Article Assignment

Attachments

716

·

N Engl J Med, Vol. 347, No. 10

·

September 5, 2002

·

www.nejm.org

T h e N e w E n g l a n d Jo u r n a l o f Me d i c i n e

WALKING COMPARED WITH VIGOROUS EXERCISE FOR THE PREVENTION
OF CARDIOVASCULAR EVENTS IN WOMEN

J

O

A

NN

E. M

ANSON

, M.D., D

R

.P.H., P

HILIP

G

REENLAND

, M.D., A

NDREA

Z. L

A

C

ROIX

, P

H

.D.,
M

ARCIA

L. S

TEFANICK

, P

H

.D., C

HARLES

P. M

OUTON

, M.D., A

LBERT

O

BERMAN

, M.D., M.P.H., M

ICHAEL

G. P

ERRI

, P

H

.D.,
D

AVID

S. S

HEPS

, M.D., M

ARY

B. P

ETTINGER

, M.S.,

AND

D

AVID

S. S

ISCOVICK

, M.D., M.P.H.

A

BSTRACT

Background

The role of walking, as compared with
vigorous exercise, in the prevention of cardiovascular
disease remains controversial. Data for women who
are members of minority racial or ethnic groups are
particularly sparse.

Methods

We prospectively examined the total
physical-activity score, walking, vigorous exercise,
and hours spent sitting as predictors of the incidence
of coronary events and total cardiovascular events
among 73,743 postmenopausal women 50 to 79 years
of age in the Women’s Health Initiative Observational
Study. At base line, participants were free of diagnosed
cardiovascular disease and cancer, and all participants
completed detailed questionnaires about physical ac-
tivity. We documented 345 newly diagnosed cases of
coronary heart disease and 1551 total cardiovascular
events.

Results

An increasing physical-activity score had
a strong, graded, inverse association with the risk of
both coronary events and total cardiovascular events.
There were similar findings among white women and
black women. Women in increasing quintiles of ener-
gy expenditure measured in metabolic equivalents
(the MET score) had age-adjusted relative risks of cor-
onary events of 1.00, 0.73, 0.69, 0.68, and 0.47, respec-
tively (P for trend, <0.001). In multivariate analyses,
the inverse gradient between the total MET score and
the risk of cardiovascular events remained strong (ad-
justed relative risks for increasing quintiles, 1.00, 0.89,
0.81, 0.78, and 0.72, respectively; P for trend <0.001).
Walking and vigorous exercise were associated with
similar risk reductions, and the results did not vary
substantially according to race, age, or body-mass in-
dex. A brisker walking pace and fewer hours spent
sitting daily also predicted lower risk.

Conclusions

These prospective data indicate that
both walking and vigorous exercise are associated
with substantial reductions in the incidence of cardio-
vascular events among postmenopausal women, irre-
spective of race or ethnic group, age, and body-mass
index. Prolonged sitting predicts increased cardio-
vascular risk. (N Engl J Med 2002;347:716-25.)

Copyright © 2002 Massachusetts Medical Society.

From the Division of Preventive Medicine, Harvard Medical School and
Brigham and Women’s Hospital, Boston (J.E.M.); the Department of Pre-
ventive Medicine, Northwestern University Medical School, Chicago (P.G.);
the Fred Hutchinson Cancer Research Center (A.Z.L., M.B.P.) and the De-
partments of Medicine and Epidemiology (D.S.S.), University of Washing-
ton, Seattle; Stanford Center for Research in Disease Prevention, Stanford,
Calif. (M.L.S.); the University of Texas Health Science Center, San Antonio
(C.P.M.); the Division of Preventive Medicine, University of Alabama at
Birmingham, Birmingham (A.O.); and the Department of Clinical and
Health Psychology (M.G.P.) and Division of Cardiology (D.S.S.), Univer-
sity of Florida, Gainesville. Address reprint requests to Dr. Manson at the
Division of Preventive Medicine, Brigham and Women’s Hospital, 900 Com-
monwealth Ave., Boston, MA 02215, or at

[email protected]

HYSICAL activity has been associated with
a reduced risk of cardiovascular disease in ep-
idemiologic studies,

1,2

but data for women
and members of minority ethnic groups have

been sparse. Moreover, the specific role of walking, the
most common form of exercise among women,

3

has
been addressed only minimally. Federal guidelines
from the Centers for Disease Control and Prevention
and the American College of Sports Medicine,

4

as well
as the Surgeon General’s Report on Physical Activity
and Health,

3

endorse at least 30 minutes of moderate-
intensity physical activity on most, and preferably all,
days of the week, in contrast to earlier guidelines that
recommended vigorous endurance exercise for at least
20 minutes three or more times per week.

