Factors Associated with Tooth Loss in Postmenopausal Women: A Community-Based Cross-Sectional Study
Abstract:
Many studies have indicated that menopause affects periodontal health and tooth loss. The possible mechanism might due to several hormonal changes and low bone mineral density (BMD) during the transition period. However, few studies have explored the role of oral hygiene in the number of remaining teeth (NRT) in postmenopausal women (PMW). The aim of this study was to explore the prevalence of and factors associated with NRT less than 20 in PMW. A community-based, cross-sectional study was conducted in coastal Yunlin County, Taiwan. NRT was calculated based on natural and filled teeth. BMD was detected by dual-energy X-ray absorptiometry at the collaborating hospital. Logistic regression analyses were used to identify the factors associated with NRT in rural PMW. Six hundred and ten rural PMW with mean age 65.4 years enrolled in this study. The mean NRT was 17.6 (standard deviation [SD] = 10.4), with 43.9% having 43.9% having <20 and 13.9% edentulous. More
than half (65.9%) reported that they seldom brushed their teeth after meals, 79.2% rarely used dental
floss, and 80% did not regularly undergo tooth scaling by a dentist. The majority of women had low
BMD, including 48.7% with osteopenia and 20.7% with osteoporosis. After adjusting for potentially
confounding variables, NRT <20 was associated with infrequent tooth scaling (odds ratio [OR] = 2.78,
95% confidence interval [CI] = 1.70–4.56) and dental floss use (OR = 2.01, 95% CI = 1.24–3.26), but not
BMD. A high prevalence of NRT <20 was found among rural PMW, but poor oral hygiene rather than
low BMD was the major contributing factor. It is an emerging issue for primary healthcare providers
and clinicians to initiate oral hygiene promotion programs for these disadvantaged women.
Introduction
Menopause is a period in which women’s health status is significantly fluctuating, with an
average onset age of 50. The global average life expectancy for women is 74 years, so the time after
menopause occupies nearly one-third of women’s lives. The secretion of estrogen rapidly declines post
menopause, and leaves postmenopausal women (PMW) at higher risk of various physical and mental
illnesses than men. This imbalanced health status and postmenopausal reduction of estrogen has
been blamed for numerous ailments, including tooth loss. Retaining at least 20 natural teeth until 80 years of age has been defined as a key element of proper general health. However, many studies
on tooth loss have focused on low bone mineral density (BMD) and periodontitis in PMW but less on
the oral hygiene behaviors owned by those with <20 remaining teeth.
In recent years, osteoporosis has received much attention on the issue of oral health in PMW.
Studies have indicated that osteoporosis, considered to be a problem of bone destruction and associated
with rapid fluctuations in estrogen, exaggerates the loss of alveolar bone, which is involved in the
retention of teeth. In fact, there are many important factors associated with tooth loss and
osteoporosis, such as advanced age, poor education, chronic diseases, tobacco and
alcohol use, oral hygiene, and unhealthy diet. However, there is still controversy in
the literature on the relationship between BMD and tooth loss in PMW.
Oral hygiene is generally ignored in the relationship between osteoporosis and tooth loss in
PMW. Oral hygiene behaviors, such as brushing, flossing, regular tooth scaling
and dental visits, have been highly recommended as the essential elements for retaining natural
teeth and maintaining oral health. Hence, we doubt that hormone reduction or osteoporosis can be the
only factors related to a number of remaining teeth (NRT) <20 in PMW. This study aimed to explore
the prevalence and factors related to NRT <20 among rural PMW and to consider the role of BMD and
oral health behaviors.
Materials and Methods
2.1. Design, Sample, and Setting
This cross-sectional descriptive study was part of a community-based health promotion program
for residents living in the rural coastal region of Taiwan. It was conducted by nursing faculties from
October 2015 to July 2016 in collaboration with the local hospitals. This study used purposive sampling
to select participants. The inclusion criteria were: (1) women with postmenopausal stage (identified by
the physicians of the research team); (2) age _50 years and less than 80 years; (3) able to walk to the
local hospital; (4) able to complete questionnaires independently or through interview; and (5) agree to
participate and sign the informed consent before enrolling in the study. The exclusion criteria were: (1)
unable to answer the questionnaires clearly; and (2) having serious disease(s), such as kidney failure
that required regular dialysis, or cancer.
2.2. Measurements
1. Demographic characteristics included age, years of education, and three self-reported chronic
diseases (diabetes mellitus, hypertension, and hyperlipidemia).
2. Bone mineral density (BMD) was measured with dual-energy X-ray absorptiometry at the
collaborating hospitals. Participants with a T-score <1.0 were classified as having osteopenia,
and a T-score <2.5 indicated osteoporosis, as recommended by WHO guidelines.
3. Number of remaining teeth (NRT) was calculated by the trained research assistants based on the
natural and filled teeth in the mouth. Removable dentures and residual roots without a crown
were excluded. Participants were divided into two categories, namely “NRT <20” and “NRT
_20”.
4. Oral health related behaviors were measured by personal habits including oral hygiene, substance
use, and eating behaviors. Firstly, oral hygiene was evaluated by three questions: “Do you
use dental floss daily?”, “Do you undergo dental checkups or tooth scaling regularly (every
6–12 months can be covered by the National Health Insurance)?”, and “Do you brush your
teeth after meals?” Answers were categorized as infrequent (“never/seldom”) versus frequent
(“usually/always”). Secondly, tobacco and alcohol use were classified as “never used” versus
“currently or formerly used”. Thirdly, eating behaviors were examined with two questions: “Does
your diet include five food groups per day, namely grains, vegetables, milk, meat, and fruits?”,
and “Do you drink at least 1500 mL of water every day? The answers “never or sometimes” were
classified as inadequate, whilst “usually or always” were coded as adequate.
2.3. Procedure and Ethical Considerations
The study received ethical approval from the Institutional Review Board (No. 104-9925B). The
research assistants were senior nursing students who had received 8 hours of training from the
investigators for how to count the NRT. All of the instruments had good reliability (inter-rater reliability
was 0.9) and validity (content validity was 0.88–0.89). Details of the interviews and measurement
procedures were based on the published literature.
2.4. Statistical Analyses
The Chi-square and t-tests were used to examine the univariate factors associated with NRT.
To identify the correctable factors related to NRT <20, a 3-level multivariate logistic regression with
forward entry method was performed. Model 1 was used to detect the correlation between each single
variable and NRT <20 without adjustment. Model 2 separately analyzed each modifiable factor by
adjusting two irreversible confounding factors – age and education. Model 3 was a complete model
considering all of the study variables to identify those associated with NRT <20. Collinearity diagnosis
was performed by the correlation analysis between these category factors by using the Cramér’s V
coe_cient. The factor which was highly correlated with other factors was excluded from the model.
The odds ratios (OR) with 95% confidence intervals (CI) were obtained. SPSS 22.0 software (IBM
SPSS Statistics for Windows, Version 22.0. Armonk, NY, USA) was used and p < 0.05 was set as
statistical significance.
3. Results
Among 651 postmenopausal women who were enrolled in this study, 41 failed to complete the
NRT measurement, and 610 participants were evaluated, with an average age of 65.4 years (SD = 7.6),
range 50–80. The mean years of education received was 3.1 (SD = 4.3). The prevalence of hypertension,
diabetes, and hyperlipidemia was 46.6%, 47%, and 20.8%, respectively. Participants also had a
high prevalence of low BMD, with mean T-scores of 1.6 _ 1.1, and 48.7% of participants having
osteopenia, and 20.7% having osteoporosis. In addition, nearly 80% of participants reported seldom
or never undergoing tooth scaling or using dental floss. Tooth brushing behavior after meals was
also unsatisfactory (65.9% seldom or never). Regarding diet, 26.9% reported they did not eat five
food groups daily, and 53.4% did not drink at least 1500 mL of water daily. The mean NRT was 17.6
(SD = 10.4) and 43.9% of the participants had NRT <20, including 13.9% who were edentulous (Table 1).
