Beverage Consumption andthe Prevalence of Tooth Loss in Pregnant Japanese Women : The Osaka Maternal and Child Health Study

Abstract

Recently, there has been growing interest in the impact of beverage consumption on dental health because changes have occurred in the types and quantities of beverage consumed. This cross-sectional study investigated the relationship between consumption frequencies of various beverages and the prevalence of tooth loss among young adult women. Study subjects were 1002 pregnant Japanese women. Tooth loss was defined as previous extraction of one or more teeth. Dietary habits were evaluated by a validated dietary history questionnaire. Logistic regression analysis was used to estimate the odds ratios and their 95% confidence intervals of tooth loss. Adjustment was made for age, gestation, parity, cigarette smoking, passive smoking at home and at work, family income, education, changes in diet in the previous one month, season when data were collected and body mass index. Of the 1002 subjects, 256 women had lost one or more teeth. Coffee consumption was independently associated with an increased prevalence of tooth loss. When subjects were divided according to consumption of coffee with or without sugar, an increased prevalence of tooth loss was found only in subjects who consumed coffee without addition of sugar. Compared with the lowest consumption of green tea, the intermediate but not the highest consumption of green tea was associated with an increased prevalence of tooth loss. There was no measurable association of intake of milk, black tea, cola, or 100% fruit juice with the prevalence of tooth loss. The findings suggest that coffee consumption might be associated with an increased prevalence of tooth loss among young adult women. 近年，嗜好飲料の種類とその摂取量が増加しており，これらの飲料の摂取が歯科疾患に与える影響について注目されている．本横断研究では，日本人の若年成人女性における各嗜好飲料の摂取頻度と歯牙喪失有症率との関連について検討した．研究対象者は1002 名の妊婦である．過去に永久歯を1本以上抜歯した経験がある場合，歯牙喪失ありと定義した．妥当性の検証された食事歴法質問調査票を用いて食習慣を評価した．解析には多変量ロジスティック回帰分析を用いた．年齢，妊娠週，子数，喫煙，家庭及び職場での受動喫煙，家計の年収，教育歴，過去1ヶ月の食事変容，回答時の季節及びbody mass indexを補正した．1002名の研究対象者のち，256 名で抜歯経験があった．コーヒー摂取は有意に歯牙喪失の有症率の高まりと関連していた．コーヒー摂取の際の砂糖使用の有無による解析では，砂糖を使用していない群でのみコーヒー摂取と歯牙喪失の有意な正の関連を認めた．緑茶摂取については，第1三分位に比較して，第2三分位でのみ歯牙喪失と有意な正の関連を認めた．牛乳，紅茶，コーラ及び100％フルーツジュースの摂取と歯牙喪失との間に関連はなかった．本研究結果より，若年成人女性においてコーヒー摂取は歯牙喪失の有症率の高まりと関連があるのかもしれない．

Full text

Page 1

Original Article
Beverage Consumption and the Prevalence of Tooth Loss in Pregnant
Japanese Women : The Osaka Maternal and Child Health Study
Keiko TANAKA1), Yoshihiro MIYAKE1), Satoshi SASAKI2), Yukihiro OHYA3),
Ichiro MATSUNAGA4), Toshiaki YOSHIDA4), Yoshio HIROTA5)and Hajime ODA3)
for the Osaka Maternal and Child Health Study Group＊
1)Department of Public Health, Faculty of Medicine, Fukuoka University
2)Department of Social and Preventive Epidemiology, School of
Public Health, The University of Tokyo
3)Division of Allergy, Department of Medical Specialties,
National Center for Child Health and Development
4)Osaka Prefectural Institute of Public Health
5)Department of Public Health, Osaka City University Graduate School of Medicine
＊Other members of the Osaka Maternal and Child Health Study Group are listed in the Appendix
Abstract
health because changes have occurred in the types and quantities of beverage consumed.
cross-sectional study investigated the relationship between consumption frequencies of various
beverages and the prevalence of tooth loss among young adult women.
pregnant Japanese women.Tooth loss was defined as previous extraction of one or more teeth.
Dietary habits were evaluated by a validated dietary history questionnaire.
analysis was used to estimate the odds ratios and their 95% confidence intervals of tooth loss.
Adjustment was made for age, gestation, parity, cigarette smoking, passive smoking at home and at
work, family income, education, changes in diet in the previous one month, season when data were
collected and body mass index.Of the 1002 subjects, 256 women had lost one or more teeth.
consumption was independently associated with an increased prevalence of tooth loss.
subjects were divided according to consumption of coffee with or without sugar, an increased
prevalence of tooth loss was found only in subjects who consumed coffee without addition of sugar.
Compared with the lowest consumption of green tea, the intermediate but not the highest consumption
of green tea was associated with an increased prevalence of tooth loss.
association of intake of milk, black tea, cola, or 100% fruit juice with the prevalence of tooth loss.
findings suggest that coffee consumption might be associated with an increased prevalence of tooth
loss among young adult women.
Recently, there has been growing interest in the impact of beverage consumption on dental
This
Study subjects were 1002
Logistic regression
Coffee
When
There was no measurable
The
Key words : coffee, cross-sectional studies, Japan, pregnant women, tooth loss
Introduction
Two major roles of nutrition in oral health are
clear : one systemic and the other local and
dietary1).Systemic nutrition affects the develop-
ment, growth, and maintenance of the oral tissues
and organs and promotes immunity and resist-
ance to infection. Meanwhile, the oral cavity is
subject to local or topical effects from whatever
enters the mouth ; therefore food choices and
dietary patterns are also essential determinants of
oral health.
Recently, there has been growing interest in the
80
Fukuoka Acta Med.
99(４)：80―89，2008
Correspondence to : Keiko Tanaka, DDS, PhD, Department of Public
Health, Faculty of Medicine, Fukuoka University, 7-45-1 Nanakuma,
Jonan-ku, Fukuoka 814-0180, Japan
Phone: + 81-92-801-1011 (ext. 3315)
Fax : + 81-92-863-8892
E-mail : k-tanaka@fukuoka-u.ac.jp