5

Although
the federal guidelines encourage a level of activity that
is safe, accessible, and feasible for most Americans

6

(at
least 75 percent of whom have less than the recom-
mended level of activity

3

), the potential benefits of
moderate-intensity activity in preventing cardiovascu-
lar events remain uncertain. Moreover, the role of time
spent in sedentary behavior, such as sitting, in predict-
ing risk remains relatively unexplored.

We therefore compared the roles of walking and
vigorous exercise in the prevention of coronary and
cardiovascular events in a large, ethnically diverse co-
hort of postmenopausal women. Using detailed assess-
ments of physical activity, we examined the magnitude
of associations between each of the measures of phys-
ical activity (the total physical-activity score, the inten-
sity of exercise [walking vs. vigorous exercise], and the
hours spent sitting) and the incidence of cardiovascular
events.

METHODS

Study Population

The study population consisted of 73,743 women who were en-
rolled in the Women’s Health Initiative Observational Study, which

P

The New England Journal of Medicine
Downloaded from nejm.org on July 2, 2021. For personal use only. No other uses without permission.

Copyright © 2002 Massachusetts Medical Society. All rights reserved.

EX E R C I S E A N D C A R D I OVA S C U L A R D I S E A S E I N WO M E N

N Engl J Med, Vol. 347, No. 10

·

September 5, 2002

·

www.nejm.org

·

717

involved a national, multicenter cohort of postmenopausal women
who were 50 to 79 years of age at entry. The Women’s Health Ini-
tiative is a prospective, ethnically and racially diverse, multicenter
clinical trial and observational study designed to address the major
causes of illness and death in postmenopausal women (see the Ap-
pendix for a list of study investigators). A total of 93,676 women
were enrolled in the observational study at 40 clinical centers be-
tween 1994 and 1998. Criteria for exclusion from the study includ-
ed the presence of any medical condition associated with predicted
survival of less than three years (e.g., class IV congestive heart fail-
ure, obstructive lung disease requiring supplemental oxygen, or se-
vere chronic liver or kidney disease), alcoholism, mental illness, or
dementia. In addition, women were excluded from the present
analyses if, at base line, they had a history of coronary heart disease,
stroke, or cancer; were nonambulatory (unable to walk at least one
block); or had missing data on the physical-activity questionnaire.
After women had been excluded for these reasons, 73,743 women
remained in the analysis. Of these women, 61,574 were non-His-
panic white, 5661 were non-Hispanic black, 2880 were Hispanic,
2288 were Asian or Pacific Islander, and 1340 were American In-
dian or of other racial or ethnic background. Race was self-assigned.
Details of the scientific rationale, design, eligibility requirements,
and base-line characteristics of the cohort have been published
elsewhere.

7

Exposure Assessment

All women enrolled in the Observational Study were required to
come for a clinic visit for base-line screening. At this visit, women
completed self-administered questionnaires related to personal and
family medical history, physical activity, smoking, diet, and other
behavioral and lifestyle-related factors. Clinical measurements in-
cluding height, weight, waist and hip circumferences, and blood
pressure were obtained by trained staff members. All women pro-
vided written informed consent, and the study protocol was ap-
proved by the institutional review board of each center.

