This study is a cross-sectional survey. We collected the data over a 10-week period from September 9, 2024, to November 17, 2024, using an Arabic-language questionnaire developed on SurveyMonkey (SurveyMonkey Inc) among adults aged 15 years and older residing in Saudi Arabia. The questionnaire was initially developed in English and translated into Arabic to suit the target population. The questionnaire link was distributed through targeted web-based advertisements on the most popular social media platforms in Saudi Arabia [6], including Instagram, X (Twitter), Snapchat, and TikTok. Participants clicking on the advertisements were redirected to a dedicated web page containing a detailed study description and consent form, followed by the questionnaire. At the start of the survey, an introductory statement provided information on the study objectives, assured confidentiality, and included the researcher’s contact information. We obtained the consent via an “Agree” button, which participants clicked to access the survey. Participants were informed that the survey would take approximately five minutes to complete. They were assured that participation is entirely voluntary, with the option to skip questions or withdraw at any time.

The study was conducted in a digital setting with an open survey accessible to anyone who clicked on the advertisements, using widely used social media platforms in Saudi Arabia (Instagram, X [Twitter], Snapchat, and TikTok) for recruitment. These platforms were chosen due to their high penetration rates and active use in Saudi Arabia [6], ensuring ease of access and effective targeting through their advertising systems. We designed the social media–targeted advertisements to be culturally appropriate and visually engaging, highlighting the voluntary nature of participation and the study’s focus on oral health–related information. The study used a simple random sampling approach facilitated by the advertisement algorithms. These algorithms were used to define the target group—adults aged 15 years and older residing in Saudi Arabia—ensuring that all individuals within this demographic had an equal chance of seeing the advertisement. According to 2022 population data from the General Authority of Statistics (GASTAT) [30] and 2023 internet penetration figures from the Communications, Space, and Technology Commission [6], the target population was estimated at 24,045,450 individuals, representing 99% of the total population in Saudi Arabia who are internet users within the targeted age group. This recruitment strategy aimed to: ensure representation of the population of interest by using an algorithm without prespecified interests, thereby targeting all individuals within the defined demographic; and maximize geographical diversity by specifying the algorithm to reach individuals residing in Saudi Arabia.

However, it is important to acknowledge that participation was voluntary, introducing self-selection bias that may affect the sample’s generalizability. Despite this, the approach was a practical and effective method for recruiting a large, diverse population.

To effectively reach a broad and diverse sample, we launched advertisement campaigns across Instagram, X (Twitter), Snapchat, and TikTok. User interactions were monitored throughout the campaigns to evaluate reach and sample representativeness. Snapchat and TikTok provided us with strategic support to optimize campaign strategies. Snapchat emphasized targeting mechanisms such as predefined and custom audiences, advanced demographics, and accurate algorithmic delivery, which enhanced the ability to reach users based on demographics, location, and language. TikTok highlighted its robust targeting tools, including lookalike or custom audiences, ensuring advertisements were delivered to relevant users while monitoring user interactions like click-through and conversion rates. Although Instagram and X (Twitter) did not provide direct support, their platforms offered advanced targeting capabilities via Ads Manager tools. These tools allowed for audience segmentation by age, location, and sex, enabling effective recruitment across diverse demographics.

To estimate the prevalence of oral health–related misconceptions, with a 5% margin of error and a 95% CI, a sample size of 384 participants was required (assuming a 50% prevalence to maximize sample size requirements). Anticipating a low participation rate typical of web-based surveys, we assumed a 2% participation rate, leading us to target the reach of approximately 19,200 users in our advertisement campaigns across all platforms to ensure the required sample size. Reach refers to the estimated number of users who would see the advertisement, with metrics from Instagram, X (Twitter), Snapchat, and TikTok helping us track both views and link clicks.

We included both male and female participants aged 15 years and older residing in Saudi Arabia and those who provided informed consent to participate in the study. The age threshold of 15 was selected based on the classification by the GASTAT in Saudi Arabia, which classifies individuals aged 15 years and older as adults for statistical and demographic purposes [31]. Additionally, adolescents in this age group are active users of social media and are likely to be exposed to oral health information digitally, making their inclusion essential to assess the study objectives.

We structured the questionnaire into five domains with a total of 29 fixed items distributed across six screens designed to align with the study’s objectives and draw on previously published and validated cross-sectional studies. Face validity was established through expert review by a panel of dental public health specialists. Experts assessed the clarity, relevance, and domain alignment of each item, as well as the overall comprehensiveness of the questionnaire within the study’s objectives. We implemented minor revisions to item wording and categorization based on their feedback (Multimedia Appendix 1). No adaptive questioning was used; all items required mandatory responses prior to proceeding to subsequent sections pages. Respondents could review and modify their answers using the “Back” button.

