Data from the KNHANES, which was conducted between 2008 and 2010 by the Division of Chronic Disease Surveillance under the Korea Centers for Disease Control and Prevention and the Korean Ministry of Health and Welfare, were used for the present study (12). KNHANES is a nationwide study of non-institutionalized civilians that uses a stratified and multi-stage probability sampling design with a rolling survey-sampling model. Sampling units were based on the population and housing consensus from the 2005 National Census Registry in Korea, which includes age, gender and geographic area (13). In total, 20,730 individuals aged ≥19 years were targeted for the survey, and 19,219 participants with complete data, including SES, were used in the present study.

All participants were asked about their lifestyle and socioeconomic characteristics by trained interviewers. Age, alcohol consumption, smoking status, residential area, daily energy intake, fat intake and marital status were considered possible confounding factors in the present study.

Participants were divided into three groups according to the amount of alcohol consumption reported per day: Non-drinker, light to moderate drinker (1–30 g/day) and heavy drinker (>30 g/day) (14). Smoking status was categorized into three groups in accordance with respondents' answers on the self-report questionnaire: (i) Non-smokers, those who had never smoked or had smoked <100 cigarettes in their life; (ii) ex-smokers, those who had smoked in the past but had stopped smoking; and (iii) current smokers, those who were smoking currently and had smoked ≥100 cigarettes in their life (15). Individuals were regarded as regular physical exercisers if they performed moderate exercise at least 5 times per week for at least 30 min per session, or performed vigorous exercise at least 3 times per week for at least 20 min per session (16).

A face-to-face interview was conducted to obtain data about place of residence (rural vs. urban) and occupation of the participants (17). Residential areas were categorized into urban areas (administrative divisions of ‘dong’) and rural areas (administrative divisions of ‘eup’ or ‘myeon’) (14). Daily food intake was assessed using the 24-h recall method, and a food frequency questionnaire was used to determine food consumed the previous day. Energy and fat intake were calculated using a food database developed for the KNHANES and the food composition table published by the National Rural Living Science Institute under the Rural Development Administration (18).

In the present study, SES was assessed by monthly household income and education level. Monthly household income levels were divided into quartiles. The first quartile included households with a monthly income <1092.4 USD, the second quartile included those with an income of 1092.4–2334.3 USD, the third included those with an income of 2334.3–3734.8 USD and the fourth included those with an income >3734.8 USD. Education levels were categorized into four groups according to the number of years of schooling completed: (i) <7 years (elementary school graduate or lower); (ii) 7–9 years (middle school graduate or some middle school); (iii) 10–12 years (high school graduate or some high school); and (iv) >13 years (university or higher) (14).

Body weight and height were measured to the nearest 0.1 kg and 0.1 cm, respectively, whilst participants were wearing light indoor clothing without shoes (19). Body mass index (BMI) was calculated using the following formula: BMI=body weight (kg)/height² (m²). Waist circumference was measured at the narrowest point between the lower border of the rib cage and the iliac crest.

In the present study, the time of day when tooth brushing was performed and the use of secondary oral products were assessed. Times of day included: Before or after breakfast, lunch, dinner and before sleeping. The frequency of daily tooth brushing was conducted according to the total number of tooth brushing incidences per day. Secondary oral products included dental floss, mouthwash, interdental brushes and electric toothbrushes (15).

The World Health Organization community periodontal index (CPI) was used to assess periodontal treatment needs and periodontal disease was defined as CPI≥3 (20). A CPI score of 3 indicated that ≥1 site had a >3.5 mm pocket in the index teeth, which are 11, 16, 17, 26, 27, 31, 36, 37, 46 and 47 according to the Federation Dentaire Internationale system (21). The mouth was divided into sextants. A CPI probe (PWHO; Osung MND Co., Ltd., Seoul, Korea) with a 0.5 mm ball tip was used. A sextant was examined only if there were ≥2 teeth present that were not scheduled for extraction. If no index teeth were present in a sextant qualifying for examination, all remaining teeth were examined and the highest score was recorded as the score for that sextant. A probing force of ~20 g was used. Trained and calibrated dentists examined the periodontal status of the participants, and the mean inter-examiner Kappa value was 089 [95% confidence interval (CI), 0.55–1.00] (15,22–23).

