Original Research Article

Frequency, Type, and Volume of Leisure-Time

Physical Activity and Risk of Coronary Heart
Disease in Young Women

Editorial, see p 300 Andrea K. Chomistek,

ScD
BACKGROUND: The inverse association between physical activity and Beate Henschel, MPH
Downloaded from http://circ.ahajournals.org/ by guest on February 14, 2017

coronary heart disease (CHD) risk has primarily been shown in studies of A. Heather Eliassen, ScD
middle-aged and older adults. Evidence for the benefits of frequency, type, Kenneth J. Mukamal, MD,
MPH
and volume of leisure-time physical activity in young women is limited.
Eric B. Rimm, ScD
METHODS: We conducted a prospective analysis among 97230 women
aged 27 to 44 years at baseline in 1991. Leisure-time physical activity was
assessed biennially by questionnaire. Cox proportional hazards models
were used to examine the associations between physical activity frequency,
type, and volume, and CHD risk.
RESULTS: During 20 years of follow-up, we documented 544 incident CHD
cases. In multivariable-adjusted models, the hazard ratio (95% confidence
interval) of CHD comparing 30 with <1 metabolic equivalent of task-
hours/wk of physical activity was 0.75 (0.570.99) (P, trend=0.01).
Brisk walking alone was also associated with significantly lower CHD risk.
Physical activity frequency was not associated with CHD risk when models
also included overall activity volume. Finally, the association was not
modified by body mass index (kg/m2) (P, interaction=0.70). Active women
(30 metabolic equivalent of task-hours/wk) with body mass index<25 kg/
m2 had 0.52 (95% confidence interval, 0.350.78) times the rate of CHD in
comparison with women who were obese (body mass index30 kg/m2) and
inactive (physical activity <1 metabolic equivalent of task-hours/wk).
CONCLUSIONS: These prospective data suggest that total volume of
leisure-time physical activity is associated with lower risk of incident CHD
Correspondence to: Andrea K.
among young women. In addition, this association was not modified by Chomistek, ScD, Department of
weight, emphasizing that it is important for normal weight, overweight, and Epidemiology and Biostatistics,
obese women to be physically active. School of Public Health, Indiana Uni-
versity Bloomington, 1025 E 7th St,
Rm C101, Bloomington, IN 47405.
E-mail achomist@indiana.edu
Sources of Funding, see p 298

Key Words:epidemiology
exercise myocardial infarction
women

2016 American Heart

Association, Inc.

290 July 26, 2016 Circulation. 2016;134:290299. DOI: 10.1161/CIRCULATIONAHA.116.021516


Clinical Perspective
What Is New?
Findings from this study indicate that physical activ-
ity is associated with lower risk of coronary heart
disease (CHD) in young women.
Exercise did not have to be strenuous to have such
associations; moderate-intensity physical activity,
including brisk walking, was associated with lower
risk of CHD.
In addition, we found that frequency of physical activ-
ity was not associated with CHD risk after adjusting
for total volume of physical activity.
We found no association between physical activity
earlier in life and CHD risk in adulthood.
Finally, the associations between physical activity
and lower CHD risk were evident regardless of body
mass index.
Downloaded from http://circ.ahajournals.org/ by guest on February 14, 2017
What Are the Clinical Implications?
For patients who are currently inactive and find join-
ing a gym intimidating, emphasizing the benefits of
walking may help them get active.
Findings from this study indicate that the frequency
of physical activity is not as important as the total
volume; thus, patients can achieve the recom-
mended 150 minutes of moderate- to vigorous-
intensity physical activity per week in as many or as
few sessions as they wish.
Our results suggest that previously inactive women
who become physically active can still decrease
their risk of CHD.
It is important for normal weight, overweight, and
obese women to be physically active.
L
eisure-time physical activity is associated with an
30% lower risk of coronary heart disease (CHD)
in women.1,2 The majority of studies, however, have
been conducted in middle-aged and older populations
because cardiovascular disease morbidity and mortality
rates are low in women <55 years.3 Manson et al4 re-
ported an inverse association between physical activity
and cardiovascular disease in 3 separate age groups
of women in the Womens Health Initiative Observational
Study, with the youngest group including women 50 to
59 years of age at baseline. However, evidence for the
benefits of exercise for CHD in younger women is very
limited.
Although CHD morbidity and mortality rates are low
in younger women, the CHD mortality rate among US
women aged 25 to 54 years has shown minimal im-
provement in the past 2 decades, in contrast to rates
in older adults that have consistently declined.5 A poten-
tial explanation may be the increases in the prevalence
of diabetes mellitus and obesity.6,7 In addition, younger
Circulation. 2016;134:290299. DOI: 10.1161/CIRCULATIONAHA.116.021516
Physical Activity and CHD in Young Women
women differ from older women with regard to their lipid
profiles, and psychosocial risk factors, as well, which
may potentially impact the association between physi-
cal activity and CHD.8,9 We have recently shown that a
healthy lifestyle that includes a healthy diet, not smoking,
normal weight, and at least 2.5 hours per week of mod-
erate- to vigorous-intensity exercise is associated with
substantially lower CHD risk in younger women.10 None-
theless, physical activity is a complex exposure because
of its many dimensions, intensity, type, and frequency,
ORIGINAL RESEARCH
all of which may be important for the prevention of CHD
in younger women.
ARTICLE
The purpose of this study was to assess the relation-
ship between the volume of total leisure-time physical
activity (in metabolic equivalent of task [MET]-hours/wk)
and CHD in young women, while also examining moder-
ate- and vigorous-intensity activity separately. In addition,
we examined whether frequency and type of exercise
were important attributes to general measures of overall
activity in relation to CHD risk. Because rises in obesity
may be a reason for the lack of decline in CHD mortal-
ity rates in young women, we investigated whether high
levels of physical activity could eliminate the adverse as-
sociation between excess weight and CHD risk. Finally,
given the high prevalence of inactivity among adolescent
girls,11 we also investigated the associations between
activity during adolescence and young adulthood and
CHD during adulthood.
METHODS
Study Population
The NHSII (Nurses Health Study II) is an ongoing cohort
study that was established in 1989 and included 116430
predominantly white registered nurses aged 25 to 42 years.
Participants completed a baseline self-administered question-
naire used to collect information on lifestyle factors, includ-
ing physical activity, health behaviors, and medical history.
Follow-up biennial questionnaires were sent to participants to
collect updated information on potential risk factors and newly
diagnosed diseases. Biennial response rates are >90%. This
study was approved by the Institutional Review Board at the
Harvard T.H. Chan School of Public Health. Informed consent
was implied by completion of the questionnaire.
Participants first reported on the frequency of physical
activity and completed a semiquantitative food frequency
questionnaire 12 in 1991, which served as the baseline for
this analysis. Women were excluded if they did not complete
the baseline physical activity questionnaire (n=15418) or
reported an inability to walk at baseline (n=69). After exclusion
of women with cardiovascular disease, cancer, or diabetes
mellitus before 1991 (n=3713), 97230 women were included
in the analysis.
Assessment of Physical Activity
Leisure-time physical activity was assessed in 1991, 1997,
2001, 2005, and 2009 through questions on average total
July 26, 2016 291
 Chomistek et al
time per week spent on various activities over the previous year.
The questions included on the 2009 NHS2 questionnaire can
be accessed at the Nurses Health Study website.13 Walking
pace, categorized as casual (<2 mph), normal (22.9 mph),
brisk (33.9 mph), or striding (4 mph), was also assessed. A
MET score was assigned to each activity based on its energy
cost.14 To calculate the amount of energy expended, the time
spent at each activity in hours per week was multiplied by its
MET score, then summed over all activities to yield total MET-
hours/wk. Moderate activities (3 METs < 6) included brisk
walking, outdoor work, yoga (beginning in 2001), and weight
training (beginning in 2001). Vigorous activities, defined as
requiring MET values 6, were jogging (>10 minutes/mile),
running (10 minutes/mile), bicycling, lap swimming, tennis/
squash/racquetball, and other aerobic exercise.
The validity and reproducibility of the physical activity ques-
tionnaire have been reported in detail elsewhere.15 In brief, the
Pearson correlation between 4 one-week diaries and the question-
naire was 0.62 for moderate and vigorous recreational activity.
Downloaded from http://circ.ahajournals.org/ by guest on February 14, 2017
In addition to time per week spent on physical activity, fre-
quency of exercise was assessed in 1991, 1993, 1995, 2005,
and 2009. This was assessed by using a single question that
read, How many times per week do you engage in physical
activity long enough to perspire heavily (including swimming)?
The responses provided were less than once/wk, once/wk, 2
to 3 times/wk, 4 to 6 times/wk, and 7 times/wk.
In 1997, participants were asked about their walking and
leisure-time activity during 5 age periods: grades 7 to 8, grades
9 to 12, ages 18 to 22, ages 23 to 29, and ages 30 to 34. For
each period, participants reported the average hours per week
they engaged in each of 3 activity categories, with examples
given for each: strenuous recreational activity (eg, running,
aerobics, swimming laps), moderate recreational activity (eg,
hiking, walking for exercise, casual cycling, and yard work), and
walking to and from school or work. Seven categories were
provided for responses ranging from 0 to 11+ hours per week.
Outcome Ascertainment
The primary end point was incident CHD, which included non-
fatal myocardial infarction and fatal CHD. Self-reported MIs
were confirmed by medical records according to World Health
Organization criteria that included symptoms plus either diag-
nostic ECG changes or elevated cardiac-specific enzymes.16
Fatal CHD was confirmed by hospital or autopsy records or if
CHD was listed as the cause of death on the death certificate
and evidence of previous CHD was available.
Statistical Analysis
All analyses were performed using SAS statistical software,
version 9.3 (SAS Institute Inc, Cary, NC). Each eligible par-
ticipant contributed person-time from the return of the 1991
questionnaire (or 1997 questionnaire for analysis of adoles-
cent physical activity) until the date of diagnosis of the first
CHD event, death, or June 2011.
Analysis of Total, Moderate-Intensity, and
Vigorous-Intensity Physical Activity
Cox proportional hazards models were used to estimate haz-
ard ratios (HR) and 95% confidence intervals (CI) for CHD
292 July 26, 2016
outcomes. Physical activity was modeled as a time-varying
exposure where simple updated levels of physical activity,
using the most recent value of activity reported, were used.
For example, events and person-time accrued between 1991
and 1997 were categorized according to exposures reported
on the 1991 questionnaire; events and person-time accrued
between 1997 and 2001 were categorized according to
exposures reported on the 1997 questionnaire; and so forth.
Participants were divided into quintiles (<1, 15.9, 614.9,
1529.9, and 30 MET-hours/wk) for total leisure-time physi-
cal activity. For moderate- and vigorous-intensity activity, cat-
egories were based on the distribution of these variables as
well as informative cut points. For example, 3 MET-hours/wk
corresponds to 1 hour of moderate or 0.5 hours of vigorous
activity and 7.5 MET-hours/wk corresponds to 2.5 hours of
moderate or 1.25 hours of vigorous activity (the current rec-
ommendations based on the Physical Activity Guidelines for
Americans).3 Tests for linear trend were computed by using
the medians for categories modeled as a continuous variable.
Analysis of Individual Types of Physical Activity
In addition, we examined each individual activity separately
while adjusting for all other activity using categories of 0, 0.1
to 0.9, and 1+ hours per week (except for walking, where cat-
egories were 0, 0.10.9, 12.4, and 2.5+ hours per week due
to a larger number of women reporting walking compared to
other types of activities). For this analysis we used hours rather
than MET-hours to be able to use the same categories for all
activities, both those requiring more METs (eg, running) and
fewer METs (eg, yard work). Yoga and weight training were not
assessed until 2001; as such, we did not have enough power
to look at these individually.
Covariables
The multivariable models were stratified by age (in months) and
calendar year and included parental history of myocardial infarc-
tion before 60 years of age (yes/no), smoking (never, former,
current: 115 cigarettes/d, current: 15 cigarettes/d), hours
per week of television watching (quartiles), Alternative Healthy
Eating Index-2010 diet score (quintiles), aspirin use (yes/no),
menopausal status (pre/postmenopausal), postmenopausal
hormone use (never, past, current), parity (none, 12 children,
34 children, 5+ children), oral contraceptive use (never, past,
current), and history of hypertension (yes/no) or hypercholes-
terolemia (yes/no) at baseline. All covariables were updated
over time, except for hypertension and hypercholesterolemia
because the incidence of these conditions may be in the causal
pathway relating physical activity to CHD. Information from pre-
vious questionnaires was used when covariable data in a given
cycle were missing.
Sensitivity Analyses
To minimize bias attributable to reverse causation in situa-
tions where preclinical cardiovascular disease may limit the
ability to exercise, in the main analysis, we stopped updating
physical activity when an individual reported difficulty climb-
ing a flight of stairs or walking. In addition, we performed
sensitivity analyses with a 2- and 4-year lag to exclude pre-
clinical cases at baseline. For example, in a 2-year lag analy-
sis, physical activity reported in 1991 would be used for the
1993 to 1995 follow-up period.
Circulation. 2016;134:290299. DOI: 10.1161/CIRCULATIONAHA.116.021516
 Physical Activity and CHD in Young Women