The first domain assessed participants’ use patterns and preferred platforms with one item [6,32-35]. Participants were asked to select their preferred social media platforms across the popular platforms in Saudi Arabia, which included X (Twitter), Snapchat, WhatsApp, Instagram, Facebook, TikTok, Telegram, YouTube, Line, and LinkedIn [6]. Responses were categorized as “more than once a day,” “once a day,” “2-3 times a week,” “4-6 times a week,” “once per week” or “once per month,” and “I don’t use this platform.”

The second domain, containing 6 items, focused on the engagement with oral health information obtained from social media [6,32,33,36,37]. Participants were asked binary (yes or no) questions, including whether they “use social media to search for oral health information,” “ever received oral health info from nondental professionals on social media,” “following dental-specific accounts on social media,” “trust in oral health information from social media,” and “influence their trust by social media profile content.” Perception of social media as a good source for oral health information was assessed with response options: “yes,” “sometimes,” and “no.” We dichotomize it to combine “yes” and “sometimes” into a single category representing participants who perceived social media as a good source of oral health information.

To capture self-reported oral health, we included a single item in the third domain using the Global Self-Ratings of Oral Health question for participants to rate their own oral health status on a 3-point Likert scale as “good,” “average,” and “poor.” [38]. For analysis, self-reported poor oral health was defined as participants selecting “poor,” while better oral health perceptions included those reporting “good” oral health. Participants who neither rated their oral health as good nor poor, reflecting a moderate perception of their oral health, included those reporting “average” status.

The fourth domain presented 14 statements on common oral health misconceptions in Saudi Arabia, derived from literature [27,28], with response options of “agree,” “disagree,” and “don’t know.” We combined responses of “disagree” and “don’t know” into a single category representing participants who do not believe in the misconceptions. The category “agree” includes those who do agree with these misconceptions.

In the fifth domain, we collected sociodemographic information across seven items: age, which was categorized into five groups: 15-24, 25-34, 35-44, 45-54, and 55+ years; sex; education, which was classified into “high school or less,” “associate or intermediate diploma,” “bachelor’s degree,” and “master’s or PhD”; occupation, which was grouped into “student,” “governmental worker,” “private worker,” and “unemployed”; marital status, “single,” “married,” or “other”; region of residence, included the 13 administrative regions of Saudi Arabia: “Riyadh,” “Makkah,” “Madinah,” “Qassim,” “Eastern Region,” “Asir Region,” “Tabuk,” “Hail,” “Northern Region,” “Jazan,” “Najran,” “Al-Baha,” and “Al-Jawf”; and nationality, “Saudi” and “Non-Saudi” within Saudi Arabia. Age and region categories were based on distributions from the GASTAT [30,39], and education levels followed the National Qualifications Framework 2023 [40].

In a preliminary phase of the study, we conducted a pilot to confirm the validity and relevance of this study prior to full implementation. We targeted a sample group representing 10% of the total study sample size, with 10 participants recruited from each of the four social media platforms, which are Instagram, X (Twitter), Snapchat, and TikTok, resulting in a total of 40 participants. Participants responded positively to the questionnaire, navigating it without difficulty. Based on feedback, we made two key adjustments to improve the survey. First, we added a “sometimes” response option to the question in the second domain, “Do you think social media is a good source for obtaining oral health information?” to provide a wider range of responses. Second, we simplified certain Arabic terms in the misconception domain to make the statements clear to participants. This pilot phase was conducted from June 17, 2024, to September 2, 2024. Participants involved in this pilot were not included in the final study sample.

Counts and percentages were reported to describe participants’ sociodemographic characteristics, their engagement with oral health information, and their self-reported oral health. We used the chi-square test to examine potential associations between the 3 domains and sex, identifying statistically significant relationships where applicable.

For the multivariable analysis, the independent variables included engagement with the oral health information domain (the manner of participants’ engagement with oral health–related information in social media measured via the six binary items, each item entered as a separate exposure in the model) and self-reported oral health (participants’ subjective evaluation of their oral health status measured via 3-point Likert scale). The dependent variables were defined as:

To assess these relationships, we used logistic regression models to present both crude (odds ratio) and adjusted odds ratio (AOR), with respective 95% CI, to examine engagement with oral health information and self-reported oral health (exposure variables) with the presence of any oral health misconceptions among social media users (outcome). Logistic regression was selected due to its appropriateness for modeling binary outcomes, allowing us to examine the likelihