International Journal of Obesity (2000) 24, 566±572
ß 2000 Macmillan Publishers Ltd All rights reserved 0307±0565/00 $15.00
www.nature.com/ijo
The effects of 18 months of intermittent vs
continuous exercise on aerobic capacity, body
weight and composition, and metabolic ®tness
in previously sedentary, moderately obese
females
JE Donnelly1*, DJ Jacobsen1, K Snyder Heelan1 , R Seip2 and S Smith3
1
Department of Health, Sport and Exercise Sciences, 104 Robinson Center, University of Kansas, Lawrence, KS 66045, USA; 2 Human
Performance Laboratory, University of Nebraska-Kearney, Kearney, NE 68849, USA; and 3 Platte Valley Medical Clinic, 3320 Avenue A,
Kearney, NE 68848, USA
OBJECTIVES: To compare the effects of 18 months of continuous vs intermittent exercise on aerobic capacity, body
weight and composition, and metabolic ®tness in previously sedentary, moderately obese females.
DESIGN: Randomized, prospective, long-term cohort study. Subjects performed continuous exercise at 60 ± 75% of
maximum aerobic capacity, 3 days per week, 30 min per session, or exercised intermittently using brisk walking for
two, 15 min sessions, 5 days per week.
MEASURES: Aerobic capacity, body weight, body composition, and metabolic ®tness (blood pressure, lipids, glucose
and insulin).
RESULTS: Signi®cant improvements for aerobic capacity of 8% and 6% were shown for the continuous and
intermittent exercise groups, respectively. Weight loss for the continuous exercise group was signi®cant at 2.1%
from baseline weight and the intermittent group was essentially unchanged. The continuous group showed a
signi®cant decrease in percentage of body fat and fat weight while the intermittent group did not. HDL cholesterol
and insulin were signi®cantly improved for both groups.
CONCLUSIONS: In previously sedentary, moderately obese females, continuous or intermittent exercise performed
long-term may be effective for preventing weight gain and for improving some measures of metabolic ®tness.
International Journal of Obesity (2000) 24, 566±572
Keywords: long-term; exercise; body weight; body composition; metabolic ®tness
Introduction
Obesity is a multi-component, chronic disease affecting approximately 35% of the adult population.1
Obesity is associated with comorbidities including
cardiovascular disease, hypertension, diabetes, orthopedic and gait abnormalities, and some cancers.2,3
Additionally, obese individuals appear to suffer from
prejudice and discrimination in the workplace and in
social relationships.4,5
Obese individuals who reduce weight generally
improve risk factors.6,7 Treatment for weight reduction varies; however, ultimately a negative energy
balance must be established if weight is to decrease.
Energy restriction may be successful for weight loss
in the short term; however, weight maintenance is not
sustained by the majority.8,9 Energy expenditure from
exercise may be capable of inducing a state of
*Correspondence: E Donnelly, Department of Health, Sport and
Exercise Sciences, 104 Robinson Center, University of Kansas,
Lawrence, KS 66045, USA.
E-mail: [email protected]
Received 8 June 1999; revised 18 November 1999; accepted
23 November 1999
negative energy balance although results from published meta-analytical studies are not impressive with
only 0.3 kg of weight loss for females and 1.3 kg for
males during 16 weeks of exercise.10 It is possible that
studies which have used exercise to induce negative
energy balance are too short in duration. Bouchard has
suggested that it may take up to 2 years for a
previously sedentary, moderately obese individual to
be able to attain enough volume of exercise to be
effective as a treatment for obesity.11
In addition to the short duration of most exercise
studies, veri®cation of exercise is frequently lacking.
Many studies using exercise to produce weight loss
have relied on unveri®ed, out-patient methodology. It
is possible that subjects do not participate in the
prescribed amount of exercise and therefore do not
achieve the desired amount of weight loss. For example, Jakicic et al found over-reporting of exercise in
moderately obese females when records were compared to accelerometers.12 Those who lost the least
weight over-reported to the greatest extent.
Exercise of suf®cient volume to potentially alter
body weight and that can be sustained by the previously sedentary, moderately overweight adult has
generally been delivered in measured doses using the
The effects of 18 months of intermittent vs continuous exercise
JE Donnelly et al
prescription criteria of frequency, intensity, and duration.13 A newer alternate recommendation is to have
the individual accumulate exercise throughout the day
by substituting physically active behaviors for sedentary behaviors.14 A frequently cited example is taking
the stairs rather than the elevator. In addition, it is
thought that accumulating exercise intermittently
many convey some advantage of convenience compared to obtaining exercise in one continuous session;
however, a direct test of this concept has not occurred.
