Shared Social Determinants and Aggregated Risk Behaviours Associated with Self-Rated Oral Health Among Schoolchildren in Jazan, Saudi Arabia

Abstract

Background

Dental and periodontal diseases share common social determinants and oral health-risk behaviours, yet these factors have not been thoroughly investigated across different contexts.

Aim

This study aimed to examine the association between aggregated common oral health-risk behaviours and their determinants with self-rated dental and gingival health in Jazan, Saudi Arabia (SA).

Methods

A cross-sectional study was conducted among a non-probability convenience sample of 687 school students in Jazan, SA. Participants completed an online questionnaire assessing behaviours related to oral hygiene, dental visits, dietary habits (sugar and fruit consumption), tobacco use, and sociodemographic and socioeconomic factors. Logistic regression modelling was performed to assess the determinants and aggregated behaviours associated with self-rated poor dental and gingival health.

Results

The median age of participating students was 15 (interquartile range: 2) years, and 67% were female. The aggregate of oral health-risk behaviours score was significantly associated with self-rated poor dental and gingival health, with each additional risk behavior associated with higher odds of poor dental health (OR = 1.29, 95% CI: 1.17–1.42) and poor gingival health (OR = 1.34, 95% CI: 1.23–1.47). Rural residence was also significantly associated with both outcomes, including poor dental health (OR = 2.44, 95% CI: 1.56–3.83) and poor gingival health (OR = 2.02, 95% CI: 1.37–2.99).

Conclusion

This study found a significant association between aggregated oral health-related risk behaviours and poor self-rated dental and gingival health among schoolchildren in Jazan region, SA. Notably, area of residence emerged as a key shared social determinant, with students living in rural areas disproportionately impacted. These findings support the shared role of oral health-risk behaviours and social factors in the development of both dental and gingival diseases.

Introduction

Oral health problems, particularly dental caries and periodontal diseases, have a substantial impact on an individual’s quality of life,^1,2 while also placing an increasing burden on healthcare budgets and workforces.^3,4 This highlights the need to control these conditions as a national priority.^5,6 Although dental caries and periodontal diseases are highly prevalent across societies and age groups, they are largely preventable.^7,8 Their multifactorial nature^9,10 provides an important basis for both short- and long-term management and prevention strategies.

Oral disease risk assessment models have become increasingly common^11,12 and have contributed to the development of local prevention policies. However, these models depend on baseline data such as sociodemographic characteristics, lifestyle habits, and health-related factors,^13,14 which may vary considerably between populations. Therefore, many efforts have been made to collect such data and better understand the factors contributing to oral disease risk.¹⁵

Prior research has investigated lifestyle and oral hygiene habits in relation to oral diseases. A study in Switzerland reported that younger children and those with migrant backgrounds were at a higher risk of caries.¹⁶ In Singapore, a national study found that although most participants brushed their teeth, only half visited the dentist at least once a year.¹⁷ A consensus statement from the European Federation of Periodontology and the European Organisation for Caries Research concluded that caries and periodontal diseases share biological determinants, particularly dental biofilm, as well as common risk factors and social determinants.¹⁸

Low socioeconomic status has also been linked to a higher risk of dental caries and periodontitis. In addition, routine oral hygiene with fluoride has been identified as a key behavioural factor affecting both conditions.¹⁸ Population-based interventions should therefore target negative behaviours such as smoking and high consumption of sugary foods and drinks. However, health risk behaviours often do not occur separately; instead, they tend to co-occur or cluster, with individuals engaging in multiple risk behaviours at the same time.^19,20 Aggregated oral health-risk behaviours have been reported among adults and children with tooth loss and untreated caries.^19–21

Studies across Arab countries highlight a substantial burden of oral health problems among children and adolescents, especially in upper-middle and lower-middle income nations. Untreated dental caries is most prevalent in deciduous teeth, while permanent teeth have an estimated 137 million unmet treatment needs, largely linked to high sugar intake in daily diets.^22–24 In Saudi Arabia (SA), the situation is especially concerning, with reported caries prevalence reaching 99% and gingival diseases affecting nearly all adolescents. Moderate gingivitis alone impacts around 45% of this age group. Poor oral health among adolescents has been associated with several modifiable lifestyle risk factors, including inadequate toothbrushing, frequent consumption of sugar-sweetened beverages and sugary foods, and limited dental care utilisation.^25–29

Evidence from different regions of SA further supports these concerns. In the Eastern Province, the frequency and amount of sweetened drink consumption were significantly associated with caries prevalence.³⁰ In Al-Madinah, severe caries were more common among children from low-income families and public schools.³¹ A national study involving major Saudi regions found that sex, toothbrushing frequency, and plaque index were significantly associated with gingivitis severity.³²

Despite this evidence, there remains limited comprehensive data on shared social determinants and aggregated oral health-related risk behaviours among schoolchildren with self-reported poor dental and gingival health in southern SA. This gap may limit efforts to address these common conditions and hinder the development of targeted oral healthcare strategies in the region. Thus, this study aimed to examine the association between aggregated common oral health-risk behaviours and their determinants with self-rated dental and gingival health among schoolchildren in Jazan, SA. We hypothesized that a higher number of aggregated oral health-related risk behaviors would be associated with greater odds of self-rated poor dental and gingival health among schoolchildren, and that these outcomes would also be patterned by shared social determinants. Self-rated oral health measures have been widely used in population-based adolescent oral health research and public health surveillance, as they provide a practical and meaningful proxy for perceived oral health status and have demonstrated associations with clinical and behavioral oral health indicators.²⁴

Materials and Methods

Study Design

This analytical cross-sectional survey was conducted during the academic year 2023–2024.

