Original Article
Click here to view the Editorial Comment by Dr. Juhn.
Asthma, other atopic conditions and risk of infections in
105 519 general population never and ever smokers
J. Helby1,2,3, B. G. Nordestgaard1,2,3,4, T. Benfield3,5 & S. E. Bojesen1,2,3,4
From the 1Department of Clinical Biochemistry; and 2The Copenhagen General Population Study, Herlev and Gentofte Hospital, Copenhagen
University Hospital, Herlev; 3Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen; 4The Copenhagen City Heart
Study, Bispebjerg and Frederiksberg Hospital, Copenhagen University Hospital, Copenhagen, Denmark; and 5Department of Infectious
Diseases, Hvidovre Hospital, Copenhagen University Hospital, Hvidovre, Denmark
Abstract. Helby J, Nordestgaard BG, Benfield T,
Bojesen SE (Copenhagen University Hospital,
Herlev; University of Copenhagen, Copenhagen,
Denmark). Asthma, other atopic conditions and
risk of infections in 105 519 general population
never and ever smokers. J Intern Med 2017;
https://doi.org/10.1111/joim.12635
Background. Individuals with atopic conditions may
have increased susceptibility to infections outside
the organs directly affected by their atopic condition.
Objective. We tested the hypothesis that atopic con-
ditions overall, and stratified by smoking history,
are associated with increased risk of hospitaliza-
tion for infections.
Methods. We collected information on smoking history
and self-reported atopic conditions from 105 519
individuals from the general population and fol-
lowed them for up to 23 years for infectious disease
hospitalizations and deaths. For asthma, we
focused on never smokers with asthma diagnosed
before age 50 (early asthma) to minimize confound-
ing by chronic obstructive pulmonary disease.
Results. During follow-up, 11 160 individuals had
infections. Never smokers with early asthma
versus no atopic conditions had significantly
Introduction
Atopic conditions are common in developed coun-
tries where around 30% of the adult general
population suffers from asthma, atopic dermatitis
or hay fever [1, 2]. Individuals with atopic condi-
tions are at increased risk of infections in the organ
directly affected, for example, asthma is associated
with increased risk of pneumonia [3, 4], whilst
atopic dermatitis is associated with increased risk
of skin infections [5]. Asthma might also be
doi: 10.1111/joim.12635
increased risks of any infection (hazard ratio
1.65; 95% confidence interval 1.40–1.94), pneu-
monia (2.44; 1.92–3.11) and any non-respiratory
tract infection (1.36; 1.11–1.67); results were
similar in ever smokers. Never smokers with any
asthma had significantly increased risks of any
infection (1.44; 1.24–1.66) and pneumonia
(1.99; 1.62–2.44). Neither atopic dermatitis
(1.00; 0.91–1.10) nor hay fever (1.00; 0.93–
1.07) was associated with risk of any infection.
In never smokers, risk estimates for any infec-
tion were comparable between asthma and dia-
betes, as were the population attributable
fractions of 2.2% for any asthma and 2.9% for
diabetes.
Conclusion. Early asthma was associated with signif-
icantly increased risks of any infection, pneumonia
and any non-respiratory tract infection in never
and ever smokers. In never smokers, risk estimates
as well as population attributable fractions for any
infection were comparable between asthma and
diabetes, suggesting that asthma may be a sub-
stantial risk factor for infections in the general
population.
Keywords: allergy, asthma, epidemiology, infectious
disease.
associated with increased risk of Escherichia coli
bloodstream infections [6, 7] and asthma, as well
as other atopic conditions, is associated with a
increased risk of severe pneumococcal disease [8–
10]. These findings suggest that individuals with
atopic conditions may suffer from impaired innate
and adaptive immune functions, leading to
increased susceptibility to non-respiratory tract
infections in individuals with asthma or hay fever
and increased susceptibility to non-skin infections
in individuals with atopic dermatitis [11]. This
ª 2017 The Association for the Publication of the Journal of Internal Medicine 1
 Asthma and risk of infections / J. Helby et al.
hypothesis has not yet been tested amongst indi-
viduals from the general population, but the ques-
tion is important as atopic conditions are highly
prevalent and could therefore represent an impor-
tant infectious disease risk factor for the individual
with an atopic condition and for society as a whole.
The hypothesis is difficult to test for asthma – in
contrast to atopic dermatitis and hay fever – as
overlapping chronic obstructive pulmonary disease
(COPD) could complicate the analysis. COPD is
common, associated with increased risk of infec-
tions [12, 13], and is difficult to distinguish from
asthma in epidemiological studies [14]. Misclassi-
fication between asthma, COPD, heart failure and
other age-related conditions may especially occur
when diagnosing older adults with respiratory
symptoms [15–17], possibly due a high burden of
comorbidities, underuse of spirometry and a high
prevalence of asthma-COPD overlap syndrome in
these individuals [15, 18, 19].
In a prospective study of 105 519 individuals from
the general population followed for up to 23 years,
we tested the hypothesis that asthma and other
atopic conditions, overall and stratified by smoking
history, are associated with increased risk of
hospitalization due to infectious diseases and with
increased risk of infection-related death. We exam-
ined asthma, atopic dermatitis and hay fever, and
for asthma we focused on never smokers with early
asthma diagnosed at age 50 years or earlier to
minimize confounding due to COPD and other
smoking- and age-related conditions.
Methods
Participants
Study participants were 105 519 individuals from
two prospective studies of the general population.
We included 9976 individuals from the Copen-
hagen City Heart Study [20, 21] enrolled between
1991 and 1994, and 95 543 individuals from the
Copenhagen General Population Study [21, 22]
enrolled between 2003 and 2013. Recruitment was
based upon the Danish Civil Registration System
and similar for the two studies. Of those invited,
61% participated in the Copenhagen City Heart
Study and 46% participated in the Copenhagen
General Population Study. All participants com-
pleted a questionnaire regarding health and life-
style, reviewed by a trained examiner to correct
possible errors. At the date of enrolment, a physical
examination and spirometry were performed as
described in detail previously [4, 23]. No
2 ª 2017 The Association for the Publication of the Journal of Internal Medicine
Journal of Internal Medicine
individuals participated in both studies. More than
99% of study participants were white and of
Danish descent. The studies were approved by
the Research Ethics Committee of the Capital
Region of Denmark (H-KF 01-144/01) and the
Research Ethics Committee of Copenhagen and
Frederiksberg (KF-100.2039/91). All participants
provided written informed consent.
Atopic conditions and diabetes
Information on atopic conditions was derived from
the questionnaire, with asthma assessed using the
question ‘Do you have asthma?’ (yes/no), hay fever
assessed using the question ‘Do food items, med-
ications, grass, flowers, animal hair or anything
else give you hayfever?’ (yes/no), and atopic der-
matitis assessed using the question ‘Do food items,
medications, grass, flowers, animal hair or any-
thing else give you dermatitis?’ (yes/no). Individu-
als with asthma were asked to answer the question
‘How many years have you had asthma?’ and age at
asthma diagnosis was calculated by subtracting
the duration of asthma in years from the individ-
ual’s age at examination. Use of asthma medication
was assessed by the question ‘Do you take medi-
cation for asthma/bronchitis (including spray/
powder) on a daily or almost daily basis?’ (yes/
no). Information on diabetes was derived from the
questionnaire using the question ‘Do you have
diabetes?’ (yes/no).
Smoking history and other covariates
Based on the questionnaire and physical examina-
tion, the following covariates were derived: smoking
history (current/former/never), cumulative smok-
ing in pack-years (with one pack year defined as 20
cigarettes or equivalent per day for 1 year), alcohol
consumption (none/moderate/heavy, with heavy
defined as >168 g per week for men and >84 g per
week for women as recommended by the Danish
National Board of Health) and body mass index
(measured weight in kilograms divided by mea-
sured height in metres squared). From the spirom-
etry, we derived forced expiratory volume in one
second (FEV1) and forced vital capacity (FVC) in
litres and as per cent of predicted [23], as well as
FEV1/FVC ratio in per cent.
Ascertainment of endpoints and COPD
The Danish Civil Registration System is 100%
complete and contains a unique identification
 Asthma and risk of infections / J. Helby et al.
number for all individuals with permanent resi-
dence in Denmark [24]. Using this number, we
retrieved information on infectious disease hospi-
talizations until November 5, 2014 from the
national Danish Patient Registry [25], which covers
all Danish hospitals. Only inpatient hospitaliza-
tions and emergency room visits with a primary
discharge diagnosis of an infectious disease were
included in the analyses. Vital status and date of
death were obtained from the Danish Civil Regis-
tration System until November 5, 2014. We also
