Years of Able Life in Older Persons-The Role of Cardiovascular Imaging
and Biomarkers: The Cardiovascular Health Study
Alshawabkeh, L. I., Yee, L. M., Gardin, J. M., Gottdiener, J. S., Odden, M. C.,
Bartz, T. M., ... & Wallace, R. B. (2015). Years of Able Life in Older Persons—The
Role of Cardiovascular Imaging and Biomarkers: The Cardiovascular Health
Study. Journal of the American Heart Association, 4(4), e001745.
doi:10.1161/JAHA.114.001745
10.1161/JAHA.114.001745
John Wiley & Sons, Ltd.
Version of Record
http://cdss.library.oregonstate.edu/sa-termsofuse
ORIGINAL RESEARCH
Years of Able Life in Older Persons—The Role of Cardiovascular
Imaging and Biomarkers: The Cardiovascular Health Study
Laith I. Alshawabkeh, MD, MSc; Laura M. Yee, MS; Julius M. Gardin, MD, MBA; John S. Gottdiener, MD; Michelle C. Odden, PhD;
Traci M. Bartz, MS; Alice M. Arnold, PhD; Kenneth J. Mukamal, MD, MPH; Robert B. Wallace, MD, MSc
Background-—As the U.S. population grows older, there is greater need to examine physical independence. Previous studies have
assessed risk factors in relation to either disability or mortality, but an outcome that combines both is still needed.
Methods and Results-—The Cardiovascular Health Study is a population-based, prospective study where participants underwent
baseline echocardiogram, measurement of carotid intima-media thickness (IMT), and various biomarkers, then followed for up to
18 years. Years of able life (YAL) constituted the number of years the participant was able to perform all activities of daily living.
Linear regression was used to model the relationship between selected measures and outcomes, adjusted for confounding
variables. Among 4902 participants, mean age was 72.65.4 years, median YAL for males was 8.8 (interquartile range [IQR], 4.3
to 13.8) and 10.3 (IQR, 5.8 to 15.8) for females. Reductions in YAL in the fully adjusted model for females and males, respectively,
were: 1.34 (95% confidence interval [CI], 2.18, 0.49) and 1.41 (95% CI, 2.03, 0.8) for abnormal left ventricular (LV)
ejection fraction, 0.5 (95% CI, 0.78, 0.22) and 0.62 (95% CI, 0.87, 0.36) per SD increase in LV mass, 0.5 (95% CI,
0.7, 0.29) and 0.79 (95% CI, 0.99, 0.58) for IMT, 0.5 (95% CI, 0.64, 0.37) and 0.79 (95% CI, 0.94, 0.65) for
N-terminal pro-brain natriuretic peptide, 1.08 (95% CI, 1.34, 0.83) and 0.73 (95% CI, 0.97, 0.5) for high-sensitivity
troponin-T, and 0.26 (95% CI, 0.42, 0.09) and 0.23 (95% CI, 0.41, 0.05) for procollagen-III N-terminal propeptide. Most
tested variables remained significant even after adjusting for incident cardiovascular (CV) disease.
Conclusions-—In this population-based cohort, variables obtained by CV imaging and biomarkers of inflammation, coagulation,
atherosclerosis, myocardial injury and stress, and cardiac collagen turnover were associated with YAL, an important outcome that
integrates physical ability and longevity in older persons. ( J Am Heart Assoc. 2015;4:e001745 doi: 10.1161/
JAHA.114.001745)
Key Words: activities of daily living • aging • biomarkers • imaging
T
he concept of “healthy” or “successful” aging has been
the subject of research for decades. It lacks a universal
definition, however, because older adults’ perception of
healthy aging is heterogeneous and might not be in complete
accord with scientific definitions.1,2 Nevertheless, maintenance of physical ability remains a consistent component in
any definition. Given that the number of persons 85 and older
is projected to double and reach 19 million by 2025,3
investigating determinants of mortality is of great importance
—but preserving physical ability, a marker of independence, is
a key goal for healthy aging. In fact, older persons rank
maintaining independence as more important than staying
alive as a health outcome.4 Cardiovascular (CV) well-being is
at the center of this paradigm.
From the Division of Cardiovascular Medicine (L.I.A.), Department of Internal Medicine (R.B.W.), Carver College of Medicine and the College of Public Health (L.I.A.,
R.B.W.), University of Iowa, Iowa City, IA; Department of Medicine, Hackensack University Medical Center, Hackensack, NJ (J.M.G.); Division of Cardiovascular Medicine,
Department of Medicine, University of Maryland School of Medicine, Baltimore, MD (J.S.G.); School of Biological and Population Health Sciences, College of Public
Health and Human Sciences, Oregon State University, Corvallis, OR (M.C.O.); Department of Biostatistics, University of Washington, Seattle, WA (L.M.Y., T.M.B., A.M.A.);
Divisions of General Medicine and Primary Care, Beth Israel Deaconess Medical Center and Harvard Medical School, Boston, MA (K.J.M.).
Accompanying Data S1 and Tables S1 through S5 are available at http://jaha.ahajournals.org/content/4/4/e001745/suppl/DC1
This article was handled independently by Holli A. DeVon, PhD, RN, as a guest editor. The editors had no role in the evaluation of the manuscript or in the decision
about its acceptance.
Correspondence to: Laith I. Alshawabkeh, MD, MSc, Division of Cardiovascular Medicine, Department of Medicine, University of Iowa Hospitals and Clinics, 200
Hawkins Dr, GH, Iowa City, IA 52242. E-mail: laithalsh@gmail.com
Received February 25, 2015; accepted March 27, 2015.
ª 2015 The Authors. Published on behalf of the American Heart Association, Inc., by Wiley Blackwell. This is an open access article under the terms of the Creative
Commons Attribution-NonCommercial License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited and is
not used for commercial purposes.
DOI: 10.1161/JAHA.114.001745
Journal of the American Heart Association
Downloaded from http://jaha.ahajournals.org/ at OREGON STATE UNIVERSITY on June 10, 2015
1
Imaging and Biomarkers, Years of Able Life
Alshawabkeh et al
Methods
Study Population
The Cardiovascular Health Study (CHS) is a community-based,
prospective, observational study that recruited adults
≥65 years to study risk factors and long-term outcomes for
CV disease (CVD). The cohort included 5888 participants
(5201 were recruited in 1989–1990; an additional 687
African Americans were recruited in 1992–1993). The institutional review board at each center approved the study, and
all participants provided informed consent. Participants
underwent baseline clinical examinations, which included
medical history, physical examination, assessment of activities of daily living (ADLs), and various imaging and laboratory
procedures. Participants were contacted every 6 months for
follow-up, alternating between a telephone interview and a
clinic visit through 1999, and only phone calls thereafter
(Figure). The design, rationale, and examination details of the
CHS have been published elsewhere.7
Imaging Procedures
Transthoracic echocardiograms were obtained in 1989–1990
for the original cohort and 1994–1995 for the AfricanAmerican cohort. Semiquantitative LV ejection fraction
(LVEF), left atrial (LA) diameter, transmitral mitral peak early
(E) velocity, peak atrial (A) velocity, and E/A ratio, LV relative
wall thickness (LV RWT), and LV mass were defined using
previously specified criteria.8
The carotid arteries were evaluated at baseline with
high-resolution B-mode ultrasonography. A composite measure that combined the maximum common carotid artery
intima media thickness (IMT) and maximum internal IMT was
obtained by averaging these 2 measurements after
Figure. Time chart of enrollment of the first cohort and second (African-American) cohort,
in addition to timing of performing the cardiovascular imaging and measurement of the
biomarkers in the Cardiovascular Health Study. CRP indicates C-reactive protein; hsTNT,
high-sensitivity cardiac troponin-T; LDL, low-density lipoprotein; NT-proBNP, N-terminal
probrain natriuretic peptide; PIIINP, procollagen III N-terminal propeptide; YAL, years of able
life.
DOI: 10.1161/JAHA.114.001745
Journal of the American Heart Association
Downloaded from http://jaha.ahajournals.org/ at OREGON STATE UNIVERSITY on June 10, 2015
2
ORIGINAL RESEARCH
Functional decline undermines quality of life and causes
substantial social and economic strain on patients and their
families. Impairment of physical ability can occur at any point
in time in individuals with chronic diseases. This varies
depending on the burden of comorbidities and many other
complex factors that are poorly understood.5 Although many
traditional risk factors correlate with mortality, untangling
their role in developing functional impairment has proven
difficult because diseases that lead to death often accelerate
functional decline.6 Thus, devising a health outcome that
combines both longevity and functional ability is more
relevant to patient interests than either alone. Furthermore,
this health outcome should preferably take into account that
some subjects move in and out of the disabled state.
Disability results from complex and heterogeneous processes in older persons. We hypothesized that CV imaging
measures of left ventricular (LV) structure and function, carotid
intima thickness, and biomarkers of inflammation, coagulation,
atherosclerosis, myocardial injury and stress, and cardiac
extracellular collagen turnover are not only associated with
mortality, but also with the likelihood of maintaining physical
ability in a large sample of community-dwelling older persons.
Imaging and Biomarkers, Years of Able Life
Alshawabkeh et al
Biomarker Assays
Biomarkers used in this study were obtained from baseline
blood samples, except for procollagen III N-terminal
propeptide (PIIINP), which was assayed from blood collected in 1996–1997. Samples were immediately centrifuged at 3000g for 10 minutes at 4°C. Aliquots of plasma
were stored in a central laboratory at 70°C. C-reactive
protein (CRP) was measured by a high-sensitivity (hs)
immunoassay, with an interassay coefficient of variation of
6.25%.
N-terminal probrain natriuretic peptide (NT-proBNP) was
measured with the Elecsys 2010 system (Roche Diagnostics,
Indianapolis, IN) with a coefficient of variation of 2% to 5%.
High-sensitivity cardiac troponin-T (hsTNT) concentrations
were measured with reagents on an Elecsys 2010 analyzer
(Roche Diagnostics, Indianapolis, IN), with an analytical
measurement range of 3 to 10 000 pg/mL. The value at
the 99th percentile cutoff from a healthy reference population
(n=616) was 13.5 pg/mL. PIIINP was determined by a
coated-tube radioimmunoassay as described previously using
commercial antisera specifically directed against the amino
terminal peptide (Orion Diagnostica, Espoo, Finland). Interand intraassay variations for determining PIIINP are both
approximately 5%. Sensitivity (lower detection limit) is
1.5 ng/mL.
