Age and Ageing 2017; 46: 994–1000 © The Author 2017.

2017. Published by Oxford University Press on behalf of the British Geriatrics Society.
doi: 10.1093/ageing/afx095 All rights reserved. For permissions, please email: journals.permissions@oup.com
Published electronically 14 June 2017

Correlates of dyspnoea and its association with

adverse outcomes in a cohort of adults aged 80
and over

Downloaded from https://academic.oup.com/ageing/article-abstract/46/6/994/3868056 by guest on 08 December 2019

ERALDA HEGENDÖRFER1,2, BERT VAES1,2, CATHARINA MATHEÏ1, GIJS VAN POTTELBERGH1,
JEAN-MARIE DEGRYSE1,2
1
Department of Public Health and Primary Care, Katholieke Universiteit Leuven, Leuven, Belgium
2
Institute of Health and Society, Université Catholique de Louvain, Brussels, Belgium
Address correspondence to: E. Hegendörfer. Tel: (+32) (0)2 764 34 60; Fax: (+32) (0)2 764 34 70.
Email: eralda.turkeshi@uclouvain.be

Abstract

Background: adults aged 80 and over, a fast growing age-group, with increased co-morbidity and frailty have not been the
focus of previous research on dyspnoea. We investigate the correlates of dyspnoea and its association with adverse out-
comes in a cohort of adults aged 80 and over.
Methods: about 565 community-dwelling adults aged 80 and over of the BELFRAIL prospective cohort had assessment of
Medical Research Council dyspnoea scale (MRC), forced expiratory volume in 1 s (FEV1), N-terminal pro-brain natriuretic pep-
tide (NT-proBNP), physical performance tests, grip strength, 15 items geriatric depression scale, activities of daily living (ADL),
body mass index (BMI) and demographics data. Kaplan–Meier survival curves, Cox and logistic multivariable regression, classifi-
cation and regression tree (CART) analysis assessed association of dyspnoea (MRC 3–5) with time-to-cardiovascular and all-
cause death (5 years), time to first hospitalisation (3 years), new/worsened ADL disability (2 years), and its correlates.
Results: participants with dyspnoea MRC 3–5 (29.9%) had increased hazard ratios for cardiovascular mortality 2.85 (95%
confidence interval 1.93–4.20), all-cause mortality 2.04 (1.58–2.64), first hospitalisation 1.72 (1.35–2.19); and increased odds
ratio for new/worsened disability 2.49 (1.54–4.04), independent of age, sex and smoking status. Only FEV1, physical per-
formance, BMI and NT-proBNP (in order of importance) were selected in the tree-based classification model for dyspnoea.
Conclusions: in a cohort of adults aged 80 and over, dyspnoea was common and an independent predictor of adverse out-
comes, with cardio-respiratory and physical performance impairments as key independent correlates. Its routine and com-
prehensive evaluation in primary care could be very valuable in caring for this age-group.

Keywords: Older people, dyspnoea, adults aged 80 and over, correlates, adverse outcomes

Introduction
Dyspnoea on exertion is a common and distressing symptom
in older adults, with prevalence ranging from 16–36% [1].
Yet, it is a non-specific and complex symptom, especially in
older adults where co-morbidity is frequent, subjective aware-
ness of dyspnoea may be reduced, and it may be attributed
to normal ageing [1–3]. In a recent systematic review on dys-
pnoea in older adults, only one study had investigated the
causes of dyspnoea in a small sample of older adults aged
60–79 year, reporting an overlap between lung disease, heart
disease, obesity and deconditioning [1, 4].
The study of dyspnoea in older adults is important as it is
a disabling symptom, with limitation of basic and instrumental
activities of daily living, as well as mobility, and is associated
with functional decline and frailty [2, 5–7]. Dyspnoea has also
been found to be an independent predictor of all-cause mor-
tality in community-dwelling older adults [8–10]. It is even
reported to be a better predictor of all-cause and specific-
cause mortality than lung function measures in patients with
chronic obstructive pulmonary disease or angina in patients
with suspected coronary heart disease [11, 12]. These findings
have led to the call for routine assessment for dyspnoea, as it

