Prevalence and risk factors of dental fluorosis among children aged 8–12 years in Shandong province of China

ABSTRACT This study was to investigate the prevalence and severity of children’s dental fluorosis (DF) in Shandong and identified the potential risk factors for DF. A total of 87 villages in Shandong were investigated to calculate the prevalence of DF and Community Fluorosis Index (CFI) in 2018–2019. Six hundred and seventy children were enrolled to identify the potential risk factors using univariate and multivariate logistic regressions. Goodman–Kruskal Gamma was used to explore the factors related to the severity of DF. In 87 villages, 1249 of 8700 (14.36%) children still have DF. The prevalence of DF in most villages was below 40% in 2018–2019. Water fluorine concentration when selected for the study and urinary fluorine concentration were related to the risk of DF (P < 0.001). Some eating habits, like lower frequency of eating fresh vegetables, eggs, and beans, were associated with the risk of DF (P < 0.001). The high water fluorine concentration, and lower frequency of eating fresh vegetables, eggs, and beans were also related to the severity of DF (P < 0.001). DF in children in Shandong province is still a common endemic disease. This study tries to provide a useful guide for the prevention and control of DF.


Introduction
Fluorine is widely dispersed in nature, such as air, water, soil, and plants, generally existing in the form of fluoride.It is also one of the essential trace elements in human body mainly distributed in bone and tooth, played an important role in bone mineralization and formation of dental enamels (Dhar and Bhatnagar 2009).In humans, the dominating route of fluoride absorption is through the gastrointestinal tract, like drinking water or food, and sometimes through the skin and respiratory tract (Jha et al. 2011).When the ingestion of fluoride is inadequate, humans may suffer from dental caries, lack of formation of dental enamel, and reduced bone mineralization (WHO 1996).However, consuming fluoride in excess also can cause damage to various bodily systems or organs.The major clinical manifestations are presented as skeletal or dental fluorosis, non-skeletal manifestations, or combinations of these maladies caused by excessive accumulation of fluoride (Wei et al. 2019).Dental fluorosis (DF) is the most common and prominent symptoms in the early stage of chronic fluorosis, and it is a developmental disturbance of enamel caused by excess intake of fluoride during tooth formation, often occurring in children (Gu et al. 2020).
DF was considered to be related to artificial fluoridation of drinking water, particularly in developing countries for declining dental caries levels (Mascarenhas 2000;Centers for Disease Control and Prevention 2001;Levy 2003).It can also be caused by the abundance of fluoride in the environment, known as endemic fluorosis, in many areas around the world like Africa, China, the Middle East, and southern Asia (WHO 2005;Yadav et al. 2019).China has one of the most serious endemic fluorosis problems in the world.The use of coal containing high fluoride as an indoor fuel source, drinking high-fluoride water, and drinking tea growing in high-fluoride areas, especially brick tea are the common causes of endemic fluorosis in China (Fung et al. 1999;Xu et al. 2017;Wei et al. 2019).Among these causes, drinking water is still the main source of human exposure to fluoride (Zhang et al. 2016).So aiming to reduce endemic fluorosis, since the 1970s, Chinese government has taken measures to change water sources in areas with high fluorine content, through water improvement and defluoridation on fluorosis.After decades of research, it has proved that the measure was effective in reducing endemic fluorosis (Wang et al. 2021).
However, despite efforts to reduce the incidence of endemic fluorosis, only skeletal fluorosis is close to elimination.DF in children is still a public health problem in high-fluoride areas in China.Recently, it was observed that fluorosis may be not only related to direct exposure to fluoride (drinking water, air, and food intake), but also affected by various indirect factors such as family economic status, dietary habits, nutrition, and personal behavior habits (Arheiam et al. 2022).Previous studies exploring risk factors for DF were performed several decades ago, and most are in developing countries, in which DF was usually caused by artificially fluoridated public drinking water, rather than the natural environment with high fluoride.Therefore, we think it is necessary to conduct research using data from recent years in China.This study tried to perform a crosssectional survey among 8-12 years old schoolchildren in Shandong province of China during 2018-2019, to investigate the current prevalence and severity of children's DF in recent years, and to explore the influencing factors of children's DF to provide certain guiding significance towards the prevention and control of DF.