retrieved information on infection-related deaths
from study enrolment until December 31, 2012
from the national Danish Register of Causes of
Death [26], which covers all deaths in Denmark
and contains death certificates with ranked diag-
noses contributing to the cause of death as certi-
fied by a physician.
Infectious diseases were classified based on the
World Health Organization International Classifi-
cation of Diseases revision 8 (ICD-8) until Decem-
ber 31, 1993 and revision 10 (ICD-10) for events
happening after this date. For the main analyses,
infectious diseases were categorized as either
pneumonia or any non-respiratory tract infection
according to the ICD-8 and ICD-10 codes shown in
Table S1. As validation of diagnoses, 141 hospital
admissions coded as infectious diseases in the
national Danish Patient Registry were compared
with detailed clinical information obtained from
hospital charts reviewed by a medical doctor [20];
in 139 of the 141 admissions (99%), the registry
data captured relevant signs and symptoms of
infectious disease, a positive culture from a sterile
site or relevant specimen and/or treatment with
antibiotics.
Information on asthma hospitalizations and hos-
pital diagnosed COPD were obtained from the
national Danish Patient Registry (asthma codes
for ICD-8: 493, ICD-10: J45-J46; COPD codes for
ICD-8: 491-492, ICD-10: J41-J44).
Statistical analysis
Statistical analyses were performed using STATA
version 13.1. All statistical tests were two-sided.
For comparisons of baseline characteristics, Pear-
son’s chi-square test was used for categorical
variables and Mann-Whitney U test was used for
continuous variables. Hazard ratios for risk of
infectious disease hospitalizations and infection-
related deaths were modelled separately using Cox
proportional hazards regression with left-truncated
age as the timescale. Follow-up began at the date of
enrolment in the study. For risk of infectious
disease hospitalization, follow-up ended on date
of event, death, emigration (n = 484) or November
5, 2014, whichever came first. For risk of infection-
related death, follow-up ended on date of death,
emigration or December 31, 2012, whichever came
first. No participants were lost to follow-up. We
observed no major violations of the proportional
hazards assumption when assessed using Schoen-
feld residuals and by plotting ln ( ln(survival))
against ln(analysis time). When examining whether
risk estimates from two groups were different, the
Z-test described by Altman and Bland was used
[27]. To account for the competing risk of death
from other causes than infections, the analyses on
risk of infection-related death were repeated using
Fine-Gray regression [28], which produced results
similar to those from the Cox model (data not
shown).
Multivariable models were adjusted for baseline
values of age, sex, smoking history, cumulative
smoking in pack-years, alcohol consumption, body
mass index, self-reported diabetes, hospital diag-
nosed COPD and study cohort. These covariates
were chosen based on previous studies which have
found them to be associated with risk of atopic
conditions and/or infections [2, 3, 12, 20, 29].
Smoking can possibly confound associations
between asthma and risk of infectious disease as
some individuals with self-reported asthma may in
reality suffer from smoking-related airway disease
[14], for example COPD, and previous studies have
found that COPD is associated with increased risk
of infectious disease hospitalization [12, 13]. As
COPD is relatively uncommon in never smokers
and in individuals younger than 50 years of age
[30], confounding by COPD as well as other smok-
ing- and age-related conditions may be reduced by
focusing the analyses on never smokers with
asthma diagnosed before 50 years of age. There-
fore, we performed analyses stratified on smoking
history and on age at asthma diagnosis, with early
asthma defined as asthma diagnosed at age
50 years or earlier, and late asthma as asthma
diagnosed after age 50 years, similarly to the
methods used in studies on asthma and risk of
invasive pneumococcal infections [10, 31]. To fur-
ther reduce confounding due to COPD, we also
adjusted the multivariable model for hospital diag-
nosed COPD at study enrolment.
ª 2017 The Association for the Publication of the Journal of Internal Medicine 3
Journal of Internal Medicine
 Asthma and risk of infections / J. Helby et al.
The population attributable fractions (PAF) for
infectious disease risk factors were calculated on
the basis of the multivariable adjusted Cox-model
using the method described by Greenland and
Drescher [32].
Information on age, sex, smoking history and study
cohort was 100% complete, whilst information on
each of the remaining covariates was more than
95% complete. Missing values were coded as
missing for categorical variables, whilst missing
values for the continuous variables cumulative
smoking (n = 2045) and duration of asthma
(n = 269) were imputed based on a linear regres-
sion on sex, age, study cohort and smoking history.
However, when excluding individuals with missing
values, all analyses produced results similar to
those presented.
Results
Baseline characteristics of participants from the
general population according to presence versus
absence of atopic conditions are shown in Table 1.
Individuals with asthma are further divided into
early asthma (diagnosed at age 50 years or earlier)
and late asthma (diagnosed after age 50). Impor-
tantly, early asthma was associated with decreased
prevalence of ever smoking and decreased cumu-
lative smoking, whilst late asthma was associated
with increased prevalence of ever smoking and
increased cumulative smoking. The prevalence of
hospital diagnosed COPD was lower in the early
asthma (8.0%) than in the late asthma group
(24.7%). When stratifying baseline characteristics
according to smoking history, the prevalence of
hospital diagnosed COPD was lower in never
smokers with early asthma (2.9%) than in never
smokers with late asthma (10.4%) (Table S2).
Risk of any infection
During a median follow-up of 6 years (range: 0–
23 years), 11 160 individuals were hospitalized at
least once due to an infectious disease cause.
Individuals with any asthma versus no atopic
conditions had significantly increased risk of any
infection (hazard ratio [HR] 1.54; 95% confidence
interval [CI] 1.44–1.65), whilst this was not seen
for individuals with atopic dermatitis alone (HR
1.00; 95% CI 0.91–1.10), individuals with hay
fever alone (1.00; 0.93–1.07), or individuals with
both atopic dermatitis and hay fever (1.08; 0.96–
1.22) (Fig. 1). When compared to individuals with
4 ª 2017 The Association for the Publication of the Journal of Internal Medicine
Journal of Internal Medicine
no atopic conditions, risk estimates for pneumo-
nia and any non-respiratory tract infections were
unchanged for individuals with atopic dermatitis
alone, individuals with hay fever alone, and
individuals with both atopic dermatitis and hay
fever (Figure S1).
Asthma and risk of specific infections
To reduce possible confounding due to COPD and
other smoking- and age-related conditions, the
analyses on asthma and risk of specific infections
were stratified on smoking history and on age at
asthma diagnosis. In never smokers, individuals
with any asthma versus no atopic conditions had
significantly increased risk of any infection (HR
1.44; 95% CI 1.24–1.66) and pneumonia (1.99;
1.62–2.44); never-smokers with early asthma ver-
sus no atopic conditions had significantly
increased risk of any infection (1.65; 1.40–1.94),
pneumonia (2.44; 1.92–3.11) and any non-respi-
ratory tract infection (1.36; 1.11–1.67) (Fig. 2).
When examining subtypes of non-respiratory tract
infections (Figures S2 and S3), the increased risk of
non-respiratory tract infections in never smokers
with early asthma was pronounced for skin infec-
tion (HR 1.29; 95% CI 0.94–1.77), urinary tract
infection (1.47; 1.02–2.12), sepsis (1.37; 0.82–
2.29), diarrhoeal disease (1.42; 0.89–2.27) and
other infections (1.42; 0.59–3.43) (Figure S3). We
found no interaction between any asthma and
smoking history or between early asthma and
smoking history on any type of infection (Figures
S2 and S3).
Asthma characteristics and risk of infections
When examining risk of any infection, pneumonia
and any non-respiratory tract infection according
to age at asthma diagnosis, risk estimates were
similar for never smokers with asthma diagnosed
before age 18, 30, 40 and 50 years (early asthma)
(Fig. 3). Likewise, risk estimates for infections were
similar in all subgroups when stratifying never
smokers with early asthma according to each of the
following characteristics: asthma hospitalization
before study enrolment (yes/no), use of asthma
medication (yes/no), FEV1/FVC ratio (<70% vs.
≥70%), FEV1 per cent of predicted (<80% vs. ≥80%),
duration of asthma (≤20 years vs. >20 years) and
presence of atopic dermatitis and/or hay fever
concurrent with asthma (yes/no) (Figures S4–S6).
Amongst the 167 never smokers with early asthma
who were hospitalized for infections, only five had
 Table 1 Baseline characteristics of individuals from the general population according to presence versus absence of atopic conditions. Individuals with
asthma are further divided into early asthma (diagnosed at age 50 years or earlier) and late asthma (diagnosed after age 50). Individuals with asthma are
classified as having ‘any asthma’ regardless of whether they also have hayfever and/or atopic dermatitis

