Physical Inactivity as a Lethal Factor in
Myocardial Infarction among Men
By CHARLES W. FRANK, M.D., EVE WEINBLATT, SAM SHAPIRO,
AND
ROBERT V. SAGER, M.D.
Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017
A RELATIVELY SEDENTARY LIFE
pattern exists for a large segment of
the United States population today. Automation, power tools, automobiles, elevators,
television sets, and electric tooth brushes have
reduced the expenditure of physical effort.
The question has been raised whether the
resultant physical inactivity is a significant
factor in the currently high incidence and
mortality rates from myocardial infarction.1
In 1953, Morris and associates2 demonstrated that bus drivers experienced a higher
age-specific incidence and 3-month mortality
from myocardial infarction than conductors.
These interesting results were based upon
relatively few patients (97), and questions remained as to whether the two groups of
workers differed in smoking habits, prevalence
of hypertension, prior coronary symptoms, or
other characteristics which might account for
the results observed. Since then a number of
studies have sought an association between
occupational classification (as an index of
physical labor) and differing parameters of
coronary disease. These include incidence
rates for myocardial infarction,3"5 angina,1 4
coronary heart disease,", prevalence rates of
coronary heart disease" and myocardial infarction7 in living populations, the prevalence of
coronary artery disease and myocardial lesions (large infarcts and small scars) in pathological studies,8* ' and mortality from coronary
heart disease in relation to occupation. 1'-12
Most of these investigations suggest that
the men in lighter jobs are more likely to
suffer from myocardial infarction or coronary
deaths, although this finding has not been
uniform. No association has been found between the prevalence of coronary atherosclerosis at autopsy and the occupational status of men dying of noncardiac causes.8 '3 All
of these studies have directed attention to
the physical activity of men in connection
with their jobs or occupation. With reduction
in work hours and increase in leisure and
vacation time, off-job activities are becoming
an increasingly important part of the total
life pattern.
The present report is one of a series from
the Health Insurance Plan of Greater New
York (HIP) study of incidence and prognosis
of coronary heart disease (CHD). Detailed
descriptions of the objectives, methodology,
and criteria for diagnosis of CHD manifestations in the HIP study appear elsewhere.13 1
The primary objective of the study is to expand knowledge about the course of disease
following the first diagnosis of a new manifestation of CHD and to explore the relationship
of certain physiological parameters and demographic or social characteristics of the patients to the course of the disease. A secondary objective is to determine the incidence
of new manifestations of CHD and its relationship to various personal characteristics.
This study is being conducted by the Health
Insurance Plan of Greater New York with the support of U. S. Public Health Service, National Institutes of Health Grant HE-05794.
Presented at the Thirty-eighth Scientific Session of
the American Heart Association, Bal Harbour,
Miami Beach, Florida, October 16, 1965.
Methods
The population under observation consists of
members of participating medical groups in HIP
(about 110,000 persons). The participants range
from 25 to 64 years of age and have been continuously enrolled in HIP for at least 2 years.
Patients whose medical records contain some suggestion that a new manifestation of CHD has
Circulation, Volu,me XXXIV. December 1966
1022
PHYSICAL INACTIVITY IN MYOCARDIAL INFARCTION
1023
Results
Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017
developed within the study period receive a baseline medical evaluation. Other sources of information include hospital charts, death certificates,
medical examiner's records and interviews with
next of kin of deceased patients. Case finding extends over a 4-year period.
At the base-line examination the patient is
questioned by a staff interviewer concerning
many personal characteristics. Included are
questions on age, educational attainment, ethnic
background, religion, smoking practices, customary physical activity on and off the job,
work status, and occupation. For characteristics
subject to change, questioning is directed at
clarifying the status just prior to the manifestation of CHD which brought the patient into the
study, as well as current status. For patients
who do not live to undergo the base-line examination, these same questions are asked of the
next of kin.
Overall 4-Week Mortality
A total of 301 men were identified as having
suffered an initial myocardial infarction in the
first 18 months of the study period (November 1, 1961, to April 30, 1963).* The overall
early mortality experience in this group of
men was similar to that described in other
studies of adult male populations.16' 17 About
one third of the men died in the 4 weeks
following diagnosis with the highest mortality
on the first day (85% of all the deaths). For
patients who were hospitalized, the mortality
within the 4-week period was 12%; half of
the hospital deaths occurred in the first 2 days.