Clinical Endpoints and Variable Definitions
The primary outcome, years of able life (YAL), was defined
as the number of years that the participant is able to
perform all 6 ADLs without difficulty: walking around the
home; getting out of bed or a chair; eating; dressing; and
bathing or showering or using the toilet. This definition
allows for recovery from difficulty. We chose ADLs because
difficulty to perform any of them makes it unlikely that a
person is able to maintain physical independence. Years
of life (YOL) was defined as years of observed life
from enrollment to death or end of follow-up (maximum
18 years).
Data on mortality and self-reported limitations in ADLs
were collected every 6 months, with some exceptions
mentioned in Data S1. Calculation and imputation of YOL
and YAL are also included in Data S1. Given that many of
the tested variables have been associated with mortality
and/or physical disability, we evaluated the secondary
outcome, the percentage of YAL:YOL, which was defined
as the percentage of observed years spent free of impairment in ADLs (Data S1).
DOI: 10.1161/JAHA.114.001745
Statistical Analysis
We excluded participants who reported any ADL difficulty at
baseline (N=436) or did not have LVEF, LA dimension, peak E
velocity, or peak A velocity measures (N=550). We did not
exclude participants with prevalent coronary heart disease
(CHD) because many of these participants had preserved ADL,
and including them is more representative of older cohorts
who have higher burden of disease. The analysis set was
therefore composed of 4902 participants. Previously imputed
data were employed for missing covariate data (Data S1).
Further exclusions were made based on the availability of
biomarker and echocardiography data: LV mass and LV RWT
measures were available for 3374 participants, NT-proBNP
was available for 3776 participants, hsTnT was available
for 3694 participants, and PIIINP was available for 3173
participants.
We used multiple linear regression procedures to model
the relationship between exposures of interest and outcome
measures YAL and YOL, stratified by sex. For each outcome,
we ran 2 sets of models. The first set was adjusted for age,
race, body mass index (BMI), and BMI-squared. The second
set was additionally adjusted for prevalent smoking, arthritis,
diabetes, cancer, glomerular filtration rate (eGFR), systolic
blood pressure (SBP), antihypertensive medication use,
stroke, congestive heart failure (CHF), and CHD, which
included myocardial infarction (MI), angina, and revascularization (see Data S1 for definitions). LVEF, LA dimension, peak
E velocity, peak A velocity, E/A ratio, LV mass, LV RWT,
carotid IMT, hsCRP, fibrinogen, low-density lipoprotein (LDL)
cholesterol, NT-proBNP, PIIINP, and hsTNT were evaluated in
separate models one at a time. Because the associations of
interest have not been shown previously, we wanted to look at
each marker individually. No significant multicollinearity was
detected in the models. Outliers did not exert significant
leverage, and in cases where inference changed, the values of
outliers were deemed reasonable and were retained in the
models.
PIIINP was evaluated at 7 and 4 years after baseline for the
original and minority cohorts, respectively (year 1996–1997)
for 3173 participants. Thus, YAL could only have a maximum
of 11 years in the PIIINP analyses, as compared with 18 years
for the other analyses.
Additionally, sensitivity analysis by the time to the first ADL
difficulty using Cox proportional hazard models was performed, adjusted for the 2 sets of covariates used in the
primary analysis, and further adjusted for incident CHF, MI,
and stroke. Compared to the primary analysis, this approach
would fail to capture persons who regain physical ability after
an acute insult (e.g., MI or stroke), but would allow us to
adjust for incident disease, which would help to rule out
associations that were solely owing to incident events. A total
Journal of the American Heart Association
Downloaded from http://jaha.ahajournals.org/ at OREGON STATE UNIVERSITY on June 10, 2015
3
ORIGINAL RESEARCH
standardization. When present, focal plaques were included in
measurement of the maximum IMT.9
Imaging and Biomarkers, Years of Able Life
Alshawabkeh et al
ORIGINAL RESEARCH
Table 1. Baseline Characteristics by Categories of Years of Able Life for Male and Female Participants in CHS
Categories of Years of Able Life
Males (n=2133)
Females (n=2769)
0 to <5
5 to <10
10 to <15
15 to 18
0 to <5
5 to <10
10 to <15
15 to 18
Characteristics at Baseline
n=586
n=660
n=397
n=490
n=551
n=814
n=614
n=790
Age, ySD
766.4
73.75.5
71.84.3
69.93.4
75.46
73.35.3
71.24.2
69.53.4
Black, n (%)
54 (9.2)
82 (12.4)
34 (8.6)
57 (11.6)
81 (14.7)
113 (13.9)
73 (11.9)
96 (12.2)
26.13.8
26.33.7
26.83.6
26.33.2
26.55.5
26.65.2
26.64.8
26.24.2
Never
172 (29.4)
201 (30.5)
133 (33.5)
190 (38.8)
298 (54.1)
450 (55.3)
353 (57.5)
472 (59.7)
Former
345 (58.9)
387 (58.6)
225 (56.7)
271 (55.3)
167 (30.3)
262 (32.2)
193 (31.4)
231 (29.2)
Current
69 (11.8)
72 (10.9)
39 (9.8)
29 (5.9)
86 (15.6)
102 (12.5)
68 (11.1)
87 (11)
CHD, n (%)
209 (35.7)
168 (25.5)
77 (19.4)
72 (14.7)
127 (23)
125 (15.4)
71 (11.6)
51 (6.5)
CHF, n (%)
58 (9.9)
29 (4.4)
6 (1.5)
5 (1)
38 (6.9)
23 (2.8)
10 (1.6)
5 (0.6)
Stroke, n (%)
52 (8.9)
33 (5)
9 (2.3)
9 (1.8)
32 (5.8)
17 (2.1)
8 (1.3)
3 (0.4)
Normal
355 (60.6)
437 (66.2)
284 (71.5)
390 (79.6)
370 (67.2)
614 (75.4)
478 (77.9)
649 (82.2)
IFG
74 (12.6)
87 (13.2)
64 (16.1)
56 (11.4)
63 (11.4)
87 (10.7)
75 (12.2)
88 (11.1)
Diabetes
2
BMI (kg/m ), meanSD
Smoking status, n (%)
Diabetes ADA status, n (%)
157 (26.8)
136 (20.6)
49 (12.3)
44 (9)
118 (21.4)
113 (13.9)
61 (9.9)
53 (6.7)
SBP
14022.7
137.921
135.320.3
132.719.4
14324.1
139.221.9
135.920.3
132.119.3
Hypertensive medication, n (%)
314 (53.6)
293 (44.4)
167 (42.1)
169 (34.5)
316 (57.4)
410 (50.4)
258 (42)
289 (36.6)
eGFR by creatinine
62.719.4
66.416.8
69.415.7
70.715.2
66.119.9
69.618.8
69.816.8
71.515.7
Arthritis, n (%)
295 (50.3)
270 (40.9)
178 (44.8)
157 (32)
347 (63)
440 (54.1)
356 (58)
344 (43.5)
Cancer, n (%)
104 (17.8)
96 (14.6)
48 (12.1)
69 (14.1)
101 (18.3)
130 (16.0)
68 (11.1)
79 (10.0)
ADA indicates American Diabetes Association; BMI, body mass index; CHD, coronary heart disease (myocardial infarction, angina, coronary artery bypass grafting, or angioplasty); CHF,
congestive heart failure; CHS, Cardiovascular Health Study; eGFR, estimated glomerular filtration rate; IFG, impaired fasting glucose; SBP, systolic blood pressure.
of 1099 participants had an incident CVD before developing
ADL difficulty.
Results
for females), never smoked, and had less prevalent CHD, CHF,
stroke, diabetes, hypertension (HTN), hypertensive medication
use, and arthritis. Additionally, subjects in the highest
category of YAL had smaller IMT and lower detectable values
of all biomarkers, except LDL, which was higher.
Baseline Characteristics
For the 4902 participants in our study, mean (SD) age was
72.65.4 years. Median YAL was 8.8 (interquartile range
[IQR], 4.3 to 13.8) years for males and 10.3 (IQR, 5.8 to 15.8)
years for females. Median YOL was 11.8 (IQR, 6.8 to 17.8)
years for males and 15.3 (IQR, 9.8 to 18) years for females.
At baseline, 18.4% of the participants had CHD (24.6% in
males and 13.5% in females) and 3.55% had CHF (4.6% in
males and 2.7% in females), whereas 13.5% of men and 4.7%
of women had abnormal LVEF.
Baseline characteristics of participants per categories of
YAL are shown in Tables 1 and 2. Compared to the lowest
category, subjects in the highest category of YAL were
younger (69.9 vs. 76 years for males and 69.5 vs. 75.4 years
DOI: 10.1161/JAHA.114.001745
Imaging
An abnormal LVEF was associated with 2.49 and 2.38 fewer
observed YAL and 3.14 and 2.51 fewer observed YOL in
females and males, respectively (P<0.001), after adjustment
for age, race, and BMI (Tables 3 and 4). This association
remained strong and statistically significant after adjustment
for chronic health conditions at baseline. Persons with
abnormal LVEF had 1.34 and 1.41 fewer observed YAL and
2.08 and 1.41 fewer observed YOL in females and males,
respectively (P<0.01). Furthermore, male participants with an
abnormal LVEF spent 3.6% less of their observed years of life
being able (P=0.01). However, this relationship was not
significant for female participants (P=0.3; Table S1).