Study design
A cross-sectional study was performed to measure the prevalence of DF in schoolchildren aged 8-12 in Shandong, and a case-control study was used to identify the potential risk factors for children's DF.This study was performed in endemic fluorosis areas in the Shandong Province of China during 2018-2019, using clinical examinations and structured questionnaires.The fluoride content in the water and urine samples was also measured using a fluoride ion-selective electrode assay.Before commencing the study, ethical approval and permissions were obtained from the Shandong Institute of Prevention and Control for Endemic Disease.The personal information of study subjects has been deleted.

Study objects and investigation content
The information on villages One county was selected from each of the 17 cities in Shandong province in this study.The fluorosis area of tea drinking type or coal burning type was excluded.Then, randomly selected six villages among these counties in each city.Those villages without a clear fluoride concentration of water supply project were excluded.Finally, a total of 87 villages were included.In each village, the baseline information of water plant were investigated; meanwhile, the concentration of fluoride in drinking water was measured by researchers during this study.

DF clinical examination
In each village, 100 children aged 8-12 years were randomly selected to determine the prevalence and severity of DF.These children were not accommodated at the school and the school has the same water source as their village.The classification of DF was determined by clinical examination to children and scored according to Dean's method (Rozier 1994).The Dean's score was based on the two most affected teeth as "0 = normal", "1 = questionable", "2 = very mild", "3 = mild", "4 = moderate" or "5 = severe"."Very mild", "mild", "moderate" and "severe" was considered having DF.And the epidemic of DF for each village used the community fluorosis index (CFI) which was calculated based on Dean's score (Dean et al. 1935).

Questionnaire on children's information
To perform a case-control study, in each village, 10 children aged 8-12 were randomly selected from the 100 children previously examined for further questionnaire survey.Through interview with their parent/guardian, the children's baseline information, parents' education level, diet and living habits were investigated.Meanwhile, the measurement of urinary fluoride concentration also needs to be carried out.The study variables ultimately include sex, age, urinary fluorine (mg/L), the highest education for parents, having brushing teeth habit or not, used fluoride toothpaste or not (including those who did not brush teeth and who did not use fluoride toothpaste), family food source, the consumption of fresh vegetables, fruit, meat, egg, bean, and milk (high frequency: eating every day; moderate frequency: eating vegetable, fruit, meat, egg, and bean once every 2 or 3 days, often or sometimes drink milk; low frequency: eating vegetable once every 5 days, eating vegetable, fruit, meat, egg and bean once every week, barely drink milk).Subjects whose variables data was incomplete were excluded, finally, 670 children were included in the analysis.

Statistical analysis
In the descriptive analysis, frequencies and percentages were used for categorical variables, while measures of central tendency and dispersion were used for quantitative variables.Categorical variables used Chi-square tests.Univariate and multivariate logistic regression were used to obtain odds ratios (ORs) with 95% CIs among subgroups for identifying potential risk factors for DF.Goodman-Kruskal Gamma was used to measure the trend coefficient between those risk factors and severity of DF (Gamma range from − 1 to 1; Gamma > 0: positive correlation; Gamma < 0: negative correlation; Gamma = 0: non-correlation).All tests were two-sided and had a significance criterion of P < 0.05.Statistical analysis was done with SPSS (version 26.0; SPSS, Chicago, IL) and Microsoft Excel 14.0.4763.0000(Microsoft, Redmond, WA).

The relationship between present water fluorine concentration and DF
Through concentration measurement of drinking water fluorine during the study, 73 villages with clear water fluorine concentrations were included in this part, and the concentration can be found in Table S2. Figure 1 shows the prevalence and CFI of DF in different water fluoride concentration groups using the median and quartiles.The Spearman's rank correlation analysis shows that water fluoride has no significant trend correlation with the prevalence of DF in children (P = 0.102), neither with CFI (P = 0.140).And there were large variations in the prevalence and CFI of DF in each water fluorine concentration group (Table S3-4).

