Distal and Proximal Influences on Self-Reported Oral Pain and Self-Rated Oral Health Status in Saudi Arabia: Retrospective Study Using a 2017 Nationwide Database

Introduction

Oral health has a major impact on overall health, medical costs, and quality of life. Major oral conditions include dental caries, periodontal disease, and tooth loss. Between 1990 and 2017, the global burden of these conditions increased by 38% [1].

There are reports of an increase in the burden of oral diseases in the Kingdom of Saudi Arabia (KSA) over the last few decades [2]. This increase is likely due to transformations in lifestyle, such as changes in dietary habits, particularly an increase in consumption of sugary foods and tobacco products [3]. Thus, oral health conditions constitute one of the major public health concerns in KSA.

Self-reported oral health status has been used as an important subjective health indicator of oral health care needs and to evaluate the individual’s quality of life [4]. Self-reported information is a cost-effective and time-saving method of data collection. Self-reported oral health can be affected by several factors, such as sociodemographic and socioeconomic factors, cultural values and beliefs, and existing oral health conditions [5].

The Multidimensional Conceptual Model of Oral Health proposed by Gilbert et al [6] states that oral diseases and related tissue damage can result in oral pain (OP) and challenges in daily living that affect self-rated oral health (SROH) status. OP can cause difficulties in chewing and sleep disturbances [7]. In addition, it can affect school and work attendance, causing a loss of a significant number of study and working hours per year [8]. Because of these concerns, OP is frequently incorporated into national health surveys. A 1989 report from the United States reported that 14.5% of adults experienced OP during the past 6 months [9], while in the United Kingdom, 28% of adults were reported to experience difficulty from OP during the past year in 1998 [10].

SROH status serves as a valuable indicator of general oral health status [6]. It is considered a comprehensive index reflecting various dimensions of oral health, including functional, psychological, and social impacts on overall well-being [11]. It has been linked to clinical oral health status, such as dental caries, tooth mobility, and tooth loss [4]. Furthermore, SROH has been found to predict future oral health outcomes, as seen in longitudinal studies assessing maternal SROH and their children's caries experience in adulthood [12].

Distal and proximal influences play significant roles in shaping oral health outcomes such as OP and SROH. Proximal influences such as oral health-related behaviors and the use of oral health services directly impact oral health [13]. On the other hand, distal influences encompass broader determinants such as socioeconomic status and access to care determinants, which also have a substantial influence on oral health outcomes [14]. Understanding the interplay between distal and proximal influences is essential for addressing oral health status among populations and developing effective interventions to improve oral health outcomes across diverse populations.

The use of conceptual frameworks for understanding determinants in oral health research can serve as a coherent map to guide researchers when inquiring about oral health conditions. Conceptual frameworks can also help researchers to include multiple factors that may explain an outcome and aid in designing statistical analyses [15]. The objective of this study was to explore how proximal and distal influences on oral health are related to both OP experience and SROH status among KSA residents by using data from a national demographic and health survey (DHS) that was conducted in 2017 in KSA. A conceptual framework was developed (Figure 1) to guide the analysis.

‎

Methods

Data Source

The original data collection was approved by the institutional review board (IRB) of the Ministry of Health of the Kingdom of Saudi Arabia (Central IRB Log #: 2019-0131M). No additional IRB approval was needed for the secondary analysis, as it qualifies under Exemption 4 of US federal regulations [45 CFR 46.104(d)(4)] due to the use of existing, nonidentifiable data. The authors have permission to use the data, which was collected with participant consent. Data analysis was conducted at the Indiana University School of Dentistry, the Department of Biostatistics at the Indiana University School of Medicine, and the Richard M. Fairbanks School of Public Health, Indianapolis. The Human Research Protection Program from the Office of Research Compliance at Indiana University determined that this secondary analysis does not require further IRB review (Protocol #: 1808825963). Neither the study principal investigator nor key personnel had any financial conflict of interest concerning this research. The data were available at the office of the Directorate of Primary Health Care Centers (Ministry of Health, Headquarters, Riyadh, KSA). The Ministry of Health used a probability multistage stratified random sampling for the DHS. Details of the sampling procedure were published previously [16]. Briefly, house-to-house visits were conducted to interview the head of a family or an eligible representative and other specific family members between February 12, 2017, and May 23, 2017. Participants answered questions related to demographic, environmental, and health-related topics. The data were received in SPSS (IBM Corp) software format, and an analysis file was created, which comprised selected variables of interest. In total, 3 parallel analyses were performed based on the age of the respondent: children, 5-14 years; adolescents, 15-24 years; and adults, ≥25 years (details are given in [14]). The analysis was done at the Biostatistics Department, Indiana University School of Medicine, Indianapolis, Indiana, United States (IRB protocol number: 1808825963).