Journal of the American Heart Association
Downloaded from http://jaha.ahajournals.org/ at OREGON STATE UNIVERSITY on June 10, 2015
4
Journal of the American Heart Association
Downloaded from http://jaha.ahajournals.org/ at OREGON STATE UNIVERSITY on June 10, 2015
4.140.76
0.690.21
0.770.27
LA dimension (mm), meanSD
Peak E velocity (m/s), meanSD
Peak A velocity (m/s), meanSD
0.360.1
1.50.4
LV RWT, meanSD
IMT (mm), meanSD
5.181.98
PIIINP (ng/mL), meanSD
4.981.70
8.90
4.95 to 14.68
118
56 to 239
120.232.5
314.767
1.75
0.88 to 3.47
1.30.3
0.350.08
177.153.4
46 (6.97)
471 (71.36)
143 (21.67)
0.750.22
0.670.18
40.71
94 (14.24)
4.711.90
6.58
2.99 to 10.07
80
38 to 153
125.733.1
315.463.1
1.72
0.92 to 2.86
1.30.3
0.340.08
169.346.8
26 (6.55)
326 (82.12)
45 (11.34)
0.720.19
0.670.15
4.020.59
40 (10.08)
6.62
2.99 to 12.23
4.941.92
—
186
83 to 378
132.638.9
330.972.6
2.46
1.22 to 4.64
1.30.4
0.370.11
147.656.7
36 (6.53)
356 (64.61)
159 (28.86)
0.890.26
0.750.24
3.870.74
54 (9.8)
5.78
2.99 to 8.99
60
35 to 119
128.932.3
304.259.3
1.38
0.68 to 2.44
1.20.3
0.340.07
162.942.2
37 (7.55)
402 (82.04)
51 (10.41)
0.710.18
0.690.15
3.940.58
35 (7.14)
4.631.55
4.48
2.99 to 7.78
134
72 to 241
133.837.9
321.961.3
1.97
0.92 to 3.45
1.20.3
0.360.08
135.742.9
37 (4.55)
601 (73.83)
176 (21.62)
0.850.22
0.740.18
3.780.63
38 (4.67)
n=814
5 to <10
4.381.29
3.45
2.99 to 5.87
104
63 to 179
132.937.2
324.764.6
1.88
0.94 to 3.25
1.10.3
0.350.07
131.233.2
25 (4.07)
492 (80.13)
97 (15.8)
0.810.22
0.740.18
3.750.61
22 (3.58)
n=614
10 to <15
—
2.99
2.99 to 4.26
82
46 to 145
137.935.8
315.459.2
1.72
0.88 to 3.00
1.10.3
0.340.07
127.030.7
34 (4.3)
658 (83.29)
98 (12.41)
0.780.19
0.730.16
3.670.57
16 (2.03)
n=790
15 to 18
A indicates atrial filling; CHS, Cardiovascular Health Study; E, early filling; hsCRP, high-sensitivity C-reactive protein; hsTNT, high-sensitivity troponin-T; IMT, carotid intima-media thickness; IQR, interquartile range; LA, left atrium; LDL, lowdensity lipoprotein; LV, left ventricular; LVEF, left ventricular ejection fraction; NT-proBNP, N-terminal probrain natriuretic peptide; PIIINP, procollagen III N-terminal Propeptide; LV RWT, left ventricular relative wall thickness.
11.25
6.70 to 18.76
hsTNT (ng/mL), median
IQR
119.936.1
LDL (mg/dL), meanSD
210
105 to 546
33475.3
Fibrinogen (mg/dL), meanSD
NT-proBNP (pg/mL), median
IQR
2.23
1.15 to 4.09
hsCRP (mg/L), median
IQR
Biomarkers
190.668.2
64 (10.92)
≥1.5
LV mass (g/m ), meanSD
339 (57.85)
0.7, 1.5
2
183 (31.23)
<0.7
E/A ratio, n (%)
119 (20.31)
Abnormal LVEF (<55%),
n (%)
Cardiovascular imaging
n=551
n=660
n=586
n=490
0 to <5
ORIGINAL RESEARCH
DOI: 10.1161/JAHA.114.001745
n=397
Females (n=2769)
15 to 18
0 to <5
10 to <15
5 to <10
Males (n=2133)
Categories of Years of Able Life
Table 2. Baseline Cardiovascular Imaging and Biomarkers by Categories of Years of Able Life for Male and Female Participants in CHS
Imaging and Biomarkers, Years of Able Life
Alshawabkeh et al
5
Journal of the American Heart Association
Downloaded from http://jaha.ahajournals.org/ at OREGON STATE UNIVERSITY on June 10, 2015
0.09)
0.26 ( 0.48,
0.89 ( 1.09,
LV RWT (per SD=0.08)
IMT (per SD=0.34)
0.69)
0.04)
0.63)
0.33)
0.5 ( 0.78,
<0.001
0.29)
0.17 ( 0.38, 0.04)
0.5 ( 0.7,
0.019
<0.001
0.22)
0.54 ( 1.36, 0.28)
Ref
0.12)
0.007
—
0.58 ( 1.04,
<0.001
0.12
0.001
0.198
—
0.013
0.027
0.001
0.097
0.045
0.92 ( 1.1,
0.31 ( 0.51,
1.09 ( 1.34,
1.63 ( 2.41,
Ref
1.2 ( 1.63,
0.45 ( 0.62,
0.24 ( 0.41,
0.32 ( 0.5,
0.73)
0.11)
0.84)
0.85)
0.76)
0.27)
0.08)
0.13)
2.36)
P Value
<0.001
0.002
<0.001
<0.001
—
<0.001
<0.001
<0.001
0.004
0.001
<0.001
1.31)
0.53 ( 0.72,
0.22 ( 0.41,
0.76 ( 1.02,
1.07 ( 1.83,
Ref
0.82 ( 1.23,
0.31 ( 0.48,
0.34)
0.03)
0.51)
0.32)
0.4)
0.14)
0.13 ( 0.28, 0.03)
0.16 ( 0.34, 0.02)
2.08 ( 2.85,
Coefficient (95% CI)
Model 2
P Value
<0.001
0.023
<0.001
0.005
—
<0.001
<0.001
<0.001
0.115
0.08
<0.001
Model 1 adjusted for age, race, and BMI. Model 2 additionally adjusted for prevalent smoking, arthritis, cancer, diabetes ADA status, eGFR, antihypertensive medication use, systolic blood pressure, congestive heart failure, stroke, and
coronary heart disease (myocardial infarction, angina, coronary artery bypass grafting, or angioplasty). A indicates atrial filling; ADA, American Diabetes Association; BMI, body mass index; CHS, Cardiovascular Health Study; CI, confidence
interval; E, early filling; eGFR, estimated glomerular filtration rate; IMT, carotid intima-media thickness; LA, left atrium; LVEF, left ventricular ejection fraction; LV RWT, left ventricular relative wall thickness; YAL, years of able life; YOL, years of
life.
0.91 ( 1.19,
1.18 ( 2.03,
Ref
LV mass (per SD=51.17)
≥1.5
0.7, <1.5
<0.001
0.51)
<0.7
0.98 ( 1.46,
<0.001
0.12)
0.15 ( 0.32, 0.03)
0.3 ( 0.49,
0.003
<0.001
E/A ratio
0.26)
0.27 ( 0.45,
0.45 ( 0.64,
Peak E velocity (per SD=0.19)
Peak A velocity (per SD=0.23)
0.2 ( 0.4, 0)
3.14 ( 3.93,
Coefficient (95% CI)
<0.001
0.17)
0.37 ( 0.57,
LA dimension (per SD=0.67)
P Value
0.002
1.34 ( 2.18,
0.49)
Coefficient (95% CI)
P Value
<0.001
1.64)
Coefficient (95% CI)
2.49 ( 3.35,
Model 1
Model 2
Model 1
Abnormal LVEF (<55%)
Cardiovascular Imaging
Years of Life
ORIGINAL RESEARCH
DOI: 10.1161/JAHA.114.001745
Years of Able Life
Table 3. Linear Regression Results of Adjusted Cardiovascular Imaging Risk Factors for YAL and YOL for Female Participants in CHS (n=2769).
Imaging and Biomarkers, Years of Able Life
Alshawabkeh et al
6
Journal of the American Heart Association
Downloaded from http://jaha.ahajournals.org/ at OREGON STATE UNIVERSITY on June 10, 2015
0.17 ( 0.39, 0.06)
0.28 ( 0.5,
Peak E velocity (per SD=0.19)
Peak A velocity (per SD=0.23)
1.27 ( 1.47,
IMT (per SD=0.34)
0.58)
0.79 ( 0.99,
<0.001
0.36)
0.13)
0.16 ( 0.42, 0.1)
0.62 ( 0.87,
0.83)
0.287
0.88 ( 1.62,
<0.001
Ref
<0.001
—
1.35 ( 1.88,
0.14 ( 0.35, 0.07)
0.01 ( 0.22, 0.2)
<0.001
2.51 ( 3.13,
<0.001
1.89)
0.85)
1.2)
1.63)
0.09)
0.02)
0.3)
1.32 ( 1.52,
1.12)
0.13 ( 0.4, 0.14)
1.11 ( 1.36,
0.225
1.98 ( 2.76,
<0.001
Ref
2.18 ( 2.73,
0.31 ( 0.53,
0.25 ( 0.47,
0.52 ( 0.73,
0.021
—
<0.001
<0.001
0.18
0.93
0.022
<0.001
0.35
<0.001
<0.001
—
<0.001
<0.001
0.006
0.032
<0.001
<0.001
P Value
0.8)
0.03)
0.47)
0.44)
1.08)
0.85 ( 1.06,
0.65)
0.13 ( 0.38, 0.12)
0.72 ( 0.97,
1.18 ( 1.92,
Ref
1.6 ( 2.12,
0.19 ( 0.4, 0.02)
0.08 ( 0.29, 0.13)
0.24 ( 0.45,
1.41 ( 2.02,
Coefficient (95% CI)
Model 2
<0.001
0.312
<0.001
0.002
—
<0.001
<0.001
0.072
0.46
0.026
<0.001
P Value
Model 1 adjusted for age, race, and BMI. Model 2 additionally adjusted for prevalent smoking, arthritis, cancer, diabetes ADA status, eGFR, antihypertensive medication use, systolic blood pressure, congestive heart failure, stroke, and
coronary heart disease (myocardial infarction, angina, coronary artery bypass grafting, or angioplasty). A indicates atrial filling; ADA, American Diabetes Association; BMI, body mass index; CHS, Cardiovascular Health Study; CI, confidence
interval; E, early filling; eGFR, estimated glomerular filtration rate; IMT, carotid intima-media thickness; LA, left atrium; LVEF, left ventricular ejection fraction; LV RWT, left ventricular relative wall thickness; YAL, years of able life; YOL, years of
life.
