1489

ORIGINAL ARTICLE
Relationship between Habit of Cigarette Smoking
and Airflow Limitation in Healthy Japanese
Individuals: The Takahata Study
Daisuke Osaka
1
, Yoko Shibata
1
, Shuichi Abe
1
, Sumito Inoue
1
, Yoshikane Tokairin
1
,
Akira Igarashi
1
, Keiko Yamauchi
1
, Tomomi Kimura
1
, Michiko Sato
1
, Hiroyuki Kishi
1
,
Noriaki Takabatake
1
, Makoto Sata
1
, Tetsu Watanabe
1
, Tsuneo Konta
1
, Sumio Kawata
2
,
Takeo Kato
3
and Isao Kubota
1
Abstract
Background Chronic obstructive pulmonary disease (COPD) is characterized by chronic airflow limitation.
The prevalence of airflow limitation in Japan is 10.9% (16.4% of males and 5.0% of females). Cigarette
smoking is well known as a major cause of COPD. However, few epidemiological studies have evaluated the
effects of cigarette smoking on pulmonary function in healthy subjects.
Methods Subjects aged 40 years or older (n=2,917), who had participated in a community-based annual
health check in Takahata, Japan, from 2004 through 2005, were enrolled in the study. The smoking histories
of these subjects were investigated using a self-reported questionnaire. Forced vital capacity (FVC), forced
expiratory volume in 1 second (FEV1), and forced expiratory flow at 25-75% of FVC (FEF25-75) were mea-
sured by standard procedures using spirometric machines.
Results There were 554 current smokers (18.6%) and 403 former smokers (13.8%). The prevalence of air-
flow limitation defined by FEV1/FVC <0.7 in this population was 10.6%, and prevalence of airflow limitation
defined by 5th percentile lower limit of normal was 6.4%. In smokers, percent predicted values of measured
spirometric parameters (%FVC, %FEV1 and %FEF25-75) decreased significantly with age, except for male
%FVC. Also, percent predicted values of measured spirometric parameters decreased significantly with in-
creasing pack-years, except for female %FEF25-75.
Conclusion Cigarette smoking increased the prevalence and severity of airflow limitation. It is concluded
that cigarette smoking increases the risk of airflow limitation in a healthy Japanese population.
Key words: pulmonary function test, airflow limitation, COPD, cigarette smoking, FEV1, community-based
health check-up, epidemiology
(Inter Med 49: 1489-1499, 2010)
(DOI: 10.2169/internalmedicine.49.3364)
Introduction
Chronic obstructive pulmonary disease (COPD) is a pre-
ventable and treatable disease, with significant extrapulmo-
nary effects that may contribute to severity in individual pa-
tients (1). The pulmonary component of COPD is character-
ized by airflow limitation that is not fully reversible. The
airflow limitation is usually progressive and is associated
with an abnormal inflammatory response of the lung to nox-
ious particles and gases. Worldwide, cigarette smoking is the
most common risk factor for COPD, although in many

Department of Cardiology, Pulmonology, and Nephrology, Yamagata University School of Medicine, Yamagata,

Department of Gastroenterol-
ogy, Yamagata University School of Medicine, Yamagata and

Department of Neurology, Hematology, Metabolism, Endocrinology, and Diabe-
tology, Yamagata University School of Medicine, Yamagata
Received for publication January 10, 2010; Accepted for publication April 21, 2010
Correspondence to Dr. Yoko Shibata, shibata@med.id.yamagata-u.ac.jp
Inter Med 49: 1489-1499, 2010 DOI: 10.2169/internalmedicine.49.3364
1490
countries, air pollution resulting from the burning of wood
and other biomass fuels has been identified as an additional
risk factor for COPD (2, 3).
Spirometry is a useful tool to diagnose and evaluate the
severity of respiratory disease. A forced vital capacity (FVC)
maneuver can be used to evaluate the condition of the air-
ways. Vital capacity (VC) and forced expiratory volume in 1
second (FEV1) are clinical markers for respiratory disease. A
percent predicted VC (%VC) value of less than 80% indi-
cates restrictive disease of the lung, while a FEV1/FVC ratio
of less than 70% is indicative of airflow limitation (4). Also,
forced expiratory flow at 25-75% of FVC (FEF25-75) is
known to be a parameter which correlates well with the total
amount of inhaled cigarette smoke in patients with
COPD (5, 6). In these patients, age and the initial value of
FEV1 were the most reliable predictors of death (7). There-
fore, it is necessary to perform pulmonary function tests in
order to diagnose and treat COPD. However, pulmonary
function testing is not widely performed even for current
heavy smokers, and many COPD patients remain undiag-
nosed (8-13).
The prevalence of airflow limitation in Japanese people
aged 40 years or older is reported to be 10.9% (16.4% of
males and 5.0% of females) (14). Cigarette smoking is well
known to be a major cause of COPD. However, there are
only a few reports of cohort studies that have evaluated the
impact of cigarette smoking on airflow limitation (15-17). In
this study, we evaluated the pulmonary function tests of
Japanese individuals who participated in a community-based
annual health check in Takahata, Yamagata. Predictive equa-
tions for FVC, FEV1 and FEF25-75 were derived from never
smoking subjects in this study. Age-related changes in %
FVC, percent predicted FEV1 (%FEV1) and percent pre-
dicted FEF25-75 (%FEF25-75) were cross-sectionally compared
between non-smokers and smokers, in order to perform an
epidemiological assessment of the effect of cigarette smok-
ing on pulmonary function.
