Siconolfietal_PaffenbargerQuestionnaire

AMERICAN JOURNAL OF EPIDEMIOLOGY

Vol. 122, No. 1

Copyright © 1985 by The Johns Hopkins University School of Hygiene and Publk Health
All rights reserved

Printed m U.SA

SELF-REPORTED PHYSICAL ACTIVITY COMPARED WITH MAXIMAL
OXYGEN UPTAKE
STEVEN F. SICONOLFI,1 THOMAS M. LASATER,1-3 ROBERT C. K. SNOW,1 AND
RICHARD A. CARLETON14

exertion; physical fitness

Higher levels of regular physical activity
have been associated with a reduced likelihood of coronary heart disease (1-3).
„ . ,,
•,. t . , , „„ , n o .
.• ~ ,
Received for publication
July 20, 1984, and in
form October 9,1984.
1
Division of Cardiology, The Memorial Hospital,

! X £ £ J 1 0 2 8 6 a (ReprintrequestetoDr'Richard

Studies by Paffenbarger and colleagues (2,
4, 5) have indicated that this reduced likelihood is independent of differences in
blood pressure, family history, presence of
excess body weight, and smoking status.
final
,
,
,
Regular exercise brings about a number of
physiologic adjustments. These include re-

duced heart rate at rest and at

a** gi ven

J
Pawtucket Heart Health Project, The Memorial
Hospital, Pawtucket, Rl.
Department of Community Health, Brown University, Providence, Rl.
4
Department of Medicine, Brown University, Providence, Rl
This work was supported in part by Grant HL
23629 from The National Heart, Lung, and Blood
Institute.

mal bodily Oxygen uptake during exercise
j s a i s o importantly influenced by habitual

i n g ' X f f ^ S S M a T £ £ K 5 s^nTn

levels of exercise. The higher levels of ox-

this project.

level of Submaximal exercise, suggestive evidence of reduced resting blood pressure,
.
, . .
.
.
,
, .
improved skeletal muscle strength and efficiency, and reduced circulating levels of
neurotransmitter bioamines (6-12). Maxi-

ygen uptake among athletes compared with
101

Downloaded from http://aje.oxfordjournals.org/ at CDC Public Health Library & Information Center on August 30, 2013

Siconoffi, S. F., T. M. Lasater, R. C. K. Snow, and R. A. Carieton (The Memorial
Hospital, Pawtucket, Rl 02860). Self-reported physical activity compared with
maximal oxygen uptake. Am J Epidemiol 1985; 122:101-5.
Maximal oxygen uptake has been used as a measure of physical fitness. This
measure increases by approximately 25% when sedentary individuals become
more physically active. Oxygen uptake measurement in the laboratory or estimation in fieldwork is complex and costly with finite risk. For the present study,
36 men and 32 women completed the Paffenbarger Physical Activity Index
Questionnaire, including a sweat-inducing physical activity frequency question,
and had measurement of oxygen uptake during pedal ergometry. Using maximal
oxygen uptake as the measure of fitness, the authors found that the Paffenbarger
Physical Activity Index, although more detailed, may be less valid than the simpler
sweat induction frequency question for estimating fitness. The correlations observed between the sweat question and oxygen uptake were 0.54 for males, 0.26
for females, and 0.46 for the total group. The correlations between the Physical
Activity Index and oxygen uptake were 0.26 for males, 0.08 for females, and 0.29
for the total group. The regression relationship (oxygen uptake = 1.92 x (sweat
days) + 23.76; standard error of estimate = 8.63 ml/kg/min) is significant for
sweat versus oxygen uptake. While the confidence interval limits the practical
ability to predict individual values, low cost, absence of risk, and population
validity suggest that fitness can be assessed rapidly and simply for epidemiologic
studies with a simple "sweat" question.

102

8ICONOLFI ET AL.