1.06)
0.15 ( 0.42, 0.12)
LV RWT (per SD=0.08)
0.8)
0.74)
1 ( 1.26,
1.58 ( 2.37,
Ref
LV mass (per SD=51.17)
≥1.5
0.7, <1.5
<0.001
1.37)
<0.001
1.93 ( 2.48,
<0.7
0.014
E/A ratio
0.06)
0.146
0.8)
0.04)
1.41 ( 2.03,
0.24 ( 0.45,
<0.001
<0.001
1.76)
0.27)
2.38 ( 3.01,
0.49 ( 0.71,
LA dimension (per SD=0.67)
Coefficient (95% CI)
Coefficient (95% CI)
P Value
Coefficient (95% CI)
P Value
Model 1
Model 2
Model 1
Abnormal LVEF (<55%)
Cardiovascular Imaging Variables
Years of Life
ORIGINAL RESEARCH
DOI: 10.1161/JAHA.114.001745
Years of Able Life
Table 4. Linear Regression Results of Adjusted Cardiovascular Imaging Risk Factors for YAL and YOL for Male Participants in CHS (n=2133).
Imaging and Biomarkers, Years of Able Life
Alshawabkeh et al
7
DOI: 10.1161/JAHA.114.001745
0.21 ( 0.35,
<0.001
0.18)
0.32 ( 0.47,
0.003
0.09)
0.26 ( 0.42,
<0.001
0.24)
0.4 ( 0.58,
PIIINP (per SD=1.79)
Model 1 adjusted for age, race, BMI, and BMI-squared. Model 2 additionally adjusted for prevalent smoking, arthritis, cancer, diabetes ADA status, eGFR, antihypertensive medication use, systolic blood pressure, congestive heart failure,
stroke, and coronary heart disease (myocardial infarction, angina, coronary artery bypass grafting, or angioplasty). ADA indicates American Diabetes Association; BMI, body mass index; CHS, Cardiovascular Health Study; CI, confidence
interval; eGFR, estimated glomerular filtration rate; hsCRP, high-sensitivity C-reactive protein; hsTNT, high-sensitivity troponin-T; LDL, low-density lipoprotein; NT-proBNP, N-terminal probrain natriuretic peptide; PIIINP, procollagen III Nterminal propeptide; YAL, years of able life; YOL, years of life.
0.07)
0.003
<0.001
<0.001
0.38)
0.82)
1.05 ( 1.29,
0.51 ( 0.63,
<0.001
<0.001
1.16)
0.5)
0.63 ( 0.75,
1.38 ( 1.61,
<0.001
<0.001
0.37)
0.83)
0.5 ( 0.64,
1.08 ( 1.34,
<0.001
<0.001
1.45 ( 1.69,
hsTNT (per 2-fold increase)
0.51)
0.65 ( 0.78,
NT-proBNP (per 2-fold increase)
0.18 (0.01, 0.36)
LDL (per SD=36.41)
0.27 ( 0.46,
Fibrinogen (per SD=65.80)
1.2)
0.245
0.09 ( 0.06, 0.25)
0.41
0.07 ( 0.09, 0.23)
0.05
0.17 (0, 0.34)
0.042
0.074
<0.001
0.15 ( 0.32, 0.01)
<0.001
0.14)
0.25 ( 0.37,
<0.001
0.15)
0.31)
0.43 ( 0.55,
0.32 ( 0.49,
0.196
<0.001
0.16)
0.12 ( 0.3, 0.06)
0.29 ( 0.42,
0.48 ( 0.61,
hsCRP (per 2-fold increase)
0.35)
0.006
<0.001
Coefficient (95% CI)
P Value
Coefficient (95% CI)
P Value
Coefficient (95% CI)
Biomarkers
0.08)
Coefficient (95% CI)
P Value
Model 2
Model 1
Model 2
Men and women above the age of 65 years who had a
favorable CV profile determined by echocardiography, carotid
IMT, or biomarkers of inflammation, atherosclerosis, myocardial injury and stress, and cardiac extracellular collagen
turnover spent more years, and a higher percentage of the
end of their lives, without difficulty in ADLs. By definition,
YAL integrates the number of years alive with the number of
years they spend without any ADL difficulty, a prime goal for
elderly persons. Furthermore, in a sensitivity analysis
adjusting for incident CHF, MI, and stroke, the results
remained statistically significant, concluding that these
variables are strongly associated with YAL irrespective of
incident CVD.
As the number of comorbidities increases, prevalence of
disability (defined as any ADL difficulty) increases.10 Having a
higher number of risk factors at middle age (smoking, HTN,
obesity, hyperlipidemia, and minor electrocardiogram [EKG]
abnormalities) has been associated with a shorter time to
disability.6,11 Furthermore, subclinical disease has been
associated with the quality of years alive beyond the age of
65. Asymptomatic CHS participants with subclinical vascular
disease defined as any common or internal IMT above the
80th percentile, maximum stenosis of the internal carotid
artery >25%, ankle-arm index ≤0.9, major EKG abnormality, or
Rose questionnaire positive for angina or claudication were
found to be less likely to be free of incident CVD, cancer,
Model 1
Discussion
Years of Life
Higher levels of hsCRP (for females only), fibrinogen (for
males only), NT-proBNP, hsTNT, and PIIINP were inversely and
strongly associated with observed YAL, YOL, and YAL:YOL
percentage (Tables 5, 6, and S2). A 2-fold increase in hsTNT in
females and males was associated with 1.08 and 0.73 fewer
YAL, 1.05 and 0.82 fewer YOL (P<0.001), and 3.5% and 1.4%
fewer YAL:YOL (P<0.02), respectively.
Years of Able Life
Biomarkers
Journal of the American Heart Association
Downloaded from http://jaha.ahajournals.org/ at OREGON STATE UNIVERSITY on June 10, 2015
ORIGINAL RESEARCH
Each SD (0.34 mm) higher carotid IMT was associated
with 0.5 and 0.79 fewer observed YAL in females and males,
respectively, in the fully adjusted model (P<0.001). Higher
LA dimension (for males only), higher peak A velocities (for
females only), E/A ratio outside of the range 0.7 to 1.5, and
higher LV mass were inversely related to YAL and YOL, but
not YAL:YOL percentage. Results for the pooled cohort of
men and women are in Data S1 (Tables S3 and S4).
Sensitivity analysis of time to first ADL difficulty adjusting
for incident CHF, MI, and stroke, in addition to the
aforementioned confounders, showed persistence of the
association between the variables and incident disability
(Table S5).
P Value
Alshawabkeh et al
Table 5. Linear Regression Results of Adjusted Biomarkers Risk Factors for YAL and YOL for Female Participants in CHS (n=2769).
Imaging and Biomarkers, Years of Able Life
8
DOI: 10.1161/JAHA.114.001745
Model 1 adjusted for age, race, BMI, and BMI-squared. Model 2 additionally adjusted for prevalent smoking, arthritis, cancer, diabetes ADA status, eGFR, antihypertensive medication use, systolic blood pressure, congestive heart failure,
stroke, and coronary heart disease (myocardial infarction, angina, coronary artery bypass grafting, or angioplasty). ADA indicates American Diabetes Association; BMI, body mass index; CHS, Cardiovascular Health Study; CI, confidence
interval; eGFR, estimated glomerular filtration rate; hsCRP, high-sensitivity C-reactive protein; hsTNT, high-sensitivity troponin-T; LDL, low-density lipoprotein; NT-proBNP, N-terminal probrain natriuretic peptide; PIIINP, procollagen III Nterminal propeptide; YAL, years of able life; YOL, years of life.
0.015
0.20 ( 0.36,
0.33 ( 0.52,
PIIINP (per SD=1.79)
0.14)
0.001
0.23 ( 0.41,
0.05)
0.007
0.31 ( 0.48,
0.13)
<0.001
0.04)
<0.001
0.58)
0.82 ( 1.06,
<0.001
1.07)
1.31 ( 1.54,
<0.001
0.73 ( 0.97,
<0.001
1.18 ( 1.42,
hsTNT (per 2-fold increase)
0.94)
0.79 ( 0.94,
0.5)
0.237
<0.001
0.65)
0.79 ( 0.94,
0.13 ( 0.09, 0.35)
0.009
<0.001
0.92)
1.05 ( 1.19,
0.31 (0.08, 0.54)
0.042
<0.001
0.65)
0.23 (0.01, 0.44)
0.001
<0.001
0.87)
1.01 ( 1.14,
LDL (per SD=36.41)
0.41 (0.17, 0.64)
0.46)
0.67 ( 0.88,
Fibrinogen (per SD=65.80)
NT-proBNP (per 2-fold increase)
<0.001
0.17)
0.37 ( 0.56,
<0.001
0.44)
0.19)
<0.001
0.39 ( 0.59,
<0.001
0.65 ( 0.86,
<0.001
0.25)
0.39 ( 0.53,
0.66 ( 0.8,
hsCRP (per 2-fold increase)
0.51)
0.22)
<0.001
0.36 ( 0.5,
<0.001
0.69 ( 0.83,
0.54)
<0.001
Model 2
Coefficient (95% CI)
P Value
Coefficient (95% CI)
Coefficient (95% CI)
Coefficient (95% CI)
Model 1
Model 2
P Value
Model 1
Biomarkers
P Value
Years of Life
Years of Able Life
Journal of the American Heart Association
Downloaded from http://jaha.ahajournals.org/ at OREGON STATE UNIVERSITY on June 10, 2015
ORIGINAL RESEARCH
chronic obstructive pulmonary disease, or new and persistent
physical disability or cognitive decline.12
With the shift toward personalized and patient-centered
care, patient preferences are playing a growing role in
clinical decision making and risk assessment. In a survey of
357 older adults living in senior centers and an assisted
living facility, with the main outcome being a person’s
prioritization of 4 health outcomes: “keeping you alive,
maintaining independence, reducing or eliminating pain, and
reducing or eliminating other symptoms (eg, dizziness,
fatigue, shortness of breath)”; 76% ranked maintaining
independence as the most important health outcome;
notably, staying alive was the least important.4 Likewise,
this pattern of preference was demonstrated in seriously ill
patients above the age of 60.13 Accordingly, we analyzed
these variables in a large sample of community dwellers and
followed them for up to 18 years.