Methods
Study population
This study formed part of the Molecular Epidemiological
Study utilizing the Regional Characteristics of 21st Century
Centers of Excellence (COE) Program and the Global COE
Program in Japan. Details of the study methodology have
been described elsewhere (18). The study was approved by
the institutional ethics committee and all participants gave
written informed consent.
This study utilized a community-based annual health
check, in which all inhabitants of Takahata town (total
population 26,026) in northern Japan, who were aged 40
years or older, were invited to participate. This region has a
resident population of 15,222 adults aged 40 years or older
(7,109 males and 8,113 females). From June 2004 through
November 2005, 1,380 males and 1,735 females (total
3,165) participated in the program and agreed to enroll in
the study. However, due to incomplete data 248 subjects
were excluded from the analysis. Data for a total of 2,917
subjects (1,325 males, 1,592 females) was entered into the
final statistical analysis.
Subjects used a self-reported questionnaire to document
their medical histories, current medications and clinical
symptoms. Current, former or never smokers were catego-
rized according to the responses on the self-reported ques-
tionnaire. Subjects who categorized themselves as never
smokers despite indicating the number of cigarettes smoked
per day or the number of years of smoking were categorized
as former smokers.
Lung function measurements
FVC, FEV1 and FEF25-75 were measured using standard
spirometric techniques, with subjects performing FVC ma-
neuvers on a CHESTAC-25 part II EX instrument (Chest
Corp., Tokyo, Japan) under guideline of the Japanese Respi-
ratory Society (JRS) (19). Bronchodilator was not adminis-
tered prior to measurements. The highest value from at least
three FVC maneuvers by each subject was used for the
analysis. Two pulmonary physicians assessed the results of
flow-volume curves by visual inspection, and they excluded
subjects with poor studied data according to JRS crite-
ria (19).
Subjects with an FEV1/FVC ratio <70% were classified as
having airflow limitation. The severity of airflow limitation
was stratified by %FEV1 in accordance with the criteria
specified by the Global Initiative for Chronic Obstructive
Lung Disease (GOLD): %FEV1 "80, mild; 50 ! %FEV1
<80, moderate; 30 ! %FEV1 <50, severe; %FEV1 <30, very
severe (1). Inconsistent with GOLD guideline (1), American
Thoracic Society (ATS) and European Respiratory Society
(ERS) recommend using 5th percentile lower limit of nor-
mal (LLN) as cutoff value for the definition of airflow ob-
struction (20, 21). We obtained predictive equation of FEV1/
FVC from this study population, and determined the equa-
tion which calculates the LLN of the subject; 5th percentile
LLN of FEV1/FVC = prediction -1.645 standard error of
estimate (SEE) (22, 23).
Statistical analysis
Students t test was used to evaluate differences in means,
and chi-square tests were used to evaluate differences in
proportions. The non-parametric Mann-Whitney U test and
Kruskal-Wallis test were used to compare variables that
were not normally distributed. Data are expressed as mean
SD. Predictive equations for FVC, FEV1, and FEF25-75 were
established by multiple linear, backward step-wise, regres-
sion. The difference of slopes in regressions was compared
by analysis of covariance (ANCOVA) between never-smoker
and former/current-smoker. All statistical analyses were per-
formed using JMP version 7.0.2 software (SAS Institute
Inc., Cary, NC, USA). Statistical significance was defined as
p<0.05.
Inter Med 49: 1489-1499, 2010 DOI: 10.2169/internalmedicine.49.3364
1491
Table 1. Comparison of the Baseline Characteristics between Men
and Women
All subjects Men Women
Number 2917 1325 1592
Age (years) 62.8 10.2 63.3 10.3 62.5 10.1
Height (cm) 156.8 9.0 163.5 6.9 151.2 6.2 **
Weight (kg) 57.9 10.3 62.8 10.0 53.9 8.6 **
BMI (kg/m
2
) 23.5 3.2 23.5 3.0 23.6 3.4
Never smoker, n (%) 1970 (67.5%) 513 (38.7%) 1457 (91.5%) **
Current smoker, n (%) 544 (18.6%) 452 (34.1%) 92 (5.8%) **
Former smoker, n (%) 403 (13.8%) 360 (27.2%) 43 (2.7%) **
Positive for the history of
pulmonary diseases, n (%)
64 (2.2%) 46 (3.5%) 18 (1.1%) **
BMI: body mass index, **: p < 0.001 v.s. Men
Results
Baseline characteristics of the participants
The baseline characteristics of the 2,917 subjects whose
data was entered into final analysis are shown in Table 1.
The mean age was 62.8 years, and there were 554 current
smokers (18.6%) and 403 former smokers (13.8%). The pro-
portion of smokers was higher among males than females.
2.2% subjects had medical history of pulmonary diseases
(male: 3.5%, female: 1.1%). However, the rate of prior diag-
nosed bronchial asthma or COPD patients in this population
was unknown, because the self-reported questionnaire did
not ask the detail of pulmonary diseases.
Age, gender, smoking history and pulmonary func-
tion
The relationships between age and the absolute values of
FVC, FEV1 and FEF25-75 are demonstrated in Fig. 1. Absolute
values of FVC, FEV1 and FEF25-75 declined with age in both
never-smokers and former/current-smokers. However, the
rates of decline of FEV1 and FEF25-75 in males, and FVC,
FEV1 and FEF25-75 in females were significantly greater in
the former/current-smoker group than in the never-smoker
group. In contrast, the difference in the rates of decline in
FVC between smoking and non-smoking male subjects did
not reach statistical significance.
Because absolute values of FVC, FEV1 and FEF25-75 were
not adjusted for age or body size of the subjects, we at-
tempted to compare the age-related changes in FVC, FEV1
and FEF25-75 using standardized values for these parameters.