The baseline characteristics of children aged 8-12 during 2018-2019
The baseline characteristics of 670 children are shown in Table 3.Among children with DF, 115 (52.5%) were boys and 104 (47.5%) were girls, and in children without DF, 235 (52.1%) were boys and 216 (47.9%) were girls.In DF groups, children aged 10 years accounted for the largest proportion (35.6%), while in non-DF groups, the proportions in these four age groups were similar.
There was no significant difference in sex distribution between DF and non-DF groups (P = 0.922); meanwhile, no significant difference was observed between the age groups (P = 0.138).Seventy-one (32.4%) children with DF have urinary fluorine concentrations between 0.485 ~ 0.710 mg/L.Among children with DF, 38 (17.4%) children were from villages with water fluorine content between 1.2 and 2.0 mg/L and 35 (16.0%) were from >2.0 mg/L, and in non-DF groups, the number was 78 (17.3%) and 46 (10.2%).

Associations of risk factors with DF among children
In Table 4 (on the left panel), the result of univariate logistic regression analysis showed that higher urinary fluorine concentration, higher water fluorine concentrations, the highest education for parents, fresh vegetable consumption, fresh fruit consumption, and egg consumption were related to the high risk of DF (P < 0.05).The result of multivariate logistic regression analysis (Table 4 right panel) presented that the high risk of DF was associated with higher urinary fluorine concentration (0.485 ~ 0.710 mg/L vs. <0.369mg/L, OR = 2.66, 95% CI: 1.56-4.54,P < 0.001).Higher water fluorine concentration (when selected into the study) was also related to a higher risk of DF (>2.0 mg/L vs. <0.8mg/L, OR = 1.81, 95% CI: 1.03-3.17,P = 0.038).The highest education for parents did not show a significant difference between children with DF and without DF (P > 0.05).In terms of dietary habits, the frequency of eating fresh vegetables, eggs, and beans was associated with the risk of DF.Compared to children with a high frequency of eating vegetables, those with moderate frequency or less frequency were more likely to have DF (OR = 3.91, 95% CI: 1.91-8.00,P < 0.001 and OR = 6.81, 95% CI: 1.92-24.16,P = 0.003).Children with a moderate frequency of eating eggs have a higher risk of DF than those eating eggs frequently (OR = 2.96, 95% CI: 1.84-4.76,P < 0.001).
Less frequency of bean consumption seems easier to cause DF in children, with OR of 2.30 (95% CI: 1.02-5.19,P = 0.044).

The relationship between risk factors and severity of DF
We further studied the relationship between the degree of DF and the above risk factors (urinary fluorine, water fluorine, fresh vegetable consumption, fresh fruit consumption, egg consumption and bean consumption) shown in Table 5.The Goodman-Kruskal Gamma analysis showed that there was no significant correlation between urinary fluoride concentration and dental fluorosis severity (P = 0.093).The water fluoride concentration was positively correlated with the severity of dental fluorosis (Gamma = 0.104, P = 0.034), indicating that the higher the concentration of fluoride in water, the more serious dental fluorosis.With the lower frequency of eating fresh vegetables, fruits, eggs, and beans, children's DF appears to be more severe (Gamma > 0, P < 0.001).