Development of the Model

Based on existing models, a multi-level conceptual framework was developed for oral health influences in KSA (Figure 1) on self-reported OP and SROH status among KSA residents. Constructs from the Multidimensional Conceptual Model of Oral Health proposed by Gilbert et al [6] and the World Health Organization Model for Oral Health Surveillance [17] were adapted by expanding the concept of proximal (direct) influences on oral health—such as diet and oral hygiene—to include distal (indirect) influences such as socioeconomic determinants.

Selection of Endogenous and Exogenous Variables

Model variables were selected after a careful review of the literature, identifying those that were both available in the survey and previously reported to influence OP and SROH status. The model variables were classified as exogenous and endogenous. Exogenous variables are those that are not affected by other variables in the model (the distal or indirect influences), while endogenous variables are affected by other variables in the model, such as proximal influences and outcome variables. In total, 13 exogenous variables were included for the adult group, and 11 and 12 exogenous variables were included for the adolescent and children groups, respectively. In total, 8 endogenous variables were included for the adult group and 7 endogenous variables were included for the adolescent and children's groups, respectively. The variables are listed in (Table 1). Exogenous variables (distal influence variables) included the age [18] of participants as a continuous variable and gender [19] as a binary variable (males or females). Citizenship [20] status was coded as a binary variable: citizens and noncitizens. Geographic regions [2] were classified into the East, West, and Central versus the North and South. Marital status was dichotomized into currently married and not married (≥25 years only). Completed education level [21] was categorized as primary, intermediate, high school, intermediate diploma, and college or higher education. In total, 5 levels of household monthly income [22] were included: lower class income (3800 Riyals or less), marginal middle-class income (3801-7699 Riyals), basic middle-class income (7700-22,900 Riyals), upper middle-class income (22,901-38,200 Riyals), and upper-class income (>38,200 Riyals). Household crowding was calculated by dividing the number of family members by the number of sleeping rooms. The responses were then grouped into 4 levels: <1, 1-2, 2-3, and >3 persons per room. Past accident experience and physical disability were assessed as binary no or yes responses. BMI was dichotomized as normal (BMI=18.5-24.9) and abnormal (BMI <18.5 and >24.9). Health insurance was expressed as a binary variable of insured versus not insured. Access to oral health services [23] in the year prior to the survey was dichotomized into “no or I do not know and yes” responses. The source of dental care was dichotomized into a government versus private clinic.

Endogenous variables (proximal influences and outcome variables) included the frequency of consuming sweets [24], individuals responded to the following question “How often do you eat sweets?” as “I don't eat at all, many times per month, once per week, many times per week, once per day, many times per day.” For soft drinks consumption frequency [25], “How often do you drink soft drinks?” responses were “I don't drink at all, many times per month, once per week, many times per week, once per day, many times per day.” For smoking status, [26] yes or no responses to the question “Do you smoke?” Frequency of tooth brushing [27,28] had 6 levels: “I have never cleaned my teeth, I clean my teeth some days but not daily, once weekly, many times per week, once daily, twice or more daily.” The frequency of dental visits was determined as from response to the question “How many times have you visited a dentist in the past year?” Valid answers include “never visited a dentist/I do not know or do not remember, did not visit the dentist in the past year, once, more than once.” For the type of visit, responses were dichotomized into visits for a complaint versus visits for routine examination and treatment [22]. For OP [6], the question was phrased as: “How many times during the past year have you felt pain in your teeth?” There were 4 levels of response: “never felt, rarely, sometimes, many times.” For SROH [6], participants were asked, “How would you describe the health of your teeth and gums?” Responses were “bad, acceptable, good, very good, excellent.”