No atopic Any asthma

Characteristics conditions Early (≤50 years) Late (>50 years) Atopic dermatitis alone Hayfever alone Atopic dermatitis and hayfever
Participants, 77 015 (73) 4785 (5) 1765 (2) 5042 (5) 13 515 (13) 3397 (3)
no. (%)
218 285 66 237 115
Age, years 59 (49–68) 52 (44–60) P = 7 9 10 69 (64–75) P = 1 9 10 55 (45–65) P = 8 9 10 54 (45–64) P = 2 9 10 52 (44–62) P = 5 9 10
(IQR)
18 14 180 96
Male sex, 36 161 (47) 1933 (40) P = 1 9 10 668 (38) P = 3 9 10 1323 (26) P = 6 9 10 6398 (47) P = 0.41 976 (29) P = 2 9 10
no. (%)
8 25 103 13
Ever smokers, 47 488 (62) 2760 (58) P = 4 9 10 1304 (74) P = 1 9 10 3111 (62) P = 0.95 7001 (52) P = 2 9 10 1879 (55) P = 1 9 10
no. (%)
17 56 197 37
Cumulative 5 (0–23) 3 (0–19) P = 3 9 10 17 (0–35) P = 4 9 10 5 (0–21) P = 0.017 1 (0–13) P = 3 9 10 2 (0–15) P = 6 9 10
smoking,
pack-years
(IQR)
16 5
Alcohol 30 148 (39) 1572 (33) P = 2 9 10 710 (40) P = 0.21 1833 (36) P = 0.0002 5068 (37) P = 6 9 10 1218 (36) P = 0.0003
consumption
>168/
Asthma and risk of infections / J. Helby et al.