In addition to using simple updated levels of physical
activity, we also conducted a secondary analysis where we
calculated the cumulative average of physical activity levels
from all available questionnaires up to the start of each 2-year
follow-up interval to represent long-term levels of exercise.17
Analysis of Joint Association and Effect
Modification
Finally, we assessed the joint association of leisure-time
physical activity and body mass index (BMI) with risk of CHD.
women developed documented incident CHD of which
254 cases occurred in women <50 years of age. We
examined total leisure-time physical activity in relation to
other potential risk factors for CHD at baseline (Table1).
Women who reported more physical activity were young-
er, had lower BMI, were less likely to smoke, watched
less television, and had a higher Alternative Healthy Eat-
ing Index-2010 - diet score.
In multivariable-adjusted models, women reporting
the highest amount of leisure-time physical activity (30

ORIGINAL RESEARCH
Participants were cross-classified into 15 groups according MET-hours/wk) were at significantly lower risk of CHD in
to the levels of physical activity (<1, 15.9, 614.9, 1529.9, comparison with women who were the least active (<1

and 30 MET-hours/wk) and BMI (<25, 2529.9, and 30 kg/
m2). The interaction was assessed using the likelihood ratio
test between the models with and without the cross-classified
physical activityBMI variables. Potential effect modification
between physical activity and other cardiovascular risk fac-
tors (smoking, age, alcohol, and diet) was similarly assessed.
Downloaded from http://circ.ahajournals.org/ by guest on February 14, 2017
ARTICLE
MET-hour/wk; hazard ratio [HR], 0.75; 95% confidence
interval [CI], 0.570.99, P for trend=0.01) (Table2).
When examined separately, moderate- and vigorous-
intensity physical activity were both inversely associated
with CHD risk. In comparison with women reporting 0
MET-hours/wk of moderate-intensity physical activity,

women reporting 15 MET-hours/wk of moderate ac-

tivity had a 33% lower risk of CHD (HR, 0.67; 95% CI,
RESULTS 0.510.87; P for trend=0.01). Similarly, women report-
The mean (SD) age of the study population at baseline ing 15 MET-hours/wk of vigorous activity had a 23%
was 36.6 (4.6) years. During 20 years of follow-up, 544 lower risk of CHD (HR, 0.77; 95% CI, 0.57 1.03) in

Table 1.Age-Standardized Characteristics According to Physical Activity at Baseline (1991), Nurses Health
Study II
Categories of Physical Activity, MET-h/wk (1991)
1 (<1) 2 (15.9) 3 (614.9) 4 (1529.9) 5 (30)
(n=19129) (n=21107) (n=20968) (n=18697) (n=17329) Ptrend
Age, y 37.2 (4.6) 36.7 (4.6) 36.5 (4.6) 36.3 (4.7) 36.1 (4.7) <0.0001
Total physical activity, MET-h/wk 0.0 (0.1) 4.0 (2.1) 10.0 (3.9) 20.2 (6.6) 46.0 (28.8)
Vigorous activity, MET-h/wk 0.0 (0.0) 0.4 (1.4) 5.9 (6.1) 15.0 (10.5) 30.4 (28.6)
Moderate activity, MET-h/wk 0.0 (0.0) 2.7 (4.4) 4.5 (7.6) 7.2 (10.0) 15.6 (28.8)
Body mass index, kg/m2 25.8 (6.3) 25.3 (5.7) 24.5 (5.0) 23.9 (4.5) 23.3 (4.1) <0.0001
Hypercholesterolemia, % 16.1 15.1 14.4 13.4 13.0 <0.0001
Hypertension, % 7.3 6.6 6.0 5.3 5.4 <0.0001
Family history of myocardial infarction, % 22.4 21.8 21.8 21.7 21.1 0.006
Current smoker, % 13.8 13.3 11.8 10.9 10.8 <0.0001
AHEI-2010* score 44.7 (10.3) 46.3 (10.2) 48.5 (10.5) 50.5 (10.8) 53.4 (11.2) <0.0001
Aspirin use, % 10.6 11.1 11.2 11.5 11.3 0.08
Current oral contraceptive user, % 9.8 10.3 10.3 11.4 11.8 <0.0001
Menopausal hormone therapy use, % 4.5 4.6 4.6 4.2 4.4 0.22
Postmenopausal, % 3.3 3.2 3.1 3.1 3.3 0.80
Parous, % 79.0 77.2 74.5 71.7 65.7 <0.0001
T.V. watching, hrs/wk 9.9 (9.7) 9.1 (8.6) 8.6 (8.1) 8.4 (7.8) 7.8 (7.8) <0.0001
All values are means (SD) for continuous variables or frequencies for categorical variables, adjusted for age (except for age), except for physical activity
variables which are medians (IQR). AHEI-2010 indicates Alternative Healthy Eating Index-2010; IQR, interquartile range; MET, metabolic equivalent of task;
and SD, standard deviation.
*The AHEI-2010 includes 11 components: high intake of vegetables, fruits, whole grains, nuts and legumes, long-chain (n-3) fats (EPA+DHA), and
polyunsaturated fatty acids; moderate intake of alcohol; and low intake of sugar-sweetened beverages and fruit juice, red and processed meat, trans fat,
and sodium. The AHEI-2010 score ranges from 0 to 110 with higher scores representing better adherence.

