Oral Health Knowledge, Attitudes, and Behaviors of Undergraduate Students Enrolled in the Faculty of Dental Medicine at the Lebanese University

ABSTRACT

Aim:

To evaluate the oral health-related knowledge, attitude, and behavior of dental students in Lebanon.

Materials and Methods:

A cross-sectional survey was conducted during the spring semester of the academic year 2024/2025. After giving informed consent, students completed a self-administered questionnaire collecting socio-demographic data, oral health knowledge, attitudes, and behaviors (KABs), and general health behaviors. The validated Arabic version of the Hiroshima University Dental Behavioral Inventory (HU-DBI) was used for the KAB questions, and a translated version of the English-modified version was used for the remaining questions.

Results:

A total of 246 students participated. The mean overall HU-DBI score was 6.71 ± 1.55. Mean KAB scores were 3.58, 1.16, and 1.97, respectively. The multiple linear regression model showed that none of the variables could statistically significantly predict the HU-DBI total score or the behavior score, while the model significantly predicted the knowledge and attitude scores. After controlling for confounders, sex and regular checkups were statistically significant predictors of the knowledge subscale, and problematic internet use was the only statistically significant predictor of the attitude subscale.

Conclusion:

An HU-DBI score within the international range was observed among Lebanese dental students, with few factors significantly predicting some subscale scores: sex and regular checkups predicted knowledge; problematic internet use predicted attitude; alas, this prediction was very weak. These findings highlight the need to use a holistic approach of improving knowledge, skills, and attitude in order to increase the chances of behavioral change and to incorporate the impact of behavioral factors on oral health into the dental curriculum so that dental students are better informed.

INTRODUCTION

Despite advancements in oral health in industrialized countries, oral diseases, such as dental caries and periodontal diseases, remain a global burden.[1,2] Oral diseases are often associated with systemic health conditions, either exacerbating the systemic condition or being influenced by it.[1,3,4] Maintaining good oral health through the use of toothbrushes, toothpastes, and interdental aids is the most effective method for preventing oral diseases.[5] However, despite the simplicity of plaque control measures, most individuals struggle to effectively comply with these measures.[6,7] This noncompliance is strongly behavioral and influenced by several factors such as personality, attitude, lifestyle, education, and socio-demographic characteristics.[6,8] Gaining sufficient knowledge about oral health and consistently practicing optimal oral hygiene are essential for maintaining adequate overall oral health.[5,6]

Dentists’ attitudes toward their own oral health significantly impact the quality of patient care.[9,10] The oral health attitudes and behaviors of dental professionals, including students, serve as indicators of their understanding of health-promoting procedures and reflect their comprehension of disease control’s value and their role in primary prevention.[9,11] To be able to effectively promote oral health among patients, family, and friends, oral healthcare professionals must develop a strong commitment to personal oral health.[11,12,13,14] By adopting positive oral health approaches, these professionals can better equip the general population with proper oral health knowledge and practices.[9,11,15]

The Hiroshima University Dental Behavioral Inventory (HU-DBI), developed by Makoto Kawamura in 1988, is a psychometric tool for evaluating dental students’ oral health attitudes and behaviors.[16] This rigorously validated tool has demonstrated strong psychometric properties and is widely adopted, especially since it is translated into multiple languages, enabling global comparisons of dental students’ oral health perspectives.[17,18,19,20] The HU-DBI was translated into Arabic and validated by Daou et al.[19] in a sample of dental students in Lebanon in 2018. Moreover, the HU-DBI can evaluate sex differences in oral health attitudes and can guide the development of programs to enhance dental students’ behaviors.[11,21]

In Lebanon, the prevalence of dental caries among children is high, and there is a lack of effective national strategies to promote oral health and, subsequently, decrease the incidence of this preventable disease.[22] Since dental professionals, including students, can better equip the general population with proper oral health knowledge and practices, the aim of this study was to evaluate the oral health-related knowledge, attitude, and behavior of students in the Faculty of Dental Medicine at the Lebanese University using the Arabic version of the questionnaire. By introducing this validated instrument to a Lebanese student population, this study will not only enable cross-cultural comparisons but will also provide valuable insights into local behavioral patterns that may influence oral health outcomes. The findings can inform the development of targeted educational interventions and public health strategies tailored to the specific needs of Lebanese youth, thereby contributing to the broader goal of improving oral health literacy and practices in the region. The main objective was to assess oral health knowledge, attitudes, and behaviors (KABs) using HU-DBI among these students. Secondary objectives were to assess the influence of sex, academic level, and clinical experience on students’ oral health KAB and to examine the relationship between oral health KAB and risk behaviors, such as tobacco use, alcohol consumption, and problematic internet use.

MATERIALS AND METHODS

STUDY DESIGN AND PARTICIPANTS

A cross-sectional survey was conducted during the spring semester of the academic year 2024/2025 among undergraduate students aged ≥18 years enrolled at the Faculty of Dental Medicine at the Lebanese University – the only public university in Lebanon. Students were classified into two categories depending on their academic year: “Pre-clinical students” enrolled in the 2nd or 3rd year, and “clinical students” enrolled in the 4th or 5th year.

The study was approved by the scientific committee of the Faculty of Dental Medicine under the number RCSC/01-04-2025, prior to its initiation. Student participation was voluntary, and participants were provided with a detailed explanation of the purpose of this study prior to filling out the questionnaire. Students also signed a written informed consent prior to their participation. All participants were kept anonymous as no personal data was collected (e.g., participant’s initials or student university identification number).

STUDY QUESTIONNAIRE AND VARIABLES

The manual self-administered survey covered three different categories: Category 1, socio-demographic characteristics (three items – age category, sex [assigned at birth] and current academic year of enrollment); Category 2, oral health KAB (20 items of the original HU-DBI questionnaire [i.e., agree/disagree questions]); Category 3, general health behavior (four items – tobacco smoking, alcohol consumption, problematic internet use, and regular dental checkups).[15] The validated Arabic version of the oral health KAB questions was used for Category 2 questions.[19] The remaining two categories, collecting data on socio-demographic characteristics and the general health behavior questions, were translated from the English-modified version of HU-DBI.[23]

Supplementary Table 1 presents the modified English version of the questionnaire, constituting the 24 items of the 2nd and 3rd categories and the scoring system employed. The higher the score, the better the oral health knowledge, attitude, and behavior, with 12 being the highest score. Items 2, 8, 10, 15, and 19 correspond to knowledge, items 6, 11, and 14 correspond to attitude, and items 4, 9, 12, and 16 correspond to behavior.[23]

DATA COLLECTION

After taking the approval from the scientific committee and the faculty administration, the study investigators contacted the class representatives (i.e., one representative for each academic year) and provided them with printed copies of the survey. We opted to selecting students to take the lead and administer the questionnaires for their peers in order to avoid any potential influence or pressure from professors. This approach provided the students with greater autonomy and ensured that responses remained free from social bias.