Setting and Sampling Frame

A non-probability convenience sample of students was recruited from middle and secondary schools in all provinces of the Jazan region in the southwestern part of SA. Jazan is one of the administrative regions of SA, located in the southwest of the country and divided into 16 provincial governorates, with headquarters in Jazan city.³³ Jazan was selected as the study site due to its particularly high prevalence of oral diseases, distinct socioeconomic disparities, and underrepresentation in national oral health research. Its combination of urban and rural populations offers a valuable context for examining the influence of social determinants on oral health outcomes.

Study Population and Recruitment

According to the 2022 Saudi Census, the Jazan region has a population of 1,404,997, representing about 4.4% of the national total.³³ Although the 2022 census does not provide specific data on adolescents, a 2016 estimate reported approximately 256,576 individuals aged 12 to 18 years in the region, indicating that adolescents constitute a significant portion of Jazan’s population.

In the Saudi Arabian education system, middle school, also known as intermediate school, covers grades 7 through 9 and typically serves students aged 12 to 15 years. Secondary school, also known as high school, includes grades 10 through 12 and generally serves students aged 15 to 18 years. The age range of 14–18 years was selected as it corresponds to middle and secondary school levels in SA, where students possess the cognitive maturity to reliably self-report behaviours and perceptions, and it has also been commonly adopted in prior national and international studies on adolescent oral health behaviours.^20,24

As of the 2023–2024 academic year, the Jazan region had 102 boys’ middle and secondary schools with approximately 29,893 students and 105 girls’ middle and secondary schools with about 28,640 students. The list of eligible middle and secondary schools in the Jazan region was obtained from the Ministry of Education to identify schools across the region and support broad geographical coverage. However, no formal probability-based or cluster sampling procedure was implemented. Both private and governmental schools from both urban and rural areas of the Jazan region were invited. Schools were invited to participate, and only those that provided formal approval were included.

Within each participating school, eligible students were invited to complete an electronic questionnaire during scheduled school hours. Eligible participants were students aged 14–18 years who were enrolled in participating middle and secondary schools during the data collection period between November 2023 and January 2024. In accordance with local ethical guidelines and standard procedures for school-based research in SA, written informed consent was obtained from parents or legal guardians through the school administrations prior to data collection. Students provided electronic assent by reviewing an information sheet at the beginning of the online survey, and indicated their voluntary agreement to participate by selecting “Agree to Participate” before accessing the questionnaire. Only students who provided this electronic assent were able to access and complete the questionnaire.

Students were excluded if they were outside the target age range, did not have parental/legal guardian consent, did not provide electronic assent, were absent during survey administration, or submitted incomplete responses for the main outcome variables. Schools serving students with special needs were not invited to participate due to financial and logistical barriers. Following school approval and parental consent, 750 students were invited to participate. Of these, 700 accessed the questionnaire. Approximately 2% of submitted questionnaires were excluded due to incomplete responses, resulting in a final analytical sample of 687 students.

Sample Size Calculation

The required sample size was calculated based on the estimated number of middle and secondary school students in the Jazan region during the 2023–2024 academic year, which was approximately 58,533 students. Using a 95% confidence level of and a 5% margin of error, the minimum required sample size was estimated to be 461 participants (Chi-squares Inc. Sample Size Calculation for Cross-sectional Studies with Percentage as Outcome).³⁴ The questionnaire remained open throughout the designated data-collection period, with the process facilitated by the schools during scheduled school hours. The final analytical sample included 687 students, exceeding the minimum required sample size of 461 participants.

Ethical Considerations

This study was approved by the Research Ethics Committee of Taibah University, College of Dentistry (Ref. TUCDRED/110923/ATAlsharif). In addition, approval from the Education Policy Research Centre, Ministry of Education was obtained in November 2023 (4,500,468,319). The survey adhered to the Declaration of Helsinki, meaning that participants were informed about the purpose of the survey and their rights to confidentiality and withdrawal without giving reasons. The obtained data were stored on a password-protected laptop, and only the study team had access to the database.

Data Collection Procedures and Variables

Surveys were conducted during scheduled school hours. An Arabic self-administered translation of the standardised WHO Oral Health Survey assessment form was used in the study.³⁵ Self-rated oral health has been shown to be a valid and reliable proxy for clinical oral health status in adolescents.²⁴ These measures are widely used in public health surveillance due to their simplicity, cost-effectiveness, and established association with both subjective and clinical oral outcomes. A pilot study was conducted with a small group of students (n = 30) to assess the clarity and cultural relevance of the Arabic-translated WHO oral health questionnaire. Minor linguistic adjustments were made based on feedback before full data collection.

Students completed the questionnaire individually using iPads during school hours. To minimize social desirability bias, students were informed that participation was voluntary, that responses were anonymous, and that no teachers, parents, or school staff would have access to individual responses. Students were seated in a manner that reduced the possibility of viewing one another’s responses. School staff facilitated access to the survey but were not involved in viewing or recording individual answers. They were asked to complete the online questionnaire via a direct link to Google Forms, which took approximately 10–15 minutes, and was preceded by an information sheet about the survey and the electronic consent button. Those who agreed to participate were able to complete the questionnaire; otherwise, the questionnaire did not open. Upon completion, the participants were instructed to submit the web form, and the data were stored in an Excel spreadsheet. The main information of the questionnaire is summarised as follows.

Behavioural Variables

Coding matrix for the aggregated oral health risk behaviour score was presented in Table 1.

Table 1.