84 g per week a,
No. (%)
11 14 5 8 6
Body mass 25.6 25.9 P = 8 9 10 26.4 P = 3 9 10 25.3 P = 7 9 10 25.3 P = 1 9 10 25.1 P = 5 9 10
index, (23.2–28.4) (23.4–29.1) (23.8–29.5) (22.9–28.4) (23.1–28.1) (22.8–28.2)
kg/m2 (IQR)
9 10
Diabetes, no. (%) 2968 (3.9) 186 (3.9) P = 0.90 118 (6.7) P = 1 9 10 172 (3.4) P = 0.11 373 (2.8) P = 4 9 10 94 (2.8) P = 0.001
223 300 8
Hospital 1182 (1.5) 385 (8.0) P = 5 9 10 436 (24.7) P < 1 9 10 63 (1.2) P = 0.11 126 (0.9) P = 6 9 10 39 (1.1) P = 0.07
diagnosed
COPD, no. (%)
300 300 13 6
Hospitalization 113 (0.1) 628 (13.1) P < 1 9 10 197 (11.2) P < 1 9 10 11 (0.2) P = 0.21 60 (0.4) P = 3 9 10 16 (0.5) P = 4 9 10
due to
asthma, no. (%)
300 300 72 50
Medication 1084 (1) 3109 (65) P < 1 9 10 1432 (81) P < 1 9 10 67 (1) P = 0.65 487 (4) P = 2 9 10 157 (5) P = 5 9 10
for asthma or
bronchitis,
no. (%)
179 300 8 32 10
FEV1/FVC 11 178 (15) 1428 (30) P = 2 9 10 840 (48) P < 1 9 10 588 (12) P = 2 9 10 1450 (11) P = 2 9 10 362 (11) P = 2 9 10
ratio <70%,
no. (%)

Journal of Internal Medicine

ª 2017 The Association for the Publication of the Journal of Internal Medicine
5
 6
Journal of Internal Medicine
Table 1 (Continued )

No atopic Any asthma

Characteristics conditions Early (≤50 years) Late (>50 years) Atopic dermatitis alone Hayfever alone Atopic dermatitis and hayfever
184 300 40 9
FEV1 <80% of 12 205 (16) 1527 (32) P = 4 9 10 923 (52) P < 1 9 10 745 (15) P = 0.04 1546 (11) P = 2 9 10 412 (12) P = 3 9 10
predicted,
no. (%)
Age at asthma 30 (14–40) 59 (55–65)

ª 2017 The Association for the Publication of the Journal of Internal Medicine
diagnosis,
years (IQR)
Duration of 20 (12–35) 8 (4–12)
asthma,
Asthma and risk of infections / J. Helby et al.

years (IQR)
Atopic dermatitis 2956 (62) 597 (34)
and/or hayfever
concurrent with
asthma, No. (%)

No. (%) is shown for categorical variables. Median (interquartile range, IQR) is shown for continuous variables. P-values are for comparisons with the
group of individuals with no atopic conditions. Pearson’s chi-square test was used for comparisons of categorical variables and Mann–Whitney U test was
used for comparisons of continuous variables.
COPD, chronic obstructive pulmonary disease; FEV1, forced expiratory volume in one second; FVC, forced vital capacity.
a
>168 g per week for men and >84 g per week for women.
 Asthma and risk of infections / J. Helby et al.
Individuals, No. Infections, No.
No atopic conditions 77015 8343
Any asthma 6550 1081
Atopic dermatitis alone 5042 480
Hayfever alone 13515 990
Dermatitis and hayfever 3397 266
Fig. 1 Risk of any infection in the general population for individuals with atopic conditions versus no atopic conditions.
Individuals with asthma are classified as having ‘any asthma’ regardless of whether they also have hayfever and/or
dermatitis. HR, hazard ratio; CI, confidence interval.
been hospitalized due to asthma within 6 months
prior to the infectious disease hospitalization.
When excluding these five individuals, all analyses
produced results similar to those presented (data
not shown).
Asthma and risk of infection-related death
In never smokers, individuals with any asthma
versus no atopic conditions had significantly
increased risk of death related to pneumonia
(HR 1.99; 95% CI 1.17–3.38), whilst risks of death
related to any infection (1.36; 0.84–2.20) and any
non-respiratory tract infection (0.67; 0.25–1.77)
were largely unchanged (Fig. 4). Amongst never
smokers with early asthma there were only six
infection-related deaths, all of which were pneu-
monia related (HR 1.93; 95% CI 0.82–4.51 for
pneumonia-related death in never smokers with
early asthma versus no atopic conditions). For
ever smokers, individuals with any asthma versus
no atopic conditions had significantly increased
risk of death related to any infection (HR 1.77;
95% CI 1.47–2.13), pneumonia (1.87; 1.51–2.31)
and any non-respiratory tract infection (1.62;
1.18–2.23); ever smokers with early asthma
versus no atopic conditions had significantly
increased risk of death related to any infection
(1.94; 1.46–2.57) and pneumonia (2.12; 1.53–
2.93).
Multivariable adjusted HR (95% CI) for any infection
1
1.00 (reference)
1.54 (1.44 to 1.65)
1.00 (0.91 to 1.10)
1.00 (0.93 to 1.07)
1.08 (0.96 to 1.22)
0.75 1 1.5 2
Hazard ratio (95% CI)
Comparing asthma to diabetes
Diabetes is a well-known risk factor for infectious
disease, so to put the magnitude of infectious
disease risk in individuals with asthma into per-
spective, we compared it with that for diabetes.
When compared with a reference group of never
smokers without diabetes or atopic conditions, risk
estimates for any infection were similar for never
smokers with any asthma (HR 1.48; 95% CI 1.28–
1.71), never smokers with early asthma (1.70;
1.44–2.01) and never smokers with diabetes
(1.79; 1.54–2.08) (Fig. 5).
When calculating the proportion of infectious dis-
ease hospitalizations that could theoretically be
prevented on the population level if a single risk
factor were completely eliminated, the population
attributable fraction (PAF) for any infection
amongst never smokers was similar for any asthma
(PAF 2.2%;95% CI 1.2–3.2%), early asthma
(2.1%;1.2–2.9%) and diabetes (2.9%;1.9–3.8%).
Discussion
In this prospective study of 105 519 general popu-
lation individuals, we found that amongst never
smokers, any asthma was associated with signifi-
cantly increased risk of any infection and pneumo-
nia, whilst early asthma diagnosed at age 50 years or
ª 2017 The Association for the Publication of the Journal of Internal Medicine 7
Journal of Internal Medicine
 Asthma and risk of infections / J. Helby et al.

