The Prevalence and Risk Factors of Dental Caries of Permanent Teeth among 12-year-old Students in Guangdong, South China

Background: Analyses of factors related to dental caries in the population have been reported in some cities and provinces of China. However, large-scale field data evaluation at a provincial level is lacking. The aim of this study was to explore the factors related to caries in the permanent dentition of 12-year-old students in Guangdong province. MATERIALS/METHODS: Equal-sized groups of urban and rural students (960 of each) in Guangdong were selected by stratified, multistage random sampling with a gender ratio of 1:1. All participants were required to undergo an oral health examination. Information concerning the subjects’ family background, dietary habits and relevant oral health behaviour was collected in a structured questionnaire. Logistic regression analysis was used to study the factors related to dental caries. The independent variables included gender, residence, socio-economic status, dietary factors, and oral health measures, among others. RESULTS: Multivariate conditional logistic regressions showed that rural residence (odds ratio (OR)=1.676, 95% confidence interval (CI): 1.350-2.080, P<0.001), female gender (OR =1.377, 95CI: 1.115-1.701, P =0.003), frequent consumption of sugary snacks (≥ once a day vs. < once a day) (OR=1.484, 95%CI: 1.123-1.962, P =0.006), and calculus (OR =1.391, 95%CI: 1.122-1.724, P =0.003) were risk factors for dental caries, while the father’s education level (>9 years vs. ≤ 9 years) (OR=0.796, 95%CI: 0.638-0.993, P =0.043) was a protective factor. CONCLUSION: female gender, poor oral hygiene and frequent consumption of sugary snacks were the main risk factors for caries of the teeth in These findings revealed the complex and dynamic influences of physiological,

3 of pain, suffering, loss of school time and economic burden of the family [2].
Dental caries is a multifactorial disease that results from the interplay among oral microflora, fermentable carbohydrates such as sugary snacks, susceptible teeth, and time [3]. Dental health is associated with child growth and development, both physical and psychological. A longitudinal study reported that caries in the primary teeth has an influence on caries in the permanent teeth [4].
Parents constituted the main influence on oral health behaviours among adolescents, while peers have an increasing impact on behaviour with maturity [5]. Given that both diet and dentition change as students progress through school and the young permanent teeth are vulnerable to caries in the age range, prevention measures are of great importance [6]. Preventive measures can be highly effective in improving students' oral hygiene is well exhibited for students, if more effective prevention measures are taken. Guangdong, located in southern China, has always been a bellwether of the Chinese economy since the Opening of China and has made remarkable achievements in economic and social development. This province is rich in medical resources relative to other provinces. However, the caries prevalence and DMFT among 12-year-old students in Guangdong remained relatively high during the national oral health epidemiological surveys in 1995 and 2005. In addition, caries in rural areas was more serious than in urban areas in Guangdong [7,8], in contrast to reports from other provinces [9]. The trend of serious caries among rural residents became clear starting in the fourth national oral health epidemiological survey. Some epidemical data concerning factors related to dental caries in the population have been reported in other cities and provinces of China [10][11][12]. However, these studies may have been insufficient to support the findings due to their small sample sizes. In contrast, the present study included large-scale samples at the whole-province level.
This survey was conducted in 2015-2016 within Guangdong Province at the request of the government to collect updated and detailed information on oral health status and related factors for the development of realistic and effective policies on dental caries prevention.

4
The Oral Health Survey scheme was approved by the Stomatological Ethics Committee of the Chinese Stomatological Association (Permit Number: 2014-003), and all of the parents of the respondents signed informed consent forms.

Sampling design
A stratified, multistage random sampling design was used to select equal-sized samples of 12-yearold participants who were representative of the province's population (Fig. 1). To ensure that the objectives of the study could be met, we used the formula N=deffµp(1-p)/δ 2 to calculate the required sample size. 'deff' was the sampling design efficiency; 'α' was the selected confidence level for statistical testing, 'p' was the expected prevalence; 'δ' was the allowable error. Considering the urban-rural and male-female stratification, the 'deff' was set to 4.5, and 'α' was set to 0.05, making µ α/2 equal to 1.96; the prevalence (p= 28.9%) of dental caries in children was estimated based on the prevalence of caries in previous national oral health surveys [13,14]. A theoretical sample size of at least 1891 participants was originally calculated. For ease of grouping and implementation, the actual sample size was adjusted to 1920. All participants were drawn from 24 schools selected from 8 districts using the probability proportional to size (PPS) sampling method. Half of the participants were from urban areas (Chancheng, Shunde, Yuexiu, and Jiangcheng), and the other half were from rural areas (Boluo, Lufeng, Raoping, and Luoding). The ratio of male to female participants was 1:1 in each school.