The class representatives then distributed the printed copies to the students. Students who wished to participate and signed the informed consent completed the survey. Participants were also given the opportunity to complete the questionnaires at their own pace and in private, rather than in front of their peers. As such, the participants took the questionnaires with them and returned the filled form to a designated box. Class representatives then collected the filled forms. This process was followed so that participants were not influenced by their professors or peers and to prevent the social desirability effect and encourage more honest and unbiased responses. The representatives handed the forms to the study investigators. Of note, questionnaires were filled out anonymously, which guarded the integrity and confidentiality of participants’ responses.

STATISTICAL ANALYSIS

We did not perform formal sample size calculations since we invited all students enrolled at the Faculty of Dentistry at the Lebanese University to participate in the study. The rationale behind such a decision was that there are only three universities with dental programs in Lebanon, all of which are located in Beirut or its suburbs. This university was selected since it is the only public university in Lebanon, and as such, the students enrolled there would represent all geographical areas and social categories in Lebanon.

Data were imported to Excel and later coded. Data were then exported to the Statistical Package for the Social Sciences (SPSS Inc., Chicago, IL, USA) version 26 for statistical analysis. Categorical variables were presented as frequencies and percentages. HU-DBI subscales and total score were reported as mean ± standard deviation (SD). There were no missing values, so no computations were made. For the univariate analysis, the chi-square, Fisher’s Exact, Mann–Whitney, and Kruskal–Wallis tests were employed as appropriate to assess associations and differences between variables, with the significance level corresponding to P value (p) ≤ 0.05. Post hoc analysis using Bonferroni correction was reported for Kruskal–Wallis tests with a significant P value. The effect size (Cohen’s d) was calculated using an online calculator for Mann–Whitney and Kruskal–Wallis tests.[24] The effect size for the chi-square test (Cohen’s ω) was calculated using the formula: ɸ = √(X²/N). Risk estimates using odds ratios (OR) and their 95% confidence intervals (95% CIs) were also reported. Multiple linear regressions were performed to determine the degree of variation of the dependent variables (HU-DBI total score, knowledge subscale, attitude subscale, and behavior subscale) that may be explained by the independent variables, controlling for confounders. All the linear regression’s assumptions were meticulously evaluated. Linearity was assessed by partial regression plots and a plot of studentized residuals against the predicted values. Independence of residuals was assessed by a Durbin–Watson statistic ranging between 1.5 and 2.5. Homoscedasticity was assessed by visual inspection of a plot of studentized residuals vs. unstandardized predicted values. Multicollinearity absence was assessed by tolerance values greater than 0.1 and variance inflation factor values less than 10. The models were further assessed to confirm that there were no studentized deleted residuals greater than ±3 SDs, no leverage values greater than 0.2, and values for Cook’s distance above 1. Meeting the assumption of normality was also assessed by a Q-Q plot. Unstandardized coefficients, their standard errors, and 95% CIs, as well as the standardized coefficients, were reported for each regression model, and a P value ≤ 0.05 was considered statistically significant.

RESULTS

STUDY PARTICIPANTS AND SOCIO-DEMOGRAPHIC CHARACTERISTICS

Of the 260 students approached, 246 filled out the forms (response rate, 94.6%). None of the questionnaires contained missing or invalid data, and as such, none was excluded from the analysis. The mean age of participants was 21.5 years, and almost two-thirds were females (n = 162; 65.9% vs. males, n = 84; 34.1%). The enrolled sample was evenly distributed over the academic years: 121 pre-clinical (years 2 and 3) and 125 clinical (years 4 and 5) [Figure 1].

Figure 1.

Socio-demographic characteristics of participants (N = 246). N = total number of participants

GENERAL HEALTH BEHAVIORS

More than two quarters of participants (n = 176; 71.5%) stated that they visit their dentist for regular dental checkups, at least once a year, with a lower percentage in females than males (68.5% vs. 77.4%). The majority of participants stated problematic internet use (n = 213, 86.6%), and only 16.3% and 12.2% reported consuming alcohol or smoking tobacco at least once a week, respectively. While no statistically significant difference was observed between pre-clinical and clinical students regarding any of these behaviors, a statistically significant difference was observed for smoking tobacco between females and males (OR, 0.399; 95% CI, 0.200–0.793; P = 0.007, Cohen’s ω = 0.017) [Tables 1 and 2].

Table 1.

General health behaviors of participants, by sex (N = 246)

Variable	Outcome	Female (N = 162)	Male (N = 84)	P value (Cohen’s ω)	P value (Cohen’s ω)	OR 95% CI	Total (N = 246)
Problematic internet use	Yes	140 (86.4%)	73 (86.9%)	0.916 (0.0067)	0.916 (0.0067)	0.959 (0.441; 2.086)	213 (86.6%)
	No	22 (13.6%)	11 (13.1%)				33 (13.4%)
Tobacco smoking	Yes	19 (11.7%)	21 (25.0%)	0.007 (0.17)	0.007 (0.17)	0.399 (0.200; 0.793)	40 (16.3%)
	No	143 (88.3%)	63 (75.0%)				206 (83.7%)
Alcohol drinking	Yes	19 (11.7%)	11 (13.1%)	0.752 (0.0198)	0.752 (0.0198)	0.882 (0.398; 1.951)	30 (12.2%)
	No	143 (88.3%)	73 (86.9%)				216 (87.8%)
Regular checkup	Yes	111 (68.5%)	65 (77.4%)	0.144 (0.093)	0.144 (0.093)	0.636 (0.346, 1.170)	176 (71.5%)
	No	51 (31.5%)	19 (22.6%)				70 (28.5%)

Data are presented as n (%), where the percentages were calculated based on n/N. All P values were calculated using the chi-square test

95% CI = 95% confidence interval; HU-DBI = Hiroshima University Dental Behavioral Inventory; n = number of participants; N = total number of participants; OR = odds ratio (disagree/agree)

Table 2.