Page 2

impact of beverage consumption on dental health
because changes have occurred in the types and
quantities of beverages consumed among children
and adolescents in the US2).
national nutrition survey in Japan3), intake of
According to a
seasonings and beverages increased from 1980 to
2000 (109.4 g/day to 182.3 g/day), although data
on changes in types or quantities of beverages
alone are not available.
Several epidemiological studies have examined
the relationship between beverages, especially
sugared ones, and oral health4)〜12).However,
the possible effect of sugared beverages on dental
health remains inconclusive.Marshall et al.4)
suggested that consumption of sugared soda pop
and sugared beverages from powder, but not milk,
100% juice and water in early childhood was
associated with increased caries risk.A
cross-sectional study in Italian 7-, 9-, and
10-year-old schoolchildren showed a significant
association of sweet drink intake with dental
caries5). In a cross-sectional study among Span-
ish schoolchildren, there was no association
between consumption of sugared soft drinks,
sugar-free soft drinks, or fruit juice and dental
caries6).Among 14-year-old children in Eng-
land, tea drinkers had a lower number of teeth
with caries than coffee drinkers and this effect
was independent of the addition of sugar7).
Previous studies have shown that the addition of
sugar in coffee or tea was associated with
increased caries or tooth loss among adults8)〜11).
On the other hand, among Swedish 68-year-old
men, consumption of coffee was inversely associ-
ated with the number of remaining teeth irrespec-
tive of the use of sugar12).The impact of
beverage consumption on dental health might not
be explained only by the presence of sugars.
The aim of this study was to evaluate the
relationship between consumption frequencies of
various selected beverages and the prevalence of
tooth loss in young Japanese women by using
baseline data from the Osaka Maternal and Child
Health Study (OMCHS).
Methods
Study population
The OMCHS is an ongoing prospective cohort
study that investigates preventive factors and
risk factors for maternal and child health, such as
allergic disorders.
described previously13)14).
Details of the OMCHS were
Eligible women for
the OMCHS were those who live in Neyagawa
City, which is one of the 43 municipalities in Osaka
Prefecture, a metropolis in Japan with a total
population of approximately 8.8 million. Among
the 3639 eligible pregnant women, 627 subjects
(17.2%) participated in the OMCHS between
November 2001 and March 2003.Eight pregnant
women who did not live in Neyagawa City but
who had become aware of the present study at an
obstetric clinic before August 2002 decided by
themselves to participate.Seventy-seven
women received explanations of the OMCHS from
public health nurses in 6 other municipalities from
August 2002 to March 2003 and were enrolled in
the OMCHS.From October 2002 to March 2003,
290 women were recruited from a university
hospital and 3 obstetric hospitals in 3 other
municipalities; it had been recommended by an
obstetrician that these women participate in the
OMCHS.Finally, a total of 1002 pregnant
women gave their fully informed consent in
writing and completed the two baseline question-
naires.The OMCHS was approved by the ethics
committees of the Osaka City University School of
Medicine and the Osaka Prefectural Institute of
Public Health.
Questionnaire
At baseline, each participant filled out two
self-administered questionnaires.The partici-
pants mailed the questionnaires to the data
management center. The questionnaires were
checked by research technicians, and missing or
illogical answers were completed by telephone
interview.
A validated self-administered diet history
Coffee Intake and Tooth Loss in Japan 81