Mortality in Relation to Factors Other than
Physical Activity
Mortality in the 4 weeks following infarction was higher in older patients (total line,
table 1), in patients with prior manifestations
The data to be presented deal with mortality experience in the 4 weeks following onset of the initial clinically documented myocardial infarction in men. The outstanding
finding is the considerably higher mortality
among men classified as "least active" physically at the time of infarction than among
the more active men. A previous report from
this study15 noted an increased risk of first
myocardial infarction among the physically
least active men.
*This includes 226 episodes satisfying the study
criteria for "highly probable" and "probable" myocardial infarction, based upon specified combinations
of electrocardiographic, clinical, and laboratory findings, and 75 men who died suddenly under circumstances suggesting that death was caused by an
initial myocardial infarction or coronary occlusion
(satisfying the study criteria for "new coronary event
leading to death").
Table 1
Level of Physical Activity at Time of First Myocardial Infarction and Early Mortality after MI among
Men by Age
% dead within 4 wk
Patients
Age (yr)
All ages
Age (yr)
Overall level of
physical activity
All
ages
< 45
45-54
55-64
All patientst
Least active
Intermediate
Most active
301
80
111
89
37
7
12
18
116
30
49
30
148
43
50
41
32.2
48.8
25.2
16.9
49.4
25.1
17.2
patients onlyt
Least active
206
27
4
Intermediate
82
66
81
17
39
22
98
30
32
32
12.1
21.6
11.0
4.5
21.3
11.6
4.4
Observed
Ageadjusted*
< 45
45-54
5 5-64
18.9
30.2
53.3
22.4
13.3
44.1
32.0
22.0
8.3
11.1
37.2
Hospitalized
Most active
51
11
12
7.4
t
9.1
0.0
7.4
17.6
7.7
0.0
17.3
23.3
15.6
9.4
*Adjusted to age distribution of all men with first myocardial infarction.
tPatients uinclassified as to overall level of physical activity are included in the totals. There were 21 such
patients of whom 15 were dead within 4 weeks; of the hospitalized patients 7 were unclassified and 2 of
these were dead within this period.
:Not computed for base of less than 10.
Circ?rlation, Volume XXXIV, December 1966
FRANK ET AL.
1024
Table 2
Mortality within Four Weeks after First Myocardial Infarction amnong Men with Selected
Characteristics
Deaths within 4 wk
Total
AgeCharacteristic
patients
Number
Observed
adjusted*
Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017
All men
Prior CHDt:
Total classified
Manifestation prior to MI
No prior manifestation
301
97
32.2
298
67
231
94
28
66
31.5
41.8
28.6
31.6
40.7
28.8
Blood pressuret§
Total classified
Elevated
Intennediate
Normal
273
52
104
117
76
23
29
24
27.8
44.2
27.9
20.5
27.6
43.4
27.0
20.8
Presence of diabetest
Total classified
Diabetic
Not diabetic
291
29
262
89
10
79
30.6
34.5
30.2
30.6
30.0
30.4
Smoking habitst
86
30.0
Total classified
30.1
287
54
30.5
31.2
Cigarette smokers
177
15
2 + packs
63
23.8
25.6
38
113
33.6
33.2
< 2 packs
32
110
29.1
Not cigarette smokers
28.7
Relative weightt **
84
30.0
Total classified
280
30.1
10
120 +
39
25.6
28.7
9
35
110-119
25.7
29.3
65
206
< 110
31.6
31.3
Occupational categoryt' tt 44
80
Total classified
274
29.2
29.4
44
135
32.6
White collar
32.9
36
139
Blue collar
25.9
26.3
*Adjusted to age distribution of all men with first myocardial infarction.
tPatients remaining unclassified as to a given characteristic are included only in the first line
of the table. The counts of these patients are:
Total unclassified
Prior CHD
Blood pressure
Presence of diabetes
Smoking habits
Amount of cigarettes
3
28
10
14
1
21
Dead within 4 wk
3
21
8
11
1
13
Relative weight
Occupational category
1
1
(men working at time of MI)
tOf the 67 patients considered to have had a manifestation of CHD before the MI, 54 either
met the study criteria for angina (evaluated patients) or provided medical record information
suggesting definite angina. The remaining patients met the study criteria for either a possible
MI or a conduction defect.
§Classification of the blood pressure level before the MI was as follows: Elevated: Medical
See page 1025 for additional footnotes.
Circulation,. Volame XXXIV,
December 1966
PHYSICAL INACTIVITY IN MYOCARDIAL INFARCTION
Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017
of CHD, and in patients with prior documented elevated blood pressure (table 2).