General health behaviors of participants, by general health behaviors (N = 246)

Variable	Outcome	Pre-clinical (N = 121)	Clinical (N = 125)	P value (Cohen’s ω)	OR 95% CI	Total (N = 246)
Problematic internet use	Yes	108 (89.3%)	105 (84.0%)	0.227 (0.077)	0.632 (0.299; 1.335)	213 (86.6%)
	No	13 (10.7%)	20 (16.0%)			33 (13.4%)
Tobacco smoking	Yes	17 (14.0%)	23 (18.4%)	0.355 (0.0589)	1.379 (0.696; 2.733)	40 (16.3%)
	No	104 (86.0%)	102 (81.6%)			206 (83.7%)
Alcohol drinking	Yes	10 (8.3%)	20 (16.0%)	0.064 (0.118)	2.114 (0.946; 4.727)	30 (12.2%)
	No	111 (91.7%)	105 (84.0%)			216 (87.8%)
Regular checkup	Yes	82 (67.8%)	94 (75.2%)	0.197 (0.082)	1.442 (0.826; 2.517)	176 (71.5%)
	No	39 (32.2%)	31 (24.8%)			70 (28.5%)

Data are presented as n (%), where the percentages were calculated based on n/N. All P values were calculated using the chi-square test

95% CI = 95% confidence interval; HU-DBI = Hiroshima University Dental Behavioral Inventory; n = number of participants; N = total number of participants; OR = odds ratio (disagree/agree)

HIROSHIMA UNIVERSITY DENTAL BEHAVIORAL INVENTORY RESPONSES

The vast majority of participants (89.4%) reported worrying about their teeth color (item 3), while less than 10% of participants agreed with disliking the color of their gums (item 7; 6.1%) and on the previous usage of plaque-disclosing agent (item 16; 7.7%). A similar agreement rate of 9.8% was obtained for using a toothbrush with hard bristles (item 17). On the contrary, brushing each tooth carefully (item 9) showed a high agreement level of 80.5%. The latter two items (items 9 and 17) correspond to the correct oral hygiene practices [Supplementary Table 2].

HU-DBI responses are presented by sex, clinical experience, and smoking status in Supplementary Tables 2,3,4, respectively. A higher percentage of females reported disagreement with bleeding gums while brushing (item 2) and agreement with worrying about the color of their gums (item 7) compared to males (item 2; 33.3% vs. 13.6%; OR, 0.314; 95% CI, 0.166–0.595; P < 0.001; item 7, 11.9% vs. 3.1%; OR, 0.236; 95% CI, 0.078–0.714; P = 0.006). On the contrary, a higher percentage of males reported agreement with worrying about the color of their teeth (item 3) and having bad breath (item 13) compared to females (item 3, 93.2% vs. 82.1%; OR, 2.984; 95% CI, 1.303–6.834; P = 0.007; item 13, 88.9% vs. 77.4%; OR, 2.338; 95% CI, 1.152–4.747; P = 0.017). Compared to clinical students, a higher percentage of pre-clinical students reported agreement with worrying about their teeth color (item 3, 95.0% vs. 84.0%; OR, 0.274; 95% CI, 0.106–0.708; P = 0.005), noticing white sticky deposits on their teeth (item 4, 33.1% vs. 18.4%; OR, 0.457; 95% CI, 0.253–0824; P = 0.008), and checking their teeth after brushing (item 12, 90.9% vs. 76.0%; OR, 0.317; 95% CI, 0.151–0.666; P = 0.002). In addition, a higher percentage of pre-clinical students reported disagreeing with not being able to help having false teeth when they are old (item 6, 43.0% vs. 30.4%; OR, 0.580; 95% CI, 0.343–0.979; P = 0.041) and never being taught professionally how to brush their teeth (item 10, 28.9% vs. 17.6%; OR, 0.525; 95% CI, 0.287–0.961; P = 0.035). More clinical students reported agreement with using a child-sized toothbrush (item 5) and thinking about cleaning their teeth well without using toothpaste (item 11) compared to pre-clinical students (item 5, 56.8% and 35.5%; OR, 2.385; 95% CI, 1.427–3.986; P = 0.001; item 11, 20.0% vs. 8.3%; OR, 2.775; 95% CI, 1.270–6.063; P = 0.008). The only significant difference observed between smokers and nonsmokers was the lower percentage of smokers disagreeing with the thought that their teeth are getting worse despite their daily brushing (10.0% vs. 24.8%; OR, 0.338; 95% CI, 0.115–0.995; P = 0.040).

Taking into consideration the academic year of enrollment, a statistically significant difference in agreement for 6/24 items was observed (Categories 2 and 3): bleeding gums while brushing (item 2; P = 0.043), worrying about teeth color (item 3; P = 0.020), using a child-sized toothbrush (item 5; P = 0.009), thinking about cleaning their teeth well without using toothpaste (item 11; P = 0.036), checking teeth in the mirror after brushing (item 12; P = 0.018), and going to the dentist/hygienist for regular checkup at least once a year (item 24; P = 0.032). The highest agreement level was observed for item 3 (98.4%) at year 2, with a steady decrease over the following three academic years (91.2%, 85.2%, and 82.8% at years 3, 4, and 5, respectively) [Supplementary Table 5].

HIROSHIMA UNIVERSITY DENTAL BEHAVIORAL INVENTORY SCORES

Mean ± SD overall HU-DBI score was 6.71 ± 1.55 (range, 2–11), with a mean knowledge score of 3.58 ± 1.11 (range, 1–5), attitudes score of 1.16 ± 0.82 (range, 0–3), and behaviors score of 1.97 ± 0.73 (range, 0–4). Female students had a non-statistically significantly slightly higher HU-DBI score (6.75 ± 1.52) than their male colleagues (6.63 ± 1.60). The sex-based differences were more pronounced in the knowledge domain of oral health, with a significant difference between both sexes (3.69 ± 1.06 vs. 3.37 ± 1.19, respectively; P = 0.047; Cohen’s d = 0.245) [Figure 2, Supplementary Table 6].

Figure 2.

Box plots of different comparisons for the HU-DBI subscales – univariate analysis. (A) Knowledge score by academic year, (B) Knowledge score by regular checkup status, (C) Attitude by clinical experience, (D) Attitude by problematic internet use, (E) Behavior by clinical experience

The overall HU-DBI score for pre-clinical participants was 6.60 ± 1.43, which was slightly lower than clinical participants (6.82 ± 1.66), albeit the difference was not statistically significant. A significant difference was obtained for the attitude and behavior subscales. For the attitude subscale, a lower mean score was reported for pre-clinical students (1.02 ± 0.74 vs. 1.30 ± 0.87 for clinical; P = 0.012; Cohen’s d = 0.3), while the opposite was observed for behavior (2.10 ± 0.68 vs. 1.85 ± 0.76, respectively; P = 0.008; Cohen’s d = 0.307) [Figure 2, Supplementary Table 6]. The overall HU-DBI score increased steadily from the second to the third and fourth academic years and then decreased in the fifth academic year, with a non-statistically significant difference. Whereas for knowledge, a statistically significant difference (P = 0.004; Cohen’s d = 0.417) was observed; the mean HU-DBI score increased from 3.22 ± 1.06 at the 2nd year to 3.77 ± 1.10 and 3.84 ± 1.02 at the 3rd and 4th years, respectively. This increase was later followed with a decrease from 4th to 5th year (3.53 ± 1.18) [Figure 2, Supplementary Table 6]. The year-to-year analysis, using the Kruskal–Wallis test, showed that this difference in the knowledge score between the academic years is statistically significant (P = 0.004). A post hoc analysis was performed to check significant score differences between the academic years. Knowledge scores showed a significant difference between 2nd and 3rd year (mean scores are 3.22 and 3.77, respectively; Bonferroni P = 0.026) and between 2nd and 4th year (mean scores 3.22 and 3.84, respectively; Bonferroni P = 0.006).