994
 Correlates of dyspnoea and its association with adverse outcomes
is easy, provides very valuable clinical information and its
management may improve the quality of life [13, 14].
No studies so far have investigated dyspnoea and its
predictive value for all-cause and cardiovascular mortality,
as well as other relevant adverse outcomes such as func-
tional decline in community-dwelling adults aged 80 and
over. This is a fast growing age-group worldwide, with a
high burden of co-morbidity and frailty, as well as a high
use of medical resources [15, 16].
The aim of this study is to investigate the prevalence and
correlates of dyspnoea, as well as its association with all-cause
and cardiovascular mortality, unplanned hospitalisation and new
or worsened disability in a cohort of adults aged 80 and over.
Methods
Study design and population
BELFRAIL is a prospective, observational, population-
based cohort study of people aged 80 years and over living
in Belgium. The study protocol and sampling methods have
been already published [16]. Between November 2008 to
September 2009, in 29 general practice centres, 567 indivi-
duals aged 80 years and older were randomly recruited,
excluding only those with severe dementia (known mini
mental status exam score <15/30), in palliative care or
medical emergencies. The study protocol was approved by
the Biomedical Ethics Committee of the Medical School of
the Université catholique de Louvain in Belgium. All partici-
pants gave informed consent.
Baseline dyspnoea
Dyspnoea was assessed with the Medical Research Council
(MRC) scale (with an additional grade 0 if not troubled by
dyspnoea) administered by the GPs at the baseline assess-
ment [16]. This scale has been widely used in studies of
dyspnoea in older adults [1, 9, 17] MRC grade 3 (‘I walk
slower than other people of same age on the level ground’)
categorised participants into those without/mild dyspnoea
(0–2) and moderate-severe dyspnoea (3–5) [7]. In the rest
of this paper, dyspnoea refers to MRC grade 3–5.
Outcome measurements
Time to all-cause and cardiovascular death, and first,
unplanned hospitalisation were used as outcome measure-
ments. The date and cause of hospitalisation (until 3.0 ±
0.25 years) and death (until 5.1 ± 0.2 years) were prospect-
ively reported by the participants’ general practitioners [16].
Disability was assessed at baseline and follow-up by the
degree of difficulty with six activities of daily living (ADL)
(see Supplementary data, Appendix 1, available in Age and
Ageing online for details), using the lowest quintile score
(20) as a cut-off [16]. New or worsened disability was con-
sidered if those without disability at baseline became
dependent at follow-up, or those dependent at baseline had
a decreased follow-up score.
Demographic and clinical variables
In addition to age and sex, the following were considered
as possible correlates of dyspnoea, based on previous
research [4, 6, 7, 18–20]: smoking status, level of education,
body mass index (BMI), forced expiratory volume in 1 s
(FEV1), N-terminal pro-brain natriuretic peptide (NT-
proBNP), 15 items geriatric depression scale (GDS-15),
physical performance tests (PPT) and grip strength.
The general practitioners recorded the age, sex, smoking
status, education level and presence of morbidities at base-
line. Recorded morbidities included respiratory (asthma or
Downloaded from https://academic.oup.com/ageing/article-abstract/46/6/994/3868056 by guest on 08 December 2019
chronic obstructive pulmonary disease) and cardiovascular
diseases (see Supplementary data, Appendix 1, available in
Age and Ageing online for details). Standardised measure-
ments of weight and height were performed during the clin-
ical research assistant’s visit.
Spirometry was recorded by two trained clinical research
assistants using a Spirobank spirometer (Medical International
Research, Rome, Italy) without reversibility testing. Two inde-
pendent researchers evaluated all spirograms based on the
acceptability and repeatability criteria of the American Thoracic
Society/European Respiratory Society [21]. Only individuals
with usable spirograms were included in this study [21, 22].
FEV1 was expressed as z-score derived from the Global Lung
Function Initiative 2012 reference equations [23]. Low FEV1
was defined as FEV1 z-score <−1.645 [24].
Serum NT-proBNP was measured with Dade-
Dimension Xpand (Siemens, Deerfield, IL, USA) using
serum samples of morning blood that were stored frozen
at −80°C until analysis. High NT-proBNP was defined as
NT-proBNP ≥400 pg/ml [25]. Physical performance was
determined based on a battery of PPT and grip strength
(see Supplementary data, Appendix 1, available in Age and
Ageing online for details). Low physical performance was
defined as being in the lowest sex-specific quintile of either
the PPT score (4 for females; 7 for males) or grip strength
(13.3 kg for females; 22.5 kg for males). The GDS-15 was
used to screen for depression in the BELFRAIL cohort,
and a score ≥5 identified participants at risk for depres-
sion [16].
Statistical analysis
Logistic regression analysis and classification and regression
tree (CART) analyses were used to investigate the potential
demographic and clinical correlates of dyspnoea. Participants
with missing FEV1 z-score were compared for statistically
significant differences in regard to study variables to those
with available FEV1 z-score.
Kaplan–Meier curves for all-cause, cardiovascular and
non-cardiovascular mortality, and hospitalisation were plot-
ted for the two dyspnoea categories, with log-rank test for
comparison. The Cox proportional hazards regression ana-
lysis estimated the hazard ratio (HR) for all-cause and car-
diovascular mortality, and unplanned hospitalisation. Odds
ratios (ORs) for new/worsened disability were estimated
with the logistic regression multivariable analysis. As there
995
E. Hegendörfer et al.

