2; all r [greater than or equal to] .82, P<.005) or V [O.sub.2] peak (Figs. 2A-2C; all r [greater than or equal to] .63, P<.005). Twelve-minute walk/run distance explained a greater percentage of variance than 9-minute distance for the following cycle ergometer variables: V[O.sub.2]peak, V[O.sub.2]peak/kilograms of lean body mass, AT, Wpeak, peak respiratory exchange ratio (Tab. 2). When [D.sub.9-12] (the distance from minute 9 to minute 12) was included in a multiple regression analysis, the distance traveled during the first 9 minutes of the test did not supply additional explanatory power for any of these variables (all P>.2; F test for analysis of concomitant variables).

[FIGURE 2 OMITTED]

When heart rate at the anaerobic threshold, as determined by the cycle ergometer test, was compared with heart rates during the walk/run test, 75% of subjects' walk/run heart rates exceeded heart rate at the AT by more than 10% at 9 minutes, and 100% of subjects' walk/run heart rates were more than 10% higher than heart rate at the AT at 12 minutes.

Discussion

In order to determine whether the 12-minute walk/run test can serve as a practical means of measuring cardiorespiratory fitness for adolescents who are overweight, we examined the ability of walk/run distance to predict results from a maximal cycle ergometer test. We found that 9-minute and 12-minute walk/run distances were positively correlated with measurements of cardiorespiratory fitness, such as V[O.sub.2]peak, AT, Wpeak, and WAT, and were negatively correlated with measurements of body adiposity, including body weight, BMI, and body fat mass. These findings appear to support and extend previous reports of the relationships between walk/run performance and cardiorespiratory fitness in children and adolescents who were not overweight (13,14,16,17,19,21) and the relationships between walk/run performance and body composition in children who were not overweight. (15,31)

We found walk/run distances to be substantially less than those reported for people who were not obese. Other researchers (14,15,19,32) have reported 12-minute walk/run distances ranging from approximately 1,800 m to 2,500 m in children and adolescents. The mean 12-minute walk/run distance we found was 1,174 m, and no subject exceeded 1,500 m. Some authors (33,34) have concluded that cardiorespiratory fitness is decreased during exercise in adults and children who are obese compared with people who are not obese. Yet, Cooper et al (35) reported that the majority of children who were obese in their study had a normal V[O.sub.2]peak compared with predicted norms. Other researchers (36,37) have concluded that cardiorespiratory fitness is not decreased in children and adolescents who are obese but that walk/ run performance may be decreased as a consequence of the increased energy cost of work associated with the load caused by adipose tissues.

Although subjects in our study had an average V[O.sub.2]peak that was less than 60% of values predicted from age and body weight, V[O.sub.2]peak was 92% of values predicted from age and height. We believe that we studied an insufficient number of subjects to determine the possibility of decreased cardiorespiratory fitness in the larger population. Given that a large proportion (~50%) of the variance in walk/run performance could be explained by variation in body mass, however, our results are consistent with suggestions that differences in walk/run performance among children who were obese and children who are not obese are largely a function of the load caused by adipose tissue.

Epstein et al (38) compared the V[O.sub.2] associated with carrying 2 different backpack loads during prolonged treadmill walking at a constant speed. They found that the oxygen cost of carrying a backpack was constant over time for a 25-kg load. Carrying a 40-kg backpack induced a continual increase in V[O.sub.2] over time, which became greater than the V[O.sub.2] for the 25-kg backpack during the latter stages of exercise. Epstein et al hypothesized that large loads, eliciting greater than 50% V[O.sub.2]peak, alter locomotion biomechanics and lead to an increase in energy costs over time and subsequent physical fatigue. The adolescents we examined who were greatly overweight all carried more than 30 kg of adipose tissue, and 70% had body fat mass in excess of 40 kg. We believe that our subjects have altered biomechanics during locomotion. This may, at least in part, account for their poor performance on the walk/run test.

Oxygen uptake varies only with work rates below the AT, whereas it varies with both work rate and time at exercise intensities above the AT. (39) Probably not coincidentally, Epstein et al (38) reported an increase in the oxygen cost of work over time at exercise intensities greater than 50% of V[O.sub.2]peak. This is the exercise intensity at which AT usually occurs. (24) In addition, some authors (39) believe the increase in V[O.sub.2] over time for a given work rate is greater in people who are less fit. Although the precise cause of this phenomenon is unknown, it may have direct implications for tests such as the walk/run test. Walk/run tests of longer duration may better discriminate cardiorespiratory fitness among people.