Disability has many causes. The variables chosen in the
analysis have been shown to be associated with development
of morbidity (that could lead to disability) or mortality. Our
study incorporated CV structural and functional assessment
and measurement of biomarkers of various systems. Furthermore, we focused our outcome into a fundamental and a
global assessment of basic physical function, which, without
any of its components, a person is markedly less likely to be
able to maintain physical independence.
Our finding that most echocardiographic parameters in the
fully adjusted model were associated with the YAL, but were
no longer significant when assessing the YAL:YOL percentage,
suggests that these variables might be associated with
longevity to a greater extent than physical ability.14 In other
words, death occurs relatively rapidly for subjects with
unfavorable measures. LA volume has been correlated with
exercise capacity.15 LV mass is known to predict incident
CHF, stroke, and CVD.16,17 Carotid IMT has been extensively
studied as a risk prediction tool and found to predict future
stroke and MI, but generally adds modest benefit when
combined with the traditional risk scores (such as the
Framingham Risk Score).18,19
Measures of inflammation and coagulation in relation to
physical function have been assessed in several observational
cohorts. hsCRP has been linked to total and CV mortality
(CVM), although modestly, and has been associated with
physical performance in older adults.20,21 Fibrinogen, among
other coagulation biomarkers, has been implicated in the
development of disability.22 The relationship between total
cholesterol and functional ability in older adults is controversial. Some studies have suggested a negative and others a
positive, relationship.23–25 LDL, however, has not been
evaluated in prospective cohorts. Our finding that LDL is
negatively associated with YAL, but not YOL or YAL:YOL
percentage—irrespective of statin use—is difficult to explain.
P Value
Alshawabkeh et al
Table 6. Linear Regression Results of Adjusted Biomarkers Risk Factors for YAL and YOL for Male Participants in CHS (n=2133).
Imaging and Biomarkers, Years of Able Life
9
Imaging and Biomarkers, Years of Able Life
Alshawabkeh et al
Low LDL could be a marker of “frailty” or could be owing to
other unmeasured comorbidities or genetic factors.
NT-proBNP predicts total and CVM, MI, stroke, and CHF.26–
28
Whereas physical activity has been shown to decrease the
likelihood of an elevation in NT-proBNP and subsequent
development of clinical CHF,29 it is demonstrated that
subclinical elevation in NT-proBNP is a marker for development of functional decline and mortality. The cardiac-specific
biomarker, troponin, measured by a high-sensitivity assay, is a
marker of chronic myocardial injury and a predictor for future
risk of CHF and CV death in community-dwelling older
adults.30 Our findings extend the value of hsTNT beyond the
traditional outcomes. hsTNT could be a marker of overall
muscular-functional decline and warrants further research.
PIIINP, a marker of collagen turnover, has been linked to the
development of death and heart failure.31,32 To our knowledge, we demonstrate, for the first time, that elevation of this
biomarker is correlated with lower disability-free survival in
community dwellers. This might be part of a phenotype of
systemic collagenous turnover that predates functional
decline before death.
Some of these variables might be associated with future
development of comorbidities, which, in turn, accelerate
functional decline. However, the relationship between these
variables and YAL remained significant even after adjusting for
incident CV outcomes, indicating that these variables are
associated with maintenance of physical ability, irrespective
of development of CVD. In fact, some have argued that,
despite developing comorbidities, some centenarians are able
to achieve exceptional age and avoid disability.5 The observed
associations could be modified by other unmeasured variables. For example, participants with abnormal LVEF could
develop other morbidities, which, in turn, lead to disability,
before developing clinical CVD. Further research is needed to
explore the role of these biomarkers and CV structural
variables in “channeling” persons into one of the pathways of
aging.
Our study has several strengths. We examined a large
sample size from a relevant cohort of older community
dwellers. The follow-up time was long and the outcomes
highly relevant for older people. Our study also has several
limitations. First, LV mass and LV RWT and the biomarkers
were not performed on the entire cohort and, in the case of
PIIINP, was performed later in the study, thus limiting the
follow-up time to 11 years. Differential absence of these
measures could theoretically have introduced bias. However,
only 550 subjects did not have these echo measures at
baseline. Second, there could be residual confounding that we
could not account for in our models. Third, although there was
no significant interaction in statin use for the LDL variable, our
power to detect a difference is limited, given that only 2.1% of
the cohort used statins at the baseline because their use in
DOI: 10.1161/JAHA.114.001745
clinical practice was not robust at the time. Fourth, upon
interpretation of the P values, multiple comparisons should be
taken into account. Nevertheless, with 30 comparisons for
the primary outcome at the P=0.05 level of significance, we
would expect 1.5 to be significant owing to chance alone.
Fifth, in the sensitivity analysis, whereas Cox regression is
focusing on first occurrence and YAL encompasses all
occurrences, because both are getting at a measure of
disability, we would expect similarities in risk factors.
Conclusion
Favorable echocardiographic measures, carotid intima thickness, and biomarkers of inflammation, coagulation, atherosclerosis, myocardial injury and stress, and extracellular
collagen turnover measured in persons above the age of 65
were associated with the number of years of able life and
independence, irrespective of development of CVD, in a large
national cohort followed for up to 18 years. Development of
predictive models and the utility of targeting these variables in
clinical interventions with the goal to improve the quality of
life of older persons above and beyond mitigation of disease
remain to be further evaluated.
Sources of Funding
This research was supported by contracts HHSN268201200
036C, HHSN268200800007C, N01 HC55222, N01HC85079,
N01HC85080, N01HC85081, N01HC85082, N01HC85083,
N01HC85086, and grant HL080295 from the National Heart,
Lung, and Blood Institute (NHLBI), with an additional contribution from the National Institute of Neurological Disorders
and Stroke. Additional support was provided by AG023629
from the National Institute on Aging. A full list of principal CHS
investigators and institutions can be found at CHS-NHLBI.org.
Disclosures
Dr Gardin received honoraria on the Speakers’ Bureau from
Gilead Sciences. All other authors have nothing to disclose.
References
1. Layte R, Sexton E, Savva G. Quality of life in older age: evidence from an Irish
cohort study. J Am Geriatr Soc. 2013;61:S299–S305.
2. Bryant LL, Corbett KK, Kutner JS. In their own words: a model of healthy aging.
Soc Sci Med. 2001;53:927–941.
3. Population Division USCB. Projections of the population by age and sex for the
United States: 2010 to 2050 (np2008-t12). 2008. Available at: http://
www.Aoa.Gov/aoaroot/aging_statistics/future_growth/future_growth.Aspx#
age. Accessed December 2014.
4. Fried TR, Tinetti ME, Iannone L, O’Leary JR, Towle V, Van Ness PH. Health
outcome prioritization as a tool for decision making among older persons with
multiple chronic conditions. Arch Intern Med. 2011;171:1856–1858.
Journal of the American Heart Association
Downloaded from http://jaha.ahajournals.org/ at OREGON STATE UNIVERSITY on June 10, 2015
10
Imaging and Biomarkers, Years of Able Life
Alshawabkeh et al
20. Oluleye OW, Folsom AR, Nambi V, Lutsey PL, Ballantyne CM. Troponin T,
B-type natriuretic peptide, C-reactive protein, and cause-specific mortality.
Ann Epidemiol. 2013;23:66–73.
6. Vita AJ, Terry RB, Hubert HB, Fries JF. Aging, health risks, and cumulative
disability. N Engl J Med. 1998;338:1035–1041.
21. Jenny NS, French B, Arnold AM, Strotmeyer ES, Cushman M, Chaves PH, Ding
J, Fried LP, Kritchevsky SB, Rifkin DE, Sarnak MJ, Newman AB. Long-term
assessment of inflammation and healthy aging in late life: the Cardiovascular
Health Study All Stars. J Gerontol A Biol Sci Med Sci. 2012;67:970–976.
7. Fried LP, Borhani NO, Enright P, Furberg CD, Gardin JM, Kronmal RA, Kuller LH,
Manolio TA, Mittelmark MB, Newman A, O’Leary DH, Psaty B, Rautaharju P,
Tracy RP, Weiler PG. The Cardiovascular Health Study: design and rationale.
Ann Epidemiol. 1991;1:263–276.
8. Gardin JM, Wong ND, Bommer W, Klopfenstein HS, Smith VE, Tabatznik B,
Siscovick D, Lobodzinski S, Anton-Culver H, Manolio TA. Echocardiographic
design of a multicenter investigation of free-living elderly subjects: the
Cardiovascular Health Study. J Am Soc Echocardiogr. 1992;5:63–72.
9. O’Leary DH, Polak JF, Wolfson SK, Bond MG, Bommer W, Sheth S, Psaty BM,
Sharrett AR, Manolio TA. Use of sonography to evaluate carotid atherosclerosis in the elderly. The Cardiovascular Health Study. CHS Collaborative
Research Group. Stroke. 1991;22:1155–1163.
10. Lunney JR, Lynn J, Foley DJ, Lipson S, Guralnik JM. Patterns of functional
decline at the end of life. JAMA. 2003;289:2387–2392.
22. McClure CK, El Khoudary SR, Karvonen-Gutierrez CA, Ylitalo KR, Tomey K,
Vopham T, Sternfeld B, Cauley JA, Harlow S. Prospective associations between
inflammatory and hemostatic markers and physical functioning limitations in
mid-life women: longitudinal results of the Study of Women’s health Across
the Nation (SWAN). Exp Gerontol. 2014;49:19–25.
23. Strandberg TE, Strandberg A, Rantanen K, Salomaa VV, Pitkala K, Miettinen TA.
Low cholesterol, mortality, and quality of life in old age during a 39-year followup. J Am Coll Cardiol. 2004;44:1002–1008.
24. Okamura T, Hayakawa T, Hozawa A, Kadowaki T, Murakami Y, Kita Y, Abbott
RD, Okayama A, Ueshima H. Lower levels of serum albumin and total
cholesterol associated with decline in activities of daily living and excess
mortality in a 12-year cohort study of elderly Japanese. J Am Geriatr Soc.