There were no differences in the mean scores for the investigated general health behaviors: tobacco and alcohol consumption, while significantly higher scores were observed for problematic internet use for the attitude score (yes vs. no, 1.11 ± 0.81 vs. 1.48 ± 0.80; P = 0.011; Cohen’s d = 0.305) and for regular dentist checkups for the knowledge score (yes vs. no, 3.67 ± 1.09 vs. 3.36 ± 1.16, respectively; P = 0.039; Cohen’s d = 0.255) [Figure 2, Supplementary Table 7].

MULTILINEAR REGRESSION ANALYSIS

The multiple linear regression model showed that none of the variables could statistically significantly predict the HU-DBI total score (F = 0.82, P = 0.57 > 0.05) or the HU-DBI behavior subscale (F = 1.33; P = 0.24) [Figure 3, Supplementary Table 8]. While the regression model statistically significantly predicted HU-DBI knowledge subscale (F = 2.145; P = 0.04; R² = 0.06) and the attitudes subscale (F = 2.17; P = 0.04; R² = 0.06).

Figure 3.

Forest plots of multivariate analysis of the HU-DBI total score and subscales. (A) HU-DBI total score, (B) Attitude subscale, (C) Behavior subscale, (D) Knowledge subscale

After controlling for confounders, sex (B = 0.39; 95% CI, 0.093–0.69; β, 0.17, P = 0.01), and regular checkups (B = 0.32; 95% CI, 0.01–0.64; β, 0.13, P = 0.04) were statistically significant predictors of the HU-DBI knowledge subscale, showing that being a female and having regular checkups resulted in higher knowledge score. Problematic internet use (B = −0.39; 95% CI, −0.69 to −0.09; β, −0.16, P = 0.01) was the only statistically significant predictor of the HU-DBI attitude subscale showing that having problematic internet use slightly lowers the scores [Figure 3, Supplementary Table 8].

DISCUSSION

Using the Arabic translated version of the HU-DBI tool, this cross-sectional survey evaluated the oral health attitudes and behaviors of dental students enrolled at the Lebanese University. The mean ± SD overall HU-DBI score was within international ranges at 6.71 ± 1.55, a mean score higher than that reported by other studies conducted on dental students in several Middle Eastern countries, including Syria (6.38 ± 1.83), Yemen (4.9), and Turkey (5.95 ± 1.65)[11,15]; however, lower than the mean score reported from Germany (7.67 ± 1.32), United Kingdom (7.33), Czech Republic (9.34 ± 1.29), and Slovakia (7.56 ± 1.73).[21,25,26] These findings highlight the differences between student behavior as well as systemic and educational differences across the different regions and countries.

When comparing between the sexes, the results of the multivariate analysis showed that sex was a very weak predictor of the knowledge subscale, where females tended to have a higher knowledge score (B = 0.39; 95% CI, 0.093–0.69; β, 0.17, P = 0.01); however, this difference was not evident in the other two domains or in the total HU-DBI score. It is important to note that there was a female predominance in this study (65.9%), mainly due to the majority of dental students at the Lebanese University being females. Moreover, significant differences were observed between the sexes for worrying about the general aesthetics, namely the color of the teeth (item 3) and gums (item 7). A higher percentage of females reported worrying about the color of their gum (11.9% vs. 3.1% for males), while more males were concerned about the color of their teeth (93.2% vs. 82.1% for females). The latter finding contradicts with what has been reported in other countries, such as the United Arab Emirates and Croatia, where females showed more interest compared with males.[27,28] Indeed, a greater demand for teeth whitening services was requested by females, highlighting their dissatisfaction with the color of their teeth.[29,30]

Dental professionals recommend plaque-disclosing agents to help motivate patients to improve and maintain their oral hygiene habits.[31] In our study, a very low percentage of participants (7.7%) agreed on using this agent previously (item 16). However, despite this low frequency, clinical students used plaque-disclosing agent (10.4%) twice as much as pre-clinical students (5%), with the highest percentage reported for the participants enrolled in the fifth academic year (12.5%). This finding indicates a significant improvement that develops as students advance through their dental education, a pattern consistently observed across multiple studies using the HU-DBI assessment tool. The percentage of participants using plaque-disclosing agents increased between freshman and senior students in studies conducted in the United Kingdom (from 33% to 78%, respectively) and Estonia (47.6% to 73.3%, respectively).[23,25]

Research showed that toothbrushing can effectively reduce plaque scores by 50% with or without toothpaste. The most recent systematic review on toothpaste efficacy provides moderate evidence that toothpaste offers no additional benefit for mechanical plaque removal during brushing.[21,32] Additionally, although small-headed “child-sized toothbrushes” are primarily recommended for children, they provide adults with superior accessibility to posterior teeth.[33] Based on these findings, the HU-DBI assessment incorporates several indicators of excellent oral hygiene awareness: agreement with brushing without toothpaste (item 11), disagreement with the notion that toothbrushing alone prevents periodontal disease (item 14), and the use of child-sized toothbrushes (item 5). In our study, a relatively high percentage of participants (46.3%) reported using child-size toothbrushes in comparison with other countries, where the percentage was <11% (3.7% in Germany; 7.9% in Saudi Arabia; 9.7% in Estonia; 10.4% in Yemen).[9,23,26,34] A very low number of Lebanese dental students believed that they could brush their teeth without using toothpaste (14.2%). A similar percentage of 13.3% was reported by a survey conducted in Saudi Arabia.[9] More than one-third of dental students in Lebanon (38.2%) disagreed with item 14, stating that toothbrushing alone prevents periodontal disease, which is a much lower frequency than reported in other countries (84.7% in the Czech Republic; 65.9% in Slovakia; 55.3% in Germany).[21,26]

According to the multivariate analysis, having problematic internet use was a very weak predictor of a slightly lower score for the attitude subscale (B = −0.39; 95% CI, −0.69 to −0.09; β, −0.16, P = 0.01). Riad et al. reported that problematic internet resulted in lower HU-DBI scores in their study, and they were the first study to report on the possible association between internet use and oral health among dental students.[15] It is important to note though that there is a shortage in studies investigating the effect of problematic internet use on oral health knowledge, attitude, and behavior. To our knowledge, few studies have investigated similar aspects, two of which are as follows: A study reported by Ghareghol et al.[35] investigated the relationship between problematic internet use and oral health-related quality of life among dental students, and another study reported by Bastani et al.[36] investigated the association between online oral health information seeking experience and knowledge, attitudes, and practices of oral health among medical students.