Predictive equations for reference values of FVC, FEV1 and
FEF25-75 based on the age, height (HT) and body weight
(BW) of the subject were derived from never-smoking sub-
jects by backward stepwise, multiple linear regression analy-
sis (Table 2). The equations for the spirometric parameters
derived from this study are indicated in Table 2. The equa-
tions for FVC and FEV1 were correlated with those of
JRS (24): FVC (male), R = 0.992, SEE = 0.070; FVC (fe-
male), R = 0.999, SEE = 0.013; FEV1 (male), R = 0.996,
SEE = 0.040; FEV1 (female), R = 0.998, SEE = 0.024.
Predictive equations for FVC, FEV1 and FEF25-75 were also
derived for male smoking subjects, using backward step-
wise, multiple linear regression analysis. However, it was
not possible to derive equations for female smokers due to
the small number of these subjects. The equations for male
smokers were: FVC (male smoker) = - 1.4425 - (0.0329
age) + (0.043 HT), R
2
= 0.543, SEE = 0.520; FEV1 (male
smoker) = 0.6852 - (0.0371 age) + (0.0245 HT) +
(0.0052 BW), R
2
= 0.555, SEE = 0.477; and FEF25-75
(male smoker) = 5.26 - (0.0613 age) + (0.0168 BW), R
2
= 0.355, SEE = 0.982.
Prevalence of airflow limitation
The data for the prevalence of airflow limitation defined
by FEV1/FVC <0.7 (AFL0.7) is summarized in Fig. 2, Ta-
bles 3 and 4. Figures 2A & 2B show the relationship be-
tween FEV1/FVC and %FEV1. 4.6% subjects with AFL0.7
had medical history of pulmonary diseases (male: 6.1%, fe-
male: 1.1%).The prevalence of airflow limitation was higher
in male subjects than in female subjects (Figs. 2C & 2D;
chi-square test: likelihood ratio = 82.58, degree of freedom
=1, p<0.001). In male subjects with AFL0.7 in former/current
smoking group were more prevalent than in never smoking
group (Table 3). In contrast, female subjects with AFL0.7 in
former/current smoking group were similar to in never
smoking group (Table 3). The prevalence of AFL0.7 also in-
creased with increasing age (Figs. 2E, 2F and Table 4).
While the prevalence of AFL0.7 in male subjects aged 4th
and 5th decades did not differ between never smoker and
former/current smoker, the prevalence of AFL0.7 in male for-
mer/current smoking subjects aged 6th decade and over was
significantly higher than in never smoking subjects (Ta-
ble 4). In female, the prevalence of AFL0.7 in former/current
smoker was only significantly higher than in never smoking
subjects in the group aged 70 and over (Table 4). The unad-
justed prevalence of AFL0.7 in the total study population was
Inter Med 49: 1489-1499, 2010 DOI: 10.2169/internalmedicine.49.3364
1492
Figure 1. Age-related changes in spirometric parameters in the Takahata population. Graphs
show the relationships between age and FVC (A, males; B, females), FEV1 (C, males; D, females)
and FEF25-75 (E, males; F, females). Open circles indicate never smokers and closed circles,
former/current smokers. Dashed line indicates the regression for never smokers and solid line, the
regression for former/current smokers. The difference in the slopes of the regression lines between
never-smokers and former/current-smokers was compared by analysis of covariance (ANCOVA),
as indicated in the inset. Spirometric parameters declined significantly with age in smokers com-
pared with never-smokers, except for FVC in males.
ANCOVA p=0.1146 ANCOVA p=0.0283
never smoker
former /
current smoker
age age
F
E
F
2
5
-
7
5
ANCOVA p<0.0001
age
F
E
F
2
5
-
7
5
ANCOVA p=0.0229
age
F
E
V
1
ANCOVA p=0.0003
F
E
V
1
ANCOVA p=0.0049
age age
A
B
C D
E
F
40 50 60 70 80 40 50 60 70 80
5
4
3
2
1
8
4
6
2
0
40 50 60 70 80 40 50 60 70 80
40 50 60 70 80 40 50 60 70 80
4
3
2
1
5
4
3
2
1
5
4
3
2
1
F
V
C
F
V
C
5
4
3
2
1
6
(L)
(L)
(L)
(L)
(L/s)
(L/s)
10.6%. The proportions of mild, moderate, severe and very
severe airflow limitation, as categorized according to the
GOLD criteria, were 42.2%, 46.4%, 10.4% and 1.0%, re-
spectively, in this population. In addition, airflow limitation
was more severe in male subjects than in female subjects
(Figs. 2C-2F; chi-square test: likelihood ratio = 84.82, de-
gree of freedom = 4, p<0.001; ordered logistic regression
analysis: likelihood ratio = 84.24, degree of freedom = 1, p<
0.001).