Circulation. 2016;134:290299. DOI: 10.1161/CIRCULATIONAHA.116.021516 July 26, 2016 293

 Chomistek et al

Table 2.Hazard Ratios (95% CI) of Coronary Heart Disease According to Quintiles of Leisure-Time Physical
Activity, 1991 to 2011, Nurses Health Study II
Total physical Quintiles of Physical Activity, MET-h/wk P Value for
activity 1 (<1) 2 (15.9) 3 (614.9) 4 (1529.9) 5 (30) Linear Trend
Cases 182 119 101 67 75
Person-years 419667 368957 394283 363488 365102
Age-adjusted 1.00 0.86 (0.681.08) 0.66 (0.520.84) 0.48 (0.360.63) 0.53 (0.410.70) <0.0001
Multivariable* 1.00 0.96 (0.761.21) 0.80 (0.631.03) 0.63 (0.470.83) 0.75 (0.570.99) 0.01
1 (0) 2 (0.12.9) 3 (37.4) 4 (7.514.9) 5 ( 15) P Value for Linear
Trend
Moderate-intensity physical activity
Cases 234 61 103 60 86
Person-years 550036 247575 438299 301637 373950
Age-adjusted 1.00 0.60 (0.450.79) 0.61 (0.480.77) 0.47 (0.350.62) 0.49 (0.380.63) <0.0001
Multivariable* 1.00 0.65 (0.490.87) 0.68 (0.540.87) 0.59 (0.440.80) 0.67 (0.510.87) 0.01
Downloaded from http://circ.ahajournals.org/ by guest on February 14, 2017

Vigorous-intensity physical activity

Cases 313 73 62 36 60
Person-years 825520 246624 259780 186773 392800
Age-adjusted 1.00 0.97 (0.751.25) 0.77 (0.591.01) 0.63 (0.440.88) 0.51 (0.380.67) <0.0001
Multivariable* 1.00 1.16 (0.891.51) 0.99 (0.751.31) 0.89 (0.631.27) 0.77 (0.571.03) 0.04
AHEI-2010 indicates Alternative Healthy Eating Index-2010; CI, confidence interval; and MI, myocardial infarction.
*The models were stratified by age (in months) and time period and included parental history of MI at or before age 60 years, smoking, TV
watching, oral contraceptive use, aspirin, AHEI-2010 score, postmenopausal status, menopausal hormone therapy use, parity, preexisting disease
(diagnosis of hypertension or hypercholesterolemia at baseline). Models for moderate- and vigorous-intensity activity include both types of activity
simultaneously.

comparison with women reporting no vigorous activity
(P for trend=0.04). When the analysis was repeated with
a 2- and 4-year lag, results for total leisure-time activ-
ity and moderate activity were attenuated and no longer
statistically significant, whereas results for vigorous ac-
tivity were similar and remained significant. In second-
ary analysis, when we used cumulative average physical
activity instead of simple updated activity, similar results
were obtained although the hazard ratios in the highest
2 categories were slightly attenuated (online-only Data
Supplement Table I).
We also examined the association between frequency
of exercise and risk of CHD (Table3). In comparison
with women reporting exercise less than once per week,
the age-adjusted HR for women reporting exercise 4+
times per week was 0.60 (95% CI, 0.460.78; P for
trend=0.0001). However, the correlation between fre-
quency and volume of exercise ranged from 0.41 to
0.44 in the questionnaire cycles where both were asked
(1991, 2005, 2009). In the multivariable model that in-
cluded volume of physical activity, the association be-
tween exercise frequency and CHD risk was attenuated
and no longer significant (HR comparing extreme cat-
egories, 0.94; 95% CI, 0.701.26, P for trend=0.84),
whereas the association for volume of physical activity
was nearly identical to that above (HR, 0.76; 95% CI,
0.561.02 comparing 30 with <1 MET-hours/wk).
Table4 shows the association between individual
activities and CHD risk. In multivariable-adjusted analy-
ses where each activity was modeled separately, but
adjusted for total volume of all other activity, aerobics,
outdoor work, and brisk walking were each significantly
inversely associated with CHD (P for trend=0.04, 0.04,
and 0.001, respectively) (Table4). Engaging in aero-
bics 1 hour/wk was associated with a 26% CHD risk
reduction (HR, 0.74; 95% CI, 0.55-0.99) and outdoor
work with a 16% CHD risk reduction (HR, 0.84; 95%
CI, 0.681.04) in comparison with women not partici-
pating in these activities (Table4). Brisk walking for
2.5 hours/wk was associated with a 35% risk reduc-
tion (HR, 0.65; 95% CI, 0.480.87) in comparison with
women who reported no brisk walking. In addition, in-
creasing speed of usual walking pace was associated
with lower CHD risk. After adjusting for hours per week
of walking, other physical activity, and covariables, the
HRs for easy/casual pace (<2 mph), normal/average
pace (22.9 mph), brisk pace (33.9 mph), and very
brisk/striding pace (4 mph) were: 1.00, 0.81 (95% CI,

294 July 26, 2016 Circulation. 2016;134:290299. DOI: 10.1161/CIRCULATIONAHA.116.021516


Table 3.Hazard ratios (95% CI) of Coronary Heart Disease According to Frequency of Physical Activity,
1991 to 2001, Nurses Health Study II
<1 Time/wk
Cases 195
Person-years 590006
Age-adjusted 1.00
Multivariable* 1.00
Multivariable* + MET-h/wk of
1.00
physical activity
AHEI-2010 indicates Alternative Healthy Eating Index-2010; CI, confidence interval; MET, metabolic equivalent of task; and MI, myocardial infarction.
*The models were stratified by age (in months) and time period and included parental history of MI at or before age 60 years, smoking, TV watching,
oral contraceptive use, aspirin, AHEI-2010 score, postmenopausal status, menopausal hormone therapy use, parity, preexisting disease (diagnosis of
hypertension or hypercholesterolemia at baseline).
Downloaded from http://circ.ahajournals.org/ by guest on February 14, 2017
0.660.98), 0.57 (95% CI, 0.440.73), and 0.33 (95%
CI, 0.170.65). Finally, results for running and tennis
also suggested an inverse association with incidence
of CHD but, most likely because of the small number of
cases in the upper categories, reductions in risk were
not statistically significant.
The joint association of physical activity and BMI on
risk of CHD is shown in the Figure. As the interaction
between physical activity and BMI was not statistically
significant (P for interaction=0.70), the inverse associa-
tion between physical activity and CHD was not modi-
fied by BMI category. In comparison with obese women
reporting <1 MET-hour/wk of physical activity, the HR
of CHD for normal weight women reporting 30 MET-
hours/wk was 0.52 (95% CI, 0.350.78). Furthermore,
among women in the highest category of exercise, those
with normal BMI had lower CHD risk than women who
were overweight or obese.
We additionally examined the association between
physical activity recalled from earlier in life and CHD
risk in adulthood (online-only Data Supplement Table
II). There was no association between physical activ-
ity during ages 12 to 22 years and risk of CHD; the
multivariable-adjusted HR comparing the highest with
the lowest category was 1.12 (95% CI, 0.831.52). We
also assessed physical activity measured at baseline
only (1991) in relation to events occurring throughout
the 20 years of follow-up, which differs from the primary
analysis where physical activity levels were updated ev-
ery 4 to 6 years during follow-up. Similar to early-life
activity, the association between baseline physical activ-
ity and risk of CHD was null; the multivariable-adjusted
HR comparing the highest with lowest quintile was 0.98
(95% CI, 0.741.29) (online-only Data Supplement Table
III). These results suggest that the favorable association
between physical activity and CHD may be best docu-
mented with levels proximal to the date of CHD inci-
dence for young and middle-aged women.
Circulation. 2016;134:290299. DOI: 10.1161/CIRCULATIONAHA.116.021516
Physical Activity and CHD in Young Women
Frequency of Physical Activity
P Value for
1 Time/wk 23 Times/wk 4 Times/wk Linear Trend
101 173 75
354486 636471 324241
0.87 (0.681.10) 0.79 (0.640.97) 0.60 (0.460.78) 0.0001
0.91 (0.711.16) 0.94 (0.761.16) 0.81 (0.611.07) 0.17
ORIGINAL RESEARCH
0.95 (0.741.22) 1.05 (0.841.30) 0.94 (0.701.26) 0.84
ARTICLE
Finally, we evaluated whether the inverse association
between physical activity and CHD was modified by other
cardiovascular risk factors (online-only Data Supplement
Table IV). Physical activity was associated with lower CHD
risk for women <50 years of age and for women 50
years of age, as well. In addition, there was no evidence
of effect modification by smoking, Alternative Healthy
Eating Index-2010 diet score, or alcohol intake.
DISCUSSION
In this large, prospective study of young US women, in-
dividuals in the highest category (30 MET-hours/wk) of
leisure-time physical activity were at a 25% lower risk
of incident CHD. In addition, both moderate-intensity
physical activity (eg, brisk walking) and vigorous-intensity
activity were associated with reduced CHD risk. Impor-
tantly, physical activity was associated with lower CHD
risk regardless of BMI group.
The mean age at baseline of participants in this study
was 36.6 years, providing a unique opportunity to ex-
amine physical activity and CHD in younger women. Ac-
cording to a recent review, the median or mean ages of
subjects in studies included in the 2008 Physical Activity
Guidelines primarily ranged from 45 to 60 years.18 Thus,
this article is a valuable contribution to the existing lit-
erature on modifiable lifestyle factors that could prevent
CHD in younger women. Primordial prevention of CHD in
this group is critical because recent data suggest that
the CHD mortality rate in women aged 25 to 54 years
may not be declining as it is in other groups, possibly
because of the increases in the prevalence of obesity
and type 2 diabetes mellitus.57
We found no association between recalled physical
activity during adolescence or early adulthood and risk
of CHD in adulthood. This finding is similar to that of
Conroy et al19 who found that, although physical activ-
ity during high school and ages 18 to 22 was associ-
July 26, 2016 295
 Chomistek et al