Coding Matrix for the Aggregated Oral Health-Risk Behavior Score

Behavioral Variable	Good Behavior Coded as 0	Risk Behavior Coded as 1
Toothbrushing frequency	Once or more daily	Less than once daily/never/sometimes
Age started brushing teeth	<7 years	≥7 years
Brushing duration	≥2 minutes	≤1 minute
Toothbrush replacement	≤3 months	>3 months
Fluoridated toothpaste use	Yes	No
Dental floss use	Yes	No
Dental attendance	Within the last 12 months	More than 12 months ago/never
Reason for dental visit	Check-up/follow-up treatment	Pain-based visit
Fruit consumption	Daily	Less than daily
Sweets consumption	Never/once weekly	Several times weekly/daily
Soft drink/soda/energy drink consumption	Once a week, several times a week, or never	Every day or several times per day
Tobacco use, any form	Non-user	Current user

Good oral-related behaviours were recorded as 0 for daily toothbrushing frequency, beginning toothbrushing before the age of 7, brushing for ≥2 minutes, a toothbrush replacement time of less than 3 months, flossing, use of fluoridated toothpaste, frequent dentist attendance (≤ 12 months), dental attendance for checkups or follow-up treatment, daily fruit consumption, infrequent consumption of sugar and sweetened soda drinks (never or ≤ once a week), and non-use of any form of tobacco. The sum of behaviour scores ranged from 0 to 12, with higher scores indicating a greater number of unfavourable behaviours. Each risk behavior was assigned equal weight in the aggregate score. This approach was used to capture the cumulative burden of multiple co-occurring risk behaviors in a simple and interpretable manner. However, equal weighting assumes that all included behaviors contribute similarly to the outcomes, although their biological and behavioral relevance may differ.

Outcome Variables

Two outcome variables of interest were self-rated dental health and self-rated gingival health. Students rated each as: very good, good, poor, or very poor. These were dichotomised for analysis into “good” (very good/good) coded as 0, and “poor” (poor/very poor) coded as 1.

Potential Confounding Covariates

Potential confounding covariates included the sociodemographic and socioeconomic factors of age, sex, parents’ educational attainment (illiterate, school or university), family income (<9000, 9000–15,999, 16,000–24,999, 25,000–34,999, 35,000–>50,000 Saudi riyals), type of school attended (government or private), area of residence (Jazan, Abu areesh, Samtah or Sabya [urban]; Ardah, Ahah al Masarhah, AlHarth, AlTwal, AlMwasam, Farasan or Damad [rural]) and home ownership (renting or owned). These categories were adapted from previously published studies conducted in similar contexts, where sociodemographic and socioeconomic factors were measured using indicators such as parental education, household income, and home ownership.^36–38

Data Analysis

The Excel spreadsheet was downloaded and imported to the Statistical Package for the Social Sciences (SPSS for Windows, version 21, IBM) for data processing and analysis. Descriptive analyses were performed to calculate frequencies and percentages for categorical data, and the median with the interquartile range (IQR) for continuous variables due to non-normal distribution (Shapiro–Wilk = p<0.05).

The distribution of self-rated dental and gingival health was examined for the whole sample and within the included behavioural factors and covariates. Hierarchical binary logistic regression was performed to examine the association between the aggregated oral health-related behaviours and self-rated oral health and gingival health, adjusting for covariates regardless of their significance in bivariate analysis. Multicollinearity was assessed using variance inflation factors (VIFs) derived from a linear regression model with the same independent variables for both self-rated oral health and gingival health, with VIF values <5 indicating no evidence of multicollinearity. All VIF values ranged from 1.024 to 1.192.

The overall model guided the analysis, showing the explanatory and outcome variables based on Dahlgren and Whitehead (1991), is presented in Figure 1 and applied in tables. Dahlgren and Whitehead’s (1991) model proposes a layered conceptual framework of health influences, which are complex and encompass upstream broader environmental, material, and social conditions (eg, housing, education, healthcare), as well as social and community support systems that shape downstream individual behaviours. Individual factors such as sex, age, and genetics are considered fixed.

Figure 1.

The overall model guided the analysis showing the explanatory and outcome variables drawn on Dahlgren and Whitehead (1991) (n=687), red arrows are the focus of this study analysis.

The adjusted analysis was presented as an adjusted odds ratio (OR) with corresponding 95% confidence intervals (CI), and a p-value <0.05 was considered statistically significant. The STROBE guidelines for reporting a cross-sectional study were followed.³⁹

Results

The study sample included 687 schoolchildren aged 14–18 years, with a median age of 15 years (IQR = 2; range: 14–18 years). Of these, 225 (33%) were male students and 462 (67%) were female students. A total of 485 students resided in urban areas and 202 resided in rural areas. Most participants attended government schools. The associations between sociodemographic and socioeconomic characteristics, individual oral health–related behaviours, and self-rated poor dental and gingival health were assessed using bivariate and fully adjusted binary logistic regression analyses (Supplementary Tables 1 and 2). Table 2 presents details of the other sociodemographic and socioeconomic factors. The proportions of students reporting self-rated poor dental and gingival health were 74% and 65%, respectively, and were distributed evenly across most sociodemographic variables, except for father’s level of education and area of residence.

Table 2.

Distribution of Sociodemographic and Socioeconomic Factors and Association with Self-Rated Poor Dental and Gingival Health Among Middle and Secondary School Students in Jazan, Saudi Arabia (N= 687)