Furthermore, regular checkups could very weakly predict the knowledge score only (B = 0.32; 95% CI, 0.01–0.64; β, 0.13, P = 0.04), while no statistical significance could be observed for the remaining domains or the total score, as shown by the multivariate analysis. These findings were slightly different from those reported by Riad et al.,[15] where a significantly higher mean HU-DBI score was reported in regular attendees for the overall score as well as all three domains. Higher scores are expected in individuals who visit the dentist regularly, since patients tend to learn more information related to oral health from their dentist which in turn will affect their attitudes and behavior. In addition, these patients will be continuously and closely monitored by their dentist, who will be able to correct any misinformation related to oral health, or unfavorable attitude or behavior.

Our study had a few limitations. One of the limitations of this study was its cross-sectional design, where longitudinal follow-up was not undertaken, and as such, we are unable to investigate temporal interference. Another limitation was related to recall bias or response bias, where participants may have provided information that did not correctly resemble their KAB. To avoid selection bias, all undergraduate students were approached and invited to participate in this study. The study included only undergraduate students in only one dental school (Lebanese University), another limitation of this study. The choice of limiting student selection to only undergraduate ones was made for easier comparison between our study and former ones, also conducted among undergraduate dental students in other countries. It is important to note that the Lebanese University was selected since it is the only public university in Lebanon and, as such, provides the best reflection for government-influenced educational programs. All other dental schools belong to the private sector and are hence influenced by other institutions/organizations. Another limitation was the uneven distribution between the sexes, which was due to the higher number of females enrolled at the faculty of Dentistry of the chosen university, and as such, could not be avoided. In addition, socioeconomic data was not collected during this study, and the effect of these data on knowledge, attitude, and behavior was not investigated. Furthermore, the survey was distributed by class representatives instead of being electronically or randomly distributed. While the multivariate regression model statistically significantly predicted the HU-DBI knowledge and attitudes subscale, the R² was 0.06 for each, indicating a “small” effect size.[37] That being said, the impact of problematic internet use on the attitudes scale should be further confirmed through longitudinal studies.

CONCLUSION

In conclusion, Lebanese dental students attained an HU-DBI score comparable to their peers in other Middle Eastern countries, with sex and regular checkups being significant predictors of the knowledge subscale, and problematic internet use a significant predictor of the attitude subscale; alas, this prediction was very weak. These findings highlight the need to use a holistic approach of improving knowledge, skills, and attitude in order to increase the chances of behavioral change. It is also important to incorporate the impact of behavioral factors on oral health into the dental curriculum so that dental students are better informed about these factors to properly design more effective future oral health educational programs. It is also important to advocate among dentists for using regular checkups as a means of prevention and oral health education, rather than only for treatment purposes.

CONFLICTS OF INTEREST

All authors certify that they have no affiliations with or involvement in any organization or entity with any financial interest or non-financial interest in the subject matter or materials discussed in this manuscript.

AUTHOR CONTRIBUTIONS

DD supervised the work and contributed to the study design. Data collection was done by AI, YH, and FFJ. Data analyses were performed by all authors. The first draft of the manuscript was written by FFJ and YH; it was corrected by DD. All authors commented on previous versions of the manuscript, and all authors read and approved the final manuscript.

ETHICAL POLICY AND INSTITUTIONAL REVIEW BOARD STATEMENT

The project was approved by the dean and the Scientific Committee of the Faculty of Dental Medicine at the Lebanese University under the number RCSC/01-04-2025. The procedures used in this study adhere to the tenets of the Declaration of Helsinki.

PATIENT DECLARATION OF CONSENT

An informed consent form was obtained from all participants prior to filling out the study questionnaire.

DATA AVAILABILITY STATEMENT

Data are available upon reasonable request.

LIST OF ABBREVIATIONS

• 95% CI: 95% confidence interval

• HU-DBI: Hiroshima University Dental Behavioral Inventory

• KABs: Knowledge, attitudes, and behaviors

• n: Number of participants

• N: Total number of participants

• P: P value

• SD: Standard deviation

• U: Mann–Whitney test

• vs: Versus

Supplementary Table 1.

The modified English version of the Hiroshima University Dental Behavioral Inventory (HU-DBI)

Item no.	Item description	Score of 1*
Item 1	I don’t worry much about visiting the dentist.	–
Item 2	My gums tend to bleed when I brush my teeth.	Disagree
Item 3	I worry about the color of my teeth.	–
Item 4	I have noticed some white sticky deposits on my teeth.	Agree
Item 5	I use a child-sized toothbrush.	–
Item 6	I think that I cannot help having false teeth when I am old.	Disagree
Item 7	I am bothered by the color of my gums.	–
Item 8	I think my teeth are getting worse despite my daily brushing.	Disagree
Item 9	I brush each of my teeth carefully.	Agree
Item 10	I have never been taught professionally how to brush.	Disagree
Item 11	I think I can clean my teeth well without using toothpaste.	Agree
Item 12	I often check my teeth in a mirror after brushing.	Agree
Item 13	I worry about having bad breath.	–
Item 14	It is impossible to prevent gum disease with tooth brushing alone.	Disagree
Item 15	I put off going to a dentist until I have a toothache.	Disagree
Item 16	I have used a dye to see how clean my teeth are.	Agree
Item 17	I use a toothbrush that has hard bristles.	–
Item 18	I don’t feel l have brushed well unless I brush with hard strokes.	–
Item 19	I feel I sometimes take too much time to brush my teeth.	Agree
Item 20	I have had my dentist tell me that I brush very well.	–
Item 21	I find myself using my smartphone/computer longer than I planned.	–
Item 22	I consume tobacco at least once a week.	–
Item 23	I drink alcohol at least once a week.	–
Item 24	I go to the dentist/hygienist for regular checkup at least once a year.	–

Category 2 includes items 1 to 20, and Category 3 includes items 21 to 24.

^*A score of 1 point is allocated to each of the “Agree” or “Disagree” answers for the items as specified in the table

Supplementary Table 2.