Butland et al (20) found greater variability in 12-minute walk/run distance compared with 2- and 6-minute test results for the same subjects. They speculated that the increased variability in 12-minute test results represented greater discrimination of exercise tolerance among subjects. Because they did not measure V[O.sub.2], they could not offer strong conclusions in this regard. Other researchers (21,22) have found that walk/run tests of longer duration are more strongly related to V[O.sub.2]peak. In our study, we found that differences in V[O.sub.2]peak were better distinguished using the 12-minute distance than the 9-minute distance (Tab. 2).

When the 9-minute distance ([D.sub.9]) was included in a regression model with [D.sub.9-12], [D.sub.9] did not explain any further variance in cardiorespiratory variables. We hypothesize that the improved ability of the 12-minute test to predict cardiorespiratory fitness may be due to the increased relative aerobic requirement associated with the longer test. We found evidence that most of our subjects performed the latter portion of the walk/run test at intensities above AT where an increase in V[O.sub.2] over time is likely to occur. Therefore, as a result of having less aerobic reserve, subjects who were less fit could be predisposed to decrease the distance traveled during the latter portion of the 12-minute test. We, however, did not perform a separate 9-minute walk/run test, and such testing might yield a different result.

We believe that it is unlikely that the total energy cost of work performed during the walk/run test would be reflected by V[O.sub.2] alone. Most subjects exceeded their heart rate at the anaerobic threshold (determined from the cycle test) during the walk/run test, indicating an anaerobic energy contribution to walk/run performance. Accordingly, Wpeak and WAT explained greater variance in walk/run distance than V[O.sub.2] peak and AT. Peak power has been shown to be a good predictor of endurance in adults. (40) Our findings suggest that power expressed relative to body mass may be a better predictor of endurance performance than V[O.sub.2] in adolescents who are overweight. One possible explanation for our results may be that both aerobic and anaerobic energy expenditure are represented by the Wpeak and WAT measurements.

Conclusion

Our results suggest that an easily obtained measure of locomotor performance (distance traveled during a 12-minute walk/run test) is related to cardiorespiratory fitness and to body composition in adolescents who are obese. The 12-minute walk/run distance is more predictive of cycle ergometry test results than the 9-minute distance, and the 12-minute test may be preferable as a measure of cardiorespiratory fitness when measurements of V[O.sub.2] are not available. Adolescents who are overweight may exceed their anaerobic thresholds during the walk/run test. In our opinion, the effect of an increase in V[O.sub.2] over time is likely to be an important contributor to the relationship between walk/run performance and cardiorespiratory fitness. Additional investigation is needed with direct measures of V[O.sub.2] during the walk/run test to confirm this hypothesis. Further investigations, using separate walk/run tests of varying duration, are needed to determine more exact relationships between walk/run test duration and cardiorespiratory fitness.

Table 1.
Subject Demographics (a)

                         All Subjects (8 M, 10 F)

Variable                 [bar]X    SD      Range

Age (y)                    14.5    2.0     12-17
Weight (kg)               119.5   42.2   72.3-194.5
Height (cm)               166.4    9.9    150-191
BMI (kg/[m.sup.2])         42.9   11.5   28.6-64.2
Percentage of fat (%)      48.6    5.3   40.3-60.4

                         African American (3 M, 5 F)

Variable                  [bar]X     SD      Range

Age (y)                  14.4        1.8     12-16
Weight (kg)             141.0       40.7   77.6-194.5
Height (cm)             167.0        9.8    150-180
BMI (kg/[m.sup.2])       49.8 (b)   10.4   34.5-64.2
Percentage of fat (%)    51.6 (b)    5.4   43.7-60.4

                          Caucasian (5 M, 5 F)

Variable                 [bar]X    SD      Range

Age (y)                    14.6    2.3     12-17
Weight (kg)               102.3   36.7   72.3-174.6
Height (cm)               166.0   10.5    156-191
BMI (kg/[m.sup.2])         36.2    8.7   28.6-57.0
Percentage of fat (%)      45.5    3.9   40.3-54.0

(a) M=male, F=female, BMI=body mass index.