Socioeconomic Characteristics		Total n (%)	Self-Reported Poor Dental Health n (%)	p-valuea	Self-Reported Poor Gingival Health n (%)	P-valuea
Overall		687(100)	513(74.7)	-	446(64.9)	-
Age/Years m(IQR)c		15[2]	15[2] vs 15(4)	0.784b	15[2] vs 15(2)	0.391b
Sex	Males	225(33)	165(73.3)	0.573	144(32)	0.72
	Females	462(67)	348(75.3)		302(68)
Area of Residence	Urban	485(71)	341(70.3)	<0.001	294(60.6)	<0.001
	Rural	202(29)	172(85.1)		152(75.2)
School Type	Government	672(98)	500(74.4)	0.280	434(64.6)	0.216
	Private	15(2)	13(86.7)		12(80.0)
Monthly Household Income	Highest (35000 – ≥50.000 SR)	51(7.4)	38 (74.5)	0.257	32 (62.7)	0.085
	Second middle (25000–34999SR)	25(6.4)	14 (65)		12 (48.0)
	Middle (16000–24999 SR)	44(6.4)	32 (72.7)		23 (52.3)
	Second lowest (9000–15999SR)	125(18.2)	97 (77.6)		79 (63.2)
	Lowest (< 9000 SR)	442(64.3)	332 (75.1)		300 (67.9)
Mother’s Education	Illiterate	199(29)	145(72.9)	0.652	123(61.8)	0.534
	School degree	376(55)	286(76.1)		250(66.5)
	University degree	112(16.3)	82(73.2)		73(65.2)
Father’s Education	Illiterate	113(16.4)	77(68.1)	0.157	61(54.0)	0.009
	School degree	427(62.2)	328(76.8)		294(68.9)
	University degree	147(21.4)	108(73.5)		91(61.9)
Mother’s Employment	Unemployed Employed	507(74) 180(26)	381(75.1) 132(73.3)	0.631	334(65.9) 112(62.2)	0.377
Father’s Employment	Unemployed Employed	292(43) 395(58)	227(77.7) 286(72.4)	0.112	193(66.1) 253(64.1)	0.579
Home Ownership	Rent Owner	147(21) 540(79)	116(78.9) 397(73.5)	0.183	98(66.7) 348(64.4)	0.617

Notes: ^aAll analysis used chi -squared test unless specified, ^bmann–Whitney U-test, values in bold signify p<0.05, ^c m(IQR)=Median (interquartile range).

The overall prevalence of aggregated oral health-related risk behaviours in the sample is presented in Table 3. The risk behaviours refer to the absence of good oral health practices, such as infrequent toothbrushing, delayed initiation of brushing, short brushing duration, infrequent toothbrush replacement, not flossing, non-use of fluoridated toothpaste, irregular dental visits, problem-oriented dental attendance, low fruit intake, frequent sugary drinks consumption, and tobacco use.

Table 3.

The Overall Prevalence of Aggregated Oral Health-Related Risk Behaviours in the Sample (n=687)

Number Oral Health Related Risk Behaviours	Number of Participants	Percentage (%)
One risk	6	0.9
Two risks	15	2.2
Three risks	50	7.3
Four risks	92	13.4
Five risks	140	20.4
Six risks	112	16.3
Seven risks	111	16.2
Eight risks	79	11.5
Nine risks	53	7.7
Ten risks	25	3.6
Eleven risks	4	0.6

Table 4 presents the distribution of all behavioural factors and their binary associations with self-rated poor dental and gingival health. The median (IQR) aggregate score for oral health-risk behaviours was 6 (2) and was statistically significantly higher (p<0.001) for students with self-reported poor dental and gingival health.

Table 4.

Distribution of Oral Health Risk-Related Behaviours Variables and Association with Self-Rated Poor Dental and Gingival Health Among Students in Jazan, Saudi Arabia (N= 687)

Oral Health Related Behaviour		Total Sample F (%)	Self-Rated Poor Dental Health F (%)	p-valuea	Self-Rated Poor Gingival Health F (%)	p-valuea
Age/years (m/IQR)b started brushing teeth	≥ 7years old <7years old	301 (43.8) 386 (56.2)	245 (81.4) 268 (69.4)	<0.001c	215 (71.4) 231 (59.8)	0.002b
Daily brushing frequency	Never/sometimes Once/twice	147 (21) 540 (79)	114 (77.6) 399 (73.9)	0.365	104 (70.7) 342 (63.3)	0.095
Brush duration	≤ 1 Minute ≥2 +Minutes	154 (22.4) 533 (77.6)	129 (83.8) 384 (72.0)	0.003	118 (76.6) 328 (61.5)	0.001
Brush replacement times	≤ 3 months > 3 months	394 (57.4) 293 (42.6)	283 (71.8) 230 (78.5)	0.047	247 (67.9) 199 (62.7)	0.156
Use of fluoridated toothpaste	No Yes	270 (39.3) 417 (60.7)	207 (76.6) 306 (73.4)	0.333	182 (67.4) 264 (63.3)	0.272
Use of brushing aids dental floss	No Yes	515 (75.0) 172 (25.0)	392 (76.1) 121 (70.3)	0.132	350 (68.5) 96 (55.8)	0.004
Frequency of dental Visits	Never/less frequent Frequent (last 3,6,12 months)	334 (48.6) 353 (51.4)	231 (69.2) 282 (79.9)	0.001	201 (60.2) 245 (69.4)	0.011
Reason for dental visits	Pain/Problems Treatment/check-up	217 (31.6) 470 (68.4)	201 (92.6) 312 (66.4)	<0.001	178 (82.0) 268 (57.0)	<0.001
Fruits consumption	Never/less frequent Everyday	387 (56.3) 300 (43.7)	306 (79.1) 207 (69.0)	0.003	276 (71.3) 170 (56.7)	<0.001
Sweets intake	Daily/several times weekly Never/once a week	468 (68.1) 219 (31.9)	362 (77.4) 151 (68.9)	0.018	326 (69.7) 120 (54.8)	<0.001
Soft drinks/soda/energy drink	Daily/several times per day Never/once a week	441 (64.2) 246 (35.8)	339 (76.9) 174 (70.7)	0.076	307 (69.6) 139 (56.5)	0.001
Tobacco use (any form)	Never/non-user Current user	579 (84.3) 108 (15.7)	431 (74.4) 82 (75.9)	0.774	373 (64.4) 73 (67.6)	0.526
Aggregate oral health risk behaviours	m(IQR)b	6 (2)	5(3)d vs c 6(3)e	<0.001	5(3)d vs c 6(3)e	<0.001

Notes: ^aAll analysis used chi-squared test unless otherwise specified, ^bm(IQR) =Median (interquartile range), ^cComparison of median (IQR) values between groups was performed using the Mann–Whitney U-test, ^de Self rated Good versus (vs) Self-rated Poor dental and gingival health, Values in bold indicate statistical significance (p<0.05).