Table 4.Hazard Ratios (95% CI) for Coronary Heart Disease Associated With Average Weekly
Hours of Individual Activities Adjusted for All Other Activity, 1991 to 2011, Nurses Health Study II
Hours/Week Spent in Individual Activities
0 0.10.9 1.0 P for Trend
Vigorous activities
Jogging

Cases 508 22 14
Age-adjusted 1.00 0.74 (0.481.13) 0.76 (0.451.31) 0.24

Multivariable* 1.00 0.88 (0.571.36) 0.97 (0.561.67) 0.84
Running

Cases 534 7 3
Age-adjusted 1.00 0.61 (0.291.29) 0.27 (0.090.85) 0.02

Multivariable* 1.00 0.71 (0.341.51) 0.40 (0.131.26) 0.11
Bicycling
Downloaded from http://circ.ahajournals.org/ by guest on February 14, 2017


Cases 415 84 45
Age-adjusted 1.00 1.04 (0.821.32) 0.90 (0.651.23) 0.51

Multivariable* 1.00 1.16 (0.921.48) 1.01 (0.741.39) 0.88
Swimming

Cases 492 35 17
Age-adjusted 1.00 1.11 (0.791.57) 1.10 (0.671.78) 0.64

Multivariable* 1.00 1.13 (0.801.60) 1.08 (0.661.77) 0.67
Tennis

Cases 534 5 5
Age-adjusted 1.00 0.57 (0.241.38) 0.46 (0.191.11) 0.06

Multivariable* 1.00 0.67 (0.281.61) 0.57 (0.241.39) 0.18
Aerobics

Cases 432 57 55
Age-adjusted 1.00 0.96 (0.721.26) 0.64 (0.480.85) 0.002

Multivariable* 1.00 1.04 (0.781.37) 0.74 (0.550.99) 0.04
Moderate activities
Outdoor work

Cases 327 87 130
Age-adjusted 1.00 0.70 (0.550.89) 0.89 (0.721.09) 0.08

Multivariable* 1.00 0.72 (0.570.91) 0.84 (0.681.04) 0.04
Hours/Week Spent in Individual Activities P for Trend
0 0.10.9 12.4 2.5
Brisk walking

Cases 427 29 35 53
Age-adjusted 1.00 0.54 (0.370.79) 0.61 (0.430.86) 0.51 (0.380.68) <0.0001

Multivariable* 1.00 0.65 (0.440.95) 0.77 (0.541.09) 0.65 (0.480.87) 0.001
AHEI-2010 indicates Alternative Healthy Eating Index-2010; CI, confidence interval; and MI, myocardial infarction.
*The models were stratified by age (in months) and time period and included total volume of other physical activity, parental history
of MI at or before age 60 years, smoking, TV watching, oral contraceptive use, aspirin, AHEI-2010 score, postmenopausal status,
menopausal hormone therapy use, parity and preexisting disease (includes a diagnosis of hypertension or hypercholesterolemia at
baseline).

296 July 26, 2016 Circulation. 2016;134:290299. DOI: 10.1161/CIRCULATIONAHA.116.021516

 Physical Activity and CHD in Young Women

In the current study, there was no association be-

1.12
1.2
1 (ref) tween weekly frequency of physical activity and CHD
1
0.90 0.85 0.92 risk once adjusted for volume of physical activity. This
0.78 0.78 is in contrast to a recent finding from the Million Women
HR for CHD

0.8 0.66* 0.76

0.59* Study, which showed that, in comparison with women
0.6 0.56*
0.49* 0.44*
0.52*
reporting strenuous activity 2 to 3 times per week, those
0.4 0.38* reporting strenuous physical activity daily were at higher
0.2 risk of CHD.26 This inconsistency could be explained, in

BMI (kg/m2)
30
0 25 - 29.9 part, by differences in analysis methods. In the Million
<1 < 25 Women Study, the analysis for frequency of physical ac-

ORIGINAL RESEARCH
1 - 5.9
6 - 14.9
15 - 29.9 30 tivity did not adjust for total volume of activity because
Physical Activity (MET-hrs/wk) the duration of activity was not assessed until 3 years

ARTICLE
after baseline. With the exception of activity frequency,
Figure. Multivariable-adjusted hazard ratios (HRs) for however, findings from the current study are similar to
coronary heart disease (CHD) for the joint association those of the Million Women Study among women 50 to
between physical activity and body mass index (BMI). 64 years of age, in particular, with regard to the benefits
The models were stratified by age and period and included of moderate physical activity for CHD.
parental history of myocardial infarction at or before age 60 Our results suggest that both moderate- and vigor-
Downloaded from http://circ.ahajournals.org/ by guest on February 14, 2017

years, smoking, TV watching, oral contraceptive use, aspirin

ous-intensity physical activity are associated with CHD
use, AHEI-2010 score, postmenopausal status, menopausal
risk reduction, similar to other studies.2730 Nonethe-
hormone therapy use, parity, and preexisting disease (in-
cludes a diagnosis of hypertension or hypercholesterolemia less, these previous studies in older men and women
at baseline). *Indicates values significant at P=0.05. P for indicate a greater magnitude of association for vigor-
interaction=0.70. AHEI-2010 indicates Alternative Healthy Eat- ous activity compared with moderate,2730 whereas we
ing Index-2010; and MET, metabolic equivalent of task. saw a modestly stronger association for moderate over
vigorous activity. The result in the present study may
ated with meeting physical activity recommendations in be a consequence of the physical activity questionnaire
used. We were unable to include an assessment of the
middle adulthood, it was not associated with the risk of
intensity at which a participant performed many of the
CHD during middle age and older. Furthermore, when
activities. So, although we categorized activities like bi-
we performed the analysis lagged 2 and 4 years, the
cycling and swimming as vigorous, some participants
association between total physical activity and CHD risk
may actually perform these activities at a much lower
was attenuated. Taken together, these findings suggest
intensity. Thus, the inability to distinguish between the
that associations between physical activity and CHD pre-
same activity performed at a truly vigorous intensity ver-
vention may be driven by most recent levels. This is con-
sus a lower intensity could have contributed random er-
sistent with evidence suggesting that exercise has acute
ror and attenuated our assessment. This is further sup-
effects on cardiovascular disease risk factors such as ported by the activity-specific hazard ratios where the
blood lipids, blood pressure, and glucose control.20 traditional vigorous activities such as running, tennis,
Specifically, physical activity is correlated with lower tri- and aerobics were more strongly inversely associated
glycerides, lower apolipoprotein B, higher high-density with CHD than activities like swimming, biking, and jog-
lipoprotein, change in low-density lipoprotein particle ging that likely have a much broader range of intensity.
size, and lower coronary artery calcium.21 Therefore, an Furthermore, walking pace was assessed in the current
important message to communicate to the public is that, study and found to be strongly inversely associated with
regardless of how inactive you may be, it is possible to risk of CHD.
experience cardiovascular benefits soon after becom- Importantly, our study suggests that leisure-time phys-
ing physically active. Nonetheless, physical activity early ical activity is associated with a reduction in CHD risk in
in life should be encouraged because it has important younger women who are normal weight, overweight, or
health benefits for children and adolescents, including obese. Although this finding has been reported in pre-
improved cardiovascular and metabolic health.3,22,23 Fur- vious studies among middle-aged and older men and
thermore, in this same cohort of women, there was a women, it is worth emphasizing, given the high preva-
suggestive inverse association between physical activity lence of overweight and obesity in young and middle-
during ages 14 to 22 years and risk of premenopausal aged US women (58.5% for women 2039 years of age,
breast cancer.24 Previous studies have also indicated 71.7% for women 4059 years of age).3133 An elevated
that past physical activity is associated with physical ac- BMI is still a significant risk factor for development of
tivity later in life.19,25 Thus, engaging in regular exercise CHD, but the increased risk associated with being over-
is important for young and old as physical activity has weight or obese is attenuated, although not completely,
important health benefits throughout life. by engaging in physical activity.