were participants with missing data on disability assessment
at follow-up, sensitivity analysis was performed considering
both worst (all those with missing data had new/worsened
disability) and best (all had no new/worsened disability
scenarios). Models were adjusted for age, sex and smoking
status and stratified for BMI, FEV1, NT-proBNP and
physical performance. They were checked for variable col-
linearity, linearity and proportionality assumptions. A two-
tailed probability value P < 0.05 was considered statistically
significant. Statistical analysis was performed with SPSS
23.0 (SPSS Inc., Chicago, IL, USA).
level or walking up a slight hill (MRC 2); 60 (10.6%) had to
walk slower than other people on the level ground (MRC 3);
89 (15.8%) had to stop for breath after walking about 100 m
or after a few minutes on level ground (MRC 4); and only 20
(3.5%) participants were too breathless to leave the house or
when getting undressed (MRC 5). Participants with dyspnoea
(MRC 3–5), had higher BMI, lower FEV1, higher NT-
proBNP, higher GDS-15 scores, lower grip strength, ADL
and PPT scores, as well as higher number of cardiovascular
diseases and higher frequency of respiratory disease com-
pared to those without dyspnoea (Table 1). Participants with
missing FEV1 z-score (44 did not perform spirometry, 21

Downloaded from https://academic.oup.com/ageing/article-abstract/46/6/994/3868056 by guest on 08 December 2019

Results
Baseline characteristics of the study population
Out of the 567 participants of the BELFRAIL cohort, 565
(99.6%) had MRC score at the baseline assessment and
were considered as this study’s population. They had a
mean age of 84.7 years and consisted of 62.8% females.
There were 126 (22.3%) participants who reported no dys-
pnoea; 118 (20.9%) were troubled by dyspnoea only on
strenuous exercise (MRC 1); 152 (26.9%) when hurrying on
had non-usable spirograms and 6 were older than 95 years,
the age limit of reference equations) had no significant differ-
ences compared to those with available FEV1, except for a
higher death rate (55.6% versus 40.8%) and dyspnoea preva-
lence (40.8% versus 28.3%).
Correlates of dyspnoea
The multivariable logistic regression analysis with all the
possible dyspnoea correlates, showed that low FEV1 (OR
3.19, 95% confidence interval (CI) 1.90–5.34), low physical

Table 1. Baseline characteristics of the study population.