ATS and ERS recommend using 5th percentile LLN value
for diagnosing COPD instead of fixed value of FEV1/FVC <
0.7 (20, 21). The prevalence of airflow limitation defined by
LLN value (AFLLLN) is summarized in Table 5. Overall
prevalence of AFLLLN was 6.4% in this population (male:
8.0%, female: 5.1%). In never smoker, the prevalence of
AFLLLN was 4.7% (male: 4.7%, female: 4.7%). In former/
Inter Med 49: 1489-1499, 2010 DOI: 10.2169/internalmedicine.49.3364
1493
Table 2. Backward Stepwise, Multiple Linear Regression Analysis
of Spirometric Parameters
Coefficient Standard
error
t p r
age -0.0288 0.0025 -11.57 <0.0001 -0.607
HT 0.0562 0.0043 13.16 <0.0001 0.643
FVC
(M)
BW 0.0088 0.0028 -3.17 0.0016 0.318
age -0.02 0.0011 -17.51 <0.0001 -0.653
HT 0.0372 0.0018 20.14 <0.0001 0.656
FVC
(F)
BW - - - 0.725 0.259
age -0.028 0.0022 -12.81 <0.0001 -0.627
HT 0.0374 0.0037 9.98 <0.0001 0.583
FEV
1
(M)
BW -0.005 0.0024 -2.03 0.043 0.315
age -0.0206 0.0010 -20.72 <0.0001 -0.665
HT 0.0273 0.0016 17.02 <0.0001 0.631
FEV
1
(F)
BW - - - 0.820 0.255
age -0.0383 0.0049 -7.85 <0.0001 -0.401
HT 0.0172 0.0074 2.33 0.020 0.279
FEF
25-75
(M)
BW - - - 0.662 0.173
age -0.0348 0.0023 -15.42 <0.0001 -0.499
HT 0.0153 0.0039 3.92 0.0001 0.375
FEF
25-75
(F)
BW 0.0036 0.0024 1.51 0.132 0.176
FVC: forced vital capacity, FEV
1
: forced expiratory volume in 1 second, FEF
25-75
:
forced expiratory flow at 25 75% of FVC, M: males, F: females, HT: height, BW:
weight
Predictive equations obtained from backward stepwise, multiple linear regression
analyses are indicated below
FVC (male) = -3.1740 - (0.0288 age) + (0.0562 HT) - (0.0088 BW), R
2
= 0.538,
standard error of estimate (SEE) = 0.493
FVC (female) = -1.7979 - (0.0201 age) + (0.0372 HT), R
2
= 0.534, SEE = 0.354
FEV
1
(male) = -1.1752 - (0.028 age) + (0.0374 HT) - (0.005 BW), R
2
= 0.501,
SEE = 0.432
FEV
1
(female) = -0.7831 - (0.0206 age) + (0.0273 HT), R
2
= 0.535, SEE = 0.308
FEF
25-75
(male) = 2.3992 - (0.0383 age) + (0.0172 HT), R
2
= 0.177, SEE = 0.971
FEF
25-75
(female) = 1.8131 - (0.0348 age) + (0.0153 HT) + (0.0036 BW), R
2
=
0.262, SEE = 0.699.
current smoker, it was 9.9% (male: 10.1%, female: 8.9%).
In people aged 6th and 7th decades, the prevalence of
AFLLLN in former/current smoker was significantly higher
than in never smoker.
Relative risk of airflow limitation
The relative risk of airflow limitation increased with in-
creasing age, male gender, a history of habitual smoking,
and increasing pack-years of smoking (Table 6).
The effects of age and amount of inhaled cigarette
smoke on pulmonary function
Using predictive values of spirometric measures obtained
from this population, we showed the relationship of age
with %FVC (Figs. 3A, 3B), % FEV1 (Figs. 3C, 3D) and %
FEF25-75 (Figs. 3E, 3F), in never-smokers and former/current-
smokers. In female former/current-smoking subjects, %FVC
(Fig. 3B), %FEV1 (Fig. 3D) and %FEF25-75 (Fig. 3F) de-
clined significantly with age, compared with never smoking
subjects. %FEV1 (Fig. 3C) and %FEF25-75 (Fig. 3E) declined
significantly with age in male former/current-smoking sub-
jects. In contrast, the difference in the rates of decline of %
FVC between male former/current-smokers and male never-
smokers did not reach statistical significance (Fig. 3A).
The association between the declines in %FVC, %FEV1
and %FEF25-75, and amount of inhaled cigarette smoke (pack-
years) were assessed. In male subjects, %FVC, %FEV1 and
%FEF25-75 declined significantly with increasing pack-years
(Figs. 4A, 4C, & 4E). In female subjects, %FVC and
%FEV1 declined significantly with increasing pack-years
(Figs. 4B & 4D), whereas %FEF25-75 did not change with in-
creasing pack-years (Fig. 4F).
Discussion
In this paper, predictive equations for FVC, FEV1 and
FEF25-75 were derived for Japanese never-smokers aged 40
years or older, living in Takahata town, and who participated
in an annual health check. Previously, JRS provides refer-
ence values for FVC and FEV1 (24), and there were good
correlations between the JRS reference values and those de-
rived in the present study. This suggests that the data ob-
tained in the present study was reliable and that the findings
are applicable to the general Japanese population. Using
these predictive values, cross-sectional comparisons of age-
related changes in %FVC, %FEV1 and %FEF25-75 between
Inter Med 49: 1489-1499, 2010 DOI: 10.2169/internalmedicine.49.3364
1494
Figure 2. Prevalence and severity of airflow limitation in the Takahata population. Graphs show
relationships between FEV1/FVC and %FEV1 in males (A) and females (B). The symbols in the
area of less than 70% for FEV1/FVC indicate individuals with airflow limitation. Among these indi
viduals, the severity of airflow limitation was stratified according to the GOLD criteria:mild:
%FEV1 80%, moderate: 50% %FEV1 <80%, severe: 30% %FEV1 <50%, and very severe:
%FEV1 <30%. The proportion of subjects in each category of severity of airflow limitation is indi
cated for males (C) and females (D). The details of severity of airflow limitation in each age group,
for males (E) and females (F) are also indicated. The prevalence of airflow limitation was higher
among male subjects (C) than female subjects (D). The prevalence of airflow limitation also in
creased with increasing age (E & F).