Univariate analysis showed that women who were older (>65 years old, p < 0.001) and had little
education (p < 0.001), diabetes (p < 0.05), low BMD (p < 0.001), and osteoporosis (p < 0.01) tended
to have NRT <20 (Table 2). Regarding the oral health related behaviors, infrequent tooth scaling
(p < 0.001), infrequent dental floss use (p < 0.001) and cigarette smoking (p < 0.05) were significantly
associated with NRT <20. Logistic regression analysis for Model 1 showed that diabetes, osteoporosis,
and smoking were associated with NRT <20 (Table 3). However, the e_ect of these variables became
insignificant after the adjustments in Models 2 and 3. In Model 3, a complete model adjusted for all
variables, infrequent tooth scaling (OR = 2.78, 95% CI = 1.70–4.56) and infrequent dental floss use
(OR = 2.01, 95% CI = 1.24–3.26) were significantly associated with the increased risk of NRT <20. The
Cramér’s V coe_cients between each pair of explanatory variables were less than 0.3 (not shown in
the table) and showed low-degree correlations. Thus, all the explanatory variables were included in
the logistic regression analyses.
Discussion
A nurse-led health promoting programwas a feature of this study and the purpose was to promote
oral health in rural PMW. This study had four main findings. First, PMW had a reduced number of
teeth, with a high prevalence of NRT < 20 and edentulism. In previous studies, the average NRT
for women aged 20 to 64 was 24, only 18.4% had NRT < 20 and the average NRT for PMW was
22–23. The present study indicated that the oral health of these participants was significantly
worse. In addition, the prevalence of NRT < 20 was not only higher than previously reported for
women under the age of 65, but also higher than that of PMW in Korea (25.9%), and the United
States (5.9%).
Second, the oral hygiene behaviors of participants were generally poor, and far from the
recommendations of the Fédération Dentaire Internationale (FDI) World Dental Federation.
In previous studies , the rare use of dental floss, never or seldom undergoing tooth scaling,
and lack of tooth brushing after meals were significantly associated with NRT <20. The high prevalence
of inadequate oral hygiene habits in PMW in this study was similar to a study in Korean PMW
(67–81%), but much higher than that found in the general population (58–70%), or in PMW
in other countries (28–52%).
Third, the present study found that osteopenia or osteoporosis was not the major factor in
increasing likelihood of NRT < 20. The study revealed that while the prevalence of osteopenia
(48.7%) was similar to some other countries, the prevalence of osteoporosis (20.7%) was lower than
other countries (24–38%). These phenomena might be due to sample variation, such as
age, occupation, and geography, but the common feature is that PMW generally have a low BMD
(prevalence >60%).
Many reports indicate that systemic BMD might be associated with alveolar bone loss or tooth
loss, due to rapid imbalance of estrogen levels a_ecting the activation of bone cells and immune
cells . However, the present study demonstrated that osteopenia or osteoporosis was not a
significant factor in NRT < 20 in PMW. Dental health interventions in PMW should focus more on
the impact of periodontal disease as a main cause of tooth loss in adults. Periodontal disease is
a product of cellular immune responses to oral inflammatory processes, which are caused by oral
infections or estrogen deficiency, and is a common oral problem in PMW. Further, the teeth
and alveolar bone have a higher chance of infection and inflammation than other bones. Based on the
principle that prevention is better than treatment, reducing the impact of inflammatory reactions on
NRT <20 is worth more attention in PMW.
Good oral hygiene habits have been recognized as reducing oral inflammatory reactions.
Fortunately, the fourth finding indicated that regular tooth scaling and use of dental floss
were two correctable factors related to NRT. This result is corroborated by studies in di_erent
populations as well as a study in Korean PMW However, in some studies evaluating
the relationship between osteoporosis and NRT in PMW, these two important oral hygiene behaviors
were not considered . A large study in South Korea explored the loss of more than eight
teeth in PMW and claimed that low bone density was associated with tooth loss in this population.
Although the researchers considered many factors, such as BMD and oral care-related behaviors, they
did not adjust for all variables in their analysis model. In summary, it is impossible to eliminate the
contribution of oral health behaviors in retaining more natural teeth.