Total population (n = 565) Moderate/severe dyspnoea (n = 169) No/mild dyspnoea (n = 396) P value
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Age (years) 84.72 (3.68) 85.16 (3.63) 84.54 (3.69) 0.07a
Female sex 356 (62.8%) 113 (66.9%) 242 (61.1%) 0.22b
BMI (kg/m2) 27.42 (4.87) 28.07 (5.22) 27.04 (4.53) 0.016a
Underweight (<18.5) 9 (1.6%) 3 (1.8%) 6 (1.5%) 0.28b
Normal (≥18.5 < 25) 167 (29.9%) 41 (25%) 126 (32.1%)
Overweight (≥25 < 30) 221 (39.6%) 65 (39.6%) 155 (39.5%)
Obese (>30) 161 (28.9%) 55 (33.5%) 105 (26.8%)
FEV1 (litres) 1.69 (0.59) 1.45 (0.53) 1.79 (0.58) <0.001a
Low FEV1 143 (28.9%) 69 (49.3%) 74 (20.9%) <0.001b
Smoker/ex-smoker 179 (31.7%) 57 (33.7%) 122 (30.8%) 0.49b
Education 0.17b
Primary 212 (37.8%) 73 (43.7%) 139 (35.3%)
Secondary 277 (49.4%) 74 (44.3%) 203 (51.5%)
College/University 72 (12.8%) 20 (12%) 52 (13.2%)
COPD/asthma 82 (14.5%) 50 (29.6%) 32 (8.1%) <0.001b
Cardiovascular diseases 2 [1,4] 3 [2,4] 2 [1,3] <0.001c
GDS-15 score 2 [1,4] 3 [2,5] 2 [1,4] <0.001c
High GDS-15 score 118 (21.3%) 55 (33.5%) 62 (16%) <0.001b
ADL score 25 [21,27] 21 [17,25] 26 [23,29] <0.001c
Low ADL score 100 (17.9%) 64 (39%) 36 (9.2%) <0.001b
Grip strength (kg) 20.9 [15.8, 27.1] 18.3 [13.7, 23.9] 21.9 [16.9, 27.7] <0.001c
Low grip strength 107 (19.6%) 50 (31.8%) 57 (14.7%) <0.001b
PPT score 9 [5, 11] 7 [4,9] 9 [6,12] <0.001c
Low PPT score 85 (15.7%) 46 (29.9%) 39 (19.2%) <0.001b
Low physical performance 153 (28.5%) 72 (47.4%) 81 (21.1%) <0.001b
NT-proBNP (pg/ml) 188.3 [93.8, 517.8] 288.1 [128.5, 812.5] 166.4 [86.2, 372.8] <0.001c
High NT-proBNP 159 (29.2%) 71 (44.1%) 88 (23%) <0.001b

Data are presented as n, mean ± SD, n (%) or median [interquartile range]. Moderate/severe dyspnoea: Medical Research Council grade 3–5; No/mild dyspnoea:
no dyspnoea or Medical Research Council grade 1–2. BMI: body mass index; FEV1: forced expiratory volume in 1 s; COPD; chronic obstructive pulmonary dis-
ease; GDS-15: 15 items geriatric depression scale, ADL: activities of daily living; PPT: physical performance test; NT-proBNP: N-terminal pro-brain natriuretic
peptide; Low FEV1: z-score < −1.645; High GDS-15 score: ≥ 5; Low ADL score: ≤20. Low grip strength/PPT score: lowest sex-specific quintile; Low physical
performance: low grip strength/PPT score; High NT-proBNP: ≥400 pg/ml.
a
P value based on Student’s t-test.
b
P value based on Chi-square test.
c
P value based on Mann–Whitney U test.

996
 Correlates of dyspnoea and its association with adverse outcomes

performance (OR 2.49, 95%CI 1.50–4.12), high NT-
proBNP (OR 2.36, 95%CI 1.45–3.85), overweight/obese
BMI (OR 2.20, 95%CI 1.29–3.75) and high GDS-15 score
(OR 1.87, 95%CI 1.09–3.22) were independently associated
with dyspnoea. Primary education level (OR 1.32 95%CI
0.83–2.11), male sex (OR 0.91 95%CI 0.48–1.72), older age
(OR 0.98 95%CI 0.91–1.05) and ever-smoker status (OR
1.71 95%CI 0.93–3.14) were not independent dyspnoea
correlates. In the complimentary CART analysis, the tree-
based classification model included only the following vari-
ables in order of tree-importance: FEV1 (100%), physical
performance (72.6%), NT-proBNP (34.4%) and BMI
(16.3%). Those with normal FEV1, physical performance,
NT-proBNP and BMI had only 4% chance of having dys-
pnoea (Supplementary data, Appendix 2, available in Age
and Ageing online for details).
Association with adverse outcomes
Mortality data were available for all the participants.
Hospitalisation data were available for 559 (98.9%) partici-
pants. During 3.0 ± 0.25 years follow-up, 288 (51.5%) had
at least one unplanned hospitalisation, and at 5.1 ± 0.2
years 242 (42.8%) had died. Data on new/worsened ADL

Downloaded from https://academic.oup.com/ageing/article-abstract/46/6/994/3868056 by guest on 08 December 2019

Figure 1. Kaplan–Meier survival analysis curves of the two groups of dyspnoea for 5 years all-cause, cardiovascular and non-
cardiovascular mortality, and unplanned hospitalisation at 3 years follow-up. MRC: Medical Research Council dyspnoea scale.

