Factors Affecting Preventive Dental Treatment of Adolescents in Korea

We conducted a multilevel analysis to identify factors affecting adolescents’ preventive dental treatment experience in South Korea. We sampled 72,435 students who participated in the 9th Korea Youth Risk Behavior Web-based Survey. The individual-level variables were divided into demographic factors, socioeconomic factors, oral health behavioral factors, and oral symptom experience factors. The regional-level variables included oral health resources, rate of students receiving oral health education at school by region, social deprivation index, and the number of private educational institutions. A higher rate of receiving oral health education in school by region was associated with increased fluoride application (1.04 times, p = 0.003). However, the number of private educational institutes per 1000 people was negatively associated with fluoride application experienced by students (0.64 times, p = 0.039). Students underwent more scaling when there were more dentists per 10,000 individuals (1.14 times, p = 0.008) and less scaling in areas with a higher social deprivation index (0.88 times, p = 0.024). To increase the access to preventive care for oral diseases among adolescents, a preventive system should be established in schools, and a primary dental care system should be established through the cooperation of the government, private dental clinics, and schools.


Introduction
Adolescence is the period during which the differentiation of permanent teeth, which must be used lifelong, is completed and is the most appropriate time to prevent oral diseases through correct oral health behaviors and preventive treatment. It is also a period with high risk for oral diseases, depending on individual factors such as the socioeconomic level of the family and eating habits, as well as environmental factors in schools and regions of residence [1][2][3]. Therefore, it is essential to develop preventive interventions in relation to social health determinants in children and adolescents in order to form correct oral health behaviors and promote oral health throughout life [4,5].
Individuals can form health behaviors influenced by their social relationships with people in their groups and the surrounding environment of the region. For example, in areas where social relations are systematically established and where social resources are abundant, local residents work together to improve their physical activities and create local spaces to reduce stress [6,7]. However, if public transportation is limited in poor areas and there are cigarette outlets in walking distance, people's smoking cessation practices can be negatively affected [8,9].
Adults living in regions with easily accessible public health services and medical facilities are less likely to have dental caries. Lee et al. showed that the lower the community welfare budget and the treatment experience by performing a multi-level analysis using national statistical data representing all Korean adolescents.

Study Population
This study analyzed secondary data using the results of an anonymous self-report questionnaire of 72,435 students (13-to 18-year-olds) from 800 middle and high schools nationwide participating in the 9th Korea Youth Risk Behavior Web-based Survey (KYRBWS) in 2013. The KYRBWS was a government-approved statistical survey conducted for students from 1st grade in middle school to 3rd grade in high school to understand the health behaviors, including oral health, smoking, and drinking, of Korean adolescents in 15 areas.
The survey was designed with stratified multistage cluster-sampling that considered regions and schools to minimize sample errors. The 800 middle and high schools from 16 cities nationwide were selected by applying the proportional distribution method [23]. Students currently enrolled in the 800 selected sample schools accessed the website with the participation number given to each of them, read the survey participation agreement that appeared on the screen, and clicked the participate button to take a survey, if they agreed to participate in the survey. The survey was conducted from 1 June to 30 July, 2013. Since 2005, the Ministry of Education, the Ministry of Health and Welfare, and the Centers for Disease Control in Korea have conducted an annual government-approved statistical survey. The KYRBWS was approved by the Institutional Bioethics Committee of the Korea Centers for Disease Control and Prevention.

Dependent Variables
The annual professional fluoride application experience and the annual scaling experience were evaluated using two questions: "Have you received professional fluoride application in the last 12 months?" and "Have you undergone scaling in the last 12 months?" The annual fluoridation experience and scaling experience were defined when each question was answered "Yes." The response "No" was excluded.