Due to the scarcity of long-term studies of exercise,
the effects on weight loss, body composition, and
metabolic ®tness are largely unknown. The purpose
of this investigation was to compare 18 months of
a traditional, prescribed program of continuous
exercise, to a program of intermittent exercise. We
hypothesized that both programs would be suf®cient
to alter body weight, body composition and measures
of metabolic ®tness (lipids, insulin, glucose, blood
pressure) in previously sedentary, moderately overweight females.
Methods
Subjects
Twenty-two subjects gave written informed consent
for this study which was approved by the University
of Nebraska at Kearney Institutional Review Board.
We chose subjects with a body mass index above 25
as these subjects were likely to have low aerobic
capacity and represented a population at risk for
continued weight gain.1,15 No subject was currently
engaged in a regular exercise program as determined
by physical activity recall questionnaire16 or had an
oxygen consumption above the `Fair Category'
according to the American Heart Association Fitness
classi®cations.17
Exercise groups
Subjects were randomized to continuous (CON,
n ˆ 11) or intermittent exercise (INT, n ˆ 11). CON
exercised for 30 min, at 60% to 75% of maximal
aerobic capacity, 3 times per week, with direct supervision in the Human Performance Laboratory, at the
University of Nebraska-Kearney. INT was instructed
to walk briskly, yet comfortably at approximately
50 ± 65% HRR, 2 times per day, 15 min per session,
5 days per week, at their home or work site. A
minimum of 2 hours elapsed between exercise sessions. By design, the two programs did not provide the
same amount of exercise. Rather, they were structured
to provide exercise according to the traditional ACSM
recommendation and the newer recommendation
issued jointly by the CDC and ACSM. Thus, these
moderately obese females received 2 different clinical
doses of exercise, delivered as they might normally
occur.
567
Adherence to exercise
For each exercise session, distance walked, heart rate
at the end of exercise, duration of exercise, and RPE
were recorded for both exercise groups. Research
assistants recorded these values after each session
completed in the Human Performance Laboratory
for CON. For INT, supervision was provided at the
subject's exercise site on a random schedule, 2 times
per week to ascertain compliance. The schedule for
visits from a research assistant was unknown to the
subject prior to arrival. This method for supervision of
outpatient exercise studies has been described previously.18
Assessments
A physical exam and health history questionnaire
were completed at baseline. All laboratory and behavioral measurements were obtained at baseline, 9
months, and 18 months. Laboratory tests included
body composition analysis, exercise tolerance test,
and blood chemistry. Following the laboratory measures, on a separate day, a 3-day diet record was
obtained.
Aerobic capacity
To measure aerobic capacity, the subject walked on a
motor-driven treadmill for 5 min to provide acclimation to the treadmill. The subject then sat quietly until
the heart rate was within 10 bpm of the resting value.
Subsequently, the subject walked to volitional exhaustion. Maximal oxygen consumption was considered
as the highest observed value19,20 using a modi®ed
Balke protocol with 3 min stages.21 Heart rates were
recorded at the end of each stage and at maximal
exertion. Prior to exercise, with the subjects seated for
5 min, blood pressure was measured until two systolic
measures were within 6 mmHg and two diastolic
measures were within 4 mmHg.22 Blood pressures
were recorded during the last 30 seconds of each
stage. Expired air was measured for oxygen and
carbon dioxide at one minute intervals using a Sensormedics MMC Horizon system calibrated before
each test according to the speci®cations of the manufacturer (Sensormedics Corp., Yorba Linda, CA).
Body composition and regional adiposity
Hydrostatic weighing (HW) at residual volume was
used to estimate percent body fat.23 Residual volume
was measured immediately before body density measurement by the method of Wilmore et al.24 Body
density was calculated by using the equation of
Goldman and Buskirk,25 and percent body fat was
calculated with equation by Brozek et al.26 To estimate regional adiposity, circumferences were taken in
triplicate and the closest 2 measures were averaged
for subsequent calculations. Circumference measures
were taken at the widest girth of the hip and the
smallest girth of the waist.27
International Journal of Obesity
The effects of 18 months of intermittent vs continuous exercise
JE Donnelly et al
568
Blood chemistry
All blood samples were analyzed by a laboratory
using procedures standardized by the Center for Disease Control. Blood samples were obtained following
an overnight, 12 hour fast. For the oral glucose
tolerance test (OGTT), blood glucose and insulin
samples (5 ml each) were drawn from an indwelling
catheter just before ingesting a 75 gm glucose drink
(fasting), and subsequently 30, 60, 90, 120, 150 and
180 min after ingestion. During the 9 and 18-month
testing periods, the OGTT was standardized to 14 ± 18
hours post exercise. Area under the curve analysis was
calculated by the trapezoidal method as described by
Allison.28 Blood samples for lipid analysis (10 ml)
were drawn only at baseline (fasting). After the
sample was drawn, the blood was spun in a centrifuge
for 5 min. Subsequently, the serum was drawn off and
placed in a small tube and then kept on ice until frozen
at ÿ70 Celsius. Serum cholesterol and triglyceride
concentrations were measured with an automated
analyzer (Du Pont Co.), using standard enzymatic
techniques. HDL was measured after removal of
VLDL and LDL from samples by precipitation with
phosphotungstate.29 Glucose was measured using an
autoanalyzer (Beckman) and insulin was measured
using a double-label antibody technique.30
Energy intake
Energy and macronutrient intakes were measured with
3-day food records (2 weekdays and 1 weekend day)
at baseline, 9 months and 18 months. Twenty-four
hour recalls were completed at 3, 6, 12, and 15
months. Subjects were instructed in recording brand
name, portion size, method of preparation and ingredients and were instructed not to purposefully reduce
energy intake. Analysis of energy and macronutrient
consumption was completed using The Food Processor1 computer program (version 4.0).31
Data analysis
Descriptive statistics, mean, minimum, maximum,
and standard deviation were calculated for all dependent variables. Descriptive statistics were also calculated for demographic variables such as age. The basic
study design was a two factor (time by treatment
group) repeated measures (time) ANOVA. In the
absence of a signi®cant interaction term, a signi®cant
main effect for time required a post-hoc analysis
(Duncans) whereas a main effect for treatment group
did not since there were only two groups.
Results
Descriptive characteristics for CON and INT for body
weight and composition at baseline, 9 months, and 18
months are shown in Table 1. There were no signi®cant differences between groups for body weight or
body composition at any period. Body weight and fat
(%fat and fat weight) decreased signi®cantly for the
continuous group across 18 months of exercise. For
INT, body weight and fat (%fat and fat weight)
decreased at 9 months and then returned to baseline
values at 18 months. Fat-free weight was unchanged
for both groups. Results for regional body fat as
measured by circumferences is shown in Table 2.
Small, non-signi®cant decreases were shown at 18
months for CON for both waist and hip circumferences with essentially no change for INT. Waist-tohip ratio showed no signi®cant differences after 18
months of exercise.
Table 3 shows differences in the exercise prescriptions. Both groups participated in over 90% of the
scheduled exercise sessions. By design, INT walked
a greater distance (819 128 km) and time
(8529 862 min) compared to CON (527 46 km;
5138 222 min). INT had a lower exercise heart rate
of 127 13 bpm compared to CON of 142 11 bpm.
Table 1 Body weight and composition
Variable
Period
n
Age (y)
Height (cm)
Weight (kg)
Baseline
Baseline
Baseline
Baseline
9 months
16 months
Baseline
9 months
16 months
Baseline
9 months
16 months
Baseline
9 months
16 months
Baseline
9 months
16 months
BMI (kgmÿ2 )
Body fat (%)
Fat free weight (kgs)
Fat weight (kg)
Continuous
Intermittent
11
54 9
164.52 4.04
81.40 5.71a
80.17 5.75a,b
79.70 5.40b
30.12 2.52
29.62 2.28
29.44 2.38
41.76 3.37a
40.72 3.49a,b
40.01 2.86b
47.39 3.74
47.48 3.88
47.79 3.76
34.01 3.74a
32.69 3.99a,b
31.90 3.28b
11
49 8
163.08 4.58
85.85 13.13a
83.91 12.43b
85.05 12.90a
32.33 5.11
31.76 5.05
32.12 5.18
42.64 4.09a
41.51 4.20b
42.10 4.99a
49.12 7.70
48.96 7.53
49.05 7.49
36.73 7.01a
34.95 6.68b
36.01 7.69a
Values are mean standard deviation. There were no between group differences.
Means with the same letter subscript are not different within group.
International Journal of Obesity
The effects of 18 months of intermittent vs continuous exercise
JE Donnelly et al
569
Table 2 Circumferences
Variable
Period
Waist (cm)
Baseline
9 months
16 months
Baseline
9 months
16 months
Baseline
9 months
16 months
Hip (cm)
Waist-to-hip ratio
Continuous
Intermittent
91.6 6.3
90.7 6.9
89.8 5.9
110.8 4.8
109.7 2.6
109.3 3.6
0.83 0.04
0.83 0.05
0.82 0.04
91.9 11.8
91.2 10.8
91.7 11.9
115.9 10.2
114.0 9.4
115.9 10.2
0.79 0.06
0.79 0.05
0.79 0.05
Values are mean standard deviation. There were no differences between or
within groups.
The exercise heart rate difference was probably due to
the exercise prescription. That is, INT was instructed
to walk briskly yet comfortably while CON exercised
in the Human Performance Laboratory on treadmills,
usually with elevated grade. The average speed of
walking for the INT group was 3.6 0.6 mph while
the CON group walked at an average speed of
3.8 0.7 mph and 1.1 1.5% incline. The ACSM
equation was used to estimate energy expenditure of
exercise. The INT group expended 3235 kJ per week
from exercise while the CON group extended 2235 kJ
per week. Thus, the INT group expended approximately 31% more energy each week than the CON
group, by design, due to the greater number of sessions and the greater time spent exercising per week.
Table 4 shows that both CON and INT had low
VO2 max at baseline of 23.6 2.8 vs 22.9 4.2
ml kgÿ1 minÿ1 , respectively. Likewise, both groups
showed modest but signi®cant increases ( 8% for
CON and 6% for INT) in maximal oxygen consumption across 18 months of exercise. In response to
exercise, resting heart rates showed a decrease of
7 bpm for CON (P < 0.05) and 3 bpm for INT (NS).
Systolic blood pressure decreased 4 mmHg for CON
(NS) and 14 mmHg for INT at 18 months (P < 0.05).
Energy and macronutrient values from the 3-day
records are shown in Table 5. Total energy did not
change signi®cantly during the duration of this study.
Likewise, macronutrient composition remained
unchanged with the exception of fat intake for INT
which was signi®cantly lower at 9 and 18 months
Table 3 Adherence to exercise protocol
Variable
n
Adherence
Session time (min)#
Total time (min)#
Weekly distance (miles)#
Weekly distance (km)#
Weekly EE (kJ)
RPE
Heart rate (beatsminÿ1 )#
Total distance (miles)#
Total distance (km)#
Supervision#
Continuous
Intermittent
11
91.9 2.3
28.9 2.4
5138 222
5.3 1.3
8.5 2.0
2235 614
14 1
142 11
330 29
527 46
Direct-100%
11
90.3 9.12
14.5 2.2
8529 862
8.1 2.3
13.0 3.6
3235 872
13 2
127 13
512 80
819 128
17.9 4.8
EEE ˆ Energy Expenditure. Values are mean standard deviation. # Signi®cant difference between groups.
compared to baseline. Data for the 24 hour recalls
was similar to the 3-day food records (not shown).
Blood chemistry results are shown in Table 6.
There was a signi®cant improvement in both CON
(18%) and INT (9%) for HDL-cholesterol from baseline to 18 months. Fasting values, and values using
area under the curve analysis remained unchanged
for both groups for glucose (mg dlÿ1 ). However,
fasting insulin (uU mlÿ1 ) decreased from baseline
to 18-months for INT (15.81 14.95; 11.90 9.38).
Insulin values for area under the curve analysis
(mU Lÿ1 10ÿ3 ) decreased signi®cantly from baseline to 18-months for both CON (14.82 5.23;
9.85 2.73) and INT (14.86 7.87; 10.79 6.90).
Discussion
We evaluated the effects of two different exercise
programs without energy restriction on aerobic capacity, body composition, and metabolic ®tness subsequent to 18 months of intervention. One group
received a traditional activity program (continuous)
where an individual exercise prescription was performed in one continuous bout under supervision in
the Human Performance Laboratory at the University
of Nebraska-Kearney. The other group received exercise accumulated in 2, 15 min sessions with at least 2
hours between sessions (intermittent). The intermittent group performed brisk walking at home or at the
workplace under supervision of research personnel
according to a random schedule.
We did not include a control group in this study.
There is ample literature regarding the risks of
sendentary behavior.2,32 Likewise, there is ample
literature which shows little change in control
groups when used in exercise intervention trials.33 ± 36
There is no reason to expect a sudden change in the
general population exercise habits in a single 18month period. Unfortunately, sedentary individuals
tend to remain sedentary. Additionally, it was our
purpose to compare two different exercise programs
and this comparison does not call for a control group.