997
E. Hegendörfer et al.

disability were available for 419 (74.2%) participants (72
had died before the second assessment, 60 refused it and 6
had incomplete ADL score). At 1.7 ± 0.21 years follow-up,
107 (25.5%) had new/worsened ADL disability.
Kaplan–Meier survival curves showed a significantly low-
er proportion of those with dyspnoea surviving all-cause, car-
diovascular and non-cardiovascular mortality at 5 years, and
being without unplanned hospitalisation at 3 years follow-up
(Figure 1). Participants with dyspnoea had higher risk of all-
cause (HR 2.04, 95%CI 1.58–2.64) and cardiovascular (HR
2.85, 95%CI 1.93–4.20) mortality, unplanned hospitalisation
(HR 1.72, 95%CI 1.35–2.19) and new/worsened ADL dis-
hospitalisation at 3 years, and new/worsened ADL disability
at around 2 years follow-up, independently of age, sex and
smoking status. These associations were significant even for
those with normal FEV1, NT-proBNP, physical performance
or BMI.
Prevalence and correlates of dyspnoea
Previous studies of dyspnoea in general populations of older
adults have reported MRC 3–5 prevalence of 8.2–32.3% [1].
Only two studies reported age-stratified data, where prevalence

Downloaded from https://academic.oup.com/ageing/article-abstract/46/6/994/3868056 by guest on 08 December 2019

ability (OR 2.49, 95%CI 1.54–4.04), even after adjustment
for age, sex and smoking (Table 2). Both worst and best
scenario analysis for the participants without follow-up ADL
scores did not significantly change these findings.
The association of dyspnoea with all-cause and cardio-
vascular mortality, and unplanned hospitalisation was statis-
tically significant for those with normal FEV1, normal or
high NT-proBNP (except for all-cause mortality), normal
physical performance and both BMI categories (Table 2).
The association with new/worsened disability was statistic-
ally significant for those with normal FEV1, normal NT-
proBNP or overweight/obese BMI (Table 2).
Discussion
Main findings
In a cohort of adults aged 80 and over, nearly 30% reported
dyspnoea MRC 3–5. Low FEV1, high NT-proBNP, low phys-
ical performance and high BMI were its key independent cor-
relates. Dyspnoea in this cohort was associated with 5-years
all-cause and cardiovascular mortality, first unplanned
for those aged 80 and over ranged 29–45.3% for males and
35–43% for females [1]. In our representative cohort of adults
aged 80 and over, 29.9% reported dyspnoea MRC 3–5.
Few studies have looked at correlates of dyspnoea MRC
3–5 in older adults, and none so far in those aged 80 and
over [6, 20]. In a population-based cohort of adults aged 70
and over in Wales, dyspnoea MRC 3–5 (32.3% of the popu-
lation) was associated with obesity, respiratory diseases, left
ventricular systolic function, lower scores of an anxiety and
depression scale, self-assessed health status, and mobility
tasks of an ADL scale [6]. In the Cardiovascular Health
Study cohort of older adults in the USA (13% were aged 80
and over), age, smoking status, hip and waist size, FEV1,
forced vital capacity and presence of cardio-respiratory dis-
eases were reported as independent correlates of dyspnoea
MRC 3–5 (10.1% of the population) [20]. In our cohort of
adults aged 80 and over, the CART analysis showed low
FEV1, low physical performance, high NT-proBNP and
BMI as independent correlates of dyspnoea.
Both FEV1 as measure of respiratory disease and NT-
proBNP as measure of cardiovascular disease have been
reported as providing important information about chronic dys-
pnoea, independent of clinical factors, in community-dwelling