Circulation. 2016;134:290299. DOI: 10.1161/CIRCULATIONAHA.116.021516 July 26, 2016 297

 Chomistek et al
Strengths of this study include its prospective design,
the detailed information on physical activity collected
multiple times during follow-up, the large number of con-
firmed CHD cases despite the relatively young age of
study participants, and minimal loss to follow-up.
Our study also has several limitations that should be
considered. Our study population, consisting of predomi-
nantly white nurses, is not representative of the general
population. Thus, we cannot necessarily generalize our re-
sults to men or other populations with different education-
al levels, incomes, or distributions of race and ethnicity.
Physical activity was self-reported, but this questionnaire
has been previously validated in this population.15 More-
over, measurement error is unlikely to bias our results be-
cause physical activity was assessed prospectively so any
reporting errors would be nondifferential with respect to
subsequent disease status. Nonetheless, the lack of asso-
ciation between physical activity during adolescence and
Downloaded from http://circ.ahajournals.org/ by guest on February 14, 2017
CHD during adulthood may be a consequence of measure-
ment error as participants had to recall physical activity
levels from 20 to 35 years earlier. As in any observational
study, the possibility of residual confounding by other life-
style characteristics must be considered; however, we
were able to adjust for many known CHD risk factors.
In conclusion, this study indicates that physical ac-
tivity is associated with a lower risk of CHD in young
women. Furthermore, exercise did not have to be strenu-
ous to have such associations; moderate-intensity physi-
cal activity, including brisk walking, was associated with
lower risk of CHD. There was no association between
physical activity earlier in life and CHD risk in adulthood,
suggesting that previously inactive women who become
physically active can still decrease their risk of CHD. Fi-
nally, the favorable associations between physical activ-
ity and lower CHD risk were evident regardless of BMI,
emphasizing that it is important for normal weight, over-
weight, and obese women to be physically active.
SOURCES OF FUNDING
This study was supported by National Institute of Health grants
UM1 CA176726 and R01 CA050385. Dr Chomistek was sup-
ported by an institutional training grant (DK007703) from the
National Institute of Diabetes and Digestive and Kidney Dis-
eases.
DISCLOSURES
None.
AFFILIATIONS
From the Department of Epidemiology and Biostatistics, School
of Public Health, Indiana University, Bloomington (A.K.C., B.H.);
Department of Epidemiology, Harvard T.H. Chan School of Pub-
lic Health, Boston, MA (A.H.E., E.B.R.); Channing Division of
298 July 26, 2016
Network Medicine, Department of Medicine, Brigham and Wom-
ens Hospital and Harvard Medical School, Boston, MA (A.H.E.,
E.B.R.); Department of Medicine, Beth Israel Deaconess Medi-
cal Center, Boston, MA (K.J.M.); and Department of Nutrition,
Harvard T.H. Chan School of Public Health, Boston, MA (E.B.R.).
FOOTNOTES
Received January 14, 2016; accepted June 16, 2016.
The online-only Data Supplement is available with this arti-
cle at http://circ.ahajournals.org/lookup/suppl/doi:10.1161/
CIRCULATIONAHA.116.021516/-/DC1.
Continuing medical education (CME) credit is available for
this article. Go to http://cme.ahajournals.org to take the quiz.
Circulation is available at http://circ.ahajournals.org.
REFERENCES
1. Sattelmair J, Pertman J, Ding EL, Kohl HW III, Haskell W, Lee IM.
Dose response between physical activity and risk of coronary
heart disease: a meta-analysis. Circulation. 2011;124:789795.
doi: 10.1161/CIRCULATIONAHA.110.010710.
2. Li J, Siegrist J. Physical activity and risk of cardiovascular disease
a meta-analysis of prospective cohort studies. Int J Environ Res
Public Health. 2012;9:391407. doi: 10.3390/ijerph9020391.
3. Physical Activity Guidelines Advisory Committee. Physical Activity
Guidelines Committee Report, 2008. Washington, DC: US Depart-
ment of Health and Human Services; 2008:683.
4. Manson JE, Greenland P, LaCroix AZ, Stefanick ML, Mouton CP,
Oberman A, Perri MG, Sheps DS, Pettinger MB, Siscovick DS.
Walking compared with vigorous exercise for the prevention of
cardiovascular events in women. N Engl J Med. 2002;347:716
725. doi: 10.1056/NEJMoa021067.
5. Wilmot KA, OFlaherty M, Capewell S, Ford ES, Vaccarino V.
Coronary heart disease mortality declines in the United States
from 1979 through 2011: evidence for stagnation in young
adults, especially women. Circulation. 2015;132:9971002. doi:
10.1161/CIRCULATIONAHA.115.015293.
6. Ford ES, Capewell S. Coronary heart disease mortality among
young adults in the U.S. from 1980 through 2002: concealed lev-
eling of mortality rates. J Am Coll Cardiol. 2007;50:21282132.
doi: 10.1016/j.jacc.2007.05.056.
7. Yang Q, Cogswell ME, Flanders WD, Hong Y, Zhang Z, Lousta-
lot F, Gillespie C, Merritt R, Hu FB. Trends in cardiovascular
health metrics and associations with all-cause and CVD mortality
among US adults. JAMA. 2012;307:12731283. doi: 10.1001/
jama.2012.339.
8. Matthews KA, Crawford SL, Chae CU, Everson-Rose SA, Sow-
ers MF, Sternfeld B, Sutton-Tyrrell K. Are changes in cardiovas-
cular disease risk factors in midlife women due to chronologi-
cal aging or to the menopausal transition? J Am Coll Cardiol.
2009;54:23662373. doi: 10.1016/j.jacc.2009.10.009.
9. Mehta LS, Beckie TM, DeVon HA, Grines CL, Krumholz HM, John-
son MN, Lindley KJ, Vaccarino V, Wang TY, Watson KE, Wenger
NK; American Heart Association Cardiovascular Disease in
Women and Special Populations Committee of the Council on
Clinical Cardiology, Council on Epidemiology and Prevention,
Council on Cardiovascular and Stroke Nursing, and Council on
Quality of Care and Outcomes Research. Acute myocardial infarc-
tion in women: a scientific statement from the American Heart
Association. Circulation. 2016;133:916947. doi: 10.1161/
CIR.0000000000000351.
10. Chomistek AK, Chiuve SE, Eliassen AH, Mukamal KJ, Willett WC,
Rimm EB. Healthy lifestyle in the primordial prevention of car-
Circulation. 2016;134:290299. DOI: 10.1161/CIRCULATIONAHA.116.021516