That is, we were not comparing the results of individual receiving the proposed intermittent or continuous
exercise programs to sedentary individuals. Rather,
International Journal of Obesity
The effects of 18 months of intermittent vs continuous exercise
JE Donnelly et al
570
Table 4 Heart rate, blood pressure, treadmill time and aerobic capacity
Variable
Period
Resting heart rate (bpm)
Resting systolic BP (mmHg)
Resting diastolic BP (mmHg)
Maximal treadmill time (min)
VO2 (ml kgÿ1 minÿ1 )
Continuous
Baseline
9 months
16 months
Baseline
9 months
16 months
Baseline
9 months
16 months
Baseline
9 months
16 months
Baseline
9 months
16 months
a
78.8 8
75 9a,b
71 10b
133 11
128 13
129 18
80 7
80 11
80 10
12.55 2.47a
16.17 3.17b
16.82 3.34b
23.6 2.8a
24.5 4.3a,b
25.6 3.4b
Intermittent
74 12
73 10
71 10
133 20a
129 15a
119 14b
81 11
82 13
81 13
13.15 3.73a
15.92 3.23b
15.35 3.71b
22.9 4.1a
24.9 3.9b
24.2 4.6b
Values are mean standard deviation. There were no between group differences.
Means with the same letter superscript are not different within group.
Table 5 Three-day food records
Variable
n
Energy (kcal)
PRO (g)
CHO (g)
Fat (g)
NA (mg)
PRO (%)
CHO (%)
Fat (%)
Period
Baseline
9 months
16 months
Baseline
9 months
16 months
Baseline
9 months
16 months
Baseline
9 months
16 months
Baseline
9 months
16 months
Baseline
9 months
16 months
Baseline
9 months
16 months
Baseline
9 months
16 months
Continuous
Intermittent
11
1940 495
1726 379
1818 528
71.4 23.7
64.3 12.0
66.1 14.3
241.6 62.9
225.6 81.1
243.1 100.1
76.0 31.9
66.4 20.9
74.6 28.0
3148 958
2661 542
2794 1091
14.6 2.4
15.0 3.4
14.6 3.8
50.7 8.6
50.6 8.4
50.4 10.0
34.7 8.2
34.3 8.1
35.1 7.5
11
2074 455
1760 466
1713 815
84.1 14.4
70.0 21.1
75.2 37.0
261.3 82.6
246.2 74.5
230.1 116.8
79.8 19.8a
60.0 19.0b
55.9 27.4b
3248 1126
3315 1101
2446 1213
16.4 2.8
15.6 3.1
17.9 3.5
49.1 7.2
54.5 4.7
52.7 7.5
34.5 6.1a
30.0 6.1b
29.4 5.8b
Values are mean standard deviation. Means with the same letter superscript are
not different within group. There were no differences between groups.
we were attempting to determine if the changes found
in the intermittent exercise group using the newer
CDC=ACSM recommendations would be similar to
those found in the continuous exercise group using the
traditional ACSM recommendations.
We are aware that the total exercise time and
energy expenditure was different between groups;
however, the primary question was the comparison
of 2 different potential intervention strategies, not the
comparison of 2 different (structure) but equal (dose)
exercise programs. By design, the INT group was
asked to perform more absolute work, but at a lower
intensity, thus, both programs represented interventions as we envisioned how they would be used in the
general adult population.