In the unadjusted logistic regression models (Table 5 and Table 6), the aggregate variable of oral health-related risk behaviours was significantly associated with poor dental and gingival health. For each additional risk behaviour, there was a 28% increase in the odds of poor dental health (odds ratio [OR] 1.28, 95% CI: 1.17–1.40) and a 32% increase in the odds of poor gingival health (OR 1.32, 95% CI: 1.12–1.44). After adjusting for potential confounding covariates, the relationship between risk behaviours and dental health increased by 1%, and the relationship between risk behaviours and gingival health increased by 2%.

Table 5.

Logistic Regression Analysis Showing the Association Between Aggregate Risk Oral Health-Related Behaviours and Self-Rated Poor Dental Health Among School Students in Jazan, Saudi Arabia (N= 687)

Explanatory Variables	Model 1					Model 2					Model 3
	β	SE	Wald χ2	OR (95% CI) ^	p-value	β	SE	Wald χ2	OR (95% CI)	p-value	β	SE	Wald χ2	OR (95%)	p-value
Age	0.024	0.066	0.138	1.03 (0.90–1.17)	0.710	−0.009	0.068	0.017	0.991 (0.87–1.13)	0.896	−0.030	0.070	0.184	0.97 (0.85–1.11)	0.668
Sex
Male				1					1					1	0.073
Female	0.109	0.186	0.342	1.12 (0.78–1.60)	0.559	0.214	0.196	1.198	1.24 (0.84–1.82)	0.274	0.363	0.202	3.223	1.44 (0.97–2.14)
Family income
Highest (35000 – ≥50.000 SR)									1					1
Second middle (25000–34999SR)						−0.712	0.534	1.778	0.49 (0.17–1.40)	0.182	−0.746	0.545	1.876	0.47 (0.16–1.38)	0.171
Middle (16000–24999 SR)						−0.044	0.505	0.007	0.96 (0.36–2.58)	0.931	−0.104	0.515	0.041	0.90 (0.33–2.47)	0.840
Second lowest (9000–15999SR)						0.222	0.408	0.296	1.25 (0.56–2.78)	0.586	0.108	0.415	0.068	1.11 (0.49–2.51)	0.794
Lowest (< 9000 SR)						0.021	0.352	0.003	1.02 (0.51–2.04)	0.953	−0.041	0.357	0.013	0.96 (0.48–1.93)	0.908
Mother’s education
University									1					1
School						0.041	0.334	0.015	1.04 (0.54–2.01)	0.901	−0.080	0.342	0.054	0.92 (0.47–1.80)	0.816
Illiterate						0.139	0.283	0.240	1.15 (0.66–2.00)	0.624	0.050	0.289	0.030	1.05 (0.59, 1.85)	0.862
Father’s education
University									1					1
School						−0.366	0.352	1.079	0.69 (0.35–1.38)	0.299	−0.337	0.360	0.878	0.71 (0.35–1.45)	0.903
Illiterate						0.067	0.267	0.063	1.07 (0.63–1.80)	0.803	0.033	0.271	0.015	1.03 (0.61–1.76)	0.349
Residency area
Urban									1					1
Rural						0.865	0.226	14.73	2.38 (1.53–3.70)	<0.001	0.892	0.230	15.096	2.44 (1.56–3.83)	<0.001
Mother employment
Employed									1					1
Unemployed						0.160	0.213	0.564	1.17 (0.77–1.78)	0.453	0.126	0.217	0.335	1.13 (0.74–1.74)	0.563
Father employment
Employed									1					1
Unemployed						0.344	0.191	3.239	1.41 (0.97–2.05)	0.072	0.291	0.194	2.240	1.34 (0.91–1.96)	0.134
House ownership
Own									1					1
Rent						0.340	0.233	2.122	1.41 (0.89–2.22)	0.145	0.209	0.238	0.773	1.23 (0.77–1.97)	0.379
Aggregated risk behaviours											0.251	0.050	25.23	1.29 (1.17–1.42)	<0.001

Notes: β= Beta coefficient; SE= estimated precision of the coefficients; ^OR(95% CI) =Odds ratio and 95% Confidence interval; Model 1 (adjust for age and gender) Hosmer Lemeshow p= 0.639 and Cox and Snell R Square (CSR²) and Nagelkerke R Square (NR²) 0.1%-0.1%; Model 2 (Adjusted for age, gender and socioeconomic variables) Hosmer Lemeshow p= 0.315 and CSR² and NR² = 4.4%-6.6%; Model 3 (Full adjusted for age gender, socioeconomic and aggregated oral health-related risk behaviours variables) Hosmer Lemeshow p=0.925, and CSR² and NR² =8.2%-12.0%; values in bold signify p<0.05.

Table 6.