Oral health examination
A special investigation team was set up to complete all the on-site investigations, including 3 oral physicians and 3 recorders who had been engaged in clinical work for more than 3 years and had passed the training conducted by the Technical Guidance Group Of National Oral Health Epidemiological Investigation. The consistency of the diagnostic standard among the investigators was calculated using Cohen's kappa statistic (0.80-0.92). Oral examinations were performed in the classrooms under artificial light using dental mirrors and ball-tipped WHO examination probes.

Questionnaire survey
All participants were asked to complete a structured questionnaire personally during the clinical examination. The questionnaire 22 questions on family background: oral behaviour; knowledge, attitude, and dietary habits related to oral health; dental experience and self-assessment of general health and oral health. The items on the questionnaire have been described in detail elsewhere [10].
The participants were divided into several groups to be examined. Our investigator administered the question to each group by reading each question to the students in the classroom; to avoid influencing the students' answers, the investigator did not elaborate on the meanings of the questions. The education levels of participants' parents were dichotomized as >9 years or ≤9 years.
Oral health knowledge and attitude scores were computed by counting the total number of correct or positive replies.

Statistical analysis
To reduce any data entry-related errors, we performed structured double data entry using Epidata 3.0 and applied validation and correction methods. All statistical analyses were performed using IBM SPSS Statistics 25.0. The prevalence of caries and the mean DMFT scores were compared between urban and rural areas and between females and males using the chi-squared test and the Wilcoxon ranksum test. The factors related to permanent dental caries were analysed by unconditional univariate logistic regression analysis and multivariate logistic regression analysis. Accordingly, a forward stepwise multivariate logistic regression model was applied to examine the risk of dental caries in relation to the independent variables whose P-values were less than 0.15 in the bivariate analyses.
Statistical significance was set at P < 0.05.