very severe
severe
moderate
mild
no airflow limitation
n =1110 (83.8%)
no airflow limitation
n =1499 (94.2%)
%FEV
1
%FEV
1
F
E
V
1
/

F
V
C
F
E
V
1
/

F
V
C
mild mild moderate moderate severe severe
v
e
r
y

s
e
v
e
r
e
A B
C
D
E F
100
70
40
50 80 100 130 30 80 100 130 50
100
70
40
30
20
10
25
15
5
0
40-49 50-59 60-69 70- 40-49 50-59 60-69 70-
(%)
(%)
(%)
(%) (%)
(%)
severe
n =7 (0.4%)
moderate
n =44 (2.8%)
mild
n =42
(2.6%)
very severe
n =3 (0.2%)
severe
n =25 (1.9%)
moderate
n =99 (7.5%)
mild
n =88
(6.6%)
never smoker
former /
current smoker
never smoker
former /
current smoker
10
8
4
6
2
0
smokers and never-smokers were then performed. In never-
smoking subjects, %FVC, %FEV1 and %FEF25-75 did not
change with age. In contrast, %FEV1 and %FEF25-75, which
are both indicators of airflow limitation, declined signifi-
cantly with age in smoking subjects. Furthermore, %FEV1
declined significantly with increasing number of pack-years
smoked. Therefore, this epidemiological study has strongly
suggested that long-term inhalation of cigarette smoke dam-
ages the airways of smokers.
In this paper, the prevalence of AFLLLN was demonstrated
(Table 5). As shown in Table 4, the prevalence of AFL0.7 in
never smoker increased with increasing age. Thus, using the
fixed value of FEV1/FVC <0.7 is questionable for examining
the existence of airflow limitation in older subjects, because
it ignores the natural age-related change of lung function.
Application of LLN value for examining the existence of
airflow limitation is thought to be able to avoid overdiag-
nosing the presence of airflow limitation in old subjects, and
underdiagnosing the presence of airflow limitation in young
subjects (22, 23). While the prevalence of AFL0.7 in never
Inter Med 49: 1489-1499, 2010 DOI: 10.2169/internalmedicine.49.3364
1495
Table 3. Comparison of the Degree of Airflow Limitation between
Never Smoker and Former/Current Smoker
male female
airflow
limitation
never smoker
former/current
smoker
never smoker
former/current
smoker
no 460 (89.7%) 650 (80.1%) 1374 (94.3%) 125 (92.6%)
mild 27 (5.3%) 61 (7.5%) 38 (2.6%) 4 (3.0%)
moderate 19 (3.7%) 80 (9.9%) 39 (2.7%) 5 (3.7%)
severe 6 (1.2%) 19 (2.3%) 6 (0.4%) 1 (0.7%)
very severe 1 (0.2%) 2 (0.3%) 0 (0%) 0 (0%)
In male subjects, the proportion of air flow limitation was significantly different
between never smoker and former/current smoker (chi-square test, p <0.0001).
In female subjects, the proportion of air flow limitation was not significantly different
between never smoker and former/current smoker (chi-square test, p =0.832).
Table 4. Comparison of the Degree of Airflow Limitation between Never Smoker and
Former/Current Smoker in Each Generation
(A)
male age (year)
40-49 50-59 60-69 70-
airflow
limitation
NS, n (%) SM, n (%) NS, n (%) SM, n (%) NS, n (%) SM, n (%) NS, n (%) SM, n (%)
no 38 (97.4) 102 (96.2) 105 (96.3) 189 (90.4) 167 (89.8) 204 (79.1) 150 (83.8) 155 (64.9)
any 1 (2.6) 4 (3.8) 4 (3.6) 20 (9.6) 19 (10.2) 54 (20.9) 29 (16.2) 84 (35.1)
mild 1 (2.6) 3 (2.8) 2 (1.8) 7 (3.3) 10 (5.4) 22 (8.5) 14 (7.8) 29 (12.1)
moderate 0 (0) 1 (0.9) 2 (1.8) 12 (5.7) 7 (3.8) 26 (10.1) 10 (5.6) 41 (17.2)
severe 0 (0) 0 (0) 0 (0) 1 (0.4) 1 (0.5) 5 (1.9) 5 (2.8) 13 (5.4)
very severe 0 (0) 0 (0) 0 (0) 0 (0) 1 (0.5) 1 (0.4) 0 (0) 1 (0.2)
P value vs NS - 0.8271 - 0.2775 - 0.0343 - 0.0004
(B)
female age (year)
40-49 50-59 60-69 70-
airflow
limitation
NS, n (%) SM, n (%) NS, n (%) SM, n (%) NS, n (%) SM, n (%) NS, n (%) SM, n (%)
no 142 (98.6) 41 (97.6) 357 (95.7) 53 (98.1) 469 (94.2) 25 (86.2) 406 (91.9) 6 (60.0)
any 2 (1.4) 1 (2.4) 16 (4.3) 1 (1.9) 29 (5.8) 4 (13.8) 36 (8.1) 4 (40.0)
mild 2 (1.4) 1 (2.4) 11 (2.9) 1 (1.9) 10 (2.0) 1 (3.4) 15 (3.4) 1 (10.0)
moderate 0 (0) 0 (0) 4 (1.1) 0 (0) 15 (3.0) 2 (6.9) 20 (4.5) 3 (30.0)
severe 0 (0) 0 (0) 1 (0.3) 0 (0) 4 (0.8) 1 (3.4) 1 (0.2) 0 (0)
very severe 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0)
P value vs NS - 0.6534 - 0.8121 - 0.2885 - 0.0020
NS: never smoker, SM: former/current smoker
Statistical analysis: chi-square test
smoking subjects aged 6th and 7th decades was over 10%,
the prevalence of AFLLLN in never smoker was less than 8%
(Tables 4 & 5). Importantly, the prevalence of AFLLLN in
smoking subjects aged 6th and 7th decades was significantly
higher than in never smokers (Table 5). In addition, three fe-
male subjects in 4th and 5th decade were underdiagnosed as
no airflow limitation by the criteria using FEV1/FVC <0.7
(Tables 4 & 5). Thus, utilization of LLN as definition for
airflow limitation should be strongly suggested instead of
fixed value of FEV1/FVC.