Men aged 55 to 64 years had a mortality
of 37%, nearly double* that of men under
age 45 (19%). The mortality of men 45 to
54 years of age fell between the two (30%).
Approximately three patients out of four
had no earlier clinical evidence of any manifestation of CHD. Angina was present
prior to the initial myocardial infarction in
about one fifth of the total group. Of the men
with prior CHD 42% were dead within 4
weeks, compared with 29% of the men for
whom the initial myocardial infarction was
the first manifestation of CHD.* When adjustment was made for the differences in age
composition of the groups with and without
prior CHD, however, the difference in mortality rates became only suggestive (significant at 0.10 level). Men known to have had
elevated blood pressure prior to infarction exhibited a greater mortality (44%) than those
classified as normal (21%)t or intermediate
(28%)* in blood pressure. Age adjustment did
not affect the differences in mortality in relation to elevated blood pressure (table 2).
Diabetes, overweight, and cigarette smoking
(including smoking of 2 or more packs a
day) did not influence early mortality. Among
men who were working at the time of initial
infarction, a slightly higher early mortality
*Difference statistically significant at the 0.05
level.
tDifference statistically significant at the 0.01
level.
1025
was observed among those in white collar
occupations (33%) than among the blue collar
workers (26%), but this difference is not statistically significant.
Mortality in Relation to Physical Activity
Figure 1 reveals a marked relationship between the overall level of physical activity and
early mortality following initial myocardial infarction in these men. The mortality increases
nearly threefold from the patients characterized as "most active" (17%) to those considered least active (49% ) ;t and there is a
-0
50r-
.a
0
401-
LU
301z
I
3-
20
LU
z
101-
LUJ
U
LUJ
O0
NO. OF PATIIENTS:
LEAST ACTIVE
80
INTERMEDIATE
III
MOST ACTIVE
89
Figure 1
Physical activity and early mortality among men after
first myocardial infarction.
records provide three or more readings dated before the MI and not obtained during periods
of hospitalization with either diastolic of 95 + or systolic of 160 +. Normal: One or more
readings dated before the MI and not during periods of hospitalization are available; all are
under 140 systolic and 90 diastolic, and there is no positive history of hypertension. Or,
no readings are available, but there is a history denying hypertension. Intermediate: Some
blood pressure readings other than during periods of hospitalization are available; readings
are borderline, or there are less than three elevated readings as defined. Or, no readings are
available and there is a history of hypertension, or no explicit denial of hypertension.
**Relative weight is defined as the ratio of patient's weight at time of MI to the mean
weight for the appropriate age-sex-height group, derived from the average weight tables of
the 1959 Build and Blood Pressure Study, Society of Actuaries (mean weight= 100).
ttOccupation was coded in accordance with definitions by U.S. Bureau of the Census (1960).
"White collar" includes professional and technical, managerial, clerical and sales categories;
craftsmen and foremen, operatives, service workers and laborers are classified as "blue collar."
t#-Restricted to men working at time of myocardial infarction.
Circulation, Volume XXXIV, December 1966
FRANK ET AL.
1026
Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017
twofold increase in mortality between the
intermediate level of physical activity (25%)
and the least active group.* Physical activity
connected with the jobt and physical activities off the jobt were both graded on a fourpoint scale, and classification into the three
overall levels of physical activity was based
on specified combinations of job-connected
and off-job activity levels (fig. 2).
Age adjustment for each of the activity
levels leaves the rates practically unchanged.
Furthermore, the inverse relationship between
early mortality and level of physical activity
is present in men 45 to 54 as well as in
those 55 to 64 years of age (table 1). Another
point of interest is that among the relatively
more active men a higher mortality is observed, as might be expected, for men aged
*Difference statistically significant at the 0.01
level.
tThe index for job-connected activity was constructed15 from answers to questions about time spent
walking and sitting on the job, the amount of walking
and use of transportation in getting to and from the
job, the frequency of lifting and carrying heavy things,
and the total hours worked. Physical activity connected with the job was assessed for 280 men (275
men working at time of myocardial infarction and
five additional men whose last job was in the 5
years preceding the myocardial infarction). Of the
21 unclassified men, 15 were dead within 4 weeks
of the infarction. Two of these deaths were men
who had been retired for more than 5 years and
were consequently not eligible for job-activity
classification. For three of the deaths the information
from next of kin was not adequate for classifying
job-connected activity, and the remaining 10 represented failures to obtain a next of kin interview.
Of the six survivors remaining unclassified, three were
evaluated but the interview information was not
adequate for assigning a classification; the remaining
three were men who refused both the base-line examination and the special interview (obtained in the
15 other men who refused examination) to establish
characteristics at time of infarction.