Downloaded from http://aje.oxfordjournals.org/ at CDC Public Health Library & Information Center on August 30, 2013

cal Activity Index Questionnaire to complete and return to the laboratory. This
questionnaire ascertains the average number of flights of stairs climbed each day, the
average number of city blocks or equivalent
walked each day, and the frequency and
duration of sports, recreational, or other
physical activities during the preceding
year and during the preceding week. An
additional question determines the number
of times per week that vigorous extended
activity sufficient to "work up a sweat" is
undertaken.
Maximal oxygen uptake (VC^maJ was
measured on a cycle ergometer (Monark,
Quinton Instruments, Seattle, WA). Subjects pedalled the ergometer at 50 rpm with
no resistance for one minute. Thereafter,
work rates were increased 150 kg-min"1 (25
watts) every minute until the subject indicated that he/she could not continue or
until it was evident that they could not
maintain the 50-rpm pedal rate. Oxygen
uptake (VO2) was measured during the last
30 seconds of each minute. Oxygen uptake
was accepted as maximal if 1) the respiratory quotient was greater than 1.0 and 2)
the oxygen uptake difference between the
final two work bouts was less than 250 ml •
min"1 (15). The highest VO2 attained was
then considered VO2mM. Oxygen uptake
was measured using a pneumotachograph
(Hewlett-Packard, Vertek Series, Lexington, MA), an S-3A Oxygen Analyzer (Applied Electrochemistry, Inc., Sunnyvale,
CA), and an LB-2 Carbon Dioxide Analyzer
MATERIALS AND METHODS
(Beckman, Schiller Park, IL). Gas analyzers were calibrated prior to each test
Thirty-six men (mean weight 78 ± 9 kg;
mean age 41 ± 14 years) and 32 women with a gas that previously had concentra(mean weight 64 ± 16 kg; mean age 42 ± tions of oxygen and carbon dioxide chemi15 years) between the ages of 20 and 70 cally analyzed by the Scholander method
years volunteered for the present study. All (16).
Relationships between estimated maxiwere hospital employees. None were chosen
because they were unusually sedentary or mal oxygen uptake and each of two other
unusually active. None were known to have parameters were compared. The routine
cardiovascular disease.
habitual physical activity was converted to
All subjects visited the Human Perform- kilocalories per week using the method deance Laboratory following a minimum of a scribed by Paffenbarger (4). Also, the retwo-hour fast. They were given the Physi- ported number of times each week of exerthe levels attainable by sedentary individuals are strongly influenced by the "training effects" produced by regular exercise
(13). Maximal oxygen uptake decreases
with advancing age and also reflects the
basal metabolic rate of the body as well as
other inborn but poorly understood factors,
in addition to the physical fitness level.
Within any individual, maximal oxygen uptake can be increased by approximately 25
per cent by shifting from a sedentary to a
very active level of habitual physical activity (14). This attribute has made maximal
oxygen uptake a valuable measure of physical fitness.
Many investigators have attempted to
assess habitual physical activity using diary, recall, or questionnaire methods. Paffenbarger (5), in his important longitudinal
study of college alumni, developed the
Physical Activity Index Questionnaire focusing on selected routine as well as extraordinary and leisure-related activities.
This questionnaire approach has obvious
face validity, assessing flights of stairs
climbed, city block equivalents walked, and
frequency of sports and recreational activity. The present study was undertaken to
test the hypothesis that the Paffenbarger
Physical Activity Index Questionnaire
yields results which reflect population fitness as assessed by maximal oxygen uptake.
Such a questionnaire may also be useful in
assessing change in fitness of a population
over time.

SELF-REPORTED ACTIVITY AND MAXIMAL OXYGEN UPTAKE

cise intense enough to cause sweating was
tabulated. Regression coefficients and
equations were derived, relating the maximal oxygen uptake, the physical activity
level, and the frequency of sweating.
RESULTS

103

50

E

40

.1
E

30 -

E

20

cise to maximal oxygen uptake for men, women, and
men and women combined are shown.

TABLE 1

Regression coefficients relating the Paffenbarger
Physical Actiuity Index (PAD, the frequency of
exercise-induced sweating (SWEAT), and maximal
oxygen uptake (VOt «,J
Total group

VO lm »vs. PAI
VO lmM vs. SWEAT
SWEAT vs. PAI
*p<0.05.
"p<0.01.