Fukuchi et al have estimated the prevalence of airflow
limitation in Japan, based on data from the Nippon COPD
Inter Med 49: 1489-1499, 2010 DOI: 10.2169/internalmedicine.49.3364
1496
Table 5. Comparison of the Prevalence of Airflow Limitation Defined by Lower Limit of Normal
Value between Never Smoker and Former/Current Smoker in Each Generation
(A) Male
age (year)
male 40-49 50-59 60-69 70-
NS SM NS SM NS SM NS SM
FEV1% aLLN 38 (97.4) 104 (98.1) 107 (98.2) 198 (94.7) 178 (95.7) 232 (89.9) 166 (92.7) 196 (82.0)
FEV1% bLLN 1 (2.6) 2 (1.9) 2 (1.8) 11 (5.3) 8 (4.3) 26 (10.1) 13 (7.3) 43 (18.0)
P value vs NS 0.7994 0.1428 0.024 0.0014
(B) female
age (year)
female 40-49 50-59 60-69 70-
NS SM NS SM NS SM NS SM
FEV1% aLLN 142 (98.6) 39 (92.9) 356 (95.4) 52 (96.3) 473 (95.0) 25 (86.2) 417 (94.3) 7 (70.0)
FEV1% bLLN 2 (1.4) 3 (7.1) 17 (4.6) 2 (3.7) 25 (5.0) 4 (13.8) 25 (5.7) 3 (30.0)
P value vs NS 0.0425 0.7761 0.0440 0.0016
Predicted equations of FEV1/FVC (FEV1%) obtained from never smokers in this study are indicated below;
Predicted FEV1% (male) = 85.76 0.112 age, standard error of estimate (SEE) = 7.94, Predicted FEV1%
(female) = 88.77 0.143 age, SEE = 6.38. NS: never smoker, SM: former/current smoker, aLLN: above
lower limit of normal, bLLN: below lower limit of normal, Statistical analysis: chi-square test
Table 6. Relative Risk of Airflow Limitation
(A) airflow limitation defined by FEV
1
/FVC < 0.7
Subject
characteristic (n)
Relative risk 95% C.I. p value
Age, years
40 - 49 (331) 1 (reference) - -
50 - 59 (745) 2.28 1.08 4.80 0.025
60 - 69 (971) 4.52 2.26 9.17 < 0.0001
70 - (870) 7.28 3.62 14.65 < 0.0001
Gender
Female (1592) 1 (reference) -
Male (1325) 2.78 2.20 3.50 < 0.0001
Smoking status
Never (1970) 1 (reference) - -
Current (544) 2.74 2.16 3.48 < 0.0001
Former (403) 2.48 1.89 3.25 < 0.0001
Pack-years *
0 (1970) 1 (reference) - -
> 0, <25 (203) 1.86 1.25 2.75 0.0023
25, <50 (299) 3.34 2.57 4.35 < 0.0001
50 (104) 3.48 2.39 5.08 < 0.0001
(B) airflow limitation defined by lower limit of normal value
Subject
characteristic (n)
Relative risk 95% C.I. p value
Age, years
40 - 49 (331) 1 (reference) - -
50 - 59 (745) 1.78 0.83 3.81 0.138
60 - 69 (971) 2.68 1.30 5.54 0.0048
70 - (870) 3.99 1.96 8.16 < 0.0001
Gender
Female (1592) 1 (reference) -
Male (1325) 1.57 1.19 2.08 0.0014
Smoking status
Never (1970) 1 (reference) - -
Current (544) 2.38 1.74 3.23 < 0.0001
Former (403) 1.73 1.18 2.54 0.0047
Pack-years *
0 (1970) 1 (reference) - -
> 0, <25 (203) 1.98 1.23 3.18 0.0044
25, <50 (299) 2.48 1.66 3.50 < 0.0001
50 (104) 3.05 1.84 5.08 < 0.0001
Data was analyzed using chi-square tests.
* Because of incomplete questionnaire report regarding the pack years of smoking, 344 subjects were excluded from the analyses.
Epidemiology study (NICE) (14). They randomly contacted
individuals from the telephone directory, and invited them to
attend the clinic for pulmonary function testing. A total of
2,343 healthy individuals were enrolled, and the prevalence
Inter Med 49: 1489-1499, 2010 DOI: 10.2169/internalmedicine.49.3364
1497
Figure 3. Age-related changes in standardized spirometric parameters in the Takahata
population. Graphs show the relationships between age and %FVC (A: males, B: females), %FEV1
(C: males, D: females) and %FEF25-75 (E: males, F: females), following standardization of the spiro-
metric measurements using predicted values. Open circles indicate never smokers and closed cir-
cles, former/current smokers. Dashed line indicates the regression for never smokers and solid line
the regression for former/current smokers. The difference in the slopes of the regression lines be-
tween never-smokers and former/current-smokers were compared by analysis of covariance (AN-
COVA), as indicated in the inset. Standardized spirometric parameters declined significantly with
age, in smokers compared with never-smokers, except for FVC in males.