No. of individuals No. of infections Multivariable adjusted hazard ratio (95% CI) P for interaction
with/without asthma with/without asthma for infections with smoking
Any asthma vs. no
atopic conditions

Any infection 0.07

Never smokers 2486 / 29527 231 / 2225 1.44 (1.24 to 1.66)
Ever smokers 4064 / 47488 850 / 6118 1.59 (1.47 to 1.72)
All individuals 6550 / 77015 1081 / 8343 1.54 (1.44 to 1.65)

Pneumonia 0.62
Never smokers 2486 / 29527 120 / 844 1.99 (1.62 to 2.44)
Ever smokers 4064 / 47488 554 / 2925 2.05 (1.86 to 2.27)
All individuals 6550 / 77015 674 / 3769 2.03 (1.85 to 2.22)

Any non-respiratory tract infection 0.09

Never smokers 2486 / 29527 130 / 1607 1.13 (0.94 to 1.36)
Ever smokers 4064 / 47488 436 / 4032 1.24 (1.12 to 1.38)
All individuals 6550 / 77015 566 / 5639 1.21 (1.10 to 1.32)

Early asthma vs. no

atopic conditions

Any infection 0.25

Never smokers 2025 / 29527 167 / 2225 1.65 (1.40 to 1.94)
Ever smokers 2760 / 47488 455 / 6118 1.71 (1.55 to 1.90)
All individuals 4785 / 77015 622 / 8343 1.68 (1.54 to 1.83)

Pneumonia 0.81
Never smokers 2025 / 29527 81 / 844 2.44 (1.92 to 3.11)
Ever smokers 2760 / 47488 275 / 2925 2.40 (2.10 to 2.74)
All individuals 4785 / 77015 356 / 3769 2.37 (2.11 to 2.66)

Any non-respiratory tract infection 0.30

Never smokers 2025 / 29527 103 / 1607 1.36 (1.11 to 1.67)
Ever smokers 2760 / 47488 254 / 4032 1.39 (1.21 to 1.58)
All individuals 4785 / 77015 357 / 5639 1.37 (1.22 to 1.53)

0.75 1 2 4
Hazard ratio (95% CI)

Fig. 2 Risk of infections in the general population for individuals with any asthma versus no atopic conditions and for early
asthma versus no atopic conditions stratified by smoking history. Early asthma was defined as asthma diagnosed at age
50 years or earlier. The sum of specific infections exceeds the number of any infections as some individuals were
hospitalized with more than one type of infection. CI, confidence interval.

earlier was associated with significantly increased could possibly be caused by an impaired immune
risk of any infection, pneumonia and any non-respi- response against bacterial infections, caused by
ratory tract infection. Results were similar in ever the tendency towards a Th2-skewed immune
smokers with early asthma. In never smokers, rela- response in individuals with allergy [11, 33].
tive risk estimates as well as population attributable Indeed, children with atopic conditions may pro-
fractions for any infection were comparable between duce fewer antibodies after pneumococcal vacci-
asthma and diabetes. Our finding that individuals nation than healthy children [34, 35], and in
with early asthma from the general population had animal models, allergic mice exhibit impaired
increased overall risk of any non-respiratory tract recruitment of neutrophils when faced with bacte-
infection is novel and corroborates earlier findings ria [36, 37]. These previous findings suggests that
from case–control studies on asthma and risk of allergy may impair innate and humoral immunity
infections due to specific pathogens. [11]. However, although there are considerable
similarities in the allergic immune response
The association between early asthma and amongst individuals with asthma, atopic dermati-
increased risk of non-respiratory tract infections tis and hay fever [38, 39], we found that neither

8 ª 2017 The Association for the Publication of the Journal of Internal Medicine
Journal of Internal Medicine
 Asthma and risk of infections / J. Helby et al.

Individuals, No. Infections, No. Multivariable adjusted hazard ratio (95% CI)
for infections

Any infection
Never smokers without atopic conditions 29527 2225 1
1.00 (reference)
Never smokers with asthma before:
age 18 708 56 1.78 (1.36 to 2.32)
age 30 1190 93 1.70 (1.37 to 2.10)
age 40 1614 124 1.63 (1.36 to 1.97)
age 50 2025 167 1.65 (1.40 to 1.94)
age 60 2294 197 1.57 (1.35 to 1.83)
age 70 2434 218 1.47 (1.27 to 1.70)
age 80 2478 229 1.44 (1.25 to 1.66)

Pneumonia
Never smokers without atopic conditions 29527 844 1
1.00 (reference)
Never smokers with asthma before:
age 18 708 26 2.71 (1.82 to 4.04)
age 30 1190 41 2.49 (1.80 to 3.44)
age 40 1614 56 2.40 (1.81 to 3.17)
age 50 2025 81 2.44 (1.92 to 3.11)
age 60 2294 99 2.30 (1.84 to 2.86)
age 70 2434 113 2.10 (1.70 to 2.59)
age 80 2478 118 1.98 (1.61 to 2.44)

Any non-respiratory tract infection

Never smokers without atopic conditions 29527 1607 1
1.00 (reference)
Never smokers with asthma before:
age 18 708 34 1.39 (0.99 to 1.96)
age 30 1190 58 1.37 (1.05 to 1.78)
age 40 1614 79 1.36 (1.08 to 1.71)
age 50 2025 103 1.36 (1.11 to 1.67)
age 60 2294 116 1.26 (1.04 to 1.52)
age 70 2434 124 1.15 (0.95 to 1.39)
age 80 2478 130 1.14 (0.94 to 1.37)

0.75 1 2 4 6
Hazard ratio (95% CI)

Fig. 3 Risk of infections in the general population for never smokers with asthma versus never smokers without atopic
conditions according to age at asthma diagnosis. The sum of specific infections exceeds the number of any infections as
some individuals were hospitalized with more than one type of infection. CI, confidence interval.

dermatitis nor hay fever were associated with
increased risk of any infection, pneumonia or any
non-respiratory tract infection. This indicates that
the increased risk of non-respiratory tract infec-
tions in individuals with early asthma may be
caused by other mechanisms not directly related
to the allergic immune response. The risk of
non-respiratory tract infections is also increased
in COPD [12], so alternatively, some common
feature of lung diseases could increase suscepti-
bility to non-respiratory tract infections. However,
these mechanistic considerations are speculative
and the exact mechanism explaining the observed
associations is currently unknown.
Our finding amongst never smokers that early
asthma was associated with increased risk of any
non-respiratory tract infection has not previously
been reported from studies of the general popula-
tion. Although no previous studies have examined
the overall risk of non-respiratory tract infections
in individuals with asthma, these results are

ª 2017 The Association for the Publication of the Journal of Internal Medicine 9
Journal of Internal Medicine
 Asthma and risk of infections / J. Helby et al.