We chose not to study subjects that were below 25
BMI or above 40 BMI. Those subjects below 25 BMI
International Journal of Obesity
may not need to reduce body weight. In our experience, subjects over 40 BMI are more likely to be on
medications and have dif®culty with exercise and
therefore, would not be good candidates for this
study.21,37
There were no differences between INT and CON
for measures of weight and body composition (Table
1). Although CON lost more weight and fat weight,
and reduced BMI compared to INT, the differences
did not reach statistical signi®cance. Table 1 shows
that INT had improved measures of weight and fat
weight at 9 months; however, showed a return to
baseline by 18 months. The reason for this is uncertain
since both INT and CON reported energy intake as
unchange. It is likely that energy intake was underreported as under-reporting of energy intake is notorious, especially in females.38
The effects of 18 months of intermittent vs continuous exercise
JE Donnelly et al
571
Table 6 Blood chemistry
Variable
Period
ÿ1
Total cholesterol (mgdl )
HDL cholesterol (mgdlÿ1 )
Fasting insulin (uUmlÿ1 )
Fasting glucose (mgdlÿ1 )
Insulin area under curve
(mULÿ1 10ÿ3 )
Glucose area under curve
(mULÿ1 10ÿ3 )
Baseline
9 months
16 months
Baseline
9 months
16 months
Baseline
9 months
16 months
Baseline
9 months
16 months
Baseline
9 months
16 months
Baseline
9 months
16 months
Continuous
Intermittent
191 56
204 51
191 44
39.2 11.4a
43.9 11.9b
46.1 9.3c
12.21 4.66
11.16 2.05
10.68 2.87
99.4 14.3
99.6 10.8
98.9 11.2
14.82 5.23a
11.48 4.11b
9.85 2.73b
27.48 7.82
26.68 5.72
26.89 5.59
190 25
216 43
203 43
41.1 11.0a
44.4 10.1b
45.0 7.5b
15.81 14.95a
13.47 11.30a,b
11.90 9.38b
98.0 15.0
100.6 12.9
101.7 12.0
14.86 7.87a
14.90 8.22a
10.79 6.90b
27.05 6.85
27.36 4.49
27.08 3.97
Values are mean standard deviation. There were no differences between groups.
Means with the same letter superscript are not different within group.
The magnitude of the weight loss from baseline is
consistent with other investigations that used exercise
without energy restriction10,39 and represents a 2.1%
decrease for CON and 1.0% decrease for INT. This
amount of weight loss is below the 5% to 10% which
has been recommended and is thought to be associated
with improvements in risk factors, especially lipids,
glucose and insulin.40 Interestingly, both INT and
CON showed improvements in HDL cholesterol and
insulin measured by area under the curve analysis.
This may indicate that exercise conveys bene®ts for
decreased risk for cardiovascular disease and diabetes
without reaching the 5% to 10% decrease in weight
from baseline values.
The amount of weight loss for CON and INt was
not great; however, neither group gained weight. Data
from NHANES III shows weight gain for adults as
they increase in age at least through the 5th decade.1
Since weight loss is dif®cult to maintain, prevention
of weight gain may be an extremely useful strategy. In
a review of weight maintenance literature, Pronk et al
concluded a common element of studies showing
weight maintenance was the use of exercise.41 From
our results, it appears that either 90 min per week of
continuous exercise at 60 to 75% of aerobic capacity,
or 150 min per week of intermittent exercise using
brisk walking, is capable of preventing weight gain in
previously sedentary, moderately obese females.
The change in measures of metabolic ®tness is
noteworthy. Treatment of obesity should be targeted
at improved health, not physical appearance. It could
be argued that weight loss is a secondary objective and
that improvement in health risk is paramount. Both
groups had signi®cant improvements in aerobic capacity. Blair et al have shown increased aerobic capacity
to be an independent predictor of mortality and may
have protective effects for certain comorbidities.42
Systolic blood pressure decreased for both groups
and the decrease reached statistical signi®cance for
INT. HDL cholesterol increased by 15% for INT and
by 9% for CON. These results exceed those reported
by Bray et al where a 5% to 10% weight loss was
induced by drug therapy and HDL cholesterol
increased 4%. Insulin measured by area under the
curve analysis showed signi®cant decreases from
baseline for both CON and INT of 34% and 28%,
respectively. This is in agreement with the ®ndings of
Leon et al where insulin was reduced by 43% in males
after 16 weeks of walking which resulted in a 6%
decrease in body weight.43 The changes in the present
study and in the study of Leon are not remarkably
different from the ®ndings of Franssila-Kallunki et al,
where an 11% decrease in weight was shown in
response to VLED and fasting insulin decreased
43% and insulin area under the curve analysis in
response to a test meal decreased 21%. Thus, exercise
in the current study was associated with changes in
metabolic ®tness which compare to studies using
more aggressive therapies and in which weight loss
was greater.
Summary
It appears that exercise may provide improvements in
aerobic capacity and metabolic ®tness similar to those
found with more aggressive therapies and greater
amounts of weight loss. Moderate exercise is infrequently associated with side effects and does not
require regular medical supervision. Except for the
cost of good walking shoes, exercise is cost ef®cient
compared to many alternative therapies. These results
suggest that long-term, intermittent or continuous
exercise may be effective to prevent weight gain and
improve some measures of metabolic ®tness with
previously sedentary, moderately overweight females.