Individual-Level Variables
Individual-level independent variables, including demographic, socioeconomic, oral health behavioral, and oral symptom experience factors, were selected from the KYRBWS. The demographic variables used were gender, school type, school year, and grade. The school year was used to reflect the age of the student; the 1st year in middle school corresponds to 13 years of age, and the 3rd year in high school corresponds to 18 years of age.
Socioeconomic variables used were the father's education level, family affluence scale (FAS), and weekly allowance. The father's education level represented the parents' education level. The FAS evaluated four questions and their response categories, and the questions and answers were as follows: "Does your family own a car?" (0 = no; 1 = yes; 2 = yes, 2 or more); "Do you have your own bedroom?" (0 = no; 1 = yes); "How many times did you travel away on holidays during the past year?" (0 = not at all; 1 = once; 2 = twice; 3 = more than twice); "How many computers does your family own?" (0 = none; 1 = one; 2 = two; 3 = more than 2).
Oral health behavior variables included snack consumption, brushing after lunch, and perceived oral health. Oral symptoms were assessed by asking, "Have you experienced tooth failure, toothache, gum disease, soft tissue disease, or bad breath during the past year?" (0 = no; 1 = yes). The oral symptom experience variable was the sum of experiences for each symptom. The summed values (range between 0 and 5) were converted to 0 and 1 (0 = no; 1 = one or more). The answer "not sure" was treated as a missing value [23,24].

Regional-Level Variables
Regional-level variables were constructed using National Statistical Data to represent the socio-economic characteristics of 16 regions, dental medical resources, preventive service beneficiaries, and the living environment of adolescents. In this study, because the KYRBWS used the status of 16 cities/provinces as individual-level variables, regional-level variables were made to match the 16 cities/provinces.
The socio-economic level of a region is determined using the social deprivation index. The social deprivation index is a composite index calculated from indicators such as residential density, housing ownership, car ownership, regional income, education level, residential facility type, and the residential environment as the average z-score provided in the population census data. This can indicate not only the lack of local financial resources but also the deficiencies of living conditions and facilities [25].
The number of dental clinics/hospitals per 10,000 population and the number of dentists per 10,000 population were used for assessing dental resources. The data used included the National Statistical Data and the statistics provided by the Korean Dental Association [26,27]. Dental resources are an indicator that can confirm access to medical use by local residents [28].
The rate of students receiving oral health education at schools by region was used to identify local priorities for oral health in adolescents. School oral health education is managed by the public health center; therefore, the higher the school oral health education beneficiary rate, the higher the interest in youth's oral health. These data were assessed by reorganizing the proportion of students who answered "Yes" for the question "Have you had oral health education at school for the past year?" in the KYRBWS data by region [23].
Private educational institutes are institutions that teach math, English, and science outside the school, and Korean middle and high school students spend time studying at private institutes after school or on weekends. Therefore, the number of private educational institutes per 1000 population was used as an index reflecting the lifestyle of Korean adolescents living in a social environment focused on the college entrance examination [26,28].

Statistical Analysis
A two-level analysis was conducted to identify factors at individual and regional levels that influenced the preventive dental treatment experience of Korean adolescents. In this study, since the dependent variable had a Bernoulli distribution as a binomial variable, multilevel logistic regression analysis was performed to convert probability to logit values, which were expressed as a linear relationship. For the independent variable, the individual-level factor variable was used as the 1-level unit, and the regional-level factor variable was used as the 2-level unit.
In terms of statistical analysis, the professional fluoride application and scaling experiences were set as dependent variables and were analyzed by dividing them into three stages, such as the null model, unconditional slope model, and conditional model. Because the null model is a process of verifying that the dependent variable differs between regions, only the dependent variable was used, and the intraclass correlation coefficient (ICC) was confirmed. Since the dependent variable was a binary variable, ICC was calculated by the formula: The unconditional slope model is a process of checking the random and fixed effects of individual-level variables and inputs individual-level independent variables with statistically significant associations between dependent variables and complex sample logistic regression. Finally, the conditional model is a process of verifying individual and regional factors affecting independent variables, and therefore, both individual-and regional-level independent variables are used. The model with the highest explanatory power for the dependent variable was finally selected and presented in the results. A likelihood ratio test was performed to test the model's suitability. Using the maximum likelihood method, -2LL (log likelihood) was used to check the deviance; the smaller the deviance value, the better the model's suitability and validity. Statistical analysis was performed using SPSS 23.0 (SPSS Inc., Chicago, IL, USA) and HLM 7.0 (Scientific Software International Inc., Chicago, IL, USA) programs.
This study obtained approval from the Institutional Bioethics Committee of the Korea Centers for Disease Control and Prevention for the use of non-identifiable secondary data and from the Bioethics Committee of Gangneung-Wonju National University (IRB No. GWNUIRB-2016-09).