diovasculardisease among youngwomen. J Am Coll Cardiol.
2015;65:4351. doi: 10.1016/j.jacc.2014.10.024.
11. Mozaffarian D. Dietary and policy priorities for cardiovascular
disease, diabetes, and obesity: a comprehensive review. Circu-
lation. 2016;133:187225. doi: 10.1161/CIRCULATIONAHA.
115.018585.
12. Willett WC, Sampson L, Stampfer MJ, Rosner B, Bain C, Witschi
J, Hennekens CH, Speizer FE. Reproducibility and validity of a
semiquantitative food frequency questionnaire. Am J Epidemiol.
1985;122:5165.
13. Nurses Health Study. http://www.nurseshealthstudy.org/sites/
default/files/questionnaires/2009long.pdf.
14. Ainsworth BE, Haskell WL, Leon AS, Jacobs DR Jr, Montoye HJ,
Sallis JF, Paffenbarger RS Jr. Compendium of physical activities:
classification of energy costs of human physical activities. Med
Sci Sports Exerc. 1993;25:7180.
15. Wolf AM, Hunter DJ, Colditz GA, Manson JE, Stampfer MJ, Cor-
sano KA, Rosner B, Kriska A, Willett WC. Reproducibility and va-
lidity of a self-administered physical activity questionnaire. Int J
Epidemiol. 1994;23:991999.
16 .Rose GA BH, Gillum R, Prineas RJ. Cardiovascular survey meth-
ods. WHO monograph series no. 56. 1982:162165.
Downloaded from http://circ.ahajournals.org/ by guest on February 14, 2017
17. Hu FB, Stampfer MJ, Rimm E, Ascherio A, Rosner BA, Spiegel-
man D, Willett WC. Dietary fat and coronary heart disease: a
comparison of approaches for adjusting for total energy intake
and modeling repeated dietary measurements. Am J Epidemiol.
1999;149:531540.
18. Shiroma EJ, Lee IM. Physical activity and cardiovascular health:
lessons learned from epidemiological studies across age, gen-
der, and race/ethnicity. Circulation. 2010;122:743752. doi:
10.1161/CIRCULATIONAHA.109.914721.
19. Conroy MB, Cook NR, Manson JE, Buring JE, Lee IM. Past physi-
cal activity, current physical activity, and risk of coronary heart
disease. Med Sci Sports Exerc. 2005;37:12511256.
20. Thompson PD, Crouse SF, Goodpaster B, Kelley D, Moyna N, Pes-
catello L. The acute versus the chronic response to exercise. Med
Sci Sports Exerc. 2001;33(6 suppl):S438S445; discussion S452.
21. Ahmed HM, Blaha MJ, Nasir K, Rivera JJ, Blumenthal RS. Effects
of physical activity on cardiovascular disease. Am J Cardiol.
2012;109:288295. doi: 10.1016/j.amjcard.2011.08.042.
22. Hay J, Maximova K, Durksen A, Carson V, Rinaldi RL, Torrance
B, Ball GD, Majumdar SR, Plotnikoff RC, Veugelers P, Boul NG,
Wozny P, McCargar L, Downs S, Lewanczuk R, McGavock J. Physi-
cal activity intensity and cardiometabolic risk in youth. Arch Pediatr
Adolesc Med. 2012;166:10221029. doi: 10.1001/archpediat-
rics.2012.1028.
Circulation. 2016;134:290299. DOI: 10.1161/CIRCULATIONAHA.116.021516
Physical Activity and CHD in Young Women
23. Owen CG, Nightingale CM, Rudnicka AR, Sattar N, Cook DG,
Ekelund U, Whincup PH. Physical activity, obesity and cardio-
metabolic risk factors in 9- to 10-year-old UK children of white
European, South Asian and black African-Caribbean origin: the
Child Heart And health Study in England (CHASE). Diabetologia.
2010;53:16201630. doi: 10.1007/s00125-010-1781-1.
24. Boeke CE, Eliassen AH, Oh H, Spiegelman D, Willett WC, Tamimi
RM. Adolescent physical activity in relation to breast cancer risk.
Breast Cancer Res Treat. 2014;145:715724. doi: 10.1007/
s10549-014-2919-5.
25. Lee IM, Paffenbarger RS Jr, Hsieh CC. Time trends in physical
ORIGINAL RESEARCH
activity among college alumni, 1962-1988. Am J Epidemiol.
1992;135:915925.
26. Armstrong ME, Green J, Reeves GK, Beral V, Cairns BJ; Million Women
ARTICLE
Study Collaborators. Frequent physical activity may not reduce vas-
cular disease risk as much as moderate activity: large prospective
study of women in the United Kingdom. Circulation. 2015;131:721
729. doi: 10.1161/CIRCULATIONAHA.114.010296.
27. Chomistek AK, Cook NR, Flint AJ, Rimm EB. Vigorous-intensity
leisure-time physical activity and risk of major chronic disease in
men. Med Sci Sports Exerc. 2012;44:18981905. doi: 10.1249/
MSS.0b013e31825a68f3.
28. Lee IM, Sesso HD, Oguma Y, Paffenbarger RS Jr. Relative intensity
of physical activity and risk of coronary heart disease. Circulation.
2003;107:11101116.
29. Tanasescu M, Leitzmann MF, Rimm EB, Willett WC, Stampfer MJ,
Hu FB. Exercise type and intensity in relation to coronary heart
disease in men. JAMA. 2002;288:19942000.
30. Swain DP, Franklin BA. Comparison of cardioprotective ben-
efits of vigorous versus moderate intensity aerobic exer-
cise. Am J Cardiol. 2006;97:141147. doi: 10.1016/j.amj-
card.2005.07.130.
31. Weinstein AR, Sesso HD, Lee IM, Rexrode KM, Cook NR, Man-
son JE, Buring JE, Gaziano JM. The joint effects of physical
activity and body mass index on coronary heart disease risk in
women. Arch Intern Med. 2008;168:884890. doi: 10.1001/
archinte.168.8.884.
32. Ogden CL, Carroll MD, Kit BK, Flegal KM. Prevalence of child-
hood and adult obesity in the United States, 2011-2012. JAMA.
2014;311:806814. doi: 10.1001/jama.2014.732.
33. Li TY, Rana JS, Manson JE, Willett WC, Stampfer MJ, Colditz GA,
Rexrode KM, Hu FB. Obesity as compared with physical activity in
predicting risk of coronary heart disease in women. Circulation.
2006;113:499506. doi: 10.1161/CIRCULATIONAHA.105.574087.
July 26, 2016 299
 Frequency, Type, and Volume of Leisure-Time Physical Activity and Risk of Coronary
Heart Disease in Young Women
Andrea K. Chomistek, Beate Henschel, A. Heather Eliassen, Kenneth J. Mukamal and Eric B.
Rimm
Downloaded from http://circ.ahajournals.org/ by guest on February 14, 2017

Circulation. 2016;134:290-299
doi: 10.1161/CIRCULATIONAHA.116.021516
Circulation is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright 2016 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/134/4/290

Data Supplement (unedited) at:

http://circ.ahajournals.org/content/suppl/2016/07/25/CIRCULATIONAHA.116.021516.DC1

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/
SUPPLEMENTAL MATERIAL

Supplemental Table 1. Hazard ratios (95% CI) of coronary heart disease according to categories of cumulative
average physical activity, 1997 2011, Nurses Health Study II
Categories of Cumulative Average Physical Activity, MET-hrs/wk
Total 1 2 3 4 5 p-value for
Physical Activity (< 1) (1 5.9) (6 14.9) (15 29.9) ( 30) linear trend
Cases 99 159 130 86 70

Person-years 250,780 434,577 489,027 410,155 326,958

Age-adjusted 1.00 0.84 0.61 0.50 0.55 <.0001

(0.65, 1.08) (0.47, 0.80) (0.37, 0.67) (0.40, 0.75)

Multivariable* 1.00 0.91 0.75 0.67 0.80 0.16

(0.71, 1.18) (0.57, 0.98) (0.50, 0.91) (0.58, 1.11)
*The models were stratified by age (in months) and time period and included parental history of MI at or before age 60 years,
smoking, T.V. watching, oral contraceptive use, aspirin, AHEI-2010 score, postmenopausal status, menopausal hormone therapy use,
parity, pre-existing disease (diagnosis of hypertension or hypercholesterolemia at baseline).
Supplemental Table 2. Hazard ratios (95% CI) of coronary heart disease according to quartiles of early life
physical activity, 1997 2011, Nurses Health Study II
Quartiles of Physical Activity, MET-hrs/wk
1 2 3 4 p-value for
(< 24.6) (24.7 41.4) (41.5 65.1) (> 65.1) linear trend
Ages 12 22 years
Cases / Person-years 86 / 263,565 82 / 261,719 95 / 266,141 87 / 260,872

Age-adjusted 1.00 1.02 1.21 1.17 0.23

(0.75, 1.38) (0.90, 1.62) (0.86, 1.57)

Multivariable* 1.00 1.00 1.17 1.06 0.58

(0.73, 1.35) (0.87, 1.57) (0.78, 1.43)

MV* + adult PA 1.00 1.02 1.21 1.12 0.37

(0.75, 1.39) (0.90, 1.63) (0.83, 1.52)
1 2 3 4 p-value for
(< 17.5) (17.6 32) (32.1 50.8) (> 50.8) linear trend
Ages 23-34 years
Cases / Person-years 103 / 263,019 85 / 263,144 68 / 263,586 94 / 262,548

Age-adjusted 1.00 0.89 0.71 0.97 0.82

(0.66, 1.18) (0.52, 0.97) (0.73, 1.29)

Multivariable* 1.00 0.96 0.82 1.10 0.55

(0.72, 1.28) (0.60, 1.11) (0.83, 1.46)

MV* + adult PA 1.00 1.01 0.90 1.23 0.18

(0.76, 1.36) (0.65, 1.23) (0.92, 1.65)
*The models were stratified by age (in months) and time period and included parental history of MI at or before age 60 years,
smoking, T.V. watching, oral contraceptive use, aspirin, AHEI-2010 score, postmenopausal status, menopausal hormone therapy use,
parity, pre-existing disease (diagnosis of hypertension or hypercholesterolemia at baseline).
Supplemental Table 3. Hazard ratios (95% CI) of coronary heart disease according to quintiles of adult physical
activity at baseline, 1991 2011, Nurses Health Study II
Quintiles of Physical Activity, MET-hrs/wk
Baseline Physical Activity 1 2 3 4 5 p-value
(< 1) (1 5.9) (6 14.9) (15 29.9) ( 30) for linear
trend
Cases 151 118 115 74 86