(b) P<.05 versus Caucasian.
Table 2.
Results of Walk/Run and Cycle Ergometry Tests

                             [bar]X      SD      Range

Anthropometrics
  Weight (kg)                 119.5     42.2    72.3-194.5
  Fat mass (kg)                59.1     25.8    30.4-108.3
  BMI (d) (kg/[m.sup.2])       42.9     11.5    28.6-64.2

Walk/run test
  [D.sub.9] (m)               883.3    162.2     545-1,120
  [D.sub.12] (m)            1,174.4    208.1     791-1,499
  [D.sub.9-12] (m)            291.2     62.6     161-428
  Mean walk/run heart
    rate (bpm)                165.9     15.4     125-188
  Peak walk/run heart
    rate (bpm)                177.5     15.9     140-200

Cycle ergometer test
  V[O.sub.2]peak (g) (mL
    [O.sub.2]/kg/min)          17.4      3.6    12.4-27.1
  V[O.sub.2]peak/kg lean
    body mass (mL
    [O.sub.2]/kg lean
    body mass/min)             33.4      4.9    25.2-45.7
  Anaerobic threshold (mL
    [O.sub.2]/kg/min)          10.6      2.1     7.2-14.0
  Peak power (W/kg)             1.43     0.42    0.9-2.1
  Power at anaerobic
    threshold (W/kg)            0.64     0.23    0.3-1.0
  Peak heart rate (bpm)       179.7      9.38    162-200
  Heart rate at anaerobic
    threshold (bpm)           139.8     11.0     121-154
  Peak respiratory
    exchange ratio              1.21     0.11   1.05-1.35
  Peak rating of
    perceived exertion         17.8      1.5      14-20

                            Correlation   Correlation   Correlation
                               With          With           With
                             [D.sub.9]    [D.sub.12]    [D.sub.9-12]
                                (a)           (b)           (c)

Anthropometrics
  Weight (kg)                 -.73 (e)      -.79 (e)       -.74 (e)
  Fat mass (kg)               -.74 (e)      -.79 (e)       -.70 (e)
  BMI (d) (kg/[m.sup.2])      -.60 (e)      -.82 (e)       -.51 (e)

Walk/run test
  [D.sub.9] (m)                              .97 (e)        .64 (e)
  [D.sub.12] (m)               .97 (e)                      .80 (e)
  [D.sub.9-12] (m)             .64 (e)       .80 (e)
  Mean walk/run heart
    rate (bpm)                 .52 (f)       .53 (f)        .39
  Peak walk/run heart
    rate (bpm)                 .47           .49 (f)        .42

Cycle ergometer test
  V[O.sub.2]peak (g) (mL
    [O.sub.2]/kg/min)          .63 (e)       .72 (e)        .75 (e)
  V[O.sub.2]peak/kg lean
    body mass (mL
    [O.sub.2]/kg lean
    body mass/min)             .50 (f)       .60 (e)        .72 (e)
  Anaerobic threshold (mL
    [O.sub.2]/kg/min)          .53 (f)       .58 (e)        .57 (f)
  Peak power (W/kg)            .82 (e)       .88 (e)        .82 (e)
  Power at anaerobic
    threshold (W/kg)           .70 (e)       .72 (e)        .60 (f)
  Peak heart rate (bpm)        .47           .50 (f)        .46
  Heart rate at anaerobic
    threshold (bpm)            .43           .44            .34
  Peak respiratory
    exchange ratio             .48 (f)       .55 (f)        .57 (f)
  Peak rating of
    perceived exertion         .17           .25            .39

(a) [D.sub.9]=distance traveled at 9 minutes.

(b) [D.sub.12]=distance traveled at 12 minutes.

(c) [D.sub.9-12]=distance traveled between 9 and 12 minutes.

(d) BMI=body mass index.

(e) P<.005, simple linear regression between variables.

(f) P<.05, simple linear regression between variables.

(g) V[O.sub.2]peak=peak oxygen uptake.

* Polar Electro Inc, 370 Crossways Park Dr, Woodbury, NY 11797.

([dagger]) Rolotape Carp, 2701 N Van Marter Dr, Spokane, WA 99206.

([double dagger]) Sensormedics Corp, 22705 Savi Ranch Pkwy, Yorba Linda, CA 92687.

([double dagger]) Sensormedics Corp, 22705 Savi Ranch Pkwy, Yorba Linda, CA 92687.

([section]) Life Measurement Instruments, 1980 Olivera Rd, Ste C, Concord, CA 94520.

([parallel]) Abacus Concepts Inc, 1984 Bonita Ave, Berkeley, CA 94704.

References

(1) Freedman DS, Srinivasan SR, Valdez RA, et al. Secular increases in relative weight and adiposity among children over two decades: the Bogalusa Heart Study. Pediatrics. 1997;99:420-426.

(2) Troiano RP, Flegal KM. Overweight children and adolescents: description, epidemiology, and demographics. Pediatrics. 1998;101 (suppl 3):497-504.

(3) Must A, Strauss RS. Risks and consequences of childhood and adolescent obesity. Int J Obes Relat Metab Disord. 1999;23(suppl 2):S2-S11.