Annual Professional Fluoride Application and Scaling Experience Rate
Professional fluoride application was more often experienced by younger adolescents, and scaling was more frequent in older adolescents (p < 0.001). In addition, the higher the grade, the more preventive treatment was experienced (p < 0.001). Regarding socio-economic factors, the higher the father's education level, and the higher the FAS level, the higher the experience rate (p < 0.05). Regarding oral health behavioral factors, students who used to brush after lunch always/almost always, when eating snacks more than three times a week, experienced more fluoride application and scaling. In addition, when students experienced oral health symptoms such as tooth fracture, toothache, gingival bleeding, soft tissue pain, and bad breath, the rate of preventive treatment was high (Table 1).

Relationship Between Personal Characteristics, Annual Professional Fluoride Application, and Scaling Experience
After adjusting for demographic factors, high-FAS-level students experienced 1.50 times more fluoride applications than those with low FAS levels (p < 0.001), and students that received an allowance of more than 20,000 won per week experienced 1.11 times more fluoride applications than those who received less than 10,000 won per week (p = 0.002). Regarding oral health behavior factors, the more frequent the tooth brushing after lunch, the more fluoride application was experienced (<0.001) ( Table 2).
After adjusting for demographic factors, students with a higher paternal education level experienced 1.28 times more scaling. In addition, students with high FAS levels were found to experience 1.90 times more scaling than those with low FAS levels (p < 0.001). Regarding oral health behavior factors, increased tooth-brushing after lunch was associated with more scaling (p < 0.001).

Multilevel Analysis of Annual Professional Fluoride Application Experience
The deviance of the final model was 178,160.92, which was lower than that of the null model (198,038.68). Regional-level factors affecting annual professional fluoride application were found to be the rate of receiving oral health education at school by region and the number of private educational institutes per 1000 people. A higher rate of students receiving oral health education at school by region was associated with increased fluoride applications (1.04 times, p = 0.003).
On the other hand, more private educational institutes per 1000 people were associated with fewer fluoride applications (0.64 times, p = 0.039). Regarding the annual experience of professional fluoride application, the annual experience of adolescents with gum bleeding was 1.19 times more than that of adolescents without gum bleeding (p < 0.001) ( Table 3).

Multilevel Analysis of Annual Scaling Experience
The deviance of the final model was 165,244.34, which was decreased compared to that of the null model (207,474.60). The regional-level factors affecting the annual scaling experience were the number of dentists per 10,000 individuals and the social deprivation index. The experience of scaling treatments was higher when there were more dentists per 10,000 individuals (1.14 times, p = 0.008), while a higher social deprivation index was linked to fewer scaling treatments (0.88 times, p = 0.024). The annual scaling experience was found to be 1.29 times higher for adolescents with gum bleeding than for those without gum bleeding (p < 0.001) ( Table 4). The data were analyzed by complex sample analysis (chi-squared test). The data were analyzed by complex sample logistic regression. Values are presented as odds ratio (95% confidence interval). OR, odds ratio; CI, confidence interval. Table 3. Multilevel analysis of annual professional fluoride application experience considering individual and regional characteristics. The data were analyzed by multilevel logistic regression. Values are presented as odds ratio (95% confidence interval). OR, odds ratio; CI, confidence interval. Table 4. Multilevel analysis of annual scaling experience considering individual and regional characteristics. The data were analyzed by multilevel logistic regression. Values are presented as odds ratio (95% confidence interval). CI, confidence interval.