*The index for off-job physical activities15 was
based on questioning tied to the time of infarction
about customary habits with respect to the frequency and nature of sports engaged in, and the
frequency of taking walks in good weather, of working around the house or apartment, and of gardening
in spring or summer.
§Difference statistically significant at the 0.05
level.
55 to 64 than for those 10 years younger, but
among the least active men those aged 45
to 54 had the higher rate. This reversal could
readily be due to chance factors, but it is
sufficiently intriguing to reexamine when
more observations become available.
If attention is focused only on men without
prior CHD or on men without elevated blood
pressure prior to infarction, the strength of
the mortality differentials in relation to physical activity level remains unchanged (fig.
3, left). Among the least active group without any earlier manifestation of CHD, 44%
were dead within 4 weeks of infarction, compared with 17% of the most active men. Mortality among men without known hypertension
was very similar (45% of the least active, 18%
of the most active). Mortality rates by physical activity level are given in table 3 and
figure 3 (right) for two groups of men defined in relation both to the presence of prior
CHD and blood pressure level prior to myocardial infarction. The least active patients
among men without either a prior manifestation or a prior elevated blood pressure show
a mortality twice that of the intermediate or
most active men in this category.§
Among patients surviving long enough to
be hospitalized both early mortality and the
Physical activity off the job
LEAST
1
2
3
,MOST
4
'-4
C¢
U)
n
c)
D
v4
C)
C4)
C.)
Figure 2
Definition of overall physical activity classes in relation to job-connected and off the job physical activity.
Circulation, Volume XXXIV, December 1966
PHYSICAL INACTIVITY IN MYOCARDIAL INFARCTION
1027
Table 3
Mortality within Four Weeks after First Myocardial Infarction, by Overall Level of
Physical Activity: Relation
to Blood Pressure
Before MI
and
prior CHD Experience
Deaths within 4 wk
Total
patients
Blood pressure and prior CHD
status; level of physical activity
Observed
Number
Ageadjusted*
Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017
No known prior CHD or elevated
blood pressure
25.9
25.7
48
187
Totalt
18
39.9
40.9
44
Least active
23.7
23.7
18
76
Intermediate
18.3
17.9
12
67
Most active
Prior CHD or elevated
blood pressure or both
32.5
36.6
34
93
Totalt
62.5
58.3
21
36
Least active
32.7
28.6
10
35
Intermediate
14.4
13.6
3
22
Most active
*Adjusted to age distribution of all men with first myocardial infarction.
tOmitted are 13 men in this group unclassified as to physical activity; 7 of these were dead
within the 4 weeks.
:Omitted are 8 men in this group unclassified as to physical activity; all of these were dead
within the 4 weeks.
clinical severity of the episodes experienced
strongly related to the level of physical
activity prior to onset of the disease. The
least active men had twice the likelihood
shown by the most active to develop a severe infarction, and they were much more
likely to succumb to these severe episodes
(tables 4 and 5; fig. 4).
were
The classification of overall physical activity
in relation to which striking differentials in
early mortality have been shown depends on
a combination of classes generated for both
job-connected and off-job physical activities.
Interest arises as to the possible role of each
of these elements separately. Table 6 illustrates the distribution of patients and mortality
Table 4
Men Hospitalized with First Myocardial Infarction by Overall Level of Physical Activity
and Clinical Severity of Episode*
Clinical severity
(% of patients)
Level of physical activity
Total classifiedt
No.
of
patients
Intermediate
195
51
78
Most active
66
Least active
Most
severe
Moderate
or slight
severity
29.7
41.2
32.1
18.2
70.3
58.8
67.9
81.8
*The "most severe" category is defined as follows:
(a) Shock is present-systolic blood pressure less than 100 mm Hg and treatment with
pressor amines or diagnosis of clinical shock by attending physician. OR,
(b) Overt cardiac failure is present-gross pulmonary edema on physical or x-ray examination or signs of systemic venous congestion. (Basilar rales alone not accepted as
demonstrating failure) OR,
(c) Ventricular tachycardia, or complete A-V block, or Stokes-Adams attacks are present.
The "moderate" and "slight" categories of severity have been combined for this presentation.
tOmitted are 7 patients unclassified as to physical activity and 4 additional patients unclassified as to clinical severity.
Circulation, Volume XXXIV, December 1966
1028
FRANK ET AL.
No prior CHD
No
known
No known prior CHD
prior
elevated blood
elevated blood
pressure
or
pressure
.0
0
0c
L/)
V-
z
U
z
Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017
LEAST INTER- MOST
ACTIVE MEDIATE ACTIVE
62
81
87
V
-1
LEAST INTER- MOST
ACTIVE MEDIATE ACTIVE
67
NO.OF PATIENTS: 44
76
r, . A-
.. . .."
LEAST INTER- NOST
ACTIVE MEDIATE ACTIVE
36
22
35
Figure 3
Physical activity and early mortality among men after first myocardial infarction. (Left) No
no known prior elevated blood pressure. (Right) No known CHD or prior elevated
blood pressure; prior CHD or elevated pressure or both.
prior CHD;
experience obtained when the total patient
population is divided into four quadrants,
rather than the three overall classes used in
the analysis above.
Men classified as relatively inactive either
on the basis of their job-related or their offjob physical activities have a higher mortality
than the relatively active men. Mortality
among men relatively inactive both on and
off the job (47%) is five times that of the
men who are relatively active in both respects
(9%). Since the job-connected and off-job
classifications are based upon different types
of information (see earlier footnotes), the relative importance of the two types of activities
cannot be judged from these data. The ap-
parently greater differential found for off-job
activity level may reflect differences in the questions asked or the conventions employed in
constructing the two indices of physical activity. Nevertheless, the level of physical activities engaged in by men while off the job,
without regard to the physical activity demands of their jobs, apparently must play a
role of importance in relation to mortality
from first myocardial infarction.
Table 5
Mortality within Four Weeks after First Myocardial Infarction among Hospitalized Men
by Level of Physical Activity at Time of MI and Clinical Severity of Episode*
Level of
physical activity
Total classifiedt
Least active
Intermediate
Most active
Most
All hospitalized
severe
men
episode
No.
% dead
in 4 wk
195
11.8
51
21.6
11.5
4.5
78
66
No.
% dead
in 4 wk
58
21
25
12
32.8
47.6
32.0
8.3
Moderate or
slight severity
% dead
No.
in 4 wk
137
30
53
54
2.9
3.3
1.9
3.7
*tSee corresponding footnotes, table 4.
Circuiation, Volume XXXIV, December 1966
PHYSICAL INACTIVITY IN MYOCARDIAL INFARCTION
"Most severe" and fatal M I's in each activity class
50[z
LU
I-
A 1 o/.
40V
0
32%
LUJ
-
30
MOST
SEVERE
c-
0
I
20
0
z
LU
10_
'FATAL
CL
Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017
01
NO. OF PATIE-NTS:
LEAST ACTIVE
51
INTERMEDIATE
78
MOST ACTIVE
66
Figure 4
Clinical severity and early mortality in relation to
level of physical activity of men hospitalized with first
myocardial infarction.
Most approaches to the study of physical
activity of men in relation to CHD have
rested on inferences made from occupational
titles. From the current data an assessment
of the job-connected physical activity of the
two broad occupational categories shows, as
expected, a higher proportion of the least
active jobs among white collar workers as
compared with blue collar workers. And the
converse is true for the most active jobs.
Nevertheless, only a small difference in mortality was observed between the two occupational groups, and this difference could readily
be due to chance factors (table 2). It is, however, of interest that a gradient in mortality*
in relation to overall level of physical activity
(combined job-connected and off-job) is
found among both white and blue collar
workers (fig. 5).
Summary and Discussion
A striking association between physical
*The difference in mortality between "least
active" and "most active" white collar workers is
statistically significant at the 0.01 level; for the corresponding groups in the blue collar category the
difference is significant at the 0.05 level.
Circulation, Volume XXXIV, December 1966
1029
inactivity and early mortality from the initial
myocardial infarction in men less than 65
years of age has been shown. The data were
derived from the first clinically manifest infarctions occurring in a population of some
55,000 male adults enrolled in the Health Insurance Plan of Greater New York during an
observation period of 18 months (301 cases).
Inquiry about customary physical activities
on and off the job has permitted delineation
of a group of "least active" men who are much
more likely to experience a clinically severe
episode and die within 4 weeks of its onset
than men who are relatively more active. More
than one fourth of the men experiencing their
first myocardial infarction were in this "least
active" group; early mortality for these men
was three times that in the group of most
active men and twice that for the intermediate
activity group. Physical activities off the job
have been shown to play an important role in
the relationship.