2008;56:529–535.
11. Daviglus ML, Liu K, Pirzada A, Yan LL, Garside DB, Feinglass J, Guralnik JM,
Greenland P, Stamler J. Favorable cardiovascular risk profile in middle age and
health-related quality of life in older age. Arch Intern Med. 2003;163:2460–
2468.
25. Schalk BW, Visser M, Deeg DJ, Bouter LM. Lower levels of serum albumin and
total cholesterol and future decline in functional performance in older persons:
the Longitudinal Aging Study Amsterdam. Age Ageing. 2004;33:266–272.
12. Newman AB, Arnold AM, Naydeck BL, Fried LP, Burke GL, Enright P, Gottdiener
J, Hirsch C, O’Leary D, Tracy R. “Successful aging”: effect of subclinical
cardiovascular disease. Arch Intern Med. 2003;163:2315–2322.
26. Di Angelantonio E, Chowdhury R, Sarwar N, Ray KK, Gobin R, Saleheen D,
Thompson A, Gudnason V, Sattar N, Danesh J. B-type natriuretic peptides and
cardiovascular risk: systematic review and meta-analysis of 40 prospective
studies. Circulation. 2009;120:2177–2187.
13. Fried TR, Bradley EH, Towle VR, Allore H. Understanding the treatment
preferences of seriously ill patients. N Engl J Med. 2002;346:1061–1066.
14. Gardin JM, McClelland R, Kitzman D, Lima JA, Bommer W, Klopfenstein HS,
Wong ND, Smith VE, Gottdiener J. M-mode echocardiographic predictors of
six- to seven-year incidence of coronary heart disease, stroke, congestive
heart failure, and mortality in an elderly cohort (the Cardiovascular Health
Study). Am J Cardiol. 2001;87:1051–1057.
15. Kusunose K, Motoki H, Popovic ZB, Thomas JD, Klein AL, Marwick TH.
Independent association of left atrial function with exercise capacity in
patients with preserved ejection fraction. Heart. 2012;98:1311–1317.
16. Bikkina M, Levy D, Evans JC, Larson MG, Benjamin EJ, Wolf PA, Castelli WP.
Left ventricular mass and risk of stroke in an elderly cohort. JAMA.
1994;272:33–36.
17. de Simone G, Gottdiener JS, Chinali M, Maurer MS. Left ventricular mass
predicts heart failure not related to previous myocardial infarction: the
Cardiovascular Health Study. Eur Heart J. 2008;29:741–747.
18. Lorenz MW, Markus HS, Bots ML, Rosvall M, Sitzer M. Prediction of clinical
cardiovascular events with carotid intima-media thickness: a systematic
review and meta-analysis. Circulation. 2007;115:459–467.
19. Simon A, Megnien J-L, Chironi G. The value of carotid intima-media thickness
for predicting cardiovascular risk. Arterioscler Thromb Vasc Biol. 2010;30:
182–185.
DOI: 10.1161/JAHA.114.001745
27. Kistorp C, Raymond I, Pedersen F, Gustafsson F, Faber J, Hildebrandt P. Nterminal pro-brain natriuretic peptide, C-reactive protein, and urinary albumin
levels as predictors of mortality and cardiovascular events in older adults.
JAMA. 2005;293:1609–1616.
28. Wallen T, Landahl S, Hedner T, Nakao K, Saito Y. Brain natriuretic peptide
predicts mortality in the elderly. Heart. 1997;77:264–267.
29. Klenk J, Denkinger M, Nikolaus T, Peter R, Rothenbacher D, Koenig W; Acti
FESG. Association of objectively measured physical activity with established
and novel cardiovascular biomarkers in elderly subjects: every step counts. J
Epidemiol Community Health. 2013;67:194–197.
30. deFilippi CR, de Lemos JA, Christenson RH, Gottdiener JS, Kop WJ, Zhan M,
Seliger SL. Association of serial measures of cardiac troponin T using a
sensitive assay with incident heart failure and cardiovascular mortality in older
adults. JAMA. 2010;304:2494–2502.
31. Barasch E, Gottdiener JS, Aurigemma G, Kitzman DW, Han J, Kop WJ, Tracy RP.
The relationship between serum markers of collagen turnover and cardiovascular outcome in the elderly: the Cardiovascular Health Study. Circ Heart Fail.
2011;4:733–739.
32. Velagaleti RS, Gona P, Sundstrom J, Larson MG, Siwik D, Colucci WS, Benjamin
EJ, Vasan RS. Relations of biomarkers of extracellular matrix remodeling to
incident cardiovascular events and mortality. Arterioscler Thromb Vasc Biol.
2010;30:2283–2288.
Journal of the American Heart Association
Downloaded from http://jaha.ahajournals.org/ at OREGON STATE UNIVERSITY on June 10, 2015
11
ORIGINAL RESEARCH
5. Terry DF, Sebastiani P, Andersen SL, Perls TT. Disentangling the roles of
disability and morbidity in survival to exceptional old age. Arch Intern Med.
2008;168:277–283.
SUPPLEMENTAL MATERIAL
Supplemental Methods
Missing outcome data
Activities of daily living (ADLs) were recorded by self-report from 1989-2011. In 1989-1999, they were recorded once
every year. Between 2000 and 2004, these activities were reported using a different, incomparable question. In 20052011, the questions were recorded each half year, again using the original question. In the entire CHS cohort of 5,888
participants, this amounted to 41.1% missing half-year values total, 50.0% of which were due to missing half years
between 1989-1999, 41.5% of which were due to the incompatible question between 2000-2004, and 8.4% of which
were due to neither. In our subset of 4,902 participants, the missing values was similar with 41.7% missing half-year
values total, 49.9% of which were due to missing half years between 1989-1999, 41.9% of which were due to the
incompatible question between 2000-2004, and 8.2% of which were due to neither.
Mortality data was complete for 100% of the participants in CHS, and thus no imputation was required for the YOL data.
Imputation of missing ADL data
We imputed missing ADL data similarly to a previously reported imputation of self-rated health.1 Observed ADL difficulty
was designated to be 0 if the person reported no difficulties and 1 if the person reported difficulty in one or more
activities of daily living (eating, walking around the home, getting out of bed, dressing, bathing, or toileting). We then
replaced the observed ADL difficulty value with the probability that a person with this observed ADL value would be
healthy (excellent, very good, or good self-rated health). Since 77% of participants with no ADL difficulties reported
being healthy, 46% participants with ADL difficulties reported being healthy, and 0% of participants of participants who
had died could report being healthy, values were set to 77, 46, and 0. These probabilities were calculated from the
entire CHS dataset of 5,888 participants.2 Missing ADL data were then imputed on this scale via linear interpolation of
each participant’s observed data over time, and were then rounded back to the original binary scale. Since all the ADL
data was missing between 2000 and 2004, we applied an additional adjustment to these years so that the average
imputed values in these years fell on a straight line between the average ADL values in 1999 and 2005.
YAL and YOL calculation
Vital status—designated to be 1 if a participant was alive and 0 if the participant had died—was determined at baseline
and each half year after baseline. This resulted in a baseline value and 36 half year values per person. These values were
then summed, minus the first and last values divided by two, to provide a trapezoidal estimate for years of life (YOL)
under the curve in half years.3 To put the value back on the yearly scale, we once again divided by two.
YAL was calculated in a similar manner, the only difference being that instead of using vital status data, imputed ADL
status was employed.
Variable Definitions
Smoking: ascertained by self report
Arthritis: diagnosis by a physician.
Cancer: ascertained by self report
Diabetes: according to the American Diabetes Association guidelines – diabetes if taking insulin or oral hypoglycemics or
if fasting glucose values are >=126. Impaired fasting glucose if fasting glucose is 110-125 and no insulin or oral
hypoglycemic medication.
eGFR was calculated based on creatinine calibrated to the Cleveland clinic: eGFRcreat(mL/min/1.73m2)=186.3*serum
creatinineˆ-1.154*ageˆ-0.203*1.212 [if black]*0.742 [if female]
Systolic blood pressure was the average zero muddler systolic blood pressure: average of the first and second corrected
reading (corrected for zero value).
Anti-hypertensive medication use: Participants were asked to bring in meds used in the last two weeks.
Antihypertensive medications: Beta-blockers, Calcium-channel blockers, Diuretics, Vasodilators, Beta-blockers with
Diuretics, Angiotensin converting enzyme inhibitors, Angiotensin converting enzyme with diuretics, Vasodilators with
Diuretics, Angiotensin Type 2 Antagonists, Angiotensin Type 2 Antagonists with Diuretics.
Prevalent Stroke, CHF and CHD were self-reported at baseline. Confirmation was sought for MI and CHF from
information collected at entry and review of hospital and physician's records; furthermore, Incident Stroke, CHF and
CHD were reviewed and adjudicated by committee.
References
1.
Diehr P, Patrick DL, Spertus J, Kiefe CI, McDonell M, Fihn SD. Transforming self-rated health and the sf-36 scales
to include death and improve interpretability. Medical Care. 2001;39:670-680
2.
Diehr P. Methods for dealing with death and missing data, and for standardizing different health variables in
longitudinal datasets: The Cardiovascular Health Study. UW Biostatistics Working Paper Series. Working Paper
390. July 2013
3.