(4) Whitaker RC, Wright JA, Pepe MS, et al. Predicting obesity in young adulthood from childhood and parental obesity. N Engl J Med. 1997; 337:869-873.

(5) Must A, Jacques PF, Dallal GE, et al. Long-term morbidity and mortality of overweight adolescents: a follow-up of the Harvard Growth Study of 1922 to 1935. N Engl J Med. 1992;327:1350-1355.

(6) Kuczmarski RJ, Flegal KM, Campbell SM, Johnson CL. Increasing prevalence of overweight among US adults: the National Health and Nutrition Examination Surveys, 1960 to 1991. JAMA. 1994;272: 205-211.

(7) Must A. Morbidity and mortality associated with elevated body weight in children and adolescents. Am J Clin Nutr. 1996;63(suppl 3): 445S-447S.

(8) Schlicker SA, Borra ST, Reagan C. The weight and fitness status of United States children. Nutr Rev. 1994;52:11-17.

(9) Gortmaker SL, Must A, Sobol AM, et al. Television viewing as a cause of increasing obesity among children in the United States, 1986-1990. Arch Pediatr Adolesc Med. 1996;150:356-362.

(10) Goran MI, Gower BA, Nagy TR, Johnson RK. Developmental changes in energy expenditure and physical activity in children: evidence for a decline in physical activity in girls before puberty. Pediatrics. 1998;101:887-891.

(11) Epstein LH, Coleman KJ, Myers MD. Exercise in treating obesity in children and adolescents. Med Sci Sports Exerc. 1996;28:428-435.

(12) Epstein LH, Goldfield GS. Physical activity in the treatment of childhood overweight and obesity: current evidence and research issues. Med Sci Sports Exerc. 1999;31 (suppl 11):S553-S559.

(13) Doolittle TL, Bigbee R. The Twelve Minute Run-Walk: a test of cardiorespiratory fitness of adolescent boys. Res Q. 1968;39:491-495.

(14) Jackson AS, Coleman AE. Validation of distance run tests for elementary school children. Res Q. 1976;47:86-94.

(15) Watson AW. Quantification of the influence of body fat content on selected physical performance variables in adolescent boys. Ir J Med Sci. 1988;157:383-384.

(16) Cormack WP, Cureton KJ, Bullock TA, Weyand PG. Metabolic determinants of 1-mile run/walk performance in children. Med Sci Sports Exerc. 1991;23:611-617.

(17) Cureton KJ, Sloniger MA, O'Bannon JP, et al. A generalized equation for prediction of VO2peak from 1-mile walk/run performance Med Sci Sports Exerc. 1995;27:445-451.

(18) Steele B. Timed walking tests of exercise capacity in chronic cardiopulmonary illness. J Cardiopulm Rehabil. 1996;16:25-33.

(19) Doolittle TL, Dominic JA, Doolittle J. The reliability of selected cardiorespiratory endurance field tests with adolescent female populations. Am Correct Ther J. 1969;23(5):135-138.

(20) Butland RJA, Pang J, Gross ER, et al. Two-, six-, and 12-minute walking tests in respiratory disease. BMJ. 1982;284:1607-1608.

(21) MacNaughton L, Croft R, Pennicott J, Long T. The 5 and 15 minute runs as predictors of aerobic capacity in high school students. J Sports Med Phys Fitness. 1990;30:24-28.

(22) Bernstein ML, Despars JA, Singh NP, et al. Reanalysis of the 12-minute walk in patients with chronic obstructive pulmonary disease. Chest. 1994;105:163-167.

(23) Washington RL, Bricker JT, Alpert BS, et al. Guidelines for exercise testing in the pediatric age group. Circulation. 1990;90:2166-2179.

(24) Wasserman K, Hansen JE, Sue DY, et al. Principles of Exercise Testing and Interpretation. Philadelphia, Pa: Lea & Febiger; 1994.

(25) Beaver WL, Wasserman K, Whipp BJ. A new method for detecting anaerobic threshold by gas exchange. J Appl Physiol. 1986;60: 2020-2027.

(26) Dempster P, Aitkens S. A new air displacement method for the determination of human body composition. Med Sci Sports Exerc. 1995;27:1692-1697.

(27) Siri WE. Body composition from fluid spaces and density: analysis of methods. In: Brozek J, Henschel A, eds. Techniques for Measuring Body Composition: Proceedings of a Conference, Quartermaster Research and Engineering Center, Natick, Massachusetts, January 22-23, 1959. Washington DC: National Academy of Sciences/National Research Council; 1961: 223-224.