0.29*
0.46"
0.57"

Men

Women

0.26
0.08
0.54" 0.26
0.54" 0.60**

Activity Index Questionnaire, this was true
only for the entire group with the larger
range of maximal oxygen uptake values and
was not demonstrated for either the male
or female subsets. A larger population
would, in all likelihood, have demonstrated
a statistically significant relationship, although one must question the importance

Downloaded from http://aje.oxfordjournals.org/ at CDC Public Health Library & Information Center on August 30, 2013

The level of fitness of this population O
compared with that of others can be assessed from the maximal oxygen uptake
figures. The mean ± standard deviation for
the entire population was 28 ± 9.7 ml • kg"1 •
min"1. For men, the mean ± standard deviation was 32.4 ± 9.8, while that for women
was 23.0 ± 6.9 ml-kg"1-min"1. For men
with an average age of 41 years, the expected average maximal oxygen consumption is approximately 34 ml kg"1-min"1,
while for women with an average age of 42
years, the level is approximately 26 mlkg"1-min"1. These data indicate that the
study population, although not randomly
0
TOTAL GROUP
selected, was at an approximately average
60 !
level of fitness when compared with values
published by the American College of
Sports Medicine (17).
E
40
The relationships among the three variables are presented in table 1. As shown, 2. 30
E
the highest correlation coefficients were
found between the maximal oxygen uptake dE_ 20
and the sweating frequency. Those for the
combined male and female group and for
men alone were statistically significant (p
0
1
2
3
4
5
6
7
< 0.01), while that for women did not reach
DAYS/WEEK OF SWEATING INDUCED BY EX£RCISE
statistical significance (p > 0.05). Although
FIGURE 1. Data points from 36 men, 32 women,
a marginally significant relationship was
and the entire group with the regression lines (solid
found between maximal oxygen uptake and lines) and the 95 per cent confidence limits (dashed
kilocalories estimated from the Physical lines) relating days per week of sweat-inducing exer-

104

SICONOLFI ET AL.

DISCUSSION

For the purposes of this study, maximal
oxygen uptake was taken as the most valid
measure of physical fitness. The procedures
necessary for measurement or estimation
of maximal oxygen uptake are complex,
costly, and carry small but finite risk (17).
They can be applied to population studies,
but the cost is relatively high.
The present study has compared two indices developed by Paffenbarger for the
epidemiologic study of regular physical activity levels. (The "sweat" question was
obtained directly from Paffenbarger and
was a part of his original questionnaire.)
Each index has face validity. One is slightly
more time consuming. Each correlates with
maximal oxygen uptake. Based on the data
from this study, however, it appears that
the frequency of activity sufficient to generate sweating relates more closely to maximal oxygen uptake than does the more
complex Physical Activity Index.
The range of maximal oxygen uptake
encountered in this study was similar to
the range in the normal population, with
the exception of the extremes encountered
in highly trained athletes. Similarly, the
range of Physical Activity Index values in
kilocalories approximates that encountered
by Paffenbarger in his work (2, 4, 5). Thus,
we feel it is unlikely that truncation of the
oxygen uptake range of the Physical Activity Index has contributed to the relatively

low correlation levels found between these
two variables.
We conclude that precise evaluation of
physical fitness in individuals requires
either direct measurement or use of validated techniques for estimating maximal
oxygen uptake (18, 19). We suggest, however, that these approaches are less feasible
for studies of large groups of people and
that information derived from questionnaires is valid. The present study strongly
suggests that brevity, low cost, and physical
activity estimates are all well served by the
very simple question, "At least once a week,
do you engage in any regular activity similar to brisk walking, jogging, bicycling, etc.,
long enough to work up a sweat?" Then, if
the answer is yes, "How many times per
week?".
The present study does not deal with
issues of changing levels of population fitness. Change in levels of physical activity
may be an important accompaniment of
successful application of health promotion
programming to large populations (20, 21).
Whether maximal oxygen uptake, the
Physical Activity Index, or the simple
"sweat" question will prove more sensitive
to changes in regular physical activity remains for future investigation to clarify.
REFERENCES