Acknowledgements
This project was supported by a grant from the
American Heart Association #9507837S.
International Journal of Obesity
The effects of 18 months of intermittent vs continuous exercise
JE Donnelly et al
572
References
1 Kuczmarski RJ, Flegal KM, Campbell SM, Johnson CL.
Increasing prevalence of overweight among US adults.
JAMA 1994; 272(3): 205 ± 211.
2 US Department of Health and Human Services P. Healthy
People 2000: National health promotion and disease prevention objectives. DHHS Publication No. (PHS) 91-50212.
Washington DC: US Government Printing Of®ce, Public
Health Service, 1990.
3 James WP. What are the health risks? The medical consequences of obesity and its health risks. Exp Clin Endocrinol
Diabetes 1998; 106(S2): 1 ± 6.
4 Stunkard AJ, LaFleur WR, Wadden TA. Stimatization of
obesity in medieval times: Asia and Europe. Int J Obes
1998; 22: 1141 ± 1144.
5 Gortmaker SL, Must A, Perrin JM, Sobol AM, Dietz WH. Social
and economic consequences of overweight in adolescence and
young adulthood. N Engl J Med 1993; 329(14): 1008 ± 1012.
6 Dattilo AM, Kris-Etherton PM. Effects of weight reduction on
blood lipids and lipoproteins: a meta-analysis. Am J Clin Nutr
1992; 56: 320 ± 328.
7 Garrison RJ, Castelli WP. Weight and thirty-year mortality of
men in the framingham study. Annals of Internal Medicine
1985; 103: 1006 ± 1009.
8 Stunkard A, McLaren-Hume M. The results of treatment for
obesity. Arch Intern Med 1959; 103: 79 ± 85.
9 Kayman S, Bruvold W, Stern JS. Maintenance and relapse
after weight loss in women: behavioral aspects. Am J Clin Nutr
1990; 52: 800 ± 807.
10 Ballor DL, Keesey RE. A meta-analysis of the factors affecting
exercise-induced changes in body mass, fat mass, and fat-free
mass in males and females. Int J Obes 1991; 15: 717 ± 726.
11 Bouchard C, Depres JP, Tremblay A. Exercise and obesity.
Obes Res 1993; 1: 133 ± 147.
12 Jakicic JM, Polley BA, Wing RR. Accuracy of self-reported
exercise and the relationship with weight loss in overweight
women. Med Sci Sports Exerc 1998; 30: 634 ± 638.
13 American College of Sports Medicine. Guidelines for Exercise
Testing and prescription. 3 ed. Philadelphia: Lea & Febiger,
1988.
14 Pate RR, Pratt M, Blair SN, et al. Physical activity and public
health: a recommendation from the Centers for Disease Control and Prevention and the American College of Sports
Medicine. JAMA 1995; 273(5): 402 ± 407.
15 Gortmaker SL, Dietz WH, Cheung LW. Inactivity, diet, and the
fattening of America. J Am Diet Assoc 1990; 90: 1247 ± 1255.
16 Sallis JF, Haskell WL, Wood PD, et al. Physical activity
assessment methodology in the ®ve-city project. Am J Epidemiol 1985; 121(1): 91 ± 106.
17 American Heart Association. Exercise testing and training of
apparently healthy individuals: a handbook for physicians.
The committee on exercise. Dallas, TX: American Heart
Association, #70-008-A, 1972.
18 Snyder KA, Donnelly JE, Jacobsen DJ, Hertner G, Jakicic JM.
The effects of long-term, moderate intensity, intermittent
exercise on aerobic capacity, body composition, blood lipids,
insulin, and glucose in overweight females. Int J Obes 1997;
21: 1180 ± 1189.
19 Ballor DL, McCarthy JP, Wilterdink EJ. Exercise intensity
does not effect the composition of diet- and exercise-induced
body mass loss. Am J Clin Nutr 1990; 51: 142 ± 146.
20 Van Dale D, Saris WHM, Schoffelen PFM, Ten Hoor F. Does
exercise give an additional effect in weight reduction regimens. Int J Obes 1987; 11: 367 ± 375.
21 Donnelly JE, Jacobsen DJ, Jakicic JM et al. Estimation of
peak oxygen consumption from a sub-maximal half mile walk
in obese females. Int J Obes 1993; 16: 585 ± 589.
22 Joint National Committee. The ®fth report of the Joint
National Committee on detection, evaluation, and treatment
of high blood pressure (JNC-V). Arch Intern Med 1993; 153:
154 ± 183.