Data Analyses

SPSS (IBM Corp) was used to perform descriptive analysis for the model variables (Table 1). R software (R Foundation for Statistical Computing) was used to perform the path analysis considering the sample weights. Missing values were not replaced or imputed in this analysis. The first step was to assess the multivariate normality of endogenous variables. Both skewness and kurtosis statistics confirmed that endogenous variables did not follow a multivariate normal distribution (P<.05). Owing to the presence of non-normal and missing data, full-information maximum likelihood estimation to perform path analysis available in the lavaan package (version 0.6.12) was used [29]. Robust standard errors (Huber-White) and scaled test statistics were calculated [30]. The software estimated the direct effect, as hypothesized in the model in (Figure 1), of each oral health influence, as well as the indirect effect for each exogenous variable on OP and SROH status through a path mediated by each proximal influence on oral health (Figure 1). For example, the effect of sex on OP was mediated by the frequency of tooth brushing. The total indirect effects on OP and SROH status reflected the effect of the path between each exogenous variable via all proximal influences on oral health. The total effects comprised the sum of the total indirect and direct effects of each distal influence on OP and SROH status.

A separate model was estimated for each age group—children 5–14 years, adolescents 15–24 years, and adults ≥25 years. Model fit was evaluated using the robust comparative fit index>0.9, robust Tucker-Lewis index>0.9, robust root mean-square-error of approximation <0.08, and robust standardized root-mean-square residual<0.08 [31].

Results

The conceptual model (Figure 1) states that OP and SROH status are directly influenced by distal and proximal influences on oral health. Furthermore, OP and SROH status are indirectly influenced by distal influences via all proximal influences except past accident experience, physical disability, and BMI, where they were indirectly influenced by OP and SROH status via only dental visit frequency, type of visit, and frequency of tooth brushing (Figure 1).

The final analysis included 29,274 adults ≥25 years of age (mean 42.2, SD 12.97), 9910 adolescents aged 15–24 (mean 20.4, SD 2.98) years, and 11,653 children aged 5 and 14 (mean 10.58, SD 2.84) years. Complete descriptive statistics are published elsewhere [16]. Table 1 presents a summary of the weighted and non-weighted estimates.

Despite 39% children, 48.5% adolescents, and 47.1% adults reporting OP in the past year, 92.9% children, 87.1% of adolescents, and 87.1% adults reported good, very good, or excellent SROH status, respectively.

The model goodness-of-fit measures showed an acceptable fit to the data, meeting the recommended values for the fit statistics (Table 2) [31].

Proximal Influences on Oral Health Effects

A higher tooth brushing frequency was strongly associated with less OP and positive SROH status in all groups. In contrast, a higher number of dental visits was associated with more OP and less favorable SROH status in all age groups. Routine examination and treatment were linked to less OP in all age groups and better SROH status in the adult and adolescent groups. Consumption of sweets was linked to greater OP in adolescents (β=0.033, P=.007) and negative SROH status in children (β=–0.086, P<.001), adolescents (β=–0.079, P<.001), and adults (β=–0.068, P<.001). Soft drinks were linked to lower OP in the adolescent group (β=–0.034, P=.005). Higher consumption of soft drinks was associated positively with SROH status (β=0.063, P<.001) in the adolescent and adult groups (β=0.068, P<.001). Smoking was associated with more OP (β=0.030, P<.001) in the adult group (Table 3).

Distal Influences on Oral Health Effects

Tables 3-5 illustrate the direct, total indirect, and total effects, respectively, of oral health influences on OP and SROH status. Tables S1, S2, and S3 in Multimedia Appendix 1 show the direct effects of the distal influences on the proximal influences and the indirect effects of each distal influence on both OP and SROH status via each proximal influence.

Age and Sex

There was a negative direct effect between age and SROH (Table 3) in both the children (β=–0.044, P=.001) and adolescent (β=–0.033, P=.008) groups. An indirect positive effect (β=0.062, P<.001) was found between age and OP in the children group (Table 4). Total effect (Table 5) of age was detected in the children group for both OP (positive relation) and SROH status (negative relation).

Among female children, a negative direct effect (β=–0.030, P=.009) was found with SROH (Table 3). Total indirect effect showed a negative association between female adults and OP (β=–0.016, P=.04; Table 4). This total indirect effect was mediated by dental visits and tooth brushing frequency, as a positive direct link was found between female sex and tooth brushing frequency in both the child (β=0.073, P<.001) and adult groups (β=0.026, P=.04; Tables S1 and S3 in Multimedia Appendix 1). Also, the total indirect effect revealed a positive association between female children (β=0.006, P=.04) and adults (β=0.006, P=.03) with SROH (Table 4). However, total effect showed a negative association between female children and SROH (β=–0.024, P=.04; Table 5).