Page 3

questionnaire was used to assess dietary habits
over the previous one month.The structure and
validity of this questionnaire have been reported
elsewhere15)16). Subjects were asked to state
how frequently they consumed each beverage
item, based on 8 categories : 6 + times /day, 4-5
times/day, 2-3 times/day, 1 time/day, 4-6 times/
week, 2-3 times/week, 1 time/week, and ＜
1time/week (except for milk, for which categories
were 2 + times /day, 1 time/day, 4-6 times/week,
2-3 times/week, 1 time/week, 2-3 times/month, 1
time/month, and ＜ 1 time/month).
The second self-administered questionnaire
inquired about age, gestation, parity, smoking
habits, passive smoking at home and at work,
family income, education, height, weight, changes
in diet in the previous one month, experience of
extraction of one or more permanent teeth
excluding third molars, and the number of
remaining teeth. Tooth loss was defined as
present via positive answer to the question,
?Have you ever experienced the extraction of
permanent teeth, excluding third molars?" Body
mass index was calculated by self-reported body
weight in kg by the square of the self-reported
height in m.
Statistical methods
The consumption of beverages was categorized
into three levels in order to represent the tertiles
as closely as possible. Potential confounding
factors, that is, age, gestation, parity, cigarette
smoking, passive smoking at home and at work,
family income, education, changes in diet in the
previous one month, season when data were
collected, and body mass index were selected.
Age was classified into 2 categories (< 31 and 31 +
years) ; gestation into 2 (< 18 and 18 + weeks);
parity into 2 (0 and 1 +) ; cigarette smoking into 3
(never, former, and current) ; passive smoking at
home into 3 (never, former, and current); passive
smoking at work into 3 (never, former, and
current) ; family income into 3 (< 4,000,000, 4,000,
000-5,999,999, and 6,000,000 + JPY/year) ; educa-
tion into 3 (< 13, 13-14, and 15 + years) ; changes
in diet in the previous one month into 3 (none or
seldom, slight, and substantial) ; and season when
data were collected into 4 (spring, summer, fall,
and winter).Body mass index was used as a
continuous variable.
Logistic regression analysis was used to
evaluate the crude odds ratios (ORs) and their
95% confidence intervals (CIs) for tooth loss
relative to intake of beverages.Multiple logistic
regression analysis was used to control for the
potential confounding effects of selected factors.
Analysis of covariance was employed to calculate
adjusted means of lost teeth according to bever-
age consumption with allowance for confounding
factors.The trend of the association was asses-
sed by a logistic regression model with assign-
ment of ordinal scores to levels of the independent
variables.Two-sided p-values less than 0.05
were considered statistically significant. Analy-
ses were carried out using the SAS software
version 9.1 (SAS Institute, Inc., Cary, NC).
Results
About half of the women were enrolled by the
18th week of gestation, and had a parity of one or
more. Slight or substantial changes in diet in the
previous one month were experienced by 702
pregnant women because of nausea gravidarum
(585 women), maternal or fetal health (107
women), and other reasons (10 women) (Table 1).
Among the 1002 enrolled women (mean age 29.8),
about one-fourth of subjects (25.5%) had lost one
or more teeth (Table 2). Table 3 provides tertile
categories of the consumption of selected bever-
ages and the prevalence values.Tooth loss was
the most frequent in the high consumption of
coffee.
Table 4 shows crude and adjusted ORs for tooth
loss (extraction of 1 + teeth) and geometric means
of lost teeth in relation to the consumption of
selected beverages.Compared with coffee con-
sumption in the lowest tertile, its intake in the
highest level was significantly associated with an
K. Tanaka et al. 82