Results
A total of 1920 students aged 12 years were investigated in this study, 827 of whom were diagnosed with caries. The overall prevalence of dental caries was 43.07%, and the mean DMFT was 1.06±1.721. There were statistically significant differences in caries prevalence by residence location and gender. Female students showed a significantly higher DMFT score and prevalence of dental caries (P<0.001) than male students, and those measures were also higher in rural areas than in urban areas (P<0.001). Whether male or female, students living in rural areas showed higher prevalence and mean DMFT scores than those living in urban areas. The prevalence of dental caries was evaluated in different groups divided by gender and residence (Table 1). The results of univariate conditional logistic regression are shown in Table 2. The risk of dental caries was significantly associated with residence, gender, whether the child was an only child, parents' education level, snack frequency, oral health education, and calculus. Almost half of the students had good toothbrushing habits (≥twice per day), while gingival bleeding or dental calculus could be found among approximately 50% of the students.
There is Where table 2 should be, check the end of the text for details.
The results of the cumulative multivariate logistic regression are shown in Table 3  people need more medical care and must pay more attention to their physical, mental and oral health. Most residents in the Guangdong area prefer sweet and sticky snacks, which may be an underlying reason for the high prevalence of caries in these years.
In this study, we found that students in rural areas had a higher level of caries prevalence and a higher DMFT score compared with those from urban areas in Guangdong, which was in line with the 8 national survey [16]. It remains us that more attention should be paid to rural students urgently.
Generally, caries of residents in urban areas are more serious in developing countries. However, caries in rural areas are becoming an increasingly serious problem in China. Preschool children and older people in China and Turkey have been similarly reported [17,18]. The difference between the urban and rural areas could be explained by dietary patterns and oral hygiene habits in some previous studies [19]. Rural residents face obstacles to recognizing the importance of oral health and obtaining timely treatment. These difficulties can be illustrated by a shortage of dental personnel, inadequate availability, and accessibility to oral health service [20], insufficient oral health knowledge and attitude [21]. Additionally, socio-economic factors such as low household income and low education level among the parents of children in rural areas might also account for this [22]. A recent study has reported that the utilization of a mobile dental vehicle can be a practical and positive strategy to support oral healthcare to rural populations to reduce inequalities to some extent [23].
The rationales behind these differences between urban areas and rural areas need more exploration in the future.
A gender difference in oral health has been widely documented over time and across cultures [24].
Female students have a higher risk of caries than male students, as reported in this survey, which has been well documented in epidemiological studies of dental caries in living human populations.
Exceptions exist but are not common. Some reports attributed the gender difference to the earlier eruption of teeth in girls, gender-based dietary preferences, and saliva osmotic pressure. A foreign study of clinical and experimental caries research confirms the impact of hormonal fluctuations [24,25], which make the oral environment significantly more cariogenic for women than for men during puberty, pregnancy, and menstruation. Caries aetiology is complex, and the quantity and quality of saliva constitute an important dimension in its occurrence and progress. Changes in saliva composition have been used to predict the susceptibility to caries. Females showed low salivary pH and similar flow rates compared to males [26,27], whereas other studies turned out just the opposite [28]. The difference in salivary levels of oral bacteria (S. mutans) has been reported in Sudanese adults [29]. These findings revealed the complex and dynamic influences of physiological, behavioural, and environmental factors on gender differences in caries incidence. A gender-based approach put forward by Arantes R to prevent dental caries might be taken into consideration later in the future [30].
Parents guide children's dietary habit and pass on oral health knowledge, attitudes and behaviour. A systemic review has shown that it is increasingly acknowledged that parents and the family act as moderators on children's oral health behaviours [31]. A qualitative study provided detail regarding parental views on the influences on children's OHB [32]. More highly educated parents are more willing to spend time and money on the exhibition of children's good oral habits [33]. In this survey, the high educational level of the father was a dominant risk factor, which indicated that the students whose father was in high educational status were relatively not easily infected with caries. This differed from some views of students in domestic and overseas related studies [34]. Nevertheless, the design of sampling and survey methods were not quite the same in these studies, which was insufficient to make a point. There has been little discussion about the impact of fathers' education background on students' caries. Interestingly, there was no significant difference between different maternal educational levels in the multivariate logistic regression model. There might be a strong correlation between maternal and paternal education. Once other factors (diet, gender, etc.) are considered, mothers' influence gradually fades away. To date, most research has focused on the association between caries and mothers' education background, with few studies discussing fathers' attributes, holding a point of view that mothers spent more time on accompanying and have more influence on students' exhibition of behaviour. A previous study had shown that sugar consumption had some relationship to the mother's education as an indirect representation of family socioeconomic status indirectly [35]. This needs further study in future research to explore the influence of parental education on children's caries.
Caries and periodontal diseases are associated with each other and share the common risk of factoradded sugar consumption [36,37]. Numerous cross-sectional epidemiological studies document the significant correlation between dental caries and sugar intake in diet [38,39], whether in longitudinal studies or cross-sectional studies. Recently, an increasing series of studies has reported that added sugar, especially the high frequency of consumption, is associated with periodontal disease [37].
Periodontal and dental status appears to correlate with nutritional habits, which are multifactorial diseases that follow similar risk patterns and develop from an interaction between chronic conditions originating early in life [40]. In our study, the high frequency of added sugar consumption (snack) was represented as a risk factor of caries among 12-year-old students in Guangdong province. In addition, a new risk factor found in the survey was calculus, which has not been assessed before.
Consequently, some relationship existed between dental caries and oral hygiene. It is suggested that an assessment of sugar intake followed by appropriate changes in the diet seems appropriate in clinical efforts to reduce the extent of caries. The mean daily added sugar varied in children, teenagers, and adults. In the UK, the recommendation for sugar consumption among the total population is below 60 g/d, while in Irish adults, the frequency of intake of added sugar intake four times per day corresponded to a mean added sugar intake of 9%, but for children and teenagers, this exceeded the WHO recommendation [39]. Hence, a rational limitation on sugar consumption should be set up according to the daily energy consumption of Chinese people. To satisfy the desire for sweets and reduce energy intake, substitute sweeteners would be helpful; these additives have been demonstrated to have no untoward effect on general health [41]. Furthermore, clear and accurate labelling for added sugar-containing packaged foods could be a useful way to remind one's awareness of limiting sugar consumption. A school-based method limiting the type of food in the school canteens is worthy of putting into effect [42].
The most obvious limitation of this study is the retrospective rather than prospective design, which introduced potential recall bias. An additional limitation is the lack of water fluoride concentration in the area tested, which has a dominant impact on the prevalence of one district. Finally, the questionnaire of this study is self-reported and might contain several potential sources of partiality, especially the participant's memory bias.
In conclusion, the prevalence of permanent teeth dental caries among 12-year-old students in Guangdong was not optimistic, especially in rural areas, female students and those who frequently consumed sugary food. Therefore, the relevant departments should modulate the direction of 11 preventive and treatment strategies on the risk factors according to this survey. More attention should be paid to female students and rural students.
Abbreviations And Acronyms We receives these fundings in support of our research for or preparation of this work.

Authors' contributions
The paper was drafted by Linmei Wu.   Figure 1 The sampling process