Person-years 374,822 415,010 412,445 368,308 340,912

Age-adjusted 1.00 0.75 0.76 0.56 0.72 0.03

(0.59, 0.96) (0.59, 0.97) (0.42, 0.74) (0.55, 0.94)

Multivariable* 1.00 0.81 0.91 0.72 0.98 0.93

(0.63, 1.03) (0.71, 1.17) (0.54, 0.95) (0.74, 1.29)
*The models were stratified by age (in months) and time period and included covariates at baseline only: parental history of MI at or
before age 60 years, smoking, oral contraceptive use, aspirin, AHEI-2010 score, postmenopausal status, menopausal hormone therapy
use, parity, and pre-existing disease (includes a diagnosis hypertension or hypercholesterolemia at baseline).
Supplemental Table 4. Hazard ratios (95% CI) of coronary heart disease according to quintiles of physical activity, stratified by cardiovascular risk
factors, 1991 2011, Nurses Health Study II
Quintiles of Physical Activity, MET-hrs/wk
1 2 3 4 5 p-value for P for interaction
(< 1) (1 5.9) (6 14.9) (15 29.9) ( 30) linear trend
Non-smoker
Cases / Person-years 127 / 374,451 80 / 332,313 77 / 359,410 52 / 334,768 61 / 337,713

Multivariable* 1.00 0.87 (0.66, 1.16) 0.80 (0.60, 1.06) 0.61 (0.44, 0.85) 0.78 (0.57, 1.07) 0.09
0.58
Current smoker
Cases / Person-years 55 / 45, 214 39 / 36,644 24 / 34,874 15 / 28,721 14 / 27,389

Multivariable* 1.00 1.28 (0.83, 1.97) 0.83 (0.50, 1.38) 0.58 (0.32, 1.05) 0.60 (0.32, 1.12) 0.02
Age < 50
Cases / Person-years 80 / 289,467 64 / 276,621 46 / 286,590 32 / 261,886 32 / 257,087

Multivariable* 1.00 1.02 (0.73, 1.42) 0.77 (0.53, 1.11) 0.63 (0.42, 0.96) 0.70 (0.46, 1.07) 0.04
0.94
Age 50
Cases / Person-years 102/ 130, 200 55 / 92,335 55 / 107,693 35 / 101,602 43 / 108,015

Multivariable* 1.00 0.90 (0.65, 1.26) 0.83 (0.60, 1.16) 0.62 (0.42, 0.91) 0.80 (0.55, 1.15) 0.15
Non-drinkers
Cases / Person-years 93 / 193,741 55/ 153,732 48 / 145,660 28 / 120,912 27 / 108,210

Multivariable* 1.00 1.00 (0.71, 1.41) 0.91 (0.64, 1.30) 0.69 (0.45, 1.06) 0.75 (0.48, 1.17) 0.10

Drinkers
0.94
Cases / Person-years 89 / 225,926 64 / 215,225 53 / 248,623 39 / 242,576 48 / 256,892

Multivariable* 1.00 0.97 (0.70, 1.34) 0.75 (0.53, 1.06) 0.62 (0.42, 0.91) 0.78 (0.54, 1.13) 0.14
AHEI < 49
Cases / Person-years 119 / 253,259 73 / 207,693 60 / 193,130 35 / 151,760 29 / 118,223

Multivariable* 1.00 1.01 (0.75, 1.36) 0.88 (0.64, 1.21) 0.68 (0.46, 0.99) 0.74 (0.49, 1.12) 0.05 0.98

AHEI 49
Cases / Person-years 51 / 149, 983 39 / 148,173 38 / 188,469 29 / 200,559 40 / 235,144

Multivariable* 1.00 0.95 (0.62, 1.45) 0.75 (0.49, 1.15) 0.59 (0.37, 0.94) 0.70 (0.46, 1.06) 0.08
*The models were stratified by age (in months) and time period and included parental history of MI at or before age 60 years, smoking, T.V. watching, oral contraceptive use,
aspirin, AHEI-2010 score, postmenopausal status, menopausal hormone therapy use, parity, pre-existing disease (diagnosis of hypertension or hypercholesterolemia at baseline).
Carolyn: Welcome to Circulation on the Run, your weekly podcast summary and backstage pass
to the journal and its editors. I'm Dr. Carolyn Lam, Associate Editor from the National
Heart Centre and Duke National University of Singapore. I am so excited to be joined in
just a moment by Dr. Andrea [inaudible 00:00:21] and Dr. Wendy Post to discuss the
feature paper this week about leisure-time physical activity and the risk of coronary
heart disease in young women. First, here's the summary of this week's issue.

The first paper, by Dr. Bohula and colleagues at the TIMI Study Group at Brigham and
Women's Hospital in Boston, Massachusetts, aim to test the hypothesis that an
atherothrombotic risk stratification tool may be useful to identify high-risk patients who
have the greatest potential for benefit from more intensive secondary preventive
therapy such as treatment with Vorapaxar following a myocardial infarction. As a
reminder, Vorapaxar is a first-in-class anti-platelet agent that inhibits thrombin-
mediated activation of platelets via the protease activator receptor 1. The authors
studied almost 8,600 stable patients with a prior myocardial infarction followed for a
median of two and a half years.

In the thrombin receptor antagonist and secondary prevention of athrothrombotic

ischemic events, TIMI 50 trial. They identified nine independent risk predictors which
were age, diabetes, hypertension, smoking, peripheral artery disease, prior stroke, prior
coronary bypass grafting, heart failure and renal dysfunction. A simple integer-based
scheme using these predictors showed a strong graded relationship with the rates of
cardiovascular death, myocardial infarction or ischemic stroke. Moreover, the net
clinical outcome was increasingly favorable with Vorapaxar across the risk groups.

In summary, this paper provides a practical strategy that could be used by clinicians to
assist with risk stratification and therapeutic decision-making regarding Veropaxar use
for secondary prevention after myocardial infarction.

The next paper is by first author Dr. [inaudible 00:02:40] and corresponding authors, Dr.
[Gerstein 00:02:43] from the Beth Israel Deaconess Medical Center and Dr. [Carr
00:02:47] from the Broad Institute of Harvard and MIT, who look at aptamer-based
proteomic profiling. Now DNA aptamers are [alu 00:02:57] nucleotides of approximately
50 base pairs in length selected for their ability to bind proteins with high specificity and
affinity. They therefore holds considerable promise for biomarker and pathway
discovery in cardiovascular diseases.

These authors applied a novel technology that uses single-stranded DNA aptamers to
measure over 1,100 proteins in a single blood sample. They applied this to a model of
planned myocardial injury and that is patients undergoing septal ablation for
hypertrophic cardiomyopathy, and they found that 217 proteins were significantly
changed in the peripheral vein blood after planned myocardial injury in this derivation
cohort. They validated 79 of these proteins in an independent cohort. Furthermore,
among 40 validated proteins that increase within one hour after myocardial injury, 23
were also elevated in patients with spontaneous myocardial infarction.

Finally, the authors applied this to archive samples from the Framingham heart study
 and showed 156 significant protein associations with the Framingham risk score. This
study is so exciting because it highlights any merging proteomics tool that captures a
large number of low abundance analytes with high sensitivity and precision, thus
providing important proof of principle for future clinical applications and this is
discussed in an excellent editorial that accompanies this paper by doctors Graham
[Malini 00:04:37], [Lau Enleui 00:04:39] from the University of Ottawa Heart Institute.

The next paper is by Dr. [Anter 00:04:51] and colleagues from the Beth Israel Deaconess
Medical Center in Boston, Massachusetts, who looked at post infarction, reentrant
ventricular tachycardia and addressed the problem that in vivo descriptions of
ventricular tachycardia circuits are currently limited by insufficient spatiotemporal
resolution. The authors therefore utilize a novel, high resolution mapping technology to
characterize the electrophysiological properties of these reentrant circuits in 15 swine.

The main finding was that the zones of slow conduction within the reentrant circuits
with the inward and outward curvatures while conduction velocity in the comment
channel isthmus itself was nearly normal. The authors further demonstrated that
entrainment mapping over estimated the true size of the isthmus. Thus, the conclusion
was that high resolution activation mapping of ventricular tachycardia may better guide
ablation therapy and ablation at zones of high curvature may be an attractive target for
ablation.

The final papers from first author, Dr. [Tang 00:06:08] and corresponding author Dr.
[Fitzgerald 00:06:10] from the University of Pennsylvania Perlman School of Medicine in
Philadelphia. These authors studied the cardiovascular consequences of prostanoid I-
receptor deletion in microsomal prostaglandin E synthase-1 deficient hyperlipidemic
mice. The clinical background to this research question is that inhibitors of
cyclooxygenase-2 or Cox-2 are well-known to relieve pain, fever and inflammation by
suppressing biosynthesis of prostacyclin and prostaglandin E2.