The possibility that the higher mortality
among the least active men might reflect a
weighting of this group with patients subject
to a higher mortality risk from other factors
has been examined. It has been found that
'White collar
@
Blue collor
70
63%
a
60LU
50-
33
40 -4.
z
LEAST INTER- MOST
ACTIVE MEDIATE ACTIVE
35
54
NO.OFPATIENTS: 44
LEAST INTER- MOST
ACTIVE MEDIATE ACTIVE
31
52
53
Figure 5
Physical activity and early mortality among men after
first myocardial infraction, by occupational category,
FRANK ET AL.
1030
Table 6
Distribution of Patients and Mortality When Patient Population is Divided into Four
Quadrants
Physical activity off the job
LEAST
-)
1
4
3
2
MOST
Total
Ht
¢
Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017
C.)i
47%*
14%
(O105 Cases)
(78 Cases)
(183 Cases)
35%
9%
23%
Ct
C)
a)
H
0z
(a1 Cases)
i( 46 Cases)
43%
((156 Cases)
*Percentage of patients dead within 4
(97 Cases)
12%
29%
(7124 Cases-)
(280 Cases)
we eks.
the relationship of mortality to physical inactivity is independent of age at infarction and
that it is present among men without any
known prior clinical manifestation of CHD
and among men without prior elevated blood
pressure. It is, of course, possible to hypothesize that subtle "subclinical" consequences
of CHD may impose limitations on the physical activities of adult men and also be responsible for the higher early mortality of
these relatively inactive men. The current data
do not provide any possibility for testing such
an hypothesis.
Another question can be posed with reference to the study's methodology. Information to classify the patient's physical activity
status at the time of onset of infarction was
obtained retrospectively. The men surviving
to the base-line examination (held, on the
average, 2 to 4 months after the infarction)
provided the information at that time; for the
same questions
were answered by their next of kin. If it were
true that survivors of their initial myocardial
infarction, in general, overstated the amount
of their customary physical activity or that
the next of kin, in general, understated the
usual activity of the men who died with the
initial attack (or a combination of these assumptions), a spurious inverse relationship
between physical activity and mortality could
result. This issue has been examined by comparing physical activity information obtained
from two sources for a group of deceased men
who were being followed for prognosis: one
source was the patient himself wvhen he
appeared for study examination; the other
was the next of kin interviewed after the
patient's death. Although the experience to
date is still very small (14 patients), no bias-
patients who succumbed the
Circulation, Volume XXXIV, December 1966
PHYSICAL INACTIVITY IN MYOCARDIAL INFARCTION
Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017
of the type postulated is suggested by this
comparison.
There is additional support for these findings from the data on survivors of their initial
myocardial infarction. Among the physically
least active men who survived the early mortality period and were evaluated at a base-line
examination, a higher proportion of the hospitalized episodes were classified "most severe"
than among the more active men. Since the
early mortality is derived overwhelmingly from
the "most severe" cases, this finding supports
the mortality data without reliance on information from the interviews with next of kin.
The current data are consistent with the
possibility that a relatively high level of physical activity may induce biological influences
which are protective against the severe complications and lethal outcome of the first
clinically recognizable myocardial infarction
in adult men. How this might operate is conjectural. "Physical conditioning" might yield a
nonspecific increased resistance to the sudden
circulatory insult. A more specific mechanism
has been suggested on the basis of animal
experimentation. Eckstein"8 demonstrated that
dogs with coronary arterial partial occlusions
develop a more effective collateral circulation
when forced to exercise than when kept at
rest. This suggests the possibility that apparently healthy adult American men, most of
whom have coronary arterial occlusive disease, benefit from exercise through the development of a more effective collateral circulation which permits them to tolerate occlusion
and infarction better than relatively inactive
men.
If physical activity exerts a protective influence by the mechanism suggested in Eckstein's work, then it would be expected that
individuals with more severe occlusive coronary arterial disease prior to the infarction
would show a relatively greater benefit from
physical exertion. Patients with prior CHD or
hypertension or both can reasonably be presumed to have occlusive disease of greater
severity, and in this study such patients did
exhibit the largest difference in mnortality experience between the "least -active" and the
Cirtulation, Volume XXXIV, December 1966
1031
relatively more active groups (table 3, fig.
3 left). It is also possible that among the men
with prior CHD or hypertension, those with
the more severe disease have become relatively inactive because of their illness. These
possibilities cannot be further explored in
the current data.