International Journal of Obesity
23 Bonge D, Donnelly JE. Trials to criteria for hydrostatic weighing at residual volume. Res Q Exerc Sport 1989; 60: 176 ± 179.
24 Wilmore JH, Vodak PA, Parr RB, Girandola RN, Billing JE.
Further simpli®cation of a method for determination of
residual lung volume. Med Sci Sports Exerc 1980; 12(3):
216 ± 218.
25 Goldman RF, Buskirk ER. Body volume measurement by
under-water weighing: description of a method. In: Brozek J,
Henschel A, eds. Techniques for measuring body composition.
Bethesda: National Academy of Sciences and National
Research Council, 1993.
26 Brozek J, Grande F, Anderson JT, Keys A. Densitometric
analysis of body composition: revision of some quantitative
assumptions. Ann NY Acad Sci 1963; 110: 113 ± 160.
27 Lohman TG, Roche AF, Martorell R. Anthropometric standardization reference manual. Champaign, III: Human
Kinetics Books, 1988.
28 Andersen RE, Wadden TA, Bartlett SJ, Zemel D, Verde TL,
Franckowiak SC. Effects of lifestyle activity vs structured
aerobic exercise in obese women. JAMA 1999; 281: 335 ± 340.
29 Burnstein M, Scholnick HR, Mor®n R. Rapid method for the
isolation of lipoproteins from human serum by precipitation
with polyanions. J Lip Res 1970; 11: 583 ± 595.
30 Morgan CR, Larner J. Immunoassay of insulin: two antibody
system: plasma insulin levels of normal, subdiabetic rats.
Diabetes 1963; 12: 115 ± 126.
31 Lee RD, Neiman DC, Rainwater M. Comparison of eight
microcomputer dietary analysis programs with the USDA
nutrient data base for standard reference. JADA 1995; 95:
858 ± 867.
32 Blair SN, Kohl HW, Paffenbarger RS, Clark DG, Cooper KH,
Gibbons LW. Physical ®tness and all-cause mortality. JAMA
1989; 262(17): 2395 ± 2401.
33 Wood PD, Haskell WL, Blair SN et al. Increased exercise
level and plasma lipoprotein concentrations: A one-year,
randomized, controlled study in sedentary, middle-aged men.
Metab 1983; 32(1): 31 ± 39.
34 Wood PD, Terry RB, Haskell WL. Metabolism of substrates:
diet, lipoprotein metabolism, and exercise. Federation Proceedings 1985; 44: 358 ± 363.
35 Milesis CA, Pollock ML, Bah MD, Ayres J, Ward A, Linnerud
AC. Effects of different duration of training on cardiorespiratory function, body composition and serum lipids. Res Q Exerc
Sport 1976; 47: 716 ± 725.
36 Gettman L, Pollock ML, Durstine JL, Ward A, Linnerud AC.
Physiological responses of men to 1, 3, and 5 day per week
training program. Res Q Exerc Sport 1976; 47: 638 ± 646.
37 Donnelly JE, Pronk NP, Jacobsen DJ, Pronk SJ, Jakicic JM.
Effects of a very-low-calorie diet and physical-training regimens on body composition and resting metabolic rate in obese
females. Am J Clin Nutr 1991; 54: 56 ± 61.
38 Martin LJ, Su WF, Jones PH, Lockwood GA, Tritchler DL, Boyd
NF. Comparison of energy intakes determined by food records
and doubly labeled water in women participating in a dietaryintervention trial. Am J Clin Nutr 1996; 63(4): 483 ± 490.
39 Epstein LH, Wing RR. Exercise and weight. Addictive Behaviors 1980; 5: 371 ± 388.
40 National Institutes of Health. Clinical guidelines on the
identi®cation, evaluation, and treatment of overweight and
obesity in adults ± the evidence report. Obes Res 1998; 6(2):
1S ± 210S.
41 Pronk NP, Wing RR. Physical activity and long-term maintenance of weight loss. Obes Res 1994; 2(4): 587 ± 599.
42 Blair SN, Kampert JB, Kohl HW, III et al. In¯uences of
cardiorespiratory ®tness and other precursors on cardiovascular disease and all-cause mortality in men and women. JAMA
1996; 276: 205 ± 210.
43 Leon AS, Conrad J, Hunninghake DB, Serfass R. Effects of a
vigorous walking program on body composition, and carbohydrate and lipid metabolism of obese young men. Am J Clin
Nutr 1979; 32: 1776 ± 1787.