Recreational physical activity was assessed by a detailed question-
naire on the frequency and duration of walking and of several other
types of activity (strenuous, moderate, and mild). Walking was as-
sessed by a series of questions about the frequency of walks out-
side the home for more than 10 minutes without stopping, the av-
erage duration of each walk, and the usual walking pace. Vigorous
exercise was defined as that in which “you work up a sweat and your
heart beats fast,” and examples included aerobics, aerobic dancing,
jogging, tennis, and swimming laps. Moderate exercise was defined
as that which was “not exhausting,” and examples included biking
outdoors, using an exercise machine (such as a stationary bicycle or
a treadmill), calisthenics, easy swimming, and popular or folk danc-
ing. Examples of mild exercise were slow dancing, bowling, and
golf. Using a standardized classification of the energy expenditure
associated with physical activities,

8

we calculated a weekly energy-
expenditure score in metabolic equivalents (MET score) for walking
and for total physical activity. Finally, participants were asked to
estimate the number of hours per day they spent engaged in seden-
tary behavior, including time spent sitting as well as lying down
or sleeping.

Reproducibility and Validation of the Physical-Activity
Assessment

A sample of participants in the Observational Study (1092 wom-
en) was recruited into a reliability study to assess the reproducibility
of selected questionnaires, including the physical-activity assessment.
The average time between base line and repeated assessments was
three months. The test–retest reliability for recreational physical
activity, including walking and strenuous activity, was assessed
(weighted kappas among all women ranged from 0.67 to 0.71).

7

The intraclass correlation coefficient for the primary summary var-
iable (total energy expenditure in MET from all recreational phys-

ical activity) was 0.77. A similar physical-activity questionnaire has
been found to be correlated with physical-activity diaries (r=0.62)
and with one-week recall of activity (r=0.79) in a cohort of female
health professionals.

9

Ascertainment of End Points

The primary end points for this study were newly diagnosed cor-
onary heart disease (nonfatal myocardial infarction or death from
coronary causes) and total cardiovascular events (myocardial infarc-
tion, death from coronary causes, coronary revascularization, angi-
na, congestive heart failure, stroke, or carotid revascularization) that
occurred after the return of the base-line questionnaire but before
August 27, 2000. Newly diagnosed cardiovascular events were iden-
tified on the basis of annual mailed follow-up questionnaires (re-
sponse rates have been above 95 percent), and permission to review
medical records was requested. Study physicians with no knowledge
of the self-reported risk-factor status reviewed the records. The di-
agnosis of nonfatal myocardial infarction was confirmed if data in
the hospital record met standardized criteria of diagnostic electro-
cardiographic changes, elevated cardiac-enzyme levels, or both.

10

Treatment with coronary or carotid revascularization was confirmed
by documentation of the procedure in the medical record. The pres-
ence of angina was confirmed by hospitalization and confirmatory
evidence on angiography, diagnostic stress test, or diagnosis by a
physician and medical treatment. The occurrence of stroke was con-
firmed by documentation in the medical record of the rapid onset
of a neurologic deficit consistent with stroke and lasting at least 24
hours or until death. The presence of congestive heart failure was
confirmed by hospitalization and diagnostic confirmatory tests.

Fatal coronary disease was considered confirmed if there was doc-
umentation in the hospital or autopsy records or if coronary disease
was listed as the cause of death on the death certificate and evidence
of previous coronary disease was available. For deaths from other
cardiovascular causes, a review of confirmatory evidence by physi-
cian-adjudicators was required.

Statistical Analysis

Our primary analyses used the detailed physical-activity assess-
ment at base line. Person-time for each woman was calculated from
the date of return of the base-line questionnaire to the date of a con-
firmed cardiovascular event, death from any cause, or August 27,
2000, whichever came first. Age-adjusted relative risks were com-
puted as the incidence rate in a specific category of activity divid-
ed by the incidence rate in the lowest quintile, with adjustment
for one-year age categories. We conducted tests of linear trend by
treating the categories as a continuous variable and assigning the
median score for each category.

11

All tests of statistical significance
were two-sided.

We used Cox proportional-hazards regression

12

to adjust simul-
taneously for potential confounding variables, including age, smok-
ing status, body-mass index (the weight in kilograms divided by the
square of the height in meters), the ratio of the waist circumfer-
ence to the hip circumference, alcohol consumption, age at meno-
pause, use of hormone-replacement therapy, parental history of pre-
mature myocardial infarction (before 55 years of age in the father
or before 65 years of age in the mother), race or ethnic group, ed-
ucation, family income, and several dietary variables. Additional
models controlled for history or absence of history of hypertension,
diabetes, and high cholesterol levels, as well as for functional status
and a summary score for mental and emotional health.