Responses of participants to HU-DBI Arabic version (N = 246) – by sex

Item	Expected value	Total	Male	Female	p value	OR 95% CI
Item 1	Agree	179 (72.8%)	116 (71.6%)	63 (75.0%)	0.571	0.841 (0.461; 1.532)
Item 2	Disagree	196 (79.7%)	22 (13.6%)	28 (33.3%)	< 0.001	0.314 (0.166; 0,595)
Item 3	Agree	220 (89.4%)	151 (93.2%)	69 (82.1%)	0.007	2.984 (1.303; 6.834)
Item 4	Agree	63 (25.6%)	39 (24.1%)	24 (28.6%)	0.443	0.793 (0.437; 1.437)
Item 5	Agree	114 (46.3%)	77 (47.5%)	37 (44.0%)	0.603	1.151 (0.677; 1.955)
Item 6	Disagree	156 (63.4%)	63 (38.9%)	27 (32.1%)	0.298	1.343 (0.770; 2.343)
Item 7	Agree	15 (6.1%)	5 (3.1%)	10 (11.9%)	0.006	0.236 (0.078; 0.714)
Item 8	Disagree	191 (77.6%)	35 (21.6%)	20 (23.8%)	0.694	0.882 (0.471; 1.650)
Item 9	Agree	198 (80.5%)	128 (79.0%)	70 (83.3%)	0.417	0.753 (0.379; 1.497)
Item 10	Disagree	189 (76.8%)	37 (22.8%)	20 (23.8%)	0.864	0.947 (0.509; 1.764)
Item 11	Agree	35 (14.2%)	19 (11.7%)	16 (19.0%)	0.119	0.565 (0.273; 1.166)
Item 12	Agree	205 (83.3%)	136 (84.0%)	69 (82.1%)	0.718	1.137 (0.566; 2.286)
Item 13	Agree	209 (85.0%)	144 (88.9%)	65 (77.4%)	0.017	2.338 (1.152; 4.747)
Item 14	Disagree	94 (38.2%)	99 (61.1%)	53 (63.1%)	0.761	0.919 (0.533; 1.584)
Item 15	Disagree	171 (69.5%)	49 (30.2%)	26 (31.0%)	0.909	0.967 (0.546; 1.713)
Item 16	Agree	19 (7.7%)	12 (7.4%)	7 (8.3%)	0.796	0.880 (0.333; 2.326)
Item 17	Agree	24 (9.8%)	15 (9.3%)	9 (10.7%)	0.715	0.850 (0.356; 2.034)
Item 18	Agree	56 (22.8%)	37 (22.8%)	19 (22.6%)	0.969	1.013 (0.540; 1.900)
Item 19	Agree	134 (54.5%)	93 (57.4%)	41 (48.8%)	0.199	1.414 (0.833; 2.400)
Item 20	Agree	185 (75.2%)	118 (72.8%)	67 (79.8%)	0.233	0.680 (0.361; 1.284)

Data are presented as n (%), where the percentages were calculated based on n/N. All P values were calculated using the chi-square test.

95% CI = 95% confidence interval; HU-DBI = Hiroshima University Dental Behavioral Inventory; n = number of participants; N = total number of participants; OR = odds ratio

Supplementary Table 3.

Responses of participants to HU-DBI Arabic version (N = 246) – by clinical status

Item	Expected value	Total	Pre-clinical	Clinical	P value	OR 95% CI
Item 1	Agree	179 (72.8%)	94 (77.7%)	85 (68.0%)	0.088	0.610 (0.345; 1.079)
Item 2	Disagree	196 (79.7%)	28 (23.1%)	22 (17.6%)	0.280	0.709 (0.380; 1.325)
Item 3	Agree	220 (89.4%)	115 (95.0%)	105 (84.0%)	0.005	0.274 (0.106; 0.708)
Item 4	Agree	63 (25.6%)	40 (33.1%)	23 (18.4%)	0.008	0.457 (0.253; 0824)
Item 5	Agree	114 (46.3%)	43 (35.5%)	71 (56.8%)	0.001	2.385 (1.427; 3.986)
Item 6	Disagree	156 (63.4%)	52 (43.0%)	38 (30.4%)	0.041	0.580 (0.343; 0.979)
Item 7	Agree	15 (6.1%)	9 (7.4%)	6 (4.8%)	0.387	0.627 (0.216; 1.820)
Item 8	Disagree	191 (77.6%)	26 (21.5%)	29 (23.2%)	0.747	1.104 (0.605; 2.012)
Item 9	Agree	198 (80.5%)	98 (81.0%)	100 (80.0%)	0.844	0.939 (0.499; 1.765)
Item 10	Disagree	189 (76.8%)	35 (28.9%)	22 (17.6%)	0.035	0.525 (0.287; 0.961)
Item 11	Agree	35 (14.2%)	10 (8.3%)	25 (20.0%)	0.008	2.775 (1.270; 6.063)
Item 12	Agree	205 (83.3%)	110 (90.9%)	95 (76.0%)	0.002	0.317 (0.151; 0.666)
Item 13	Agree	209 (85.0%)	106 (87.6%)	103 (82.4%)	0.254	0.663 (0.326; 1.348)
Item 14	Disagree	94 (38.2%)	77 (63.6%)	75 (60.0%)	0.557	0.857 (0.512; 1.435)
Item 15	Disagree	171 (69.5%)	38 (31.4%)	37 (29.6%)	0.759	0.918 (0.534; 1.581)
Item 16	Agree	19 (7.7%)	6 (5.0%)	13 (10.4%)	0.110	2.225 (0.817; 6.058)
Item 17	Agree	24 (9.8%)	11 (9.1%)	13 (10.4%)	0.729	1.161 (0.499; 2.702)
Item 18	Agree	56 (22.8%)	32 (26.4%)	24 (19.2%)	0.175	0.661 (0.362; 1.206)
Item 19	Agree	134 (54.5%)	64 (52.9%)	70 (56.0%)	0.625	1.134 (0.686; 1.873)
Item 20	Agree	185 (75.2%)	90 (74.4%)	95 (76.0%)	0.769	1.091 (0.611; 1.946)

Data are presented as n (%), where the percentages were calculated based on n/N. All P values were calculated using the chi-square test.

95% CI = 95% confidence interval; HU-DBI = Hiroshima University Dental Behavioral Inventory; n = number of participants; N = total number of participants; OR = odds ratio

Supplementary Table 4.