Logistic Regression Analysis Showing the Association Between Aggregate Risk Oral Health-Related Behaviours and Self-Rated Poor Gingival Health Among School Students in Jazan, Saudi Arabia (N= 687)

Explanatory Variables	Model 1					Model 2					Model 3
	β	SE	Wald χ2	OR (95% CI) ^	p-value	β	SE	Wald χ2	OR (95% CI)	p-value	β	SE	Wald χ2	OR (95%)	p-value
Age	0.047	0.060	0.625	1.05 (0.93–1.18)	0.429	0.017	0.063	0.073	1.02 (0.90–1.15)	0.788	0-0.007	0.065	0.011	0.99 (0.88–1.13)	0.915
Sex
Male				1					1					1
Female	0.068	0.170	0.160	1.07 (0.77–1.49)	0.689	0.159	0.179	0.789	1.17 (083–1.67)	0.375	0.331	0.187	3.120	1.39 (0.96–2.01)	0.077
Family income
Highest (35000 –≥50.000 SR)									1					1
Second middle (25000–34999SR)						−0.616	0.512	1.45	0.54 (0.20–1.47)	0.23	0.171	0.329	0.269	0.52 (0.18–146)	0.211
Middle (16000–24999 SR)						−0.563	0.450	1.56	0.57 (0.24–1.38)	0.21	−0.275	0.376	0.535	0.49 (0.20–1.23)	0.130
Second lowest (9000–15999SR)						−0.116	0.363	0.10	0.89 (0.44–1.82	0.75	−0.709	0.469	2.289	0.76 (0.36–1.59)	0.465
Lowest (< 9000 SR)						0.242	0.318	0.58	1.27 (0.68–2.38)	0.45	−0.660	0.528	1.563	1.19 (0.62–2.26)	0.604
Mother’s education
University									1					1
School						−0.354	0.308	1.316	0.70 (0.38–1.29)	0.251	−0.493	0.318	2.409	0.61 (0.33–1.14)	0.121
Illiterate						−0.188	0.261	0.517	0.83 (0.50–1.38)	0.472	−0.298	0.666	1.217	0.74 (0.44–1.26)	0.270
Father’s education
University									1					1
School						−0.524	0.320	2.674	0.59 (0.32–1.11)	0.102	−0.529	0.330	2.567	0.59 (0.31–1.13)	0.109
Illiterate						0.101	0.242	0.176	1.11 (0.69–1.78)	0.675	0.053	0.248	0.046	1.01 (0.65–1.72)	0.831
Residency area
Urban									1					1
Rural						0.669	0.192	12.079	1.95 (1.34–2.85)	0.001	0.705	0.199	12.598	2.02 (1.37–2.99)	<0.001
Mother employment
Employed									1					1
Unemployed						0.247	0.195	1.607	1.28 (0.87–1.87)	0.205	0.202	0.200	1.018	1.22 (0.83–1.81)	0.313
Father employment
Employed									1					1
Unemployed						0.114	0.173	0.432	1.12 (0.80–1.57)	0.511	0.052	0.178	0.085	1.05 (0.74–1.49)	0.771
House ownership
Own									1					1
Rent						0.022	0.206	0.011	1.02 (0.68–1.53)	0.916	−0.128	0.214	0.357	0.88 (0.58–1.34)	0.550
Aggregated risk behaviours											0.296	0.047	40.415	1.34 (1.23–1.47)	<0.001

Notes: β= Beta coefficient; SE= estimated precision of the coefficients; ^OR=Odds ratio and 95% Confidence interval; Model 1 (adjust for age and gender) Hosmer Lemeshow p= 0.413 and Cox and Snell R Square (CSR²) and Nagelkerke R Square (NR²) 0.1%-0.2%; Model 2 (Adjusted for age, gender and socioeconomic factors) Hosmer Lemeshow p= 0.967and CSR² and NR² = 4.8%-6.6%; Model 3 (Full adjusted for age gender, socioeconomic and aggregated oral health-related risk behaviours) Hosmer Lemeshow p=0.925 and CSR² and NR² =10.8%-14.2%. Values in bold signify p<0.05.

Schoolchildren residing in rural areas were 2.44 times more likely to report poor dental health (Table 5, p<0.001) and twice as likely to report poor gingival health (Table 6, p<0.001) compared to those in urban areas.

Discussion

To our knowledge, this is the first study to examine the association between aggregated common oral health-risk behaviours and their determinants and self-rated dental and gingival health in a sample of 14–18-year-old schoolchildren in Jazan, SA. The analysis revealed that oral health-risk behaviours were significantly associated with self-reported poor dental and gingival health. While statistical significance indicates the likelihood that the association is not due to chance, the strength of association is best understood through the magnitude of the OR. In this study, the OR for self-rated poor dental health was 1.29 (95% CI: 1.17–1.42), and for poor gingival health it was 1.34 (95% CI: 1.23–1.47). According to Chen et al (2010), ORs in this range indicate a small-to-moderate effect size, which is meaningful in public health, especially when behaviours accumulate over time.⁴⁰ These findings suggest a consistent and measurable relationship between the accumulation of risk behaviours and perceived oral health, reinforcing the need for integrated behavioural interventions in oral health promotion.

Because the aggregate oral health-risk behaviour score was the main exposure in this study, the findings should be interpreted in terms of cumulative behavioural burden rather than the effect of any single behaviour alone. Each additional risk behaviour was associated with increased odds of self-rated poor dental and gingival health, indicating that adolescents who accumulate several unfavourable behaviours may experience a compounded increase in perceived oral health risk compared with peers reporting fewer risk behaviours. This supports the importance of multi-component oral health promotion strategies that target several behaviours simultaneously, rather than focusing on isolated practices. This is also consistent with previous evidence showing that health-risk behaviours rarely occur in isolation but tend to co-occur or cluster within the same individuals.^41,42 In the oral health field, similar clustering patterns have been reported among adults with tooth loss¹⁹ and among children with untreated caries,²⁰ supporting the relevance of examining accumulated behaviours rather than isolated risk factors.