13

The total
MET score, the MET score for walking, time spent in vigorous ex-
ercise, walking pace, and hours spent sitting and lying down or
sleeping were analyzed separately. Differences in the results for ac-
tivity according to race (white women vs. black women), age, and
body-mass index were assessed. Secondary analyses excluded data
for the first year of follow-up in order to minimize potential bias
caused by the presence of subclinical disease.

The New England Journal of Medicine
Downloaded from nejm.org on July 2, 2021. For personal use only. No other uses without permission.

Copyright © 2002 Massachusetts Medical Society. All rights reserved.

718

·

N Engl J Med, Vol. 347, No. 10

·

September 5, 2002

·

www.nejm.org

T h e N e w E n g l a n d Jo u r n a l o f Me d i c i n e

RESULTS

During up to 5.9 years of follow-up (mean, 3.2
years; total, 232,971 person-years), we documented
345 newly diagnosed cases of coronary disease (287
nonfatal myocardial infarctions and 58 deaths from
coronary causes), 309 strokes, and 1551 first cardio-
vascular events among the 73,743 women 50 to 79
years of age who completed a detailed physical-activ-
ity questionnaire, were ambulatory, and were free of
cardiovascular disease and cancer at base line. The
base-line characteristics of the cohort and the distri-

bution of physical-activity profiles and other risk fac-
tors have been described elsewhere.

7

The total physical-activity score (in MET-hours per
week) at base line had a strong inverse relation with
the risk of coronary heart disease during the follow-up
period (Table 1). In age-adjusted analyses, the rela-
tive risk declined with increasing quintiles of the to-
tal MET score (1.00, 0.73, 0.69, 0.68, and 0.47, re-
spectively; P for trend <0.001). Risk reductions for
increasing categories of walking (P for trend=0.004)
were similar to those for increasing categories of vig-

*Coronary heart disease includes nonfatal myocardial infarction and fatal coronary disease. Multivariate models included age (as a continuous variable),
smoking status (0, 1 to 14, 15 to 24, or »25 cigarettes per day), race or ethnic group (non-Hispanic white, non-Hispanic black, Hispanic, Asian, or other), level
of education (5 categories), family income (7 categories), body-mass index (<25.0, 25.0 to 29.9, or »30.0), waist-to-hip ratio (as a continuous variable), level
of alcohol intake (0, 1 to 4, 5 to 14, or »15 g per day), parental history of premature myocardial infarction, age at menopause, use or nonuse of hormone-
replacement therapy, percentage of calories from saturated fat, number of servings of fruit and vegetables per day, and dietary fiber intake (g per day). MET
denotes metabolic equivalent, and CI confidence interval.

T

ABLE

1.

R

ELATIVE

R

ISKS

OF

C

ARDIOVASCULAR

D

ISEASE

A

CCORDING

TO

Q

UINTILE

OF

T

OTAL

P

HYSICAL

-A

CTIVITY

S

CORE

AND

C

ATEGORIES

OF

W

ALKING

AND

V

IGOROUS

E

XERCISE

.*

C

ATEGORY

Q

UINTILE

OF

T

OTAL

MET-H

R

PER

W

K

P V

ALUE

FOR

T

REND

1
(

LOWEST

) 2 3 4
5

(

HIGHEST

)

Total exercise

Total MET score (MET-hr/wk)
Median
Range

0
0–2.4

4.2
2.5–7.2

10.0
7.3–13.4

17.5
13.5–23.3

32.8
»23.4

Coronary heart disease
No. of cases 92 70 68 70 45
No. of person-years 44,989 45,329 46,003 49,338 47,312
Age-adjusted relative risk (95% CI) 1.00 0.73 (0.53–0.99) 0.69 (0.51–0.95) 0.68 (0.50–0.93) 0.47 (0.33–0.67) <0.001

Total cardiovascular disease
No. of cases 396 342 304 281 228
No. of person-years 44,448 44,836 45,550 48,948 46,972
Age-adjusted relative risk (95% CI) 1.00 0.83 (0.71–0.95) 0.72 (0.62–0.84) 0.63 (0.54–0.74) 0.55 (0.47–0.65) <0.001
Multivariate relative risk (95% CI) 1.00 0.89 (0.75–1.04) 0.81 (0.68–0.97) 0.78 (0.66–0.93) 0.72 (0.59–0.87) <0.001

Walking

Energy expenditure (MET-hr/wk)
Median
Range

0
None

1.5
0.1–2.5

3.8
2.6–5.0

7.5
5.1–10.0

16.7
>10.0

Coronary heart disease
No. of cases 133 64 52 47 49
Age-adjusted relative risk (95% CI) 1.00 0.71 (0.53–0.96) 0.60 (0.44–0.83) 0.54 (0.39–0.76) 0.61 (0.44–0.84) 0.004

Total cardiovascular disease
No. of cases 550 322 249 236 194
Age-adjusted relative risk (95% CI) 1.00 0.88 (0.77–1.01) 0.70 (0.60–0.81) 0.66 (0.57–0.77) 0.58 (0.49–0.68) <0.001
Multivariate relative risk (95% CI) 1.00 0.91 (0.78–1.07) 0.82 (0.69–0.97) 0.75 (0.63–0.89) 0.68 (0.56–0.82) <0.001

Vigorous exercise

Energy expenditure (min of strenuous
exercise/wk)

Median
Range

0
None

30
1–60

90
61–100

140
101–150

210
>150

Coronary heart disease
No. of cases 269 35 13 14 14
Age-adjusted relative risk (95% CI) 1.00 1.12 (0.79–1.60) 0.56 (0.32–0.98) 0.73 (0.43–1.25) 0.58 (0.34–0.99) 0.008

Total cardiovascular disease
No. of cases 1220 125 78 61 67
Age-adjusted relative risk (95% CI) 1.00 0.87 (0.72–1.04) 0.73 (0.58–0.92) 0.69 (0.53–0.89) 0.60 (0.47–0.76) <0.001
Multivariate relative risk (95% CI) 1.00 0.91 (0.73–1.12) 0.81 (0.63–1.06) 0.85 (0.64–1.13) 0.76 (0.58–1.00) 0.01

The New England Journal of Medicine
Downloaded from nejm.org on July 2, 2021. For personal use only. No other uses without permission.

Copyright © 2002 Massachusetts Medical Society. All rights reserved.

EX E R C I S E A N D C A R D I OVA S C U L A R D I S E A S E I N WO M E N

N Engl J Med, Vol. 347, No. 10

·

September 5, 2002

·

www.nejm.org

·

719

Figure 1.

Age-Adjusted Relative Risks of Cardiovascular Disease According to Quintile of Total MET Score in Subgroups Defined by
Race, Age, and Body-Mass Index (BMI).
The reference category is the lowest quintile of MET score.

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.00

0.81

0.58
0.48

0.68

Black women (n=5661)

1.00

0.82
0.73

0.63
0.55

White women (n=61,574)

A
g

e
-A

d
ju

st
e

d
R

e
la

ti
v

e
R

is
k

o
f

C
a

rd
io

v
a

sc
u

la
r

D
is

e
a

se

0.0

0.2

0.4

0.6

0.8

1.0

1.2

70–79 yr (n=15,856)50–59 yr (n=24,803) 60–69 yr (n=32,127)

0.63
0.54

0.45

1.00

0.68
0.63

0.56
0.50

1.00

0.79
0.86

0.75

0.64

1.00
0.93

1.00

0.75
0.84

0.61 0.60

1.00

0.82

0.67 0.71

0.58

1.00 0.96

0.73
0.80

0.65

0.0

0.2

0.4

0.6

0.8

1.0

1.2

Race

Age

Body-Mass Index

BMI »30.0 (n=16,806)BMI 25.0–29.9 (n=24,590)BMI <25.0 (n=30,583)

Quintile of total MET score

1 2 3 4 5
Lowest Highest

P for trend <0.001 P for trend=0.02

P for trend <0.001 P for trend <0.001 P for trend <0.001

P for trend <0.001 P for trend <0.001 P for trend=0.007

Total Physical Activity

The New England Journal of Medicine
Downloaded from nejm.org on July 2, 2021. For personal use only. No other uses without permission.

Copyright © 2002 Massachusetts Medical Society. All rights reserved.

720

·

N Engl J Med, Vol. 347, No. 10

·

September 5, 2002

·

www.nejm.org

T h e N e w E n g l a n d Jo u r n a l o f Me d i c i n e

Figure 2.

Age-Adjusted Relative Risks of Cardiovascular Disease According to Energy Expenditure from Walking (MET-Hr/Wk)
in Subgroups Defined by Race, Age, and Body-Mass Index (BMI).
The reference category is the lowest category of energy expenditure from walking.

1.00

0.78

0.49 0.51 0.46

1.00
0.91

0.70
0.66

0.49

1.00

0.89
0.78

0.72 0.72

1.00
0.95

0.71
0.75

0.67

1.00
0.90

0.71 0.75
0.64

1.00

0.79 0.78

0.56 0.59

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.00

0.67
0.73

0.65

1.00

0.85

0.70
0.66

0.55

1.00

0.0

0.2

0.4

0.6

0.8

1.0

1.2

0.0

0.2

0.4

0.6

0.8

1.0

1.2

70–79 yr (n=15,856)50–59 yr (n=24,803)

BMI <25.0 (n=30,583)

Black women (n=5661)White women (n=61,574)

60–69 yr (n=32,127)

Race

Age

Body-Mass Index

BMI »30.0 (n=16,806)BMI 25.0–29.9 (n=24,590)

Energy expenditure from walking
(MET-hr/wk)

0 0.1–2.5 2.6–5.0 5.1–10.0 >10.0

A
g

e
-A

d
ju

st
e

d
R

e
la

ti
v

e
R

is
k

o
f

C
a

rd
io

v
a

sc
u

la
r

D
is

e
a

se

Walking

P for trend <0.001 P for trend=0.11

P for trend <0.001 P for trend <0.001 P for trend=0.004

P for trend=0.003 P for trend <0.001 P for trend <0.001

The New England Journal of Medicine
Downloaded from nejm.org on July 2, 2021. For personal use only. No other uses without permission.

Copyright © 2002 Massachusetts Medical Society. All rights reserved.

EX E R C I S E A N D C A R D I OVA S C U L A R D I S E A S E I N WO M E N

N Engl J Med, Vol. 347, No. 10

·

September 5, 2002

·

www.nejm.org

·

721

orous exercise (activities with MET scores of 6 or
higher; P for trend=0.008) (Table 1).

Reductions in the risk of total cardiovascular events
with increasing categories of total MET scores, walk-
ing, and vigorous exercise were similar to those for the
risk of coronary disease (Table 1). Women who either
walked or exercised vigorously at least 2.5 hours per
week had a risk reduction of approximately 30 percent.
Similar reductions in the risk of cardiovascular events
with an increasing MET score were observed for white
women and for black women (for other racial and eth-
nic groups, the samples were not large enough to be
analyzed separately), as well as for women in different
categories of age or body-mass index (Fig. 1). The rel-
ative risk of cardiovascular disease in the highest quin-
tile of MET score as compared with the lowest quin-
tile was 0.55 (95 percent confidence interval, 0.47 to
0.65) among white women and 0.48 (95 percent con-
fidence interval, 0.25 to 0.93) among black women.
Moreover, increasing categories of walking were in-
versely associated with the risk of cardiovascular events
in each of these subgroups (Fig. 2). Women who en-
gaged in both walking and vigorous exercise had great-
er reductions in cardiovascular risk than those who
did either one alone (the age-adjusted relative risk for
those in the highest category of each was 0.37 [95
percent confidence interval, 0.25 to 0.57]) (Fig. 3).

In multivariate analyses, after simultaneous control
for age, race or ethnic group, smoking status, body-
mass index, waist-to-hip ratio, socioeconomic status,
several dietary factors, and other covariates, physical
activity remained a powerful predictor of the subse-
quent risk of cardiovascular events (Table 1). For in-
creasing quintiles of the total MET score, the relative
risks were 1.00, 0.89, 0.81, 0.78, and 0.72, respec-
tively (P for trend <0.001). Increasing categories of
walking were associated with similar reductions in risk
(relative risks, 1.00, 0.91, 0.82, 0.75, and 0.68, respec-
tively; P for trend <0.001), which were also similar
to the risk reductions with vigorous exercise (Table 1)
and remained unchanged after simultaneous inclusion
of walking and vigorous exercise in the model. These
results were not substantially altered after further con-
trol for biologic variables that could be considered to
be in the causal pathway, such as hypertension, hyper-
cholesterolemia, and diabetes, and for the summary
score for mental and emotional health

13

(the relative
risks of cardiovascular events with increasing total
MET scores were 1.00, 0.92, 0.87, 0.83, and 0.77,
respectively; P for trend=0.008). When we exclud-
ed data from the first year of follow-up (to minimize
potential bias caused by the influence of subclinical
disease on the activity level), the results were not ma-
terially altered (the multivariate relative risk of car-

Figure 3.

Joint Association of Walking and Vigorous Exercise with the Age-Adjusted Relative Risk of Cardiovascular Disease.
RR denotes relative risk.

Energy Expenditure from Walking (MET-hr/wk)

A
g

e
-A

d
ju

st
e

d
R

e
la

ti
v

e
R

is
k

o
f

C
a

rd
io

v
a

sc
u

la
r

D
is

e
a

se

0.0

0.2

0.4

0.6

0.8

1.0

2.6–10.0«2.5 >10.0

0 min/wk
of vigorous
exercise

1–100 min/wk
of vigorous
exercise

>100 min/wk
of vigorous
exercise

0.71

0.78

1.00

0.49

0.72
0.70

0.37

0.43

0.67

P<0.05 for all comparisons with
reference group (RR=1.00)

The New England Journal of Medicine
Downloaded from nejm.org on July 2, 2021. For personal use only. No other uses without permission.

Copyright © 2002 Massachusetts Medical Society. All rights reserved.

722 · N Engl J Med, Vol. 347, No. 10 · September 5, 2002 · www.nejm.org

T h e N e w E n g l a n d Jo u r n a l o f Me d i c i n e

diovascular disease for women in the highest quintile
of total MET score was 0.76 [95 percent confidence
interval, 0.61 to 0.96; P for trend=0.02]).

Walking pace was also an important determinant
of reduction in cardiovascular risk (Fig. 4). As com-
pared with women who never or rarely walked (the
reference category, with a relative risk of 1.00), wom-
en who walked at faster paces of 2 to 3 miles per hour
(mph) (3.2 to 4.8 km per hour), 3 to 4 mph (4.8 to
6.4 km per hour), and more than 4 mph had relative
risks of cardiovascular disease of 0.86, 0.76, and 0.58,
respectively (P for trend=0.002), according to multi-
variate models that included control for time spent
walking.

Finally, we assessed the relation between hours
spent sitting, as well as hours spent lying down or
sleeping, and the risk of cardiovascular events. After
we accounted for age and recreational energy expend-
iture (total MET score), the relative risk of cardiovas-
cular disease was 1.38 (95 percent confidence interval,
1.01 to 1.87) among women who spent 12 to 15
hours per day lying down or sleeping and 1.68 (95

percent confidence interval, 1.07 to 2.64) among
women who spent at least 16 hours per day sitting, as
compared with those who spent less than 4 hours per
day. Other durations of sitting or lying down were not
significantly associated with cardiovascular risk.

DISCUSSION

These prospective data from an ethnically diverse
cohort of postmenopausal women indicate that both
walking and vigorous exercise are associated with sub-
stantial reductions in the incidence of cardiovascular
events. In contrast, prolonged time spent sitting pre-
dicts increased risk. We observed similar magnitudes
of risk reduction with walking and vigorous exercise,
and the results were similar among white women and
black women as well as among women in different age
groups and categories of body-mass index. These find-
ings extend those of previous analyses from predom-
inantly white populations14-17 to a racially and ethni-
cally diverse cohort of women in the United States.
The results also lend further support to current fed-
eral exercise guidelines that endorse moderate-inten-

Figure 4. Multivariate Relative Risk of Cardiovascular Disease and Relative Risk Adjusted for Age and Walking Time, According to
Walking Pace.
Multivariate relative risks were adjusted for age, time spent walking, smoking status (0, 1 to 14, 15 to 24, or »25 cigarettes per day),
race or …

0 replies

Leave a Reply

Want to join the discussion?
Feel free to contribute!

Leave a Reply

Your email address will not be published. Required fields are marked *