Responses of participants to HU-DBI Arabic version (N = 246) – by smoking status

Item	Expected value	Total	Nonsmoker	Smoker	P value	OR 95% CI
Item 1	Agree	179 (72.8%)	151 (73.3%)	28 (70.0%)	0.668	0.850 (0.404; 1.787)
Item 2	Disagree	196 (79.7%)	40 (19.4%)	10 (25.0%)	0.422	1.383 (0.625; 3.062)
Item 3	Agree	220 (89.4%)	184 (89.3%)	36 (90.0%)	0.898	1.076 (0.350; 3.310)
Item 4	Agree	63 (25.6%)	54 (26.2%)	9 (22.5%)	0.622	0.817 (0.366; 1.827)
Item 5	Agree	114 (46.3%)	95 (46.1%)	19 (47.5%)	0.872	1.057 (0.536; 2.083)
Item 6	Disagree	156 (63.4%)	76 (36.9%)	14 (35.0%)	0.820	0.921 (0.453; 1.871)
Item 7	Agree	15 (6.1%)	10 (4.9%)	5 (12.5%)	0.064	2.800 (0.903; 8.687)
Item 8	Disagree	191 (77.6%)	51 (24.8%)	4 (10.0%)	0.040	0.338 (0.115; 0.995)
Item 9	Agree	198 (80.5%)	166 (80.6%)	32 (80.0%)	0.932	0.964 (0.413; 2.251)
Item 10	Disagree	189 (76.8%)	47 (22.8%)	10 (25.0%)	0.764	1.128 (0.514; 2.475)
Item 11	Agree	35 (14.2%)	27 (13.1%)	8 (20.0%)	0.253	1.657 (0.692; 3.972)
Item 12	Agree	205 (83.3%)	172 (83.5%)	33 (82.5%)	0.877	0.932 (0.381; 2.280)
Item 13	Agree	209 (85.0%)	173 (84.0%)	36 (90.0%)	0.330	1.717 (0.573; 5.147)
Item 14	Disagree	94 (38.2%)	128 (62.1%)	24 (60.0%)	0.799	0.914 (0.457; 1.827)
Item 15	Disagree	171 (69.5%)	65 (31.6%)	10 (25.0%)	0.410	0.723 (0.334; 1.567)
Item 16	Agree	19 (7.7%)	16 (7.8%)	3 (7.5%)	1.000*	0.963 (0.267; 3.471)
Item 17	Agree	24 (9.8%)	18 (8.7%)	6 (15.0%)	0.222	1.843 (0.682; 4.978)
Item 18	Agree	56 (22.8%)	48 (23.3%)	8 (20.0%)	0.649	0.823 (0.355; 1.905)
Item 19	Agree	134 (54.5%)	112 (54.4%)	22 (55.0%)	0.942	1.026 (0.519; 2.026)
Item 20	Agree	185 (75.2%)	157 (76.2%)	28 (70.0%)	0.405	0.728 (0.345; 2.395)

Data are presented as n (%), where the percentages were calculated based on n/N. All P values marked with asterisk (*) were calculated using Fisher’s exact test. All other P value were calculated using the chi-square test.

95% CI = 95% confidence interval; HU-DBI = Hiroshima University Dental Behavioral Inventory; n = number of participants; N = total number of participants; OR = odds ratio

Supplementary Table 5.

Responses to HU-DBI Arabic version (N = 246) – by academic year

Item	Expected Value	Year 2	Year 3	Year 4	Year 5	P value
Item 1	Agree	46 (71.9%)	48 (84.2%)	40 (65.6%)	45 (70.3%)	0.134
Item 2	Disagree	44 (68.8%)	49 (86.0%)	53 (86.9%)	50 (78.1%)	0.043
Item 3	Agree	63 (98.4%)	52 (91.2%)	52 (85.2%)	53 (82.8%)	0.020
Item 4	Agree	21 (32.8%)	19 (33.3%)	10 (16.4%)	13 (20.3%)	0.066
Item 5	Agree	21 (32.8%)	22 (38.6%)	34 (55.7%)	37 (57.8%)	0.009
Item 6	Disagree	38 (59.4%)	31 (54.4%)	43 (70.5%)	44 (68.8%)	0.207
Item 7	Agree	4 (6.2%)	5 (8.8%)	3 (4.9%)	3 (4.7%)	0.798*
Item 8	Disagree	47 (73.4%)	48 (84.2%)	46 (75.4%)	50 (78.1%)	0.522
Item 9	Agree	50 (78.1%)	48 (84.2%)	49 (80.3%)	51 (79.7%)	0.860
Item 10	Disagree	43 (67.2%)	43 (75.4%)	52 (85.2%)	51 (79.7%)	0.106
Item 11	Agree	3 (4.7%)	7 (12.3%)	13 (21.3%)	12 (18.8%)	0.036
Item 12	Agree	59 (92.2%)	51 (89.5%)	46 (75.4%)	49 (76.6%)	0.018
Item 13	Agree	54 (84.4%)	52 (91.2%)	47 (77.0%)	56 (87.5%)	0.166
Item 14	Disagree	23 (35.9%)	21 (36.8%)	26 (42.6%)	24 (37.5%)	0.873
Item 15	Disagree	41 (64.1%)	42 (73.7%)	44 (72.1%)	44 (68.8%)	0.664
Item 16	Agree	5 (7.8%)	1 (1.8%)	5 (8.2%)	8 (12.5%)	0.151*
Item 17	Agree	8 (12.5%)	3 (5.3%)	6 (9.8%)	7 (10.9%)	0.582
Item 18	Agree	21 (32.8%)	11 (19.3%)	15 (24.6%)	9 (14.1%)	0.074
Item 19	Agree	31 (48.4%)	33 (57.9%)	39 (63.9%)	31 (48.4%)	0.226
Item 20	Agree	46 (71.9%)	44 (77.2%)	49 (80.3%)	46 (71.9%)	0.628
Item 21	Agree	56 (87.5%)	52 (91.2%)	50 (82.0%)	55 (85.9%)	0.523
Item 22	Agree	6 (9.4%)	11 (19.3%)	13 (21.3%)	10 (15.6%)	0.287
Item 23	Agree	4 (6.2%)	6 (10.5%)	7 (11.5%)	13 (20.3%)	0.101
Item 24	Agree	37 (57.8%)	45 (78.9%)	44 (72.1%)	50 (78.1%)	0.032

Data are presented as n (%), where the percentages were calculated based on n/N. All P values marked with asterisk (*) were calculated using Fisher’s exact test. All other P value were calculated using chi-square test.

HU-DBI = Hiroshima University Dental Behavioral Inventory; n = number of participants; N = total number of participants

Supplementary Table 6.

Knowledge, attitudes, behaviors, and total HU-DBI scores – Sex, academic year, experience (N = 246)

Variable	Outcome	Knowledge		Attitude		Behavior		Total
		Mean ± SD score	P value (Cohen’s d)	Mean ± SD score	P value (Cohen’s d)	Mean ± SD score	P value (Cohen’s d)	Mean ± SD score	P value (Cohen’s d)
Sex	Female	3.69 ± 1.06	0.047 (0.245)	1.12 ± 0.80	0.312 (0.12)	1.94 ± 0.68	0.413 (0.094)	6.75 ± 1.52	0.532 (0.078)
	Male	3.37 ± 1.19		1.24 ± 0.86		2.02 ± 0.82		6.63 ± 1.60
Academic year	2nd year	3.22 ± 1.06	0.004 (0.417)	1.00 ± 0.71	0.063 (0.266)	2.11 ± 0.72	0.066 (0.265)	6.33 ± 1.39	0.062 (0.27)
	3rd year	3.77 ± 1.10		1.04 ± 0.78		2.09 ± 0.63		6.89 ± 1.42
	4th year	3.84 ± 1.02		1.34 ± 0.83		1.80 ± 0.79		6.98 ± 1.62
	5th year	3.53 ± 1.18		1.25 ± 0.91		1.89 ± 0.74		6.67 ± 1.69
Experience	Pre-clinical	3.48 ± 1.11	0.102 (0.201)	1.02 ± 0.74	0.012 (0.3)	2.10 ± 0.68	0.008 (0.307)	6.60 ± 1.43	0.199 (0.161)
	Clinical	3.68 ± 1.11		1.30 ± 0.87		1.85 ± 0.76		6.82 ± 1.66

All the P values were calculated using Mann–Whitney U test and Kruskal–Wallis test.