Whatever mechanisms may account for the
gradient of early mortality in relation to levels
of physical activity, it is of some interest that
the strength of the relationship here discussed
is greater than the increase in risk for incidence of first myocardial infarction among the
"least active" men which was previously reported from this study.15 Incidence of first
myocardial infarction among least active men
was roughly one and a half times that among
the most active men, while a threefold differential in early mortality is found between
the patients in these two categories. Conversely, the first incidence data from the study
showed a markedly increased risk for myocardial infarction among cigarette smokers
(a twofold risk for all cigarette smokers and
a sixfold risk for heavy cigarette smokers in
comparison with nonsmokers), but no influence
on early mortality with initial infarction has
been found in relation to smoking habits.
There is then the distinct possibility that certain factors, identified in this and other studies
as increasing the risk of myocardial infarction,
act biologically in a manner different from
that of factors influencing early mortality.
These data also have implications relevant
to statistics on the prevalence of myocardial
infarction in relation to physical activity and
in relation to smoking habits. The high immediate mortality of physically inactive men
with first myocardial infarction removes these
men disproportionately from any population
of survivors. In a prevalence survey, the result might well be to attenuate or even eliminate an association between physical inactivity
and myocardial infarction. The situation is different for cigarette smoking. Here, because of
the lack of differential in mortality soon after
the onset of infarction, the prevalence figures
would- reflect the- strong association between
FRANK ET AL.
1032
Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017
cigarette smoking and the incidence of myocardial infarction.
The descending gradient of early mortality
from first myocardial infarction with increasing levels of physical activity has been shown
to hold within the two occupational categories
of white and blue collar workers. No significant difference in mortality was shown between the two occupational groups, and such
a broad classification, by itself, contributes
little to the understanding of the role of physical activities in CHD. Actually, the current analysis suggests that physical activities
engaged in by men while off the job play a
highly important role.
Many questions arise from these findings
for which the current data cannot provide
answers. There may well be important differences between different types of physical
activities: a sustained, endurance type of effort in contrast to briefer and more intense
exertion; walking or climbing in contrast to
heavy lifting or straining, commonly accompanied by a Valsalva type maneuver; and sudden severe exertion in the habitually inactive
or untrained man in contrast to similar effort
on the part of men in better condition. The
desirability of examining these issues with
more refined investigative tools is clear.
In this study the level of customary physical
activity was classified as of the time period
preceding the initial myocardial infarction.
No historical information on changes in patterns of physical activity throughout adult
life was sought. One cannot judge from the
data presented whether the relatively inactive
adult male who moderately increases his customary level of physical activity thereby acquires the advantage of the lower mortality
from first infarction shown by the other men.
Nor is it possible to predict how soon such
a postulated advantage might appear. The
importance of obtaining answers to these
questions is evident. Favorable answers
would project the exciting possibility that
moderate increases in the physical activities
of the least active members of an adult male
population without prior myocardial infarction
might significantly reduce the incidence of, and
the morbidity and mortality from, this manifestation.
Acknowledgment
Thanks are due to the following HIP mnedical
groups and examining physicians for their continuing
cooperation in this study: Astoria Medical Group
(Dr. A. Bianco), Bay Ridge Medical Group (Dr. M.
Burns), Brooklyn Medical Group (Dr. S. Wagner),
Central Manhattan Medical Group (Dr. G. Ramirez),
Central Medical Group of Brooklyn (Dr. H. B. Orenstein), Clinton Medical Group (Dr. J. S. Idriss), Flatbush Medical Group (Dr. N. Shaftel), Jamaica Medical Group (Drs. J. Ohnysty and J. H. Schwab), Kings
Highway Medical Group (Dr. I. A. Schiller), Metropolitan-Hudson Medical Group (Dr. A. A. Schaye),
Montefiore Hospital Medical Group (Dr. L. Wertheimer), New York Medical Group (Dr. T. R. Fink),
Queens Boulevard Medical Group (Dr. I. Wecksell),
Queensboro Medical Group (Dr. G. Mirrer), Washington Heights Medical Group (Dr. K. E. Lauer).
Credit is also due to Dr. George E. Seiden, the
study's electrocardiographer, and to the entire field
and central study staff.
References
1. Fox, S. M., AND SKINNER, J. S.: Physical activity
and cardiovascular health. Amer J Cardiol 14:
731, 1964.
2. MoRRIs, J. N., HEADY, J. A., RAFFLE, P. A. B.,
ROBERTS, C. G., AND PARKS, J. W.: Coronary
heart disease and physical activity of work.
Lancet 2: 1053, 1953.
3. CHAPMAN, J. M., GOERKE, L. S., DIXON, WV.,
LOVELAND, D. B., AND PHILLIPS, E.: Clinical
status of a population group in Los Angeles
under observation for two to three years.