The current study’s contribution lies in its use of an aggregated measure of behaviours including oral hygiene practices, dental attendance, reason for dental visits, fruit consumption, sugar intake, sweetened soda consumption and smoking. Most studies investigating the relationship between health-related behaviours and dental or gingival health have examined the impact of multiple separate individual behaviours on dental or gingival health within the same population.^3,13,16,20 It is expected that specific risk factors for dental caries, such as sugar consumption, poor oral hygiene and lack of dental visits, are associated with a higher number of untreated dental caries due to their direct role on disease development. However, this analysis found that the accumulation of health-related risk behaviours was also associated with poor self-rated dental health, including those with direct (age of brushing commencement, brushing duration, toothbrush replacement frequency, flossing), indirect (tobacco use) or inconclusive (fruit consumption) relationships to dental caries. This observation aligns with a United Kingdom study that found that children with a higher number of oral health-risk behaviours were more likely to have untreated dental caries and showed a gradient-like increase with each additional risk behaviour.²⁰ Similarly, previous research has shown that poor diet, inadequate oral hygiene and irregular dental attendance are important contributors to dental caries risk,^13,20,30 and that these behaviours may operate cumulatively. Therefore, the present findings extend existing evidence by showing that the accumulation of oral health-risk behaviours is also reflected in adolescents’ self-rated dental health in a Saudi regional context.

In addition to the aggregate score, the individual behaviour analyses showed that pain- or problem-based dental attendance was one of the strongest predictors of both self-rated poor dental and gingival health. Students who visited the dentist because of pain or dental problems had substantially higher odds of reporting poor oral health than those who attended for check-ups or follow-up care. This finding suggests that dental attendance in this population may be largely symptom-driven rather than preventive, reflecting delayed care-seeking, limited access to routine preventive services, or low perceived need for regular dental visits. This aligns with evidence showing that dental visiting patterns are important determinants of oral health outcomes.⁴³ Therefore, school-based and community-based interventions should aim to shift dental care-seeking from problem-oriented attendance toward regular preventive dental visits.

In the individual behavior analyses, tobacco use was not significantly associated with either self-rated poor dental or gingival health. This should not be interpreted as evidence against the established role of tobacco in oral and periodontal disease. Rather, the null association may reflect the relatively low prevalence of reported tobacco use in this adolescent sample, limited cumulative exposure at this age, and possible underreporting due to social desirability bias. Tobacco was retained in the aggregate index because of its strong biological and public health relevance to oral health risk.

Earlier studies have reported that certain health behaviours tend to cluster in adults and children and are directly linked to tooth loss, untreated caries and irregular dental visits.^19–21 However, limited evidence has specifically examined this clustering in relation to gingival health. In the present study, the aggregate score based on 12 oral health-risk behaviours was significantly associated with self-reported poor gingival health. This finding is important because the literature on behavioural clustering has focused more heavily on dental outcomes such as untreated caries and tooth loss,^19,20 while gingival health has more often been examined in relation to individual behaviours. By demonstrating an association between accumulated risk behaviours and self-rated gingival health, the present study suggests that gingival outcomes may also be shaped by a broader behavioural risk profile rather than by single behaviours alone.

Many studies have also established the association between self-reported or clinically assessed gingival health and factors such as improper oral hygiene practices,⁴⁴ smoking,⁴⁵ infrequent dental visits⁴³ and socioeconomic status,⁴⁴ but have examined each of these factors individually. Considering that oral health-risk behaviours contribute to periodontal inflammation and eventually tooth loss; several studies have explored the association between clusters of such behaviours and adult tooth loss. These studies, which have examined behaviours related to oral hygiene,⁴⁶ smoking, BMI, dental attendance and medical insurance,¹⁹ have confirmed this association. For example, Kolawole et al reported associations between oral hygiene measures and periodontal status among schoolchildren,⁴⁴ while Ayo-Yusuf et al identified smoking as a longitudinal predictor of self-reported gingivitis among adolescents.⁴⁵ These findings support the biological and behavioural plausibility of including oral hygiene, tobacco use and dental attendance in the aggregate risk score used in the present study.

Beyond behavioural accumulation, place of residence emerged as the main shared sociodemographic determinant of both outcomes. Schoolchildren residing in rural areas were 2.44 times more likely to report poor dental health and twice as likely to report poor gingival health compared to those in urban areas. This highlights the commonality of social determinants for both oral health diseases¹⁸ and suggests that the determinants of these aggregated oral risk behaviours in the present study sample extend beyond individual socioeconomic status, such as parental education or family income. The observed rural disadvantage is also in line with studies showing that oral health inequalities are socially patterned and that lower socioeconomic position, limited healthcare access and disadvantaged neighbourhood conditions are associated with poorer oral health outcomes.^47–52

Previous studies have also shown that oral health-risk behaviours are clustered among those with lower education, income and social status, suggesting that those at the bottom of the social hierarchy are more likely to engage in unhealthy behaviours.^{20,21,49,53,54} This finding suggests that oral health-risk behaviours –such as poor dental hygiene practices, poor diet, smoking and infrequent dental visits– are not evenly distributed across the population but tend to be more prevalent in certain groups.⁵⁰ In Saudi Arabia, previous research has also linked socioeconomic indicators, including parental education and income, with dental caries and oral health practices among children and adolescents.^31,36–38 Together, these findings suggest that behavioural risk accumulation should be understood within broader social and environmental contexts rather than interpreted solely as individual choice. A recent geospatial analysis also showed that neighbourhood-level economic advantage was associated with lower edentulism, further supporting the role of structural and area-level determinants in oral health inequalities.⁵¹ It is noteworthy to consider the complexities of social determinants of health in different contexts, and SA is not different.⁵²

The findings from this study have significant implications for public health, particularly in shaping efficient and cost-effective policies and interventions that target aggregated oral hygiene and lifestyle risk behaviours among schoolchildren in SA. Effective health education that begins in childhood has the potential to instil lifelong habits that prevent chronic diseases, improve quality of life and reduce healthcare costs. Specifically, tailored interventions can address the distinct needs highlighted by residency differences in oral health practices, dental attendance, dietary habits and smoking habits. Given that similar risk behaviours are shared across dental and gingival outcomes, interventions that target multiple behaviours simultaneously may be more efficient than programmes focused on single behaviours. This approach is consistent with the common risk factor approach, which supports integrated health promotion strategies addressing shared behavioural and social determinants of disease.¹⁸