Table 2. Association of dyspnoea with time to all-cause and cardiovascular death, unplanned hospitalisation and new/
worsened disability.
All-cause mortality Cardiovascular mortality Unplanned hospitalisation New/worsened disability
HR (95%CI) HR (95%CI) HR (95%CI) OR (95%CI)
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
All cases
Unadjusted 2.22 (1.72–2.87) 3.06 (2.09–4.50) 1.74 (1.37–2.22) 2.60 (1.63–4.17)
Adjusted 2.04 (1.58–2.64) 2.85 (1.93–4.20) 1.72 (1.35–2.19) 2.49 (1.54–4.04)
Stratified
Low FEV1 1.34 (0.87–2.06) 1.99 (1.00–3.95) 1.26 (0.81–1.95) 1.02 (0.45–2.34)
Normal FEV1 1.99 (1.35–2.94) 2.92 (1.67–5.11) 2.04 (1.44–2.88) 4.95 (2.46–9.96)
High NT-proBNP 1.46 (0.98–2.18) 2.05 (1.14–3.70) 1.86 (1.23–2.81) 1.83 (0.77–4.38)
Normal NT-proBNP 2.07 (1.45–2.96) 2.82 (1.61–4.94) 1.47 (1.06–2.03) 3.08 (1.65–5.74)
Low physical performance 1.15 (0.75–1.75) 1.42 (0.76–2.65) 1.42 (0.94–2.13) 2.16 (0.94–4.96)
Normal physical performance 2.44 (1.70–3.51) 3.84 (2.20–6.72) 1.74 (1.25–2.42) 1.98 (0.96–4.09)
Overweight/obese 2.12 (1.54–2.92) 2.95 (1.84–4.74) 1.73 (1.28–3.81) 2.41 (1.37–4.23)
Normal/underweight 2.15 (1.36–3.41) 2.99 (1.43–6.27) 1.89 (1.21–2.94) 2.42 (0.92–6.42)

HR: hazard ratio from Cox regression; OR: odds ratio from logistic regression; 95%CI: 95% confidence interval. Reference group: no/mild dyspnoea. All cases
analysis: Adjusted for age, sex, smoking status; number of cases in analysis: 565 mortality, 559 hospitalisation, 419 new/worsened disability. Stratified analysis:
adjusted for age, sex, smoking status. Low FEV1: forced expiratory volume in 1 s z-score < −1.645; High NT-proBNP: N-terminal pro-brain natriuretic peptide ≥
400 pg/ml; Low physical performance: grip strength or physical performance test < lowest sex-specific quintile; Overweight/Obese: body mass index ≥ 25 kg/m2;
Normal/Underweight: <25 kg/m2.

998
 Correlates of dyspnoea and its association with adverse outcomes

adults 45–84 years old [18]. Poor performance on a single chair
stand test has also been reported as independently associated
with moderate-severe dyspnoea in community-dwelling adults
65–80 years old [7]. We confirmed these findings in our cohort
of adults aged 80 and over, using a broader measurement of
physical performance (battery of PPT and grip strength). We
found that low physical performance was strongly associated
with dyspnoea independent of low FEV1 and high NT-
proBNP. One suggested mechanism of the association between
low physical performance and dyspnoea in older adults is
through the age-related sarcopenia that includes weakness of
ambulation and respiratory muscles [7]. Also, deconditioning
ADL at follow-up. Yet, sensitivity analysis on the effect of
these missing data did not show significant differences from
the reported findings. We also used the MRC scale that mea-
sures only one dimension of dyspnoea, its impact, and not its
sensory-perceptual and affective distress dimensions [2].
Although the MRC scale has been widely used in older adults,
future research should consider multi-dimensional scales of
dyspnoea, as well as its dynamics over time [1, 13].
Conclusion

Downloaded from https://academic.oup.com/ageing/article-abstract/46/6/994/3868056 by guest on 08 December 2019