130
100
70
40
130
100
70
40
160
130
100
70
40
%
F
E
V
1
%
F
V
C
never smoker former / current smoker
%
F
V
C
%
F
E
V
1
%
F
E
F
2
5
-
7
5
%
F
E
F
2
5
-
7
5
A
B
C D
E F
130
100
70
40
300
200
100
300
200
100
0
ANCOVA p=0.2338
age
40 50 60 70 80
ANCOVA p=0.0055
ANCOVA p=0.0001 ANCOVA p=0.0004
ANCOVA p<0.0001 ANCOVA p=0.0058
(%)
(%)
(%) (%)
(%) (%)
age
40 50 60 70 80
age
40 50 60 70 80
age
40 50 60 70 80
age
40 50 60 70 80
age
40 50 60 70 80
of airflow limitation in the Japanese population was esti-
mated at 10.9%. It was also reported that 56% of subjects
had mild airflow limitation, while 38% had moderate, 5%
severe, and 1% very severe airflow limitation. Airflow limi-
tation was significantly more prevalent among males than
females (16.4% vs. 5.0%; p<0.001), and among older sub-
jects (3.5% in those aged 40-49 years vs. 24.4% in those
aged > 70 years; p<0.001). The present study showed simi-
lar results with respect to the overall prevalence of AFL0.7,
and the greater prevalence of AFL0.7 among males and older
subjects.
In contrast, the severity of airflow limitation seemed to
differ between the NICE study and the present study. In the
present study, the proportion of subjects with moderate air-
flow limitation appeared to be greater than that in the NICE
study (Fig. 2). The smoking rate in the Japanese population
in 2005 was approximately 39.3% among males and 11.3%
among females (25, 26), while the smoking rates in Taka-
hata were 34.1% among males and 5.8% among females.
Thus, it is unlikely that individuals living in Takahata are as
Inter Med 49: 1489-1499, 2010 DOI: 10.2169/internalmedicine.49.3364
1498
Figure 4. The effect of cigarette smoking on standardized spirometric parameters in subjects in
the Takahata population. Graphs show the relationships between standardized spirometric pa-
rameters and pack-years of smoking. %FVC (A: males and B: females) and %FEV1 (C: males and
D: females) were significantly correlated with pack-years. %FEF25-75 was significantly correlated
with pack-years in male subjects (E), but not in female subjects (F).
r=-0.11423
p=0.0082
r=-0.16068
p=0.0002
r=-0.15623
p=0.0003
A
C
E
0 20 40 60 80 100120
140
120
100
80
40
60
0 20 40 60 80 100120
140
120
100
80
40
60
20
0 20 40 60 80 100120
200
100
0
r=-0.35997
p=0.0020
r=-0.28921
p=0.0144
r=0.038893
p=0.7474
B
D
F
0 10 20 30 40
130
100
70
0 10 20 30 40
130
100
70
0 10 20 30 40
180
140
100
60
20
%
F
V
C
%
F
E
V
1
%
F
E
F
2
5
-
7
5
(%)
(%)
(%)
exposed to cigarette smoke compared with populations liv-
ing in other areas of Japan. Takahata town has an aging
population, and therefore the age-distribution in the present
study also differed from that of the NICE study (Table 3).
Since the prevalence of airflow limitation is higher in older
persons, this is a possible explanation for the differences in
the severity of airflow limitation between subjects in the
NICE study and those in the present study. This explanation
is partly supported by the report by Omori et al (11). They
demonstrated the prevalence of AFL0.7 on hospital-based
medical check-up in Japanese subjects aged 40-69, and dem-
onstrated the severity of airflow limitation between smokers
and nonsmokers according to the age decade of the subjects,
whereas NICE study lacked this analysis. The proportion of
moderate airflow limitation in smokers increased with in-
creasing decades of subjects in their paper, although the pro-
portion of mild airflow limitation in smokers was still the
highest in the subjects aged 60-69 (11).
The annual rate of decline in FEV1 in the Japanese popu-
lation can be estimated from the equation for FEV1 derived
in the present study. In never-smokers, the estimated annual
rate of decline in FEV1 was 28 mL for males and 20 mL for
females, while in male former/current smokers it was 37
mL. Fletcher and Peto investigated the effect of cigarette
smoking on the decline in FEV1 in male workers in Lon-
don (15). Recently, analysis of data from the Framingham
Offspring Cohort also demonstrated that the decline in FEV1
was more rapid in smokers than in never-smokers (16). To
date, no epidemiological analysis has been performed on the
impact of cigarette smoking on pulmonary function in the
Inter Med 49: 1489-1499, 2010 DOI: 10.2169/internalmedicine.49.3364
1499
Japanese population. This is the first study to demonstrate
the decline in pulmonary function (FEV1, as well as FVC
and FEF25-75) in Japanese subjects, by statistical analysis of
data from spirometric measurements performed as part of a
community-based annual health check.
There are a couple of limitations to this study. Self-
reported questionnaires may not always provide accurate
data, due to misunderstanding of the questionnaire or failure
to accurately recall smoking histories. The questionnaire
asked the existence of history of pulmonary diseases and
current therapy for pulmonary disease, but did not ask the
details. Thus, the prevalence of bronchial asthma or medica-
tion for the disease was not revealed in this questionnaire.
In addition, because the cigarette smoking rate in Japanese
females was low, the number of female subjects enrolled in
this study was only 135 (Table 1). However, despite these
limitations, the study clearly demonstrated the impact of
cigarette smoking on pulmonary function in a Japanese
population.
In conclusion, cigarette smoking enhances the prevalence
of airflow limitation. Cigarette smoking was also shown to
be a factor that correlated with the decline in %FEV1 in
both males and females. The present study clearly demon-
strated that long-term cigarette smoking increases the risk of
airflow limitation in healthy individuals. The findings from
this epidemiological study have demonstrated that early ces-
sation of smoking is necessary in order to prevent the devel-
opment of COPD among smokers.