Chewing and speech ability, attendance at a dental checkup within a year and incidence of caries in permanent teeth were evaluated.

Data were presented as either mean ± standard error or as percentage (standard error), as indicated. Statistical analyses were performed using the survey procedure of the SAS 9.2 statistical software package for Windows (SAS Institute, Inc., Cary, NC, USA) to account for the complex sampling design. Two-sided P<0.05 were considered to indicate a statistically significant difference. Analysis of variance was applied to examine the relationship between the daily frequency of tooth brushing, use of secondary oral products per day, and SES.

Multivariate logistic regression analyses were applied to examine the odds ratios and 95% CI of tooth brushing ≥3 per day and the use of secondary oral products according to the SES. There are four logistic regression models. Model 1 was adjusted with age, gender and BMI. Adjustments were then made for the same variables as Model 1 plus smoking, drinking, and regular physical exercise (Model 2). Adjustments were made for the variables in Model 2 plus energy intake, fat intake, and periodontal treatment needs (CPI≥3) (Model 3). In addition, adjustments were made for the variables in Model 3 plus education or income (Model 4).

Table I describes the characteristics of the study population, whom were grouped according to the daily frequency of tooth brushing and the use of secondary oral products. Tooth brushing frequency and the use of secondary products differed significantly depending on various factors. Individuals who brushed ≥3 times per day were younger with lower BMI, lower waist circumference and less periodontal treatment needs. In current smokers, 35.8% brushed less than twice, 30.4% brushed twice and 24.2% brushed ≥3 times per day. Similar trends were detected in the use of secondary oral products (Table I). It should be noted that the characteristics of the study population differed between the groups categorized by the number of times tooth brushing was carried out and the use of secondary oral products.

Participants who brushed ≥3 times per day according to age are presented in Fig. 1. The incidence of individuals who brushed ≥3 times per day decreased with increasing age. The use of secondary oral products according to age is presented in Fig. 2. Participants aged between 30 and 39 years exhibited the highest percentage of secondary oral product usage. The results suggest that age was associated with oral health behaviors.

It was demonstrated that the time of day that tooth brushing occurred differed according to socioeconomic factors (Fig. 3). At the majority of the time points, participants in the highest quartile of income were most likely to brush their teeth. Similarly, individuals with the highest education were more likely to brush their teeth at the most of the time points. Socioeconomic factors were also associated with the use of secondary oral products (Fig. 4). Participants in the highest quartile of income and highest education were more likely to use dental floss, interdental brushes, mouthwash and electric toothbrushes. Participants who brushed ≥3 times per day exhibited the highest proportion of individuals in the highest quartile of income, and individuals who used more than one type of secondary oral product had a higher proportion in the highest quartile of income and education level, as compared with the other groups (Table II). Furthermore, participants in the highest quartile of income were most likely to have had a dental examination within the past year, as compared with the other quartiles (Table III). Similarly, individuals who had completed a higher level of education (high school graduate or higher) were more likely to have had a dental examination within the past year, as compared with the lower levels of education, which were defined as less than a high school degree. The results suggested that socioeconomic factors are associated with oral health behaviors.

Tables IV and V present the adjusted odds ratios and 95% CIs from multiple logistic regression analyses. The association detected between the frequency of tooth brushing and SES remained after all adjustment models (Table IV). Adjusted odds ratios and their 95% CIs were 1.086 (0.941–1.253), 1.158 (1.006–1.334), and 1.264 (1.094–1.460) for the second, third and fourth quartiles of income, respectively (Model 4). Adjusted odds ratios and their 95% CIs for the middle school, high school, and university or higher graduate groups were 1.475 (1.268–1.716), 1.765 (1.537–2.026), and 2.686 (2.286–3.155), respectively (Model 4). The association between the use of secondary oral products and SES remained after adjustment (Table V). Adjusted odds ratios and their 95% CIs were 1.266 (1.092–1.469), 1.599 (1.371–1.865), and 1.835 (1.559–2.161) for the second, third and fourth quartiles of income, respectively (Model 4). Adjusted odds ratios and their 95% CIs for the middle school, high school, and university or higher graduate groups were 2.231 (1.849–2.692), 3.669 (3.085–4.364), and 5.736 (4.734–6.951), respectively (Model 4). The results demonstrate an association between SES and oral health behaviors in a large sample of the Korean population.