HU-DBI = Hiroshima University Dental Behavioral Inventory; N = total number of participants; SD = standard deviation

Supplementary Table 7.

Knowledge, attitudes, behaviors, and total HU-DBI scores – General health behaviors (N = 246)

Variable	Outcome	Knowledge		Attitude		Behavior		Total
		Mean ± SD score	P value (Cohen’s d)	Mean ± SD score	P value (Cohen’s d)	Mean ± SD score	P value (Cohen’s d)	Mean ± SD score	P value (Cohen’s d)
Tobacco smoking	Nonsmoker	3.56 ± 1.11	0.327 (0.12)	1.14 ± 0.79	0.817 (0.027)	1.98 ± 0.72	0.820 (0.026)	6.68 ± 1.55	0.488 (0.087)
	Smoker	3.70 ± 1.16		1.25 ± 0.98		1.93 ± 0.80		6.88 ± 1.56
Alcohol drinking	Nondrinker	3.54 ± 1.11	0.110 (0.197)	1.15 ± 0.80	0.787 (0.032)	2.00 ± 0.72	0.241 (0.134)	6.69 ± 1.54	0.526 (0.079)
	Drinker	3.87 ± 1.11		1.23 ± 0.97		1.80 ± 0.76		6.90 ± 1.65
Problematic internet use	No	3.67 ± 1.02	0.697 (0.048)	1.48 ± 0.80	0.011 (0.305)	1.82 ± 0.85	0.268 (0.127)	6.97 ± 1.65	0.182 (0.167)
	Yes	3.57 ± 1.13		1.11 ± 0.81		2.00 ± 0.71		6.67 ± 1.53
Regular checkup	No	3.36 ± 1.16	0.039 (0.255)	1.14 ± 0.84	0.867 (0.02)	2.03 ± 0.70	0.416 (0.093)	6.53 ± 1.55	0.224 (0.152)
	Yes	3.67 ± 1.09		1.16 ± 0.81		1.95 ± 0.74		6.78 ± 1.55

All the P values were calculated using Mann–Whitney U test and Kruskal–Wallis test.

HU-DBI = Hiroshima University Dental Behavioral Inventory; N = total number of participants; SD = standard deviation

Supplementary Table 8.

Multiple linear regression of the HU-DBI total score and subscales

Predictors	B (Unstandardized coefficients)	95.0% confidence interval for B		Standard error of B	Beta (Standardized Coefficients)	P value	Collinearity statistics
		Lower bound	Upper bound				Tolerance	VIF
HU-DBI total score
(Constant)	4.49	−0.38	9.36	2.47		0.07
Academic year	0.01	−0.29	0.31	0.15	0.01	0.95	0.34	2.98
Sex	0.21	−0.21	0.63	0.21	0.06	0.33	0.95	1.06
Age	0.09	−0.16	0.33	0.13	0.07	0.50	0.34	2.94
Problematic internet use	−0.33	−0.91	0.25	0.29	−0.07	0.26	0.98	1.02
Tobacco smoking	0.24	−0.31	0.79	0.28	0.06	0.39	0.93	1.08
Alcohol drinking	0.12	−0.49	0.73	0.31	0.02	0.70	0.95	1.05
Regular checkups	0.26	−0.18	0.70	0.22	0.08	0.24	0.96	1.04
Knowledge subscale
(Constant)	2.53	−0.91	5.96	1.74		0.15
Academic year	0.08	−0.13	0.29	0.11	0.08	0.44	0.34	2.98
Sex	0.39	0.09	0.69	0.15	0.17	0.01	0.95	1.06
Age	0.00	−0.18	0.17	0.09	0.00	0.97	0.34	2.94
Problematic internet use	−0.11	−0.52	0.30	0.21	−0.03	0.59	0.98	1.02
Tobacco smoking	0.19	−0.20	0.58	0.20	0.06	0.33	0.93	1.08
Alcohol drinking	0.25	−0.18	0.68	0.22	0.07	0.25	0.95	1.05
Regular checkups	0.32	0.01	0.64	0.16	0.13	0.04	0.96	1.04
Attitude subscale
(Constant)	−0.69	−3.22	1.84	1.28		0.59
Academic year	−0.01	−0.16	0.15	0.08	−0.01	0.95	0.34	2.98
Sex	−0.07	−0.29	0.15	0.11	−0.04	0.52	0.95	1.06
Age	0.10	−0.02	0.23	0.07	0.17	0.11	0.34	2.94
Problematic internet use	−0.39	−0.69	−0.09	0.15	−0.16	0.01	0.98	1.02
Tobacco smoking	0.11	−0.17	0.40	0.14	0.05	0.44	0.93	1.08
Alcohol drinking	0.01	−0.31	0.33	0.16	0.00	0.96	0.95	1.05
Regular checkups	0.00	−0.23	0.23	0.12	0.00	0.98	0.96	1.04
Behavior subscale
(Constant)	2.65	0.37	4.93	1.16		0.02
Academic year	−0.07	−0.21	0.07	0.07	−0.11	0.33	0.34	2.98
Sex	−0.11	−0.31	0.09	0.10	−0.07	0.27	0.95	1.06
Age	−0.01	−0.13	0.10	0.06	−0.03	0.80	0.34	2.94
Problematic internet use	0.17	−0.10	0.45	0.14	0.08	0.20	0.98	1.02
Tobacco smoking	−0.06	−0.32	0.20	0.13	−0.03	0.65	0.93	1.08
Alcohol drinking	−0.14	−0.43	0.14	0.15	−0.06	0.33	0.95	1.05
Regular checkups	−0.07	−0.27	0.14	0.10	−0.04	0.54	0.96	1.04

Model = “Enter” method in SPSS statistics.

For HU-DBI total score: Durbin–Watson = 1.65; R square = 0.02; ANOVA: F = 0.82, P = 0.57. For knowledge subscale: Durbin–Watson = 1.84; R square = 0.06; ANOVA: F = 2.145; P = 0.04. For attitude subscale: Durbin–Watson = 1.94; R square = 0.06; ANOVA: F = 2.17; P = 0.04. For behavior subscale: Durbin–Watson = 1.91; R square = 0.04; ANOVA: F = 1.33; P = 0.24. HU-DBI = Hiroshima University Dental Behavioral Inventory; VIF = Variance inflation factor

Funding Statement

Nil.