1. Morris JN, Chave SPW, Adam C, et al. Vigorous
exercise and leisure-time activity and the incidence of coronary heart disease. Lancet
1973;l:333-9.
2. Paffenbarger RS Jr, Laughlin ME, Gima AS, et
al. Work activity of longshoremen as related to
death from coronary heart disease and stroke. N
Engl J Med 1970;282:1109-14.
3. Taylor HL, Klepetar E, Keys A, et al. Death rates
among physically active and sedentary employees
of the railroad industry. Am J Public Health
1962;52:1697-1707.
4. Paffenbarger RS Jr, Wing AL, Hyde RT. Physical
activity as an index of heart attack risk in college
alumni. Am J Epidemiol 1978;108:161-75.
5. Paffenbarger RS Jr, Wolf PA, Notkin J, et al.
Chronic disease in former college students. I.
Early precursors of fatal coronary heart disease.
Am J Epidemiol 1966;83:314-28.
6. Astrand PO, Rodahl K. Textbook of work physiology. 2nd ed. New York: McGraw-Hill Book
Company, 1977:394-6.
7. Clausen JP, Trap-Jensen J. Heart rate and arterial blood pressure during exercise in patients with

Downloaded from http://aje.oxfordjournals.org/ at CDC Public Health Library & Information Center on August 30, 2013

of such weak relationships for predictive
utility.
The two self-report measures of habitual
physical activity were correlated in men, in
women, and in the total group. Those with
the highest correlations, days of sweating
versus maximal oxygen uptake, are illustrated in figure 1. The relatively large 95
per cent confidence limits restrict precision
of prediction for any individual. The relationships, however, particularly in men, indicate potential utility for estimating fitness in population studies.

SELF-REPORTED ACTIVITY AND MAXIMAL OXYGEN UPTAKE

8.

9.
10.

12.
13.
14.

15.

16.
17.

18.

19.

20.
21.

terminations. In: Larson LA, ed. Fitness, health
and work capacity: international standards for
assessment. New York: Macmillan Co, Inc,
1974:324-5.
Scholander PS. Analyzer for accurate estimation
of respiratory gases in one-half cubic centimeter
samples. J Biol Chem 1947;167:235-50.
American College of Sports Medicine. Guidelines
for graded exercise testing and exercise prescription. 2nd ed. Philadelphia: Lea & Febiger,
1980:11-56.
Siconolfi SF, Garber CE, Lasater TM, et al. A
simple, valid step test for estimating maximal
oxygen uptake in epidemiologic studies. Am J
Epidemiol 1985;121:382-90.
Siconolfi SF, Cullinane EM, Carleton RA, et al.
Assessing VOi™ in epidemiologic studies: modification of the Astrand-Rbyming test. Med Sci
Sports Exerc 1982;14:335-8.
Lasater TM, Elder JP, Carleton RA, et al. The
Pawtucket Heart Health Program, prospects and
process. Proc Acad Behav Med Res (In press).
Elder JP, Abrams DB, Beaudin P, et al. Behavioral community psychology and the prevention
of heart disease: community level applications of
the Pawtucket Heart Health Program. Proc Acad
Behav Med Res (In press).

Downloaded from http://aje.oxfordjournals.org/ at CDC Public Health Library & Information Center on August 30, 2013

11.

angina pectoris: effects of training and of nitroglycerin. Circulation 1976;53:436-^12.
Hartley LH, Mason JW, Hogan RP, et al. Multiple
hormonal responses to graded exercise in relation
to physical training. J Applied Physiol 1972;
33:602-6.
Blackburn H. Non-pharmacological treatment of
hypertension: discussion. Ann NY Acad Sci
1978;304:236-42.
Denolin H, Mallion J. Hypertension and exercise.
In: Drunner D, Jokl E, eds. Medicine and sports.
Vol 10. Role of exercise in internal medicine.
Basel: S. Karger, 1977:97-117.
McArdle WD, Katch FI, Katch VL. Exercise physiology, energy, nutrition, and human performance.
Philadelphia: Lea & Febiger, 1981:203-4.
Sannerstedt R. Rehabilitation in arterial hypertension. Adv Cardiol 1978;24:164-73.
Ekblom B. Effects of physical training on oxygen
transport system in man. Acta Physiol Scand
(Suppl) 1969;328:9-45.
Saltin B, Bloomqvist B, Mitchell JH, et al. Responses to submaximal and maximal exercise
after bed rest and training. Circulation
1968;38(Suppl 7):l-55.
Wyndham CH. The validity of physiological de-

105