Diehr P, Psaty BM, Patrick DL. Effect size and power for clinical trials that measure years of healthy life. Statistics
in Medicine. 1997;16:1211-1223
Table S1. Linear regression results of adjusted cardiovascular imaging risk factors for percentage of YOL spent in an able state for male and female participants in
CHS
% Years of Life Without ADL Difficulty
Males (n=2,133)
Females (n=2,769)
Model 1
Model 2
Model 1
Model 2
Coefficient
P
Coefficient
P
Percent
P
Percent
P
Cardiovascular Imaging
(95% CI)
value
(95% CI)
value
(95% CI)
value
(95% CI)
value
Abnormal LVEF (<55%)
-5
-3.6
-0.8
2.1
(-7.7, -2.3)
<0.001
(-6.3, -0.8)
0.012 (-4.8, 3.2) 0.679
(-1.9, 6.2) 0.299
LA Dimension
-0.8
-0.5
-0.8
-0.4
(per SD= 0.67)
(-1.7, 0.2)
0.104
(-1.4, 0.5)
0.336 (-1.8, 0.1) 0.086
(-1.3, 0.5) 0.401
Peak E velocity
-0.4
-0.1
-0.6
-0.3
(per SD= 0.19)
(-1.4, 0.6)
0.422
(-1.1, 0.8)
0.82
(-1.4, 0.2) 0.156
(-1.2, 0.5) 0.419
Peak A velocity
-0.2
0.1
-0.9
-0.5
(per SD= 0.23)
(-1.2, 0.8)
0.678
(-0.9, 1)
0.848
(-1.7, 0)
0.057
(-1.4, 0.4) 0.277
E/A ratio
0.078
0.444
0.702
0.786
<0.7
-2.2
-1.1
-1
0.1
(-4.6, 0.3)
0.08
(-3.5, 1.3)
0.369 (-3.2, 1.3)
0.4
(-2.1, 2.3) 0.939
0.7, <1.5
Ref
--Ref
-Ref
-≥1.5
-2.9
-1.8
-0.2
1.4
(-6.3, 0.5)
0.096
(-5.2, 1.6)
0.305 (-4.2, 3.8)
0.92
(-2.5, 5.3) 0.488
LV Mass
-1.1
-0.3
-0.9
0.4
0.535
(per SD=51.17)
(-2.2, -0.1)
0.037
(-1.4, 0.8)
0.567 (-2.2, 0.4) 0.198
(-0.9, 1.7)
LV RWT
-0.6
-0.7
0
0.2
0.687
(per SD=0.08)
(-1.7, 0.6)
0.331
(-1.8, 0.4)
0.212
(-1.1, 1)
0.933
(-0.8, 1.2)
IMT
-2.5
-1.5
-1.9
-1.1
0.03
(per SD= 0.34)
(-3.4, -1.5)
<0.001
(-2.5, -0.6)
0.001
(-2.9, -1) <0.001 (-2.1, -0.1)
Model 1 adjusted for age, race, and BMI. Model 2 additionally adjusted for smoking, arthritis, cancer, diabetes ADA status, eGFR, antihypertensive medication
use, systolic blood pressure, congestive heart failure, stroke, and coronary heart disease (myocardial infarction, angina, coronary artery bypass grafting, or
angioplasty). ADL, Activities of Daily Living; LVEF, Left ventricular Ejection Fraction; LA, Left Atrium; SD, Standard Deviation; E, Early Filling; A, Atrial Filling; RWT,
Relative Wall Thickness; IMT, Carotid Intima-Media Thickness.
Table S2. Linear regression results of adjusted biomarkers risk factors for percentage of YOL spent in an able state for male and female participants in CHS
% of Years of Life Without ADL Difficulty
Males (n=2,133)
Females (n=2,769)
Model 2
Model 1
Model 1
Model 2
Coefficient
P
Coefficient
P
Percent
P
Percent
P
Biomarkers
(95% CI)
value
(95% CI)
value
(95% CI)
value
(95% CI)
value
hsCRP
-1
-0.4
-1.4
-0.9
(per 2-fold increase)
(-1.6, -0.4)
0.002
(-1, 0.3)
0.282
(-2, -0.8) <0.001 (-1.6, -0.3) 0.003
Fibrinogen
-1.5
-0.9
-0.5
-0.2
(per SD= 65.80)
(-2.4, -0.6)
0.001
(-1.8, 0)
0.049
(-1.3, 0.4)
0.316
(-1.1, 0.7)
0.683
LDL
0.7
0.4
0.8
0.6
(per SD= 36.41)
(-0.3, 1.7)
0.152
(-0.6, 1.3)
0.472
(0, 1.6)
0.058
(-0.2, 1.4)
0.135
NT-proBNP
-1.6
-1.4
-1.4
-1
(per 2-fold increase)
(-2.3, -1)
<0.001 (-2.1, -0.7)
<0.001
(-2, -0.8) <0.001 (-1.6, -0.3) 0.004
hsTNT
-2
-1.4
-4.4
-3.5
(per 2-fold increase)
(-3, -0.9)
<0.001 (-2.5, -0.3)
0.014
(-5.5, -3.2) <0.001 (-4.7, -2.2) <0.001
PIIINP
-1.9
-1.4
-3.2
-2.2
(per SD =1.79)
(-3.2, -0.5)
0.008 (-2.8, -0.09)
0.037
(-4.6, -1.8) <0.001 (-3.6, -0.84) 0.002
Model 1 adjusted for age, race, BMI, and BMI^2. Model 2 additionally adjusted for smoking, arthritis, cancer, diabetes ADA status, eGFR, antihypertensive
medication use, systolic blood pressure, congestive heart failure, stroke, and coronary heart disease (myocardial infarction, angina, coronary artery bypass
grafting, or angioplasty). hsCRP, high sensitivity C-Reactive Protein; IQR, Inter-Quartile Range; LDL, Low Density Lipoprotein cholesterol; NT-proBNP, N-Terminal
pro-Brain Natriuretic Peptide; hsTNT, high sensitivity Troponin-T; PIIINP, Procollagen III N-terminal Propeptide.
Table S3. Linear regression results of adjusted cardiovascular imaging risk factors for YAL, YOL, and percentage of YOL spent in an able state for all participants in
CHS (n=4,902).
Years of Able Life
Years of Life
% Years of Life Without ADL Difficulty
Model 1
Model 2
Model 1
Model 2
Model 1
Model 2
Cardiovascular Imaging
Coefficient
P
Coefficient
P
Coefficient
Coefficient
P
Percent
P
Percent
P
P value
Variables
(95% CI)
value
(95% CI)
value
(95% CI)
(95% CI)
value
(95% CI)
value
(95% CI)
value
Abnormal LVEF (<55%)
-2.43
<0.001
-1.4
<0.001
-2.72
<0.001
-1.66
<0.001
-3.7
0.001
-1.6
0.173
(-2.93, -1.94)
(-1.89, -0.91)
(-3.19, -2.24)
(-2.13, -1.19)
(-6, -1.5)
(-3.9, 0.7)
LA Dimension
-0.43
<0.001
-0.23
0.002
-0.41
<0.001
-0.2
-0.9
0.013
-0.5
0.156
(per SD= 0.67) (-0.58, -0.28)
(-0.37, -0.08)
(-0.55, -0.27)
(-0.33, -0.06)
0.005 (-1.5, -0.2)
(-1.2, 0.2)
Peak E velocity
-0.23
0.001
-0.09
0.197
-0.25
<0.001
-0.1
-0.5
0.098
-0.2
0.433
(per SD= 0.19) (-0.37, -0.09)
(-0.22, 0.05)
(-0.38, -0.11)
(-0.23, 0.02)
0.115 (-1.2, 0.1)
(-0.9, 0.4)
Peak A velocity
-0.37
<0.001
-0.23
0.001
-0.38
<0.001
-0.26
-0.6
0.074
-0.2
0.453
(per SD= 0.23) (-0.52, -0.23)
(-0.37, -0.09)
(-0.52, -0.25)
(-0.39, -0.12) <0.001 (-1.2, 0.1)
(-0.9, 0.4)
E/A ratio
<0.001
<0.001
<0.001
<0.001
0.132
0.882
<0.7
-1.39
<0.001
-0.91
<0.001
-1.63
-1.16
<0.001
-1.4
0.084
-0.4
0.635
<0.001
(-1.75, -1.03)
(-1.26, -0.57)
(-1.97, -1.28)
(-1.49, -0.84)
(-3.1, 0.2)
(-2, 1.2)
0.7, <1.5
------Ref
Ref
Ref
Ref
Ref
Ref
≥1.5
-1.38
<0.001
-0.72
0.01
-1.8
-1.13
<0.001
-1.7
0.215
-0.3
0.827
<0.001
(-1.96, -0.8)
(-1.27, -0.17)
(-2.35, -1.25)
(-1.65, -0.6)
(-4.3, 1)
(-2.9, 2.3)
LV Mass (per SD=51.17)
-0.96
<0.001
-0.57
<0.001
-1.10
<0.001
-0.74
<0.001
-1
0.017
0.0
0.964
(-1.15, -0.77)
(-0.75, -0.38)
(-1.27, -0.92)
(-0.92, -0.57)
(-1.9, -0.2)
(-0.9, 0.8)
LV RWT (per SD=0.08)
-0.22
0.013
-0.16
0.052
-0.23
0.004
-0.18
0.019
-0.3
0.492
-0.2
0.67
(-0.39, -0.05)
(-0.32, 0)
(-0.4, -0.07)
(-0.34, -0.03)
(-1, 0.5)
(-0.9, 0.6)
Carotid intima thickness
<0.001
<0.001
-1.12
<0.001
-0.71
<0.001
<0.001
<0.001
-1.08
-0.65
-2.2
-1.3
(per SD= 0.34) (-1.23, -0.94)
(-1.26, -0.98)
(-0.84, -0.57)
(-0.8, -0.51)
(-2.9, -1.5)
(-2, -0.6)
Model 1 adjusted for age, race, sex, and BMI. Model 2 additionally adjusted for smoking, arthritis, cancer, diabetes ADA status, eGFR, antihypertensive
medication use, systolic blood pressure, congestive heart failure, stroke, and coronary heart disease (myocardial infarction, angina, coronary artery bypass
grafting, or angioplasty). LVEF, Left ventricular Ejection Fraction; LA, Left Atrium; SD, Standard Deviation; E, Early Filling; A, Atrial Filling; RWT, Relative Wall
Thickness; IMT, Carotid Intima-Media Thickness.
Table S4. Linear regression results of adjusted biomarkers risk factors for YAL, YOL, and percentage of YOL spent in an able state for all participants in CHS
(n=4,902).