Page 4

increased prevalence of tooth loss, showing a clear
dose-response relationship.Adjustment for
selected confounders under investigation attenu-
ated this positive association, but the association
remained significant (adjusted OR = 1.52, 95% CI
= 1.04-2.22, P for linear trend = 0.03).However,
coffee consumption was unrelated to the number
of lost teeth; the adjusted geometric means of lost
teeth were 2.2, 2.4, and 2.4 teeth among pregnant
women in the lowest, intermediate, and highest
tertile of coffee consumption, respectively.Com-
pared with the lowest consumption of green tea,
the intermediate but not the highest consumption
of green tea was associated with an increased
prevalence of tooth loss.This association
changed little after adjustment for confounders
(adjusted OR = 1.53, 95% CI = 1.10-2.13).There
was no significant association between green tea
consumption and the number of lost teeth.
There was no measurable association of intake of
milk, black tea, cola, or 100% fruit juice with the
prevalence and the number of tooth loss.
When subjects were divided according to
consumption of coffee with or without sugar, an
increased prevalence of tooth loss was only
apparent in subjects without the addition of sugar,
after adjustment for confounders under study
(Table 5).We also found a clear dose-response
Coffee Intake and Tooth Loss in Japan83
Table 3
Tertile categories of the consumption of
selected beverages among the 1002 pregnant
women, OMCHS, Japan
Variable
Consumption category
LowIntermediateHigh
Milk
Frequency
Prevalence
Coffee
Frequency
Prevalence
Green tea
Frequency
Prevalence
Black tea
Frequency
Prevalence
Cola
Frequency
Prevalence
100% fruit juice
Frequency
Prevalence
< 4 times/week
104/363 (28.7%)
4 − 7 times/week
130/542 (24.0%)
2 + times/day
22/97 (22.7%)
< 1 time/week
85/405 (21.0%)
1 − 6 times/week
79/318 (24.8%)
1 + time/day
92/279 (33.0%)
< 4 times/day
112/507 (22.1%)
4 − 5 times/day
95/311 (30.6%)
6 + times/day
49/184 (26.6%)
< 1 time/week
110/456 (24.1%)
1 − 7 times/week
87/326 (26.7%)
2 + times/day
59/220 (26.8%)
< 1 time/week
140/563 (24.9%)
1 time/week
42/183 (23.0%)
2 + times/week
74/256 (28.9%)
< 1 time/week
102/440 (23.2%)
1 time/week
57/205 (27.8%)
2 + times/week
97/357 (27.2%)
Table 1
Distribution of the baseline characteristics of
the 1002 pregnant women, OMCHS, Japan
Number (%) or mean (SD)
Age (years)
< 31
31 +
Gestation (weeks)
< 18
18 +
Parity
0
1 +
Cigarette smoking
Never
Former
Current
Passive smoking at home
Never
Former
Current
Passive smoking at work
Never
Former
Current
Family income (yen/year)
< 4,000,000
4,000,000 − 5,999,999
6,000,000 +
Education (years)
< 13
13 − 14
15 +
Change in diet in the previous one month
None or seldom
Slight
Substantial
Season when data were collected
Spring
Summer
Fall
Winter
Body mass index (kg/m²)
Factor
580 (57.9)
422 (42.1)
508 (50.7)
494 (49.3)
489 (48.8)
513 (51.2)
697 (69.6)
121 (12.1)
184 (18.4)
284 (28.3)
224 (22.4)
494 (49.3)
344 (34.3)
538 (53.7)
120 (12.0)
301 (30.0)
403 (40.2)
298 (29.7)
323 (32.2)
413 (41.2)
266 (26.6)
300 (29.9)
435 (43.4)
267 (26.7)
318 (31.7)
162 (16.2)
223 (22.3)
299 (29.8)
21.4 (2.8)
Table 2
Distribution of tooth loss among the 1002
pregnant women, OMCHS, Japan
Number of teeth lostNumber of subjects (%)
0
1
2
3
4
746 (74.5)
85 (8.5)
57 (5.7)
29 (2.9)
43 (4.3)
42 (4.2) 5 +