Citizenship, Regions, and Education Levels

Among non-Saudi citizens, the direct (β=–0.081, P<.001) and total (β=–0.085, P<.001) effects showed a negative association with OP in the children’s age group but a positive association in the adult group (β=0.023, P=.02) through the total indirect effect.

Adolescents and adults from the north and south regions were linked to less OP than those from the east, west, and central regions through the total (β=–0.033, P=.01 and β=–0.045, P<.001) effects pathways (Table 5). However, OP was positively linked to children from the southern and northern regions through the total indirect pathway (β=0.027, P<.001).

Higher education level was associated with positive SROH status directly (β=0.046, P=.003) and through the total direct effect pathway (β=0.050, P=.002). Furthermore, higher education was linked with greater tooth brushing frequency and less OP indirectly via tooth brushing frequency (β=–0.004, P<.001) and positive SROH status (β=0.008, P<.001) (Table S3 in Multimedia Appendix 1).

Monthly Household Income and Insurance

In the children group, higher household income was associated with less OP through the direct pathway (β=–0.030, P=.02), but higher income was positively associated with greater OP through the total indirect pathway (β=0.024, P=.003). In addition, higher income was associated with less OP (β=–0.002, P=.03) and a positive SROH status (β=0.006, P=.007) when mediated by tooth brushing frequency (Table S1 in Multimedia Appendix 1).

In the children group, having health insurance was associated with less OP through direct (β=–0.36, P=.03), total indirect (β=–0.023, P=.03), and total effect pathways (β=–0.058, P=.003). On the other hand, insurance was associated positively with OP through the direct pathway in both adolescents (β=0.051, P=.004) and adults (β=0.042, P<.001). Moreover, in adults, insurance was linked to less OP indirectly via dental visit frequency (β=–0.028, P=.01).

Access to Oral Health Services

In the children group, lack of access to oral health care was linked to more OP and negative SROH status through direct, total indirect, and total effects (Tables 2-4).

Private clinic visits as the regular source of dental care were associated with more OP and worse SROH status in the children group through the direct (for OP: β=0.053, P=.001 and for SROH status: β=–0.136, P<.001, respectively) and through total effect pathways (for OP: β=0.039, P=.03; for SROH status: β=–0.124, P<.001). However, private clinic visits as the regular source of dental care were associated with better SROH status through a total indirect effect (β=0.012, P=.009). Private clinic as the regular source of dental care was associated with less OP and better SROH status in the children group through the indirect effect via tooth brushing frequency (for OP: β=–0.005, P=.003; for SROH status: β=0.017, P<.001).

Past Accident Experience and Physical Disability

In the adult group, past accident experience was negatively associated with SROH status through the direct (β=–0.043, P<.001) and total effects pathways (β=–0.036, P=.002) and positively associated with SROH status through the total indirect pathway (β=0.008, P=.01).

Physical disability was linked to OP and negative SROH status in the child group through the total-effects pathway (Table 4). In the adult group, physical disability was associated with greater OP and worse SROH status through the direct and total effect pathways (Tables 2 and 4).

BMI

Normal BMI was linked to better SROH status through the direct pathway (β=0.051, P<.001) and through the total effect pathway (β=0.047, P<.001). In the adolescent group, normal BMI was linked to less OP and better SROH status through the direct and total effects pathways (Tables 2 and 4). In the adult group, normal BMI was linked to less OP through the total indirect and total effect pathways (Tables 3 and 4).

Discussion

Principal Findings

The findings of this study partially support the adapted conceptual framework for distal and proximal influences on self-reported OP and SROH status in Saudi Arabian residents (Figure 1). All proximal influences were associated with OP, except sweets and soft drink consumption in the children and adult groups. Unexpectedly, the consumption of soft drinks was associated with less OP in the adolescent group, although OP was positively associated with greater consumption of sweets. Moreover, all proximal influences were associated with SROH status except soft drink consumption and type of dental visit in the children’s age group and smoking in the adult group. In addition, greater consumption of sweets and a higher number of dental visits were negatively associated with SROH status in the adolescent and children groups, but soft drink consumption was positively linked to SROH status in the adolescent and adult groups. Regarding distal influences, the majority of them showed an association with both OP and SROH status through the direct, indirect, and total effect pathways.

This study indicated that the prevalence of OP in Saudi Arabia is high. Data from 20 different countries in a meta-analysis illustrated that OP prevalence in children and adolescents was 36.2% [32]. This is almost half of the prevalence found in this study in similar age groups (60.2% for children and 67.8% for adolescents in KSA). Furthermore, the findings from this study differed from National Health and Nutrition Examination Survey (NHANES) 2015-2018 data [33] on adult US residents where OP was associated with education and income level, while in KSA it was not.