No. of individuals No. of deaths Multivariable adjusted hazard ratio (95% CI) for P for interaction
with/without asthma with / without asthma infection related death with smoking
Any asthma vs. no
atopic conditions

Any infection 0.27

Never smokers 2357 / 27896 21 / 260 1.36 (0.84 to 2.20)
Ever smokers 3918 / 45603 154 / 1111 1.77 (1.47 to 2.13)
All individuals 6275 / 73499 175 / 1371 1.69 (1.42 to 2.01)

Pneumonia 0.90
Never smokers 2357 / 27896 18 / 168 1.99 (1.17 to 3.38)
Ever smokers 3918 / 45603 119 / 807 1.87 (1.51 to 2.31)
All individuals 6275 / 73499 137 / 975 1.86 (1.53 to 2.27)

Any non-respiratory tract infection 0.08

Never smokers 2357 / 27896 5 / 111 0.67 (0.25 to 1.77)
Ever smokers 3918 / 45603 50 / 408 1.62 (1.18 to 2.23)
All individuals 6275 / 73499 55 / 519 1.44 (1.06 to 1.94)

Early asthma vs. no

atopic conditions

Any infection 0.07

Never smokers 1926 / 27896 6 / 260 1.12 (0.49 to 2.59)
Ever smokers 2648 / 45603 58 / 1111 1.94 (1.46 to 2.57)
All individuals 4574 / 73499 64 / 1371 1.76 (1.35 to 2.29)

Pneumonia 0.45
Never smokers 1926 / 27896 6 / 168 1.93 (0.82 to 4.51)
Ever smokers 2648 / 45603 44 / 807 2.12 (1.53 to 2.93)
All individuals 4574 / 73499 50 / 975 2.02 (1.49 to 2.74)

Any non-respiratory tract infection

Never smokers 1926 / 27896 0 / 111
Ever smokers 2648 / 45603 18 / 408 1.55 (0.94 to 2.55)
All individuals 4574 / 73499 18 / 519 1.23 (0.75 to 2.00)

0.25 0.5 1 2 4 8
Hazard ratio (95% CI)

Fig. 4 Risk of infection-related death in the general population for individuals with any asthma versus no atopic conditions
and for early asthma versus no atopic conditions stratified by smoking history. Early asthma was defined as asthma
diagnosed at age 50 years or earlier. The sum of deaths related to specific infections exceeds the number of deaths related
to any infection as some individuals had more than one type of infection listed on their death certificate. CI, confidence
interval.

supported by case–control studies on specific increased risk of respiratory as well as non-respi-

pathogens, reporting that asthma was associated
with increased risk of Escherichia coli bloodstream
infections [6], invasive pneumococcal disease [10,
31] and herpes zoster [40]. When comparing
asthma to diabetes, which is a well-known infec-
tious disease risk factor [3, 20, 41], the relative risk
estimates for any infection in never smokers were
similar for any asthma, early asthma and diabetes.
Likewise, the population attributable fractions for
any infection in never smokers were similar for any
asthma, early asthma and diabetes. These findings
imply that asthma may be a substantial risk factor
for infections in the general population, and that
health care professionals should be aware of the
ratory tract infections in individuals with early
asthma.
Neither atopic dermatitis nor hay fever were associ-
ated with increased risk of any infection, pneumonia
or any non-respiratory tract infection, so we cannot
confirm the association between dermatitis and
increased risk of lower respiratory tract infections
reported in a cross-sectional study of 1008 individ-
uals [42]. Another case–control study with 174 cases
found that atopic dermatitis was associated with
increased risk of serious pneumococcal disease,
defined as pneumococcal pneumonia, pneumococ-
cal sepsis or other invasive pneumococcal infections

10 ª 2017 The Association for the Publication of the Journal of Internal Medicine
Journal of Internal Medicine
 Asthma and risk of infections / J. Helby et al.

Individuals, No. Infections, No. Multivariable adjusted hazard ratio P for difference
(95% CI) for infections with diabetes
Any infection
Never smokers without diabetes or atopic conditions 28486 2043 1.00 (reference)
1
Never smokers with any asthma 2486 231 1.48 (1.28 to 1.71) 0.07
Never smokers with early asthma 2025 167 1.70 (1.44 to 2.01) 0.65
Never smokers with diabetes 1295 218 1.79 (1.54 to 2.08)

Pneumonia
Never smokers without diabetes or atopic conditions 28486 770 1.00 (reference)
1
Never smokers with any asthma 2486 120 2.05 (1.66 to 2.54) 0.41
Never smokers with early asthma 2025 81 2.46 (1.92 to 3.15) 0.07
Never smokers with diabetes 1295 89 1.79 (1.42 to 2.27)

Any non-respiratory tract infection

Never smokers without diabetes or atopic conditions 28486 1469 1.00 (reference)
1
Never smokers with any asthma 2486 130 1.18 (0.97 to 1.42) 0.0003
Never smokers with early asthma 2025 103 1.44 (1.17 to 1.77) 0.05
Never smokers with diabetes 1295 163 1.89 (1.60 to 2.25)

0.5 1 2 4
Hazard ratio (95% CI)

Fig. 5 Risk of infections in the general population for never smokers with any asthma, never smokers with early asthma
and never smokers with diabetes versus never smokers without diabetes or atopic conditions. Early asthma was defined as
asthma diagnosed at age 50 years or earlier. The sum of specific infections exceeds the number of any infections as some
individuals were hospitalized with more than one type of infection. CI, confidence interval.

[9]. These results are not directly comparable to ours,
as we do not have information on the specific
causative pathogens for most infectious disease
cases, but our results suggest that dermatitis is not
associated with an overall increased risk of serious
non-skin infections. However, it is likely that indi-
viduals with hay fever and/or atopic dermatitis are
more frequently prescribed antibiotics in the outpa-
tient setting without requiring hospitalization, as
previous studies have found increased risk of upper
respiratory tract infections in individuals with hay
fever and increased risk of skin infections in individ-
uals with atopic dermatitis [5, 43]. Hypothetically,
frequent exposure to antibiotics amongst individu-
als with hay fever and atopic dermatitis may decrease
risk of more serious infections that would require
hospitalization, and this may explain why hay fever
and atopic dermatitis were not associated with
increased risk of hospitalization for any infection.
as alcohol, body mass index, diabetes and COPD.
This made it possible to minimize confounding due
to COPD and other smoking- and age-related
conditions by stratifying the analyses on smoking
history and age at asthma diagnosis.
It was not possible for us to examine risk of
infections in individuals using specific types of
asthma medications such as inhaled corticos-
teroids, as we only have information on whether
or not the participants took any type of daily or
almost daily medication for asthma. However, we
found no indication that the increased risk of any
infection, pneumonia and any non-respiratory
tract infection in never smokers with early asthma
was caused by use of asthma medication as risk
estimates were similar when comparing individu-
als with asthma taking asthma medication to those
who did not.