Page 5

relationship between consumption of coffee with-
out sugar and the prevalence of tooth loss.A
significant interaction was observed in the
association of coffee consumption in the in-
termediate but not the highest level with the
prevalence of tooth loss between the intake of
coffee with and without sugar (P = 0.04 and 0.18
for homogeneity of OR for the intermediate and
the highest consumption, respectively).
Discussion
We found a positive association between coffee
consumption and the prevalence of tooth loss,
although there was no association between coffee
intake and the number of lost teeth. We have no
immediate explanation for this observed associa-
tion.Some inflammatory mediators would elicit
clinical signs of inflammation and connective
tissue destruction, as well as attachment loss with
pocketing and bone loss17).A cross-sectional
study in Greece showed that compared with
coffee nondrinkers, subjects who consumed 200
ml or more coffee per day had higher concentra-
tions of serum interleukin 6, C-reactive protein,
and tumor necrosis factor α18).Alternatively,
coffee consumption may be regarded as a
reflector of an unhealthy lifestyle.In a survey of
a population sample of 25 to 60-year-old Au-
strians, coffee intake was positively associated
with the consumption of main fat sources (meat
and sausages), wine, and lemonades and inversely
with intake of fruit and milk and performance of
physical activity19).
We were surprised to observe that an increased
prevalence of tooth loss was only found in subjects
who did not use sugar in the current study.The
findings are in partial agreement with previous
studies that indicated no relationship between
sugar in coffee and the number of remaining
teeth12)and caries experience7), but at variance
with other studies that found a positive associa-
tion of the use of sugar in coffee with tooth loss9)11)
and root caries10).According to a systematic
review regarding the relationship between sugar
consumption and dental caries experience, sugar
consumption is a moderate-to-weak risk factor
for caries in people who expose to fluoride such as
toothpaste and professional applications20).The
increase of fluoride in the oral environment might
reduce the detrimental effect of sugar on oral
diseases, although both the total amount and the
frequency of sugar are important elements in the
K. Tanaka et al.84
Table 4
Odds ratios for tooth loss (extraction of 1 +
teeth) and geometric means of lost teeth
according to levels of beverage consumption
among the 1002 pregnant women, OMCHS,
Japan
BeverageCrude odds ratio
(95%CI)
Adjusted odds ratio
(95%CI)
1)
Adjusted means
of lost teeth
(95%CI)
1)
Milk
Low
Intermediate
High
P for linear trend
Coffee
Low
Intermediate
High
P for linear trend
Green tea
Low
Intermediate
High
P for linear trend
Black tea
Low
Intermediate
High
P for linear trend
Cola
Low
Intermediate
High
P for linear trend
100% fruit juice
Low
Intermediate
High
P for linear trend
1.00
0.79 (0.58 − 1.06)
0.73 (0.42 − 1.22)
0.10
1.00
0.79 (0.58 − 1.08)
0.79 (0.45 − 1.34)
0.17
2.3 (2.0 − 2.6)
2.4 (2.1 − 2.7)
2.2 (1.8 − 3.0)
0.91
1.00
1.24 (0.88 − 1.77)
1.85 (1.31 − 2.62)
0.0005
1.00
1.25 (0.86 − 1.80)
1.52 (1.04 − 2.22)
0.03
2.2 (1.8 − 2.5)
2.4 (2.0 − 2.8)
2.4 (2.1 − 2.8)
0.35
1.00
1.55 (1.13 − 2.14)
1.28 (0.86 − 1.88)
0.07
1.00
1.53 (1.10 − 2.13)
1.23 (0.81 − 1.84)
0.11
2.3 (2.0 − 2.6)
2.3 (2.0 − 2.7)
2.5 (2.0 − 3.1)
0.52
1.00
1.15 (0.83 − 1.59)
1.15 (0.80 − 1.66)
0.39
1.00
1.30 (0.92 − 1.83)
1.20 (0.81 − 1.76)
0.25
2.3 (2.0 − 2.6)
2.3 (1.9 − 2.7)
2.5 (2.0 − 3.0)
0.51
1.00
0.90 (0.60 − 1.33)
1.23 (0.88 − 1.71)
0.29
1.00
0.89 (0.59 − 1.34)
1.16 (0.81 − 1.64)
0.49
2.2 (2.0 − 2.5)
2.0 (1.6 − 2.5)
2.7 (2.3 − 3.2)
0.12
1.00
1.28 (0.87 − 1.86)
1.24 (0.90 − 1.71)
0.19
1.00
1.25 (0.84 − 1.85)
1.30 (0.93 − 1.83)
0.12
2.4 (2.1 − 2.8)
2.3 (1.9 − 2.8)
2.2 (1.9 − 2.6)
0.37
1)Adjusted for the categories of age (< 31 and 31 + years),
gestation (< 18 and 18 + weeks), parity (0 and 1 +),
cigarette smoking (never, former, and current), passive
smoking at home (never, former and current), passive
smoking at work (never, former, and current), family
income (< 4,000,000, 4,000,000 − 5,999,999, and 6,000,000
+ JPY/year), education (< 13, 13 − 14, and 15 + years),
changes in diet in the previous one month (none or
seldom, slight, and substantial), season when data were
collected (spring, summer, fall, and winter), and for body
mass index as a continuous variable.