Amer J Public Health 47: 33, 1957.
4. ZUKEL, W. J., ET AL.: Short-term community
study of the epidemiology of coronary heart
disease. Amer J Public Health 49: 1630,
1959.
5. BRUNNER, D., AND MANELIS, G.: Myocardial infarction among members of communal settlements in Israel. Lancet 2: 1049, 1960.
6. STAMLER, J., LINDBERG, H. A., BERKSON, D. M.,
SHAFFER, A., MILLER, W., AND POINDEXTER.
A.: Prevalence and incidence of coronary
heart disease in strata of a Chicago industrial corporation. J Chronic Dis 11: 405, 1960.
7. BROWN, R. G., DAVIDSON, L. A. G., MCKEOWN,
T., AND WHITFIELD, A. G. W.: Coronary artery
disease: Influences affecting its incidence in
males in the seventh decade. Lancet 2: 1073,
1957.
8. MORRIS, J. N., AND CRAWFORD, M.: Coronary
heart disease and physical activity of work.
Brit Med J 2: 1485, 1958.
Circulation, Volume XXXIV, December 1966
PHYSICAL INACTIVITY IN MYOCARDIAL INFARCTION
9. SPAIN, D. M., AND BRADESS, V. A.: Occupational physical activity and the degree of coronary
atherosclerosis in "normal" men. Circulation
22: 239, 1960.
10. BRESLOW, L., AND BUELL, P.: Mortality from
coronary heart disease and physical activity
of work in California. J Chronic Dis 11:
421, 1960.
11. KAHN, H. A.: Relationship of reported coronary
heart disease mortality and physical activity
of work. Amer J Public Health 53: 1058, 1963.
12. TAYLOR, H. L., KLEPETAR, E., KEYES, A., PARLIN,
W., BLACKBURN, H., AND PUCHNER, T.:
Death rates among physically active and
sedentary employees of the railroad industry.
Amer J Public Health 52: 1697, 1962.
Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017
13. SHAPiRO, S., WEINBLATT, E., FRANK, C. W.,
SAGER, R. V., AND DENSEN, P. M.: The
H.I.P. study of incidence and prognosis of
coronary heart disease: Methodology. j
Chronic Dis 16: 1281, 1963.
14. FRANK, C. W., WEINBLATT, E., SHAPIRo, S.,
Circulation, Volume XXXIV, December 1966
1033
SEMEN, G. E., AND SAGER, R. V.: The H.I.P.
study of incidence and prognosis of coronary
heart disease: Criteria for diagnosis. J. Chronic
Dis 16: 1293, 1963.
15. SHAPIRO, S., WEINBLATT, E., FRANK, C. W.,
AND SAGER, R. V.: The H.I.P. study of incidence and prognosis of coronary heart disease: Preliminary findings on incidence of myocardial infarction and angina. J Chronic Dis
18: 527, 1965.
16. PELL, S., AND D'ALONZO, C. A.: Immediate
mortality and five-year survival of employed
men with a first myocardial infarction. New
Eng J Med 270: 915, 1964.
17. KANNEL, W. B., BARRY, P., AND DAWBER, T. R.:
Immediate mortality in coronary heart diseaseThe Framingham study. Fourth World Congress of Cardiology, Mexico City, Proceedings
A and B, 1963.
18. ECKSTEIN, R. XV.: Effect of exercise and coronary artery narrowing on coronary collateral
circulation. Circulation Research 5: 230,
1957.
Physical Inactivity as a Lethal Factor in Myocardial Infarction among Men
CHARLES W. FRANK, EVE WEINBLATT, SAM SHAPIRO and ROBERT V.
SAGER
Downloaded from http://circ.ahajournals.org/ by guest on June 18, 2017
Circulation. 1966;34:1022-1033
doi: 10.1161/01.CIR.34.6.1022
Circulation is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 1966 American Heart Association, Inc. All rights reserved.
Print ISSN: 0009-7322. Online ISSN: 1524-4539
The online version of this article, along with updated information and services, is
located on the World Wide Web at:
http://circ.ahajournals.org/content/34/6/1022.citation
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles
originally published in Circulation can be obtained via RightsLink, a service of the Copyright
Clearance Center, not the Editorial Office. Once the online version of the published article for
which permission is being requested is located, click Request Permissions in the middle column of
the Web page under Services. Further information about this process is available in the Permissions
and Rights Question and Answer document.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Circulation is online at:
http://circ.ahajournals.org//subscriptions/