The present study has a number of strengths, including the inclusion of both male and female participants and representation from both governmental and private schools, which enhances the diversity and representativeness of the sample. This study, while comprehensive, has several limitations that should be considered. We acknowledge that self-rated measures may be influenced by participants’ knowledge and perceptions. However, self-rated dental and gingival health have been widely used in epidemiological studies and are considered valid indicators of perceived oral health status, especially in large-scale surveys where clinical examination is not feasible due to logistical, financial, or ethical constraints. We also acknowledge that although the questionnaire was based on a standardized WHO instrument and pilot-tested in the present context, a full psychometric validation study was not conducted specifically in this Jazan school population. The cross-sectional design also limits the ability to establish causality. Moreover, the use of a convenience sample of students limits the generalisability of the findings, as it may not accurately represent the broader population of Jazan school children as well as those who did not enroll in middle and secondary schools during the study. Additionally, the higher proportion of adolescents from public school may introduce bias. To mitigate this limitation, the study included a wide range of demographic variables to provide a detailed analysis of the data, enhancing the understanding of different subgroup behaviours within the convenience sample. Additionally, schools serving students with special needs were not included due to financial and logistical constraints. Therefore, the findings may not be generalisable to adolescents with special educational or healthcare needs, who may have distinct oral health risks and access-to-care barriers. Furthermore, the absence of intra-oral clinical examinations, the use of a self-administered electronic questionnaire, and the exclusive focus on adolescents, may also limit the generalisability of the findings to other age groups and may be affected by recall bias and social desirability particularly for tobacco use. The aggregate score assigned equal weight to all 12 behaviors. This simplifying assumption may have diluted or amplified the observed associations, as some behaviors, such as tobacco use or frequent sugar intake, may have stronger biological links to oral disease than others, such as toothbrush replacement timing or fruit consumption. Future studies could explore weighted indices or latent class approaches. Furthermore, the logistic regression models (Table 5 and Table 6) demonstrated modest explanatory power for self-rated dental health (Cox and Snell R² = 8.2%; Nagelkerke R² = 12.0%) and self-rated gingival health (Cox and Snell R² = 10.8%; Nagelkerke R² = 14.2%), suggesting that additional unmeasured factors may influence these outcomes. This is consistent with the multifactorial nature of self-reported oral health and the use of an aggregate oral health behavior measure within a single-area convenience sample. Finally, there are additional behavioural factors not covered by the survey that could influence dental and gingival health. It is noteworthy to mention that the comparisons of the present study findings should be considered with caution when compared with different age groups and aggregate oral health-related risk behaviours. Future research is required to consider optimal cut-off points for aggregate oral health risk behaviours for both dental and gingival diseases to facilitate results interpretation and presentation, given the cost of dichotomizing or trichotomising continuous variables, as highlighted by Altman and Royston.⁵⁵

Conclusion

This study demonstrated a significant association between aggregated oral health-related risk behaviors and self-reported poor dental and gingival health among schoolchildren in the Jazan region of SA. The area of residency emerged as a key shared social determinant, with students living in rural areas disproportionately impacted. These findings highlight the need for comprehensive and context-sensitive public health interventions that target multiple interrelated risk behaviors rather than isolated factors.

Future oral health promotion efforts should prioritize multi-component school-based programs that incorporate behavioral education, regular dental screenings, and parental engagement—particularly in underserved rural areas. Integrating oral health education into the broader school curriculum and involving community leaders may also improve health literacy and foster sustainable behavior change. Furthermore, national strategies should consider addressing structural disparities such as access to care and socioeconomic inequalities to reduce the oral health burden among adolescents in high-risk settings. Additionally, oral health promotion could be integrated into existing school health, science, or physical education modules through structured behavior-tracking activities, supervised toothbrushing education, dietary awareness sessions, tobacco-prevention messages, and referral pathways for preventive dental check-ups. Such integration may support sustainable behavior change while reducing reliance on symptom-driven dental care.

Practical Relevance

Scientific Rationale for the Study

Among the school children in southern SA reporting poor dental and gingival health, there was no comprehensive study of social determinants and aggregate oral-related risk behaviours. Dental and gingival diseases share modifiable behavioral and social determinants. In southern Saudi Arabia, evidence on how these behaviors accumulate among adolescents and relate to self-rated dental and gingival health remains limited.

Principal Findings

This study showed that each additional oral health-risk behavior was associated with higher odds of self-rated poor dental and gingival health. Rural residence was also a shared determinant of both outcomes, suggesting that behavioral risks and structural inequalities may jointly shape adolescent oral health.

Implications and Recommendations

These findings support the need for multi-component school-based oral health programs that combine oral hygiene education, dietary counseling, tobacco prevention, preventive dental referral pathways, and parental engagement. Oral health promotion could be integrated into existing school health, science, or physical education modules through structured behavior-tracking activities, supervised toothbrushing education, dietary awareness sessions, and referral pathways for preventive dental check-ups. Public health practitioners should prioritize rural schools and strengthen links between schools, primary healthcare centers, and dental services to promote preventive rather than problem-driven dental attendance.

Funding Statement

No external funding was received for this study.

Data Sharing Statement

The data are available from Prof. Alla Alsharif upon request.

Author Contributions

All authors made a significant contribution to the work reported, whether that is in the conception, study design, execution, acquisition of data, analysis and interpretation, or in all these areas; took part in drafting, revising or critically reviewing the article; gave final approval of the version to be published; have agreed on the journal to which the article has been submitted; and agree to be accountable for all aspects of the work.

Disclosure

All authors declare no conflicts of interest in this work.