Amongst the strengths of this study are the
prospective general population design, the large
sample size of 105 519 individuals and the avail-
ability of detailed information on self-reported
atopic conditions as well as smoking history and
other possible infectious disease risk factors such
Our study is limited by only having information on
infections treated in hospitals, which makes us
unable to assess whether asthma was also associ-
ated with increased risk of less severe cases of
infections that are typically treated by general
practitioners. Theoretically, our results may

ª 2017 The Association for the Publication of the Journal of Internal Medicine 11
Journal of Internal Medicine
 Asthma and risk of infections / J. Helby et al.
therefore be biased if general practitioners have a
lower threshold for admitting patients with asthma
to a hospital instead of treating them in an outpa-
tient setting. However, as asthma was also associ-
ated with increased risk of infection-related death,
it is less likely that our findings on increased risk of
infections in individuals with asthma are caused
solely by a lower threshold for hospital admission.
Another potential limitation is that the study was
conducted in a Danish population with a high
prevalence of atopic conditions and a relative low
burden of serious infections, as in most other
developed countries [44, 45]. Thus, our results
may not necessarily be applicable to other popula-
tions with different patterns of allergy and infec-
tious diseases.
Taken together, in this prospective study of
105 519 general population individuals followed
for up to 23 years, we found that amongst never
smokers, any asthma was associated with signifi-
cantly increased risks of any infection and pneu-
monia, whilst early asthma was associated with
significantly increased risks of any infection, pneu-
monia and any non-respiratory tract infection. In
never smokers, relative risk estimates as well as
population attributable fractions for any infection
were comparable between asthma and diabetes,
suggesting that asthma may be a substantial risk
factor for infections in the general population.
Conflict of interest statement
The authors have declared no conflicts of interest.
Acknowledgements
We thank the participants and staff of the Copen-
hagen City Heart Study and the Copenhagen
General Population Study for their important con-
tributions.
Funding
This work was supported by the Danish Council for
Independent Research, the research foundation for
health research of the Capital Region of Denmark
and by Herlev and Gentofte Hospital, Copenhagen
University Hospital.
Author contributions
J. Helby, B.G. Nordestgaard and S.E. Bojesen were
involved in the conception and design. J. Helby,
12 ª 2017 The Association for the Publication of the Journal of Internal Medicine
Journal of Internal Medicine
B.G. Nordestgaard, T. Benfield and S.E. Bojesen
participated in collection of data and assembly of
databases. J. Helby, B.G. Nordestgaard, T. Ben-
field and S.E. Bojesen analysed and interpreted
data. J. Helby wrote the manuscript. J. Helby, B.G.
Nordestgaard, T. Benfield and S.E. Bojesen were
involved in revision and final approval of the
manuscript.
References
1 Gupta R, Sheikh A, Strachan DP, Anderson HR. Burden of
allergic disease in the UK: secondary analyses of national
databases. Clin Exp Allergy 2004; 34: 520–6.
2 Meggs WJ, Dunn KA, Bloch RM, Goodman PE, Davidoff AL.
Prevalence and nature of allergy and chemical sensitivity in a
general population. Arch Environ Health 1996; 51: 275–82.
3 Almirall J, Bolibar I, Serra-Prat M et al. New evidence of risk
factors for community-acquired pneumonia: a population-
based study. Eur Respir J 2008; 31: 1274–84.
4 Colak Y, Afzal S, Nordestgaard BG, Lange P. Characteristics
and prognosis of never-smokers and smokers with asthma in
the Copenhagen general population study. A prospective
cohort study. Am J Respir Crit Care Med 2015; 192: 172–81.
5 Leung DY. Infection in atopic dermatitis. Curr Opin Pediatr
2003; 15: 399–404.
6 Bang DW, Yang HJ, Ryoo E et al. Asthma and risk of non-
respiratory tract infection: a population-based case-control
study. BMJ Open 2013; 3: e003857.
7 Laupland KB, Gregson DB, Church DL, Ross T, Pitout JD.
Incidence, risk factors and outcomes of Escherichia coli
bloodstream infections in a large Canadian region. Clin
Microbiol Infect 2008; 14: 1041–7.
8 Juhn YJ, Kita H, Yawn BP et al. Increased risk of serious
pneumococcal disease in patients with asthma. J Allergy Clin
Immunol 2008; 122: 719–23.
9 Jung JA, Kita H, Yawn BP et al. Increased risk of serious
pneumococcal disease in patients with atopic conditions other
than asthma. J Allergy Clin Immunol 2010; 125: 217–21.
10 Talbot TR, Hartert TV, Mitchel E et al. Asthma as a risk factor
for invasive pneumococcal disease. N Engl J Med 2005; 352:
2082–90.
11 Juhn YJ. Risks for infection in patients with asthma (or other
atopic conditions): is asthma more than a chronic airway
disease? J Allergy Clin Immunol 2014; 134: 247–57.
12 Inghammar M, Engstrom G, Ljungberg B, Lofdahl CG, Roth A,
Egesten A. Increased incidence of invasive bacterial disease in
chronic obstructive pulmonary disease compared to the
general population – a population based cohort study. BMC
Infect Dis 2014; 14: 163.
13 Benfield T, Lange P, Vestbo J. COPD stage and risk of
hospitalization for infectious disease. Chest 2008; 134: 46–
53.
14 Toren K, Brisman J, Jarvholm B. Asthma and asthma-like
symptoms in adults assessed by questionnaires. A literature
review. Chest 1993; 104: 600–8.
15 Gibson PG, McDonald VM, Marks GB. Asthma in older
adults. Lancet 2010; 376: 803–13.
 Asthma and risk of infections / J. Helby et al.
16 Abramson MJ, Perret JL, Dharmage SC, McDonald VM,
McDonald CF. Distinguishing adult-onset asthma from
COPD: a review and a new approach. Int J Chron Obstruct
Pulmon Dis 2014; 9: 945–62.
17 Jorge S, Becquemin MH, Delerme S et al. Cardiac asthma in
elderly patients: incidence, clinical presentation and out-
come. BMC Cardiovasc Disord 2007; 7: 16.
18 Soriano JB, Visick GT, Muellerova H, Payvandi N, Hansell AL.
Patterns of comorbidities in newly diagnosed COPD and
asthma in primary care. Chest 2005; 128: 2099–107.
19 Gershon AS, Victor JC, Guan J, Aaron SD, To T. Pulmonary
function testing in the diagnosis of asthma: a population
study. Chest 2012; 141: 1190–6.