However, suppression of these prostaglandins particularly prostacyclin by Cox-2

inhibitors or deletion of the I-prostanoid receptor for prostacyclin is known to accelerate
atherogenesis and enhance thrombogenesis in mice. In contrast, deletion of the
microsomal prostaglandin E synthase1 has been shown to suppress PGE2 but increase
biosynthesis of prostacyclin. It therefore confers analgesia while attenuating
atherogenesis and does not predispose mice to thrombogenesis. Therefore, possibly
contributing to cardiovascular efficacy.

In this particular study, therefore, the authors sought to determine the relative
contribution of suppressing PGE2 versus augmenting prostacyclin to the impact of
depletion of microsomal prostaglandin E synthase-1 in hyperlipidemic mice. The main
findings were that augmentation of prostacyclin is the dominant contributor to the
favorable thrombogenic profile of microsomal prostaglandin E synthase-1 depletion in
these atherosclerotic mice while suppression of PGE2 accounted for the protective
effects in atherosclerosis and the exciting clinical take-home message is that inhibitors
of the microsomal prostaglandin E synthase-1 may be less likely to cause cardiovascular
adverse effects than NSAIDS or specific inhibition of Cox-2. Those were the highlights of

COTR134_04 Page 2 of 6
 this week. Now for our feature paper.

Our feature paper today is entitled "The frequency, [type 00:08:41] and volume of
leisure time physical activity and risk of coronary heart disease in young women" and I
am so excited to be joined by two lovely ladies today to discuss this paper. First, the first
and corresponding author Dr. Andrea [Comastick 00:08:58] from the School of Public
Health of Indiana University Bloomington and Dr. Wendy Post, associate editor from the
Johns Hopkins University. Welcome Andrea and Wendy.

Andrea: Hi. Thanks.

Wendy: Thank you so much for having us.

Carolyn: I am just so excited that we are talking about a paper about women being discussed by
women. What more could you ask for? I have to say this is a first for Circulation on the
Run, which is why Im just so excited, so lets get straight into it.

Andrea, maybe I could just ask you to start by sharing the story of how you and your
team came up with some new questions and new data because Im sure a lot of
listeners are thinking theres a lot of data on exercise and how good it is for
cardiovascular health in women already.

Andrea: Yeah, that's a great question. When we started talking about conceptualizing this paper,
the first thing was to focus on younger women. Most of the previous work on physical
activity and heart disease has been in older adults and that's primarily because it's older
adults that have heart attacks. Its hard to get a large enough study of young women
that has enough coronary heart disease events to be able to study this. We were
fortunate where we had a large cohort in the nurses health study too of women and
because its been followed for over 20 years, we had enough events to be able to
examine this association.

We did want to think about, "Okay, what can we add?" because theres a lot of
information about just overall physical activity and health, so what can we do
differently? Im pretty familiar with the physical activity guidelines and really tried to
look at what in the guidelines currently and then what could we add? What could be of
interest when they start revising the guidelines which is actually going to happen very
soon.

That was when we started focusing on, "Okay, instead of looking at just overall activity,
look at intensity, comparing moderate and vigorous." We also wanted to look at
frequency of physical activity and looking at frequency but also adjusted for a total time
or total amount of physical activity that somebody does. Then we are also, the third
thing was that we thought was important was looking at adolescent physical activity.

We know that kids, unfortunately as they get older and get into their teenage years,
their activity declines quite a bit. Looking at how this physical activity during
adolescence earlier life impact coronary heart disease risk in adulthood. Those were the

COTR134_04 Page 3 of 6
 three main things that we were focusing on when we first conceptualized the paper.

Carolyn: Nice. Tell us, what did you find?

Andrea: We did find that exercise is just as beneficial in younger woman as it is in older adults,
which is great. We also found that moderate intensity exercise is just as beneficial as
vigorous intensity exercise, which I think is a really important message to get out there. I
think a lot of people, especially those that are really inactive to begin with are
completely intimidated about the fact of trying to think about going to a gym or trying
to jog or run a marathon or something like that.

I think really emphasizing that moderate intensity activity is beneficial and we found
that walking was actually the most beneficial activity that we looked at in our study, that
brisk walking was really really good for everybody and really lowered risk of coronary
heart disease.

Carolyn: Hooray.

Andrea: Yeah, and the other thing we found which might be of interest for those that are also
extremely busy, especially this target population where a lot of people are moms and
working was that frequency didn't seem to matter, that as long as people were
exercising for a couple hours a week that they should be that they could accumulate it in
a couple times a week or they could do it more frequently, four or five times a week. It
didn't seem to matter.

Carolyn: Thats cool. You know what? I think a lot of these things we'll also discuss at the Editorial
Board when we're looking at this paper. Wendy, we promised that we would give a
backstage pass to the Editorial Board and The Journal, so could you share a little bit
about what we talked about there?

Wendy: Well, the Editorial Board was really excited about this paper. We loved the emphasis on
young women and the important public health message about how we need to get out
there and move and exercise to reduce our risk for cardiovascular disease. As was
mentioned, there have been previous studies that also show the benefit of exercise but
the Editorial Board especially liked the large sample size, the long duration of follow-up,
the number of events that had been accrued that allowed for sophisticated analyses,
adjustment for confounders and the very rigorous study design and excellent statistical
methods that have been used in this study and so many other studies from the nurses
health study, but I think we particularly just loved the message. The message was great.

We need to get out there and move. We need to tell our patients, especially young
women, that now we have data that if you start exercising now, it will help in the future
but also the study showed that if you hadn't exercised much in early life thats starting
to exercise more proximal to the event was also important as well.

Carolyn: Thank you Wendy. I also remember that we talked about the lack of interaction with
body mass index, and I thought that was a great message. Andrea, could you maybe

COTR134_04 Page 4 of 6
 share a little bit about that?

Andrea: Yeah, this is something that previous investigators have looked at the interaction
between body mass index and exercise. Unfortunately, weve all found the same thing
so it doesnt seem to matter whether women are normal weight or overweight or obese
that they still get benefit when they exercise, and I think thats really encouraging. I
know a lot of people might start to exercise because they really want to drop some
weight but just trying to emphasize even if the numbers on the scale aren't changing,
that exercise still has all these really great benefits for heart disease and also for many
other diseases.

Carolyn: Exactly. Can I just ask both of you and maybe Ill start with Andrea, what will you do
different now both as a woman and as a clinician seeing women now that you know
what you do from your data?

Andrea: Well, Im not a clinician. Im an epidemiologist so unfortunately I dont get to see

patients and counsel them although I do try to talk to community members as a public
health person and really get in the community on board with what were talking about. I
just try to tell people, I actually talked to a group of people last week, and just trying to
say, "Anything is better than nothing and just trying to even start with some short
walks." Again, just emphasizing you dont have to go to a gym or you dont have to be
doing anything that's super strenuous but just do stuff that feels good and just try to get
your heart rate up a little bit like going out for a brisk walk. I think that's my main
message that I try to tell everybody is at least start with something and get moving a
little bit.

Carolyn: I love that. Wendy?

Wendy: I like to emphasize the data about brisk walking. I thought that was great because many
of our patients dont want to join a gym, dont have the time to join a gym so just
getting out and walking is fabulous exercise and now we have the data here that in
young women that after 20 years of follow-up, brisk walking was associated with I think
it was a 35% reduction in risk for cardiovascular disease during follow-up.

In addition, I liked the message about the total amount of time that you spend
exercising in a week is whats important. It doesnt matter whether you divide that into
seven days a week to get to that same amount of time or whether you do it in bursts of
three days a week, and I think thats particularly important for the many women who
have so many different responsibilities and may not have time every day to go out and
exercise. The days that you do have time, just exercise a little bit more those days, so
lots of really important messages for our patients and for ourselves.

Carolyn: I really couldnt agree more and just from my point of view, because I see a lot of
patients in Asia and I do acknowledge just like you did, Andrea, in your paper that your
data are predominantly in white populations. Still one of the messages I like to get out
to the women I see is we have very skinny women and when I see younger women, and
I really like emphasizing that, "Hey, just because youre not struggling with an obesity

COTR134_04 Page 5 of 6
 issue or just because youre young, it doesnt mean you dont need to exercise and that
we all should just get moving." Thank you very, very much for that Andrea.

Andrea: Oh, no. It's my pleasure and thank you for having me come on today and talk about this.

Carolyn: Thank you too, Wendy. Do you have any other comments?

Wendy: No, but congratulations on your publication, Andrea.

Andrea: Oh, thank you so much, Wendy. I was really happy to get the message that guys were
excited about it. Thank you so much.

Carolyn: Youve been listening to Circulation on the Run. Thank you for joining us this week and
please tune in next week.

COTR134_04 Page 6 of 6