Discussion
Due to the research found that water fluorine is related to endemic fluorosis, since the 1970s, the Chinese government has taken measures to improve water quality and reduce fluoride in response to the epidemic of endemic fluorosis (Zhao and Wang 2011).After decades of efforts, endemic fluorosis has been controlled at a low prevalence level in China and the measures were proved effective.Although the number of patients with skeletal fluorosis is rare now, the prevalence of DF still should not be ignored.Through the study, we performed in Shandong, we found that there were almost no DF-free villages among the villages investigated during 2018-2019.Although the prevalence of DF is concentrated below 40% in most villages, some villages still have a particularly high prevalence of DF, up to 62.4%.By CFI, the epidemic of DF in most villages were negative or acceptable, but seven villages have mild or moderate epidemic levels.Additionally, when measuring the concentration of fluoride in drinking water in this research, 14 out of 87 villages still had water fluoride above 1.2 mg/L, while the Chinese researchers believe that the concentration of fluoride in water should be under 1.2 mg/L (Wei et al. 2019).Therefore, reducing water fluoride and preventing DF is still a very important task for the government.
In recent years, with the development of the economy and life level, people pay more attention to their appearance.Although DF does not cause pathological damage, it does a negative impact on the individual image.It was found that DF sometimes caused esthetic displeasure in children, which may lead to psychological and behavioral complications in affected individuals (Pagliari Tiano et al. 2009;Buzalaf et al. 2013).So our study explored the risk factors and try to provide a useful guide for the prevention and controlling of DF.In this study, it was observed that water fluoride concentration is closely related to the formation of DF, which was consistent with the previous studies (Iheozor-Ejiofor et al. 2015;Goodarzi et al. 2016).Whether endemic fluorosis or artificial fluoridation of water, drinking water is the most important way of exposure to fluorine.In this research, we also found that urinary fluoride level was associated with the prevalence of DF.In children, almost 35% of ingested fluoride is excreted in the urine which is the main route for removal of fluoride from the body (Villa et al. 2010).It also found that there was a strong linear relationship between total daily fluoride intake and daily urinary fluoride excretion (Villa et al. 2010).Therefore, as a harmless and easily available biological material, urinary fluoride content may play a role in estimates of total daily fluoride intake and predictions of risk of fluorosis.Interestingly, we also observed that at higher urinary fluoride concentrations (>0.71 mg/L), the effect of developing DF became weakened.Meanwhile, there seems to be no trend relationship between urinary fluoride concentration and DF severity.This may be explained in part that children with higher urinary fluoride were often living in a high-fluoride environment for generations or exposed to the extremely high concentration of fluorine instantly, so the body may have a certain tolerance to fluorine.
Many previous studies have found that children's nutritional levels may affect the development of DF (Villa et al. 2010;Buzalaf and Whitford 2011).However, Sah et al. considered that there was no significant effect of nutritional status on the development of DF (Sah et al. 2022).In this study, we found that diet habits not only had a potential influence on the occurrence of DF, but also related to the severity of DF.The high frequency of eating fresh vegetables was beneficial to reduce the occurrence and severity of DF.It may be explained that fresh vegetables contain a large number of beneficial minerals and vitamins, which may be able to counteract the toxic side effects of fluoride on the body.Certain minerals, such as boron, calcium, magnesium, and aluminum, can bind to fluorine in the gastrointestinal tract and hinder its absorption by the body (Elsair et al. 1980;Franke et al. 1985).Some other minerals, like selenium and copper can affect the activity of enzymes and reduce the toxicity of fluorine in the human body (Chen et al. 2022).Additionally, a vegetarian diet can result in more alkaline urine which, in turn, leads to increased urinary fluoride excretion (Whitford 1990).We also found that higher consumption of beans and eggs protected children from DF.This may be because beans and eggs are high-quality proteins that can be better absorbed by the human body and help children grow and develop well.Beans and eggs are also rich in minerals, especially calcium, which is essential for fluorosis antagonism (Ekstrand and Ehrnebo 1979;Shulman and Vallejo 1990).Previous studies have found that the use of fluoride toothpaste may promote the development of DF, but we did not observe this phenomenon (Ahokas et al. 1999).We suspect that, on the one hand, the fluorine content of fluoride toothpaste was difficult to estimate, whether it actually causes the occurrence of DF is unclear (Levy et al. 1995(Levy et al. , 1998)).On the other hand, the period of permanent dentition development is the critical time of DF, around age 3-6, during which it is not clear whether fluoride toothpaste is being used (Cangussu et al. 2002).
This study was based on a large cross-sectional study, which can better reflect the prevalence of dental fluorosis among school-age children in Shandong Province, and even represent eastern China.However, this study still has some limitations.Due to the large number of missing data, we did not include some potential influencing factors, such as family income, the age when children start brushing their teeth, and the times a week they brush their teeth.At the same time, due to the limited conditions, we believe that children's daily dietary intake of fluorine and other trace elements should be further studied.

Conclusion
DF in children in Shandong province is still one of the most common endemic diseases.Urinary fluorine concentration, water fluorine concentration, and some eating habits like eating fresh vegetables, eggs, and beans were associated with the risk of DF in children, and those factors should be considered to guide the endemic prevention work.

Disclosure statement
No potential conflict of interest was reported by the authors.

Figure 1 .
Figure 1.The prevalence and CFI of DF in different water fluoride groups.

Table 1 .
The prevalence of DF and information on water supply project in 87 villages during 2018-2019.

Table 2 .
The epidemic of children's DF in 17 cities of Shandong during 2018-2019.

Table 5 .
The relationship between severity of DF and some risk factors (urinary fluorine concentration, water fluorine concentration and eating habits).