Discussion
This study identified the individual-and regional-level factors that influence preventive dental treatment in Korean adolescents by using government-approved National Statistical Data that is representative of Korean adolescents.
The annual professional fluoride application experience was found to be 1.04 times higher in areas with a higher rate of receiving school oral health education by region. These findings were consistent with previous studies that showed that enhancing children's access to medical care in community-based programs was associated with increased preventive care [29].
In Korea, school health education is conducted in two ways. First, as part of the oral health program under the Oral Health Act of Korea, dental hygienists working at public health centers visit elementary, middle, and high schools to provide oral health education to students in schools [30]. The health promotion activities of a community may vary depending on the policies of the local government and the characteristics of local residents [31]. If there is a problem with the health condition and behavior of the local adolescents or if the residents' demand for youth health is high, the health center will be interested in the students' health and will implement the health programs. Second, through a 'health' class, which is one of the school curricula, a school nurse provides oral health education [32]. The middle and high schools in Korea consider a university entrance-oriented education as the top priority and conduct classes according to this orientation. The Ministry of Education in Korea sets the health class as an elective or liberal arts class, and allows it to be conducted at the discretion of the principal [32]. In fact, high school students only do self-study for subjects not included in university entrance exams, which focus on subjects such as Korean, math, English, science, and social science.
Despite this social environment, the high proportion of oral health education observed in local schools means that principals who oversee schools' class curriculum are interested in and consider students' oral health important. The high percentage of individuals who have benefited from local schools' oral health education means that local schools and health public centers have a high interest in adolescents' oral health. Therefore, these results may indicate that public health centers' and schools' interest in oral health of the students can significantly affect adolescents' preventive treatment experience.
In this study, adolescents living in regions with more private educational institutes experienced 36% fewer professional fluoride applications. The results of this study suggest that the greater the commitment to learning in a region, the lower the interest in and the priority given to preventive care, and the less likely that such care is to be experienced. In a previous study, which was analyzed using the Korean National Health and Nutrition Survey data, the main cause of the unmet dental care needs in adolescents was that "the school could not be vacated (male and female average 35.94%);" this finding supports the results of this study [33]. In Korea, adolescents spend more than half of their daily life studying, which implies that their physical and mental health is poor. According to the Organization for Economic Co-operation and Development (OECD)'s Program for International Student Assessment, 23% of Korean adolescents spend at least 60 h per week studying, both in and out of school (OECD average: 13%) [34]. Moreover, after school, students are engaged in an additional 2-5 h of extra-curricular activities. Because of this academic-oriented lifestyle, Korean adolescents have high levels of stress with regard to poor grades, with a low percentage of leisure time and physical activity [35]. Additionally, higher academic stress leads to increased unhealthy behaviors, such as smoking and drinking [36].
The World Health Organization (WHO) encourages schools to promote health by creating a physical environment in which the health risks of students can be addressed, including regular preventive dental care [37]. In 2009, Korea started a health promotion school project, and as of 2013, it was operative in 85 schools [38]. This accounts for only 0.7% of the total number of schools, and as it mainly focuses on general health, oral health intervention has been insufficient. In order to solve the problems of academic stress, unhealthy behavior, and unmet dental care among the youth, it is necessary to activate the WHO-recommended health promotion school program.
Adolescents in regions with more dentists per 10,000 individuals were found to experience 1.14 times more scaling. Additionally, with a higher social deprivation index, 12% fewer scaling treatments were experienced. This was consistent with previous studies, which reported that regions with lower socioeconomic levels had less access to medical resources because dental resources were concentrated in large metropolitan regions with high socioeconomic standards [13]. The number of dentists in the region was significantly associated with the annual scaling experience, but not with the annual professional fluoride application experience.