The findings from this study indicated that dental visit frequency was positively associated with OP and negatively associated with SROH status, while routine visits were associated with less pain and better SROH status in all age groups. It would be expected that a higher frequency of dental visits would be associated with better oral health; however, it appears that Saudi residents visit the dentist mainly when they have a complaint, such as OP, and do not normally visit the dentist for a routine check-up. A similar trend has been previously reported in national [34] and some subnational studies in Saudi Arabia [35-37]. The lack of interest in routine oral examinations and treatment visits (9.0% of children, 6.2% of adolescents, and 5.4% of adults reported routine dental visits) can be explained by the habitually optimistic view of the Saudi population about their oral health, as 92.9% of children, 87.1% of adolescents, and adults self-rated their oral health as good to excellent despite roughly two-thirds of them having reportedly experienced OP during the previous 6 months. Infrequent routine visits and optimistic SROH status combined with less frequent tooth brushing is a critical finding of this study because the conclusion of these combined observations is that Saudi residents do not recognize risk factors that can adversely affect their oral health. This could be due to the cultures and beliefs in KSA, which are reinforced by the family and community [38]. Oral health officials and policy makers should consider raising public awareness about the importance of oral hygiene and routine dental clinic visits to improve oral health in the general population.

This study had several important limitations. Its cross-sectional nature, like all similar DHS studies, limits the ability to investigate distal and proximal influences on OP and SROH status over time, hindering the assessment of causal relationships between the predictive factors considered and the study outcome. On the other hand, this study is the first to assess the association between a number of predictive factors and oral health outcomes using a conceptual framework to guide the analysis. To our knowledge, no longitudinal study has measured changes in oral health status in Saudi Arabia, which is needed to determine oral health risk factors specific to this country’s population.

Another major limitation in this study is the subjectivity of a self-reported survey over objective clinical evaluation, which may introduce response bias such as recall bias [39] in the responses (such as dental visit frequency in the past year) or social desirability bias [40] (such as not revealing smoking status). Such biases may have influenced the response to the question related to the consumption of soft drinks, particularly in the adolescent group. Adolescents who consume more soft drinks might underreport their OP due to a perception that admitting to pain could lead to restrictions on their soft drink consumption by parents or guardians [41], which in turn may have resulted in the apparent association of increased soft drink consumption with lower OP and better SROH status, which goes against clear evidence of an association between higher soft drink consumption and negative oral health outcomes in a number of cross-sectional and longitudinal studies [42]. Furthermore, self-reported OP and SROH status may not reflect exact oral health clinical status because they express a person’s perception of their OP and oral health, which can be influenced by psychosocial and cultural factors. For example, NHANES 2003-2004 data from the United States showed that Latinos reported better SROH status than White individuals, while Latino individuals had more oral disease and lower access to and use of dental care [43]. Although using SROH indicators over a clinical measurement may have introduced some reporting bias, it is still a convenient, cost-effective, and expedited method to assess oral health status at a national level with a large sample and has shown a positive association with clinical oral health status [20]. Self-reported oral health measures have been used in many countries and national surveys, such as NHANES [33,43].

Despite these limitations, this study has a number of strengths, including the use of data from the 2017 KSA DHS, which applied a random sampling design with a large sample size, over a wide geographic distribution, and broad age range, which ensured the representative nature of the findings to the entire KSA, a heterogeneous country characterized by areas of very high income within a developing country. The second advantage is the use of the conceptual framework to guide the analysis, which enabled the use of path analysis. This is a preferred approach to delineate complex relationships, including both the direct and indirect effects of numerous predictive factors, over traditional multiple regression methods. This enabled testing the direct and indirect effects of different oral health influences on SROH status and OP. In this regard, the conceptual framework in this study is a significant contribution to the understanding of oral health influences in Saudi Arabia. However, the inclusion of other important oral health influences, such as coping skills and social support constructs, could have increased the explanatory power of the study [44].

Conclusions

Although OP is prevalent among Saudi residents, they still have a positive view of their oral health. Frequent tooth brushing, routine dental visits, and reduced sweet consumption are associated with less OP and better SROH. However, more frequent dental visits seem to address complaints rather than preventive care. Future research should investigate why residents have a positive perception of oral health despite high levels of OP and negative outcomes.