Nearly 30% of our representative cohort of adults aged 80
due to low physical activity, which is further worsened by dys-
and over reported dyspnoea MRC 3–5 that was independ-
pnoea, may exacerbate the age-related sarcopenia and then dys-
ently associated with higher risk for adverse outcomes and
pnoea on exertion [7]. These mechanisms, and interventions
correlated to cardiopulmonary and physical performance
that modify either or both sarcopenia and dyspnoea, such as
impairments. As it is easy to assess and provides valuable
exercise programs, need to be further investigated [7, 26, 27].
clinical information for improving the focus and prioritisa-
Our findings in adults aged 80 and over emphasise the
tion of care, dyspnoea should be actively searched for and
importance of a more comprehensive approach to the evalu-
managed as a multi-factorial symptom in this age-group.
ation of dyspnoea in this age-group, looking beyond cardio-
respiratory diagnoses and considering the multiple, age-related
impairments that may influence dyspnoea [7, 17, 28, 29]. Key points
• Nearly 30% of our representative cohort of adults aged
Dyspnoea and adverse outcomes 80 and over reported dyspnoea MRC 3–5.
• Dyspnoea was independently associated with higher risk
Previous studies in older adults have reported dyspnoea as an
for mortality, hospitalisation and disability.
independent predictor of all-cause mortality [8–10]. We
• Cardiopulmonary and physical performance impairments
extended these findings into adults aged 80 and over, and
were the main independent correlates of dyspnoea.
investigated association to unplanned hospitalisation and
• Dyspnoea is not only a symptom of common cardio-
new/worsened disability that are more relevant outcomes for
respiratory diseases but also a potential marker of frailty
this age-group. We found that these associations were pre-
and adverse outcomes.
sent, and for some outcomes event stronger, in participants
with normal FEV1, NT-proBNP or physical performance.
Thus, dyspnoea is a symptom that provides very valuable
prognostic information regarding development of adverse
outcomes in adults aged 80 and over, even in those with nor- Supplementary data
mal cardio-respiratory function or physical performance.
Supplementary data mentioned in the text are available to
Our findings support the recent call for a systematic and
subscribers in Age and Ageing online.
routine assessment of dyspnoea in older adults, not only as
a symptom of high burden chronic diseases such as cardio-
respiratory ones, but also as a geriatric syndrome and Conflicts of interest
potential marker of frailty and adverse outcomes [9, 13, 14].
Pro-active evaluation for dyspnoea is particularly import- None declared.
ant in the older adults, as its perception and spontaneous
report is reduced in this age-group [3, 14]. Previous research Funding
has reported poor early recognition of symptoms in older
adults due to impaired awareness and interpretation of symp- The BELFRAIL study was supported by an unconditional
toms, including dyspnoea [30]. They may be unaware of its grant from Fondation Louvain, Brussels, Belgium (grant
increasing severity due to altered sensory perception, become number B40320084685). E.H. has received a scholarship
accustomed to it or assign it to ‘normal’ ageing [3, 30]. from ERAWEB 2, Erasmus Mundus program of the
European Union at Katholieke Universiteit Leuven.
Strengths and limitations
References
The strengths of this study are its prospective cohort design,
with comprehensive and standardised assessments, and 1. van Mourik Y, Rutten FH, Moons KG et al. Prevalence and
almost complete data on mortality and hospitalisation follow- underlying causes of dyspnoea in older people: a systematic
up. Some of the limitations are the missing data on FEV1 and review. Age Ageing 2014; 43: 319–26.