20 Benfield T, Jensen JS, Nordestgaard BG. Influence of dia-
betes and hyperglycaemia on infectious disease hospitalisa-
tion and outcome. Diabetologia 2007; 50: 549–54.
21 Zacho J, Benfield T, Tybjaerg-Hansen A, Nordestgaard BG.
Increased baseline C-reactive protein concentrations are asso-
ciated with increased risk of infections: results from 2 large
Danish Population Cohorts. Clin Chem 2016; 62: 335–42.
22 Thomsen M, Ingebrigtsen TS, Marott JL et al. Inflammatory
biomarkers and exacerbations in chronic obstructive pul-
monary disease. JAMA 2013; 309: 2353–61.
23 Lokke A, Marott JL, Mortensen J, Nordestgaard BG, Dahl M,
Lange P. New Danish reference values for spirometry. Clin
Respir J 2013; 7: 153–67.
24 Pedersen CB, Gotzsche H, Moller JO, Mortensen PB. The
Danish Civil Registration System. A cohort of eight million
persons. Dan Med Bull 2006; 53: 441–9.
25 Lynge E, Sandegaard JL, Rebolj M. The Danish National
Patient Register. Scand J Public Health 2011; 39: 30–3.
26 Helweg-Larsen K. The Danish register of causes of death.
Scand J Public Health 2011; 39: 26–9.
27 Altman DG, Bland JM. Interaction revisited: the difference
between two estimates. BMJ 2003; 326: 219.
28 Fine JP, Gray RJ. A Proportional hazards model for the
subdistribution of a competing risk. J Am Stat Assoc 1999;
94: 496–509.
29 Hansen EF, Rappeport Y, Vestbo J, Lange P. Increase in
prevalence and severity of asthma in young adults in Copen-
hagen. Thorax 2000; 55: 833–6.
30 Fabricius P, Lokke A, Marott JL, Vestbo J, Lange P. Preva-
lence of COPD in Copenhagen. Respir Med 2011; 105: 410–7.
31 Klemets P, Lyytikainen O, Ruutu P et al. Risk of invasive
pneumococcal infections among working age adults with
asthma. Thorax 2010; 65: 698–702.
32 Greenland S, Drescher K. Maximum likelihood estimation of
the attributable fraction from logistic models. Biometrics
1993; 49: 865–72.
33 James KM, Peebles RS Jr, Hartert TV. Response to infections
in patients with asthma and atopic disease: an epiphe-
nomenon or reflection of host susceptibility? J Allergy Clin
Immunol 2012; 130: 343–51.
34 Lee HJ, Kang JH, Henrichsen J et al. Immunogenicity and
safety of a 23-valent pneumococcal polysaccharide vaccine in
healthy children and in children at increased risk of pneu-
mococcal infection. Vaccine 1995; 13: 1533–8.
35 Arkwright PD, Patel L, Moran A, Haeney MR, Ewing CI, David
TJ. Atopic eczema is associated with delayed maturation of
the antibody response to pneumococcal vaccine. Clin Exp
Immunol 2000; 122: 16–9.
36 Beisswenger C, Kandler K, Hess C et al. Allergic airway
inflammation inhibits pulmonary antibacterial host defense.
J Immunol 2006; 177: 1833–7.
37 Habibzay M, Saldana JI, Goulding J, Lloyd CM, Hussell T.
Altered regulation of Toll-like receptor responses impairs
antibacterial immunity in the allergic lung. Mucosal Immunol
2012; 5: 524–34.
38 Mrabet-Dahbi S, Maurer M. Does allergy impair innate
immunity? Leads and lessons from atopic dermatitis. Allergy
2010; 65: 1351–6.
39 Spergel JM. From atopic dermatitis to asthma: the atopic
march. Ann Allergy Asthma Immunol 2010; 105: 99–106.
40 Kwon HJ, Bang DW, Kim EN et al. Asthma as a risk factor for
zoster in adults: A population-based case-control study.
J Allergy Clin Immunol 2016; 137: 1406–12.
41 Muller LM, Gorter KJ, Hak E et al. Increased risk of common
infections in patients with type 1 and type 2 diabetes mellitus.
Clin Infect Dis 2005; 41: 281–8.
42 Rantala A, Jaakkola JJ, Jaakkola MS. Respiratory infections
in adults with atopic disease and IgE antibodies to common
aeroallergens. PLoS ONE 2013; 8: e68582.
43 Juhn YJ, Frey D, Li X, Jacobson R. Streptococcus pyogenes
upper respiratory infection and atopic conditions other than
asthma: a retrospective cohort study. Prim Care Respir J
2012; 21: 153–8.
44 To T, Stanojevic S, Moores G et al. Global asthma prevalence
in adults: findings from the cross-sectional world health
survey. BMC Public Health 2012; 12: 204.
45 Lozano R, Naghavi M, Foreman K et al. Global and regional
mortality from 235 causes of death for 20 age groups in 1990
and 2010: a systematic analysis for the Global Burden of
Disease Study 2010. Lancet 2012; 380: 2095–128.
Correspondence: Stig E. Bojesen, Department of Clinical Bio-
chemistry, Herlev and Gentofte Hospital, Copenhagen University
Hospital, Herlev Ringvej 75, 2730 Herlev, Denmark.
(fax: +45 3868 3311; e-mail: stig.egil.bojesen@regionh.dk).
Supporting Information
Additional Supporting Information may be found in
the online version of this article:
Table S1. Categorization of infectious diseases
according to the World Health Organization Inter-
national Classification of Diseases revision 8 (ICD-
8) and revision 10 (ICD-10).
Table S2. Baseline characteristics stratified by
smoking history for individuals with no atopic
conditions, any asthma, early asthma and late
asthma.
Figure S1. Risk of infections in the general popu-
lation for individuals with atopic dermatitis alone
versus no atopic conditions, hayfever alone versus
no atopic conditions and atopic dermatitis and
hayfever combined versus no atopic conditions.
ª 2017 The Association for the Publication of the Journal of Internal Medicine 13
Journal of Internal Medicine
 Asthma and risk of infections / J. Helby et al.
Figure S2. Risk of any infection, pneumonia, any
non-respiratory tract infection and subtypes of
non-respiratory tract infections in the general
population for individuals with any asthma versus
no atopic conditions stratified by smoking history.
Figure S3. Risk of any infection, pneumonia, any
non-respiratory tract infection and subtypes of
non-respiratory tract infections in the general
population for individuals with early asthma ver-
sus no atopic conditions stratified by smoking
history.
Figure S4. Risk of any infection in the general
population for never smokers with early asthma
14 ª 2017 The Association for the Publication of the Journal of Internal Medicine
Journal of Internal Medicine
versus never smokers without atopic conditions
stratified by asthma characteristics.
Figure S5. Risk of pneumonia in the general
population for never smokers with early asthma
versus never smokers without atopic conditions
stratified by asthma characteristics.
Figure S6. Risk of any non-respiratory tract
infection in the general population for never
smokers with early asthma versus never smokers
without atopic conditions stratified by asthma
characteristics.