Years of Able Life
Years of Life
% Years of Life Without ADL Difficulty
Model 2
Model 1
Model 2
Model 1
Model 1
Model 2
Coefficient
P
Coefficient
P
Coefficient
Coefficient
Percent
P
Percent
P
Biomarkers
P value
P value
(95% CI)
value
(95% CI)
value
(95% CI)
(95% CI)
(95% CI)
value
(95% CI)
value
hsCRP
-0.56
<0.001
-0.33
<0.001
-0.55
<0.001
-0.32
-1.2
<0.001
-0.7
0.003
(per 2-fold increase)
(-0.66, -0.46)
(-0.42, -0.23)
(-0.64, -0.45)
(-0.41, -0.23) <0.001 (-1.6, -0.7)
(-1.1, -0.2)
Fibrinogen
-0.46
<0.001
-0.25
<0.001
-0.47
<0.001
-0.26
-0.9
0.003
-0.5
0.091
(per SD= 65.80)
(-0.6, -0.32)
(-0.38, -0.12)
(-0.61, -0.34)
(-0.39, -0.13) <0.001 (-1.6, -0.3)
(-1.2, 0.1)
LDL
0.27
<0.001
0.2
0.16
0.018
0.12
0.8
0.018
0.5
0.12
0.003
(per SD= 36.41)
(0.13, 0.41)
(0.07, 0.34)
(0.03, 0.3)
(-0.01, 0.24) 0.073
(0.1, 1.4)
(-0.1, 1.1)
NT-proBNP
-0.82
<0.001
-0.63
<0.001
-0.84
<0.001
-0.65
-1.5
<0.001
-1.1
<0.001
(per 2-fold increase)
(-0.92, -0.72)
(-0.73, -0.53)
(-0.93, -0.75)
(-0.74, -0.55) <0.001
(-1.9, -1)
(-1.6, -0.6)
hsTNT
-1.3
<0.001
-0.9
<0.001
-1.34
<0.001
-0.94
-3.1
<0.001
-2.3
<0.001
(per 2-fold increase)
(-1.47, -1.13)
(-1.07, -0.72)
(-1.5, -1.18)
(-1.1, -0.78) <0.001 (-3.8, -2.3)
(-3.1, -1.4)
PIIINP
-0.37
<0.001
-0.29
<0.001
-0.31
<0.001
-0.25
<0.001
-2.6
<0.001
-2.1
<0.001
(per SD =1.79)
(-0.50, -0.24)
(-0.41, -0.17)
(-0.42, -0.20)
(-0.36, -0.15)
(-3.6, -1.6)
(-3.0, -1.1)
Model 1 adjusted for age, race, sex, BMI, and BMI^2. Model 2 additionally adjusted for smoking, arthritis, cancer, diabetes ADA status, eGFR, antihypertensive
medication use, systolic blood pressure, congestive heart failure, stroke, and coronary heart disease (myocardial infarction, angina, coronary artery bypass
grafting, or angioplasty). hsCRP, high sensitivity C-Reactive Protein; IQR, Inter-Quartile Range; LDL, Low Density Lipoprotein; NT-proBNP, N-Terminal pro-Brain
Natriuretic Peptide; hsTNT, high sensitivity Troponin-T; PIIINP, Procollagen III N-terminal Propeptide.
Table S5. Cox proportional hazards regression results of cardiovascular imaging and biomarkers risk factors on time to first ADL difficulty amongst participants in
CHS (n=4902).
All
Abnormal LVEF (<55%)
LA Dimension (per SD=0.67)
Peak E velocity (per
SD=0.19)
Peak A velocity (per
SD=0.23)
E/A ratio
<0.7
0.7, <1.5
>=1.5
LV Mass (per SD=51.17)
LV RWT (per SD=0.08)
Carotid intima thickness
(per SD=0.34)
hsCRP (per 2-fold increase)
Fibrinogen (per SD=65.80)
LDL (per SD=36.41)
NT-proBNP (per 2-fold
increase)
hsTNT (per 2-fold increase)
PIIINP (per SD=1.79)
Model 1
HR
P
(95% CI)
value
1.2
(1.08, 1.34) 0.001
1.04
(1, 1.07)
0.024
1.03
(1, 1.06)
0.064
1.05
(1.01, 1.08) 0.006
1.13
(1.04, 1.22)
1.06
(0.93, 1.21)
1.09
(1.04, 1.14)
1.05
(1.01, 1.09)
1.13
(1.1, 1.17)
1.07
(1.05, 1.1)
1.05
(1.02, 1.08)
0.93
(0.9, 0.95)
1.1
(1.07, 1.12)
1.19
(1.15, 1.23)
1.11
(1.05, 1.17)
0.002
0.351
<0.001
0.015
<0.001
<0.001
0.002
<0.001
<0.001
<0.001
<0.001
Males
Model 2
HR
P
(95% CI)
value
1.22
(1.09, 1.36) <0.001
1.03
(1, 1.07)
0.05
1.03
(0.99, 1.06) 0.118
1.03
(1, 1.07)
0.071
1.08
(1, 1.17)
1.05
(0.92, 1.2)
1.07
(1.02, 1.12)
1.04
(1, 1.08)
1.1
(1.06, 1.13)
1.05
(1.03, 1.07)
1.03
(1, 1.07)
0.94
(0.91, 0.97)
1.09
(1.07, 1.12)
1.15
(1.11, 1.2)
1.1
(1.05, 1.16)
0.055
0.484
0.003
0.07
<0.001
<0.001
0.03
<0.001
<0.001
<0.001
<0.001
Females
Model 1
HR
P
(95% CI)
value
1.23
(1.08, 1.41) 0.002
1.04
(0.99, 1.09) 0.131
1.03
(0.98, 1.08) 0.292
1.06
(1.01, 1.12) 0.018
Model 2
Model 1
HR
P
HR
P
(95% CI)
value
(95% CI)
value
1.28
1.14
(1.12, 1.48) <0.001 (0.95, 1.37) 0.16
1.03
1.04
(0.98, 1.08) 0.221 (0.99, 1.09) 0.09
1.03
1.04
(0.98, 1.08) 0.306 (0.99, 1.08) 0.09
1.04
1.04
(0.98, 1.09) 0.171
(1, 1.09)
0.06
Model 2
HR
P
(95% CI)
value
1.11
(0.92, 1.34) 0.297
1.03
(0.99, 1.08) 0.144
1.03
(0.99, 1.08) 0.104
1.03
(0.99, 1.08) 0.135
1.19
(1.06, 1.35)
1.12
(0.99, 1.26)
1.05
(0.95, 1.17)
1.08
(0.9, 1.29)
1.12
(1.06, 1.19)
1.05
(0.99, 1.12)
1.19
(1.14, 1.24)
1.11
(1.07, 1.15)
1.1
(1.05, 1.15)
0.91
(0.87, 0.96)
1.12
(1.08, 1.15)
1.17
(1.11, 1.23)
1.11
(1.03, 1.18)
0.004
0.391
<0.001
0.107
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
0.004
1.08
(0.9, 1.3)
1.1
(1.04, 1.17)
1.04
(0.98, 1.11)
1.14
(1.09, 1.19)
1.08
(1.04, 1.11)
1.07
(1.03, 1.12)
0.94
(0.89, 0.98)
1.12
(1.08, 1.16)
1.13
(1.07, 1.2)
1.1
(1.03, 1.18)
0.078
0.391
0.001
0.233
<0.001
<0.001
0.002
0.01
<0.001
<0.001
0.006
1.08
(0.98, 1.2)
1.04
(0.86, 1.26)
1.06
(0.99, 1.13)
1.05
(1, 1.1)
1.09
(1.04, 1.14)
1.05
(1.02, 1.08)
1.01
(0.96, 1.05)
0.93
(0.9, 0.97)
1.08
(1.05, 1.11)
1.21
(1.15, 1.28)
1.11
(1.02, 1.21)
0.12
0.69
0.08
0.04
<0.001
0.001
0.82
<0.001
<0.001
<0.001
0.01
1.01
(0.83, 1.22)
1.03
(0.96, 1.09)
1.04
(0.99, 1.09)
1.06
(1.01, 1.11)
1.03
(1, 1.06)
1
(0.96, 1.04)
0.95
(0.91, 0.98)
1.07
(1.04, 1.11)
1.18
(1.11, 1.24)
1.1
(1.01, 1.2)
0.342
0.932
0.447
0.149
0.014
0.038
0.986
0.005
<0.001
<0.001
0.028
Model 1 adjusted for age, race, sex, and BMI. Model 2 additionally adjusted for smoking, arthritis, cancer, diabetes ADA status, eGFR, antihypertensive
medication use, systolic blood pressure, congestive heart failure (CHF), stroke, and coronary heart disease (CHD), in addition to incident CHF, stroke and
myocardial infarction. HR, Hazard Ratio; LVEF, Left ventricular Ejection Fraction; LA, Left Atrium; SD, Standard Deviation; E, Early Filling; A, Atrial Filling; RWT,
Relative Wall Thickness; IMT, Carotid Intima-Media Thickness. hsCRP, high sensitivity C-Reactive Protein; IQR, Inter-Quartile Range; LDL, Low Density Lipoprotein;
NT-proBNP, N-Terminal pro-Brain Natriuretic Peptide; hsTNT, high sensitivity Troponin-T; PIIINP, Procollagen III N-terminal Propeptide.
Years of Able Life in Older Persons−−The Role of Cardiovascular Imaging and Biomarkers:
The Cardiovascular Health Study
Laith I. Alshawabkeh, Laura M. Yee, Julius M. Gardin, John S. Gottdiener, Michelle C. Odden, Traci
M. Bartz, Alice M. Arnold, Kenneth J. Mukamal and Robert B. Wallace
J Am Heart Assoc. 2015;4:e001745; originally published April 23, 2015;
doi: 10.1161/JAHA.114.001745
The Journal of the American Heart Association is published by the American Heart Association, 7272 Greenville Avenue,
Dallas, TX 75231
Online ISSN: 2047-9980
The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://jaha.ahajournals.org/content/4/4/e001745
Data Supplement (unedited) at:
http://jaha.ahajournals.org/content/suppl/2015/04/27/jah3943.DC1.html
Subscriptions, Permissions, and Reprints: The Journal of the American Heart Association is an online only Open
Access publication. Visit the Journal at http://jaha.ahajournals.org for more information.
Downloaded from http://jaha.ahajournals.org/ at OREGON STATE UNIVERSITY on June 10, 2015