999
E. Hegendörfer et al.
2. Parshall MB, Schwartzstein RM, Adams L et al. An official
American Thoracic Society statement: update on the mechan-
isms, assessment, and management of dyspnea. Am J Respir
Crit Care Med 2012; 185: 435–52.
3. Petersen S, von Leupoldt A, Van den Bergh O. Geriatric dys-
pnea: doing worse, feeling better. Ageing Res Rev 2014; 15:
94–9.
4. Pedersen F, Mehlsen J, Raymond I et al. Evaluation of dys-
pnoea in a sample of elderly subjects recruited from general
practice. Int J Clin Pract 2007; 61: 1481–91.
5. Barberger-Gateau P, Nejjari C, Tessier JF et al. Assessment
of disability and handicap associated with dyspnoea in elderly
subjects. Disabil Rehabil 1995; 17: 83–9.
6. Ho SF, O’Mahony MS, Steward JA et al. Dyspnoea and quality
of life in older people at home. Age Ageing 2001; 30: 155–9.
7. Vaz Fragoso CA, Araujo K, Leo-Summers L et al. Lower
extremity proximal muscle function and dyspnea in older per-
sons. J Am Geriatr Soc 2015; 63: 1628–33.
8. Pesola GR, Ahsan H. Dyspnea as an independent predictor
of mortality. Clin Respir J 2016; 10: 142–52.
9. Ahmed T, Steward JA, O’Mahony MS. Dyspnoea and mor-
tality in older people in the community: a 10-year follow-up.
Age Ageing 2012; 41: 545–9.
10. Sorlie PD, Kannel WB, O’Connor G. Mortality associated
with respiratory function and symptoms in advanced age. The
Framingham study. Am Rev Respir Dis 1989; 140: 379–84.
11. Nishimura K, Izumi T, Tsukino M et al. Dyspnea is a better
predictor of 5-year survival than airway obstruction in
patients with COPD. Chest 2002; 121: 1434–40.
12. Abidov A, Rozanski A, Hachamovitch R et al. Prognostic sig-
nificance of dyspnea in patients referred for cardiac stress
testing. N Engl J Med 2005; 353: 1889–98.
13. Banzett RB, O’Donnell CR. Should we measure dyspnoea in
everyone? Eur Respir J 2014; 43: 1547–50.
14. Bousquet J, Dinh-Xuan AT, Similowski T et al. Should we
use gait speed in COPD, FEV1 in frailty and dyspnoea in
both? Eur Respir J 2016; 48: 315–9.
15. World Health Organization. World Report on Ageing and
Health. Switzerland: WHO Press, 2015.
16. Vaes B, Pasquet A, Wallemacq P et al. The BELFRAIL
(BFC80+) study: a population-based prospective cohort study
of the very elderly in Belgium. BMC Geriatr 2010; 10: 39.
17. Miner B, Tinetti ME, Van Ness PH et al. Dyspnea in
community-dwelling older persons: a multifactorial geriatric
health condition. J Am Geriatr Soc 2016; 64: 2042–50.
18. Oelsner EC, Lima JA, Kawut SM et al. Noninvasive tests for
the diagnostic evaluation of dyspnea among outpatients: the
Multi-Ethnic Study of Atherosclerosis lung study. Am J Med
2015; 128: 171–80.e5.
1000
19. Gronseth R, Vollmer WM, Hardie JA et al. Predictors of dys-
pnoea prevalence: results from the BOLD study. Eur Respir
J 2014; 43: 1610–20.
20. Enright PL, Kronmal RA, Higgins MW et al. Prevalence
and correlates of respiratory symptoms and disease in
the elderly. Cardiovascular health study. Chest 1994; 106:
827–34.
21. Miller MR, Hankinson J, Brusasco V et al. Standardisation of
spirometry. Eur Respir J 2005; 26: 319–38.
22. Turkeshi EVB, Andreeva E, Mathei C, Adriaense W, van
Pottelbergh G, Degryse J. Short-term prognostic value of
forced expiratory volume in 1 second divided by height cubed
in a prospective cohort of people 80 years and older. BMC
Downloaded from https://academic.oup.com/ageing/article-abstract/46/6/994/3868056 by guest on 08 December 2019
Geriatr 2015; 15: 15.
23. Stanojevic S. GLI-2012 Excel Sheet Calculator. Version 4 [soft-
ware]. Updated 25 May 2014. Available from: www.
lungfunction.org (accessed 18 June 2015)
24. Vaz Fragoso CA, McAvay G, Van Ness PH et al. Aging-
related considerations when evaluating the forced expiratory
volume in 1 second (FEV1) over time. J Gerontol A Biol Sci
Med Sci 2015; 71: 929–34.
25. Dickstein K, Cohen-Solal A, Filippatos G et al. ESC
guidelines for the diagnosis and treatment of acute and
chronic heart failure 2008: the Task Force for the diagnosis
and treatment of acute and chronic heart failure 2008 of the
European Society of Cardiology. Developed in collaboration
with the Heart Failure Association of the ESC (HFA) and
endorsed by the European Society of Intensive Care
Medicine (ESICM). Eur J Heart Fail 2008; 10: 933–89.
26. Ries AL, Bauldoff GS, Carlin BW et al. Pulmonary rehabilita-
tion: joint ACCP/AACVPR evidence-based clinical practice
guidelines. Chest 2007; 131(5 Suppl):4S–42S.
27. Cruz-Jentoft AJ, Baeyens JP, Bauer JM et al. Sarcopenia:
European consensus on definition and diagnosis: report of
the European working group on sarcopenia in older people.
Age Ageing 2010; 39: 412–23.
28. Marcus BS, McAvay G, Gill TM et al. Respiratory symptoms,
spirometric respiratory impairment, and respiratory disease in
middle-aged and older persons. J Am Geriatr Soc 2015; 63:
251–7.
29. Pedersen F, Raymond I, Mehlsen J et al. Prevalence of dia-
stolic dysfunction as a possible cause of dyspnea in the eld-
erly. Am J Med 2005; 118: 25–31.
30. Riegel B, Dickson VV, Cameron J et al. Symptom recognition
in elders with heart failure. J Nurs Scholarsh 2010; 42:
92–100.
Received 13 December 2016; editorial decision 3 May
2017