In fact, evidence indicates that dental flossing, regular dental checkups, and tooth scaling are the
most e_ective ways to preserve NRT. This is even more important in PMW, since the decline of
estrogen during menopause reduces the ability to maintain the balance between beneficial and harmful
bacteria in the oral environment, and the plaque biofilm accumulated by oral bacteria strengthens
the inflammatory response in the body, increasing susceptibility to progressive periodontal disease.
Consequently, the risk of tooth loss . The best way to reduce this susceptibility is to remove the
plaque under the gum-line by regular oral care. Previous literature has also proved that
dental plaque and gingivitis can be significantly improved by brushing and flossing compared with
brushing alone , and even using interdental brush is superior to dental floss. However, women
were less concerned about oral health when compared with menopausal symptoms, osteoporosis, and
other chronic diseases . In this study, there were 65.9–80% of PMW who never or seldom used
dental floss, brushed their teeth after meals, or underwent tooth scaling, even though this is covered by
the national health insurance system every 6–12 months. In order to encourage regular use of dental
health services, the insured is only required to pay minimal registration and co-payment fees (3–5
USD). Perhaps many women in the study population believe that osteoporosis and periodontal disease
are inevitable in the postmenopausal stage, which might be misunderstood that the tooth loss in
PMW is the fate of high susceptibility to osteoporosis, thus their reliance on excessive medical services
to retain teeth and lack of awareness of the benefits of oral hygiene. Further studies are necessary to
explore the prevailing level of knowledge and attitudes toward use of dental health services.
Good oral health is an indicator of general health, and can reduce medical expenses and improve
nutrition, communication, self-esteem, and quality of life. Oral hygiene education could
contribute significantly to improving oral health. It is urgent to strengthen the practice of dental
flossing and regular tooth scaling among PMW to achieve the global goal of retaining at least 20 teeth
through age 80.
Limitations
The advantages of this study were the discovery of the high prevalence of NRT < 20 and the
importance of oral hygiene for maintaining NRT in PMW, but there were still some limitations. First,
hormone therapy and osteoporosis medication data were not collected. Some studies have pointed out
that estrogen use can prevent from tooth loss in PMW , while the widely used bisphosphonate
drugs for the treatment of osteoporosis have been reported to have potential risks of osteonecrosis of
the jaw. According to other surveys, only about one-third of Taiwanese PMW have used or are
currently treated with hormones or prescription drugs for osteoporosis . Therefore, women’s
oral health may be worse than the survey indicated in this study, if up to one-third of the participants
are assumed to receive some benefit from estrogen therapy. Secondly, the trained research assistants
only counted the number of teeth, and there was no assessment of other oral parameters, such as
dental caries, periodontal disease, oral hygiene index, and plaque index. Therefore, the NRT data. reported here do not address other potential oral problems. Finally, the recruitment of a single region
and non-random sampling might limit the generalization of the findings.
Conclusions
For postmenopausal women in rural areas in Taiwan, the prevalence of NRT < 20 and edentulism
is significantly higher than that reported in previous studies in other regions. Poor oral hygiene
behavior was found to be common. The occurrence of NRT < 20 in postmenopausal women was
most strongly related to infrequent flossing and tooth scaling, rather than age, education level, or
osteoporosis. Low BMD was not a primary factor in NRT < 20, and tooth loss cannot be strictly
attributed to hormonal changes in menopause. Awareness of oral hygiene is the most important issue
for retaining dentition in postmenopausal women.
Author Contributions: Methodology, formal analysis, and writing—original draft preparation, review and
editing, M.-Y.P. and T.-C.H.; investigation and data creation, P.-H.C.; conceptualization, project administration,
and supervision, M.-Y.C.
Funding: This research was funded by Formosa Plastic Group, grant number FCRPF610011.
Acknowledgments: We appreciate all participants who agreed to join this study and all sta_ members from the
Chang Gung Memorial Hospital, Yunlin branch, for their invaluable support.
Conflicts of Interest: The authors declare no conflict of interest.