Acknowledgement
We thank Taiko Aita, and Eiji Tsuchida for their excellent
technical assistance.
Funding: This study was supported by a grant-in-aid from the
Global COE program of the Japan Society for the Promotion of
Science and grants-in-aid for Scientific Research from the Minis-
try of Education, Culture, Sports, Science and Technology, Japan
(18590835, 18790530, 19590880, and 20590892).
References
1. Global Initiative for Chronic Obstructive Lung Disease. Global
strategy for the diagnosis, management and prevention of chronic
obstructive pulmonary disease. Revised 2006. Available at http://
www.goldcopd.org/
2. Pauwels RA, Rabe KF. Burden and clinical features of chronic ob-
structive pulmonary disease (COPD). Lancet 364: 613-620, 2004.
3. Rabe KF, Beghe B, Luppi F, Fabbri LM. Update in chronic ob-
structive pulmonary disease 2006. Am J Respir Crit Care Med
175: 1222-1232, 2007.
4. Pierce R. Spirometry: an essential clinical measurement. Aust Fam
Physician 34: 535-539, 2005.
5. Morris JF. Spirometry in the evaluation of pulmonary function.
West J Med 125: 110-118, 1976.
6. Wang IM, Stepaniants S, Boie Y, et al. Gene expression profiling
in patients with chronic obstructive pulmonary disease and lung
cancer. Am J Respir Crit Care Med 177: 402-411, 2008.
7. Anthonisen NR, Wright EC, Hodgkin JE. Prognosis in chronic ob-
structive pulmonary disease. Am Rev Respir Dis 133: 14-20,
1986.
8. Fukahori S, Matsuse H, Takamura N, et al. Prevalence of chronic
obstructive pulmonary diseases in general clinics in terms of
FEV1/FVC. Int J Clin Pract 63: 269-274, 2009.
9. Kamimura T, Koga T, Oshita Y, et al. Prevalence of previously un-
diagnosed airflow limitation in patients who underwent preopera-
tive pulmonary function test. Kurume Med J 53: 53-57, 2006.
10. Kojima S, Sakakibara H, Motani S, et al. Effects of smoking and
age on chronic obstructive pulmonary disease in Japan. J Epide-
miol 15: 113-117, 2005.
11. Omori H, Nonami Y, Mihara S, Marubayashi T, Morimoto Y, Ai-
zawa H. Prevalence of airflow limitation on medical check-up in
Japanese subjects. J UOEH 29: 209-219, 2007.
12. Takemura H, Hida W, Sasaki T, Sugawara T, Sen T. Prevalence of
chronic obstructive pulmonary disease in Japanese people on
medical check-up. Tohoku J Exp Med 207: 41-50, 2005.
13. Tan WC, Ng TP. COPD in Asia: where East meets West. Chest
133: 517-527, 2008.
14. Fukuchi Y, Nishimura M, Ichinose M, et al. COPD in Japan: the
Nippon COPD epidemiology study. Respirology 9: 458-465, 2004.
15. Fletcher C, Peto R. The natural history of chronic airflow obstruc-
tion. Br Med J 1: 1645-1648, 1977.
16. Kohansal R, Martinez-Camblor P, Agusti A, Buist AS, Mannino
DM, Soriano JB. The natural history of chronic airflow obstruc-
tion revisited: an analysis of the Framingham offspring cohort.
Am J Respir Crit Care Med 180: 3-10, 2009.
17. Kojima S, Sakakibara H, Motani S, et al. Incidence of chronic ob-
structive pulmonary disease, and the relationship between age and
smoking in a Japanese population. J Epidemiol 17: 54-60, 2007.
18. Konta T, Hao Z, Abiko H, et al. Prevalence and risk factor analy-
sis of microalbuminuria in Japanese general population: the Taka-
hata study. Kidney Int 70: 751-756, 2006.
19. The Committee of Pulmonary Physiology JRS. Guidelines for pul-
monary function tests: Spirometry, flow-volume curve, diffusion
capacity of the lung. Tokyo, 2004 (in Japanese).
20. Pellegrino R, Brusasco V, Viegi G, et al. Definition of COPD:
based on evidence or opinion? Eur Respir J 31: 681-682, 2008.
21. Pellegrino R, Viegi G, Brusasco V, et al. Interpretative strategies
for lung function tests. Eur Respir J 26: 948-968, 2005.
22. Cerveri I, Corsico AG, Accordini S, et al. Underestimation of air-
flow obstruction among young adults using FEV1/FVC <70% as a
fixed cut-off: a longitudinal evaluation of clinical and functional
outcomes. Thorax 63: 1040-1045, 2008.
23. Swanney MP, Ruppel G, Enright PL, et al. Using the lower limit
of normal for the FEV1/FVC ratio reduces the misclassification of
airway obstruction. Thorax 63: 1046-1051, 2008.
24. Japanese Respiratory Society. Available at http://www.jrs.or.jp/
quicklink/glsm/guideline/nopass_pdf/spirogram.pdf.
25. Katanoda K, Marugame T, Saika K, et al. Population attributable
fraction of mortality associated with tobacco smoking in Japan: a
pooled analysis of three large-scale cohort studies. J Epidemiol
18: 251-264, 2008.
26. Ministry of Health Labour and Welfare. Measures for national
health promotion. Available at http://www.mhlw.go.jp/english/wp/
wp-hw2/part2/p3_0024.pdf.
2010 The Japanese Society of Internal Medicine
http://www.naika.or.jp/imindex.html