In 2013, regarding the ranking of the total cost of medical care benefits by disease among adolescents aged 15 to 19 years, dental caries was reported to rank fourth, whereas gingivitis and periodontal disease ranked sixth [39]. This report means that many adolescents visited the dental clinic when they had oral symptoms, such as dental caries and gingivitis, and paid for and received treatment from the dentist. In addition, Korea's national health insurance mainly covers treatment items. Preventive measures, such as professional fluoride application, are not covered by the national health insurance; thus, dental visits for treatments are more frequent. Therefore, the number of dentists only affects the scaling experience, which is considered to be the effect of Korea's treatment-oriented medical system. It is also thought that the impact of the number of dentists in the region on preventive care is small and contributes to improving access to dental visits for treatment.
The WHO encourages schools to run disease prevention and health promotion programs because schools are well placed to promote lifelong healthy behaviors and serve as a platform to access families and communities [40]. In the USA, school-based health centers offer comprehensive services to students and families at minimal or no cost [41]. In Georgia, USA, legislation has been enacted to allow dental hygienists to provide preventive care in schools if there are few dentists in an area and the distance from dental services is significant [42]. In Korea, the School Oral Health Center is currently installed in only 2.8% of 6145 elementary and special schools nationwide, centered on vulnerable socio-economic areas, and middle and high schools do not have this facility [43]. There are no full-time dentists and dental hygienists in school dental clinics, and dentists and dental hygienists working at the local public health center regularly go on business trips to schools, providing students with preventive care such as oral examinations, fluoride applications (solution, gel, or varnish), sealants, scaling, and education. As of 2019, 1130 dental hygienists and 420 dentists are employed in South Korea's public health sector [26]. If school dental clinics are expanded and operated in the same way as they are now, it is expected that there will be a shortage of dental workforce. Therefore, the government should support the expansion of oral health programs by seeking ways to utilize the dental workforce in the public sector with related associations.
In order to promote the oral health of middle and high school students, active participation and changes in public health centers, associations, and government are necessary. To promote school-based oral health, the related government agencies, such as the Ministry of Education and the Ministry of Health and Welfare, should promote cooperation and joint research with specialized institutions, such as the Dentist Association and the Dental Hygiene Association, to develop standardized school-based education programs. Furthermore, the developed programs should be piloted, and the results should be assessed to accumulate evidence. In addition, they should also work on legal measures to enable dentists or dental hygienists to provide comprehensive oral care services to students full time in schools, similar to the practice in developed countries. Public health center officials should be interested in the oral health of adolescents and implement oral health promotion programs for adolescents.
This study was limited in that several factors could not be considered, given that it utilized secondary data, and actual dental treatment data were not used. In addition, this study used self-completed surveys on dental treatment experience, which could be confounded by re-call and subjective interpretations. To identify the factors related to adolescents' oral health behavior, it is necessary to conduct a multilevel analysis that encompasses three levels (individual, school, and region levels) by adding school-level factors, as teenagers spend most of their time in school. Nevertheless, this study was meaningful in that it identified factors that affect the preventive treatment experience of Korean adolescents at individual and regional levels by using National Statistics Data that are representative of Korean adolescents.

Conclusions
This study confirmed that adolescents' access to preventive dental treatment was influenced not only by individual level factors but also by regional factors. To improve access to preventive care for Korean adolescents, it is necessary to establish a primary dental care system through the cooperation of government, private dental clinics, and schools, so that students can regularly receive dental care. In socioeconomically vulnerable regions with low access to dental care, dental teams should be utilized through the school environment to create a system that provides students with continuous oral health care, and to establish a medical delivery system where health centers and private dentists can collaborate to provide treatment if necessary.