(28) Rao CJ. Linear Statistical Inference and Its Applications. 2nd ed. New York, NY: John Wiley & Sons Inc; 1973:290.

(29) Cooper DM, Weiler-Ravell DJ. Gas exchange response to exercise in children. Am Rev Respir Dis. 1984;129(2 pt 2):S47-S48.

(30) Cooper DM, Weiler-Ravell D, Whipp BJ, Wasserman K. Aerobic parameters of exercise as a function of body size during growth in children. J Appl Physiol. 1984;56:628-634.

(31) Cureton KJ, Baumgartner TA, McManis BG. Adjustment of one mile walk test scores for skinfold thickness in youth. Ped Exerc Sci. 1991;3:152-167.

(32) Maksud MG, Coutts KG. Application of the Cooper twelve-minute run-walk test to young males. Res Q. 1971;42:54-59.

(33) Zanconato S, Baraldi E, Santuz P, et al. Gas exchange during exercise in obese children. Eur J Pediatr. 1989;148:614-617.

(34) Reybrouck T, Mertens L, Schepers D, et al. Assessment of cardiorespiratory exercise function in obese children and adolescents by body mass-independent parameters. Eur J Appl Physiol Occup Physiol. 1997;75: 478-483.

(35) Cooper DM, Poage J, Barstow TJ, Springer C. Are obese children truly unfit? Minimizing the confounding effect of body size on the exercise response. J Pediatr. 1990;116:223-230.

(36) Rowland TW. Effects of obesity on aerobic fitness in adolescent females. Am J Dis Child. 1991;145:764-768.

(37) Maffeis C, Schena F, Zaffanello M, et al. Maximal aerobic power during running and cycling in obese and non-obese children. Acta Paediatr. 1994;83:113-116.

(38) Epstein Y, Rosenblum J, Burstein R, Sawka MN. External load can alter the energy cost of prolonged exercise. Eur J Appl Physiol Occup Physiol. 1988;57:243-247.

(39) Gaesser GA, Poole DC. The slow component of oxygen uptake kinetics in humans. In: Holloszy JO, ed. Exercise and Sport Sciences Reviews. Vol 24. Baltimore, Md: Williams & Wilkins; 1996: chap 2.

(40) Balmer J, Davison RC, Bird SR. Peak power predicts performance power during and outdoor 16.1-km cycling time trial. Med Sci Sports Exerc. 2000;32:1485-1490.

B Drinkard, PT, MSPT, CCS, is Senior Physical Therapist, Rehabilitation Medicine Department, Warren Grant Magnuson Clinical Center, National Institutes of Health (NIH), Bethesda, MD 20892 (USA) ([e-mail]). Address all correspondence to Mr Drinkard.

J McDuffie, PhD, is Postdoctoral Fellow, Unit on Growth and Obesity, Developmental Endocrinology Branch, National Institute of Child Health and Human Development (NICHHD), NIH.

S McCann, BA, is Research Trainee, Unit on Growth and Obesity, Developmental Endocrinology Branch, and Division of Digestive Diseases and Nutrition, National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK), NIH.

G Uwaifo, MD, is Clinical Fellow, Unit on Growth and Obesity, Developmental Endocrinology Branch.

J Nicholson, MD, is Clinical Research Trainee, Unit on Growth and Obesity, Developmental Endocrinology Branch.

JA Yanovski, MD, PhD, is Chief, Unit on Growth and Obesity, Developmental Endocrinology Branch.

Dr Yanovski and Mr Drinkard are commissioned officers in the US Public Health Service.

Mr Drinkard, Dr McDuffie, and Dr Yanovski provided concept/research design. Mr Drinkard, Dr McDuffie, Dr Uwaifo, and Dr Yanovski provided writing. Mr Drinkard, Dr McDuffie, Dr Nicholson, and Dr Yanovski provided data collection, and Dr Yanovski, Dr McDuffie, and Ms McCann provided data analysis. Dr Yanovski, Ms McCann, and Dr Uwaifo provided project management. Dr Yanovski provided fund procurement, subjects, and institutional liaisons. Mr Drinkard provided facilities/equipment. The authors thank Susan Mihans, PT, for her assistance in conducting the study.

This study was approved by the National Institute of Child Health and Human Development Institutional Review Board.

This work was supported by a grant from the National Institute of Child Health and Human Development (Z-01-HD-04-00641) to Dr Yanovski.

This work was presented at the October 2000 meeting of the North American Association for the Study of Obesity, Long Beach, Calif.

The opinions expressed in this article are those of the authors and not necessarily those of the US Public Health Service or the National Institutes of Health.

This article was submitted October 31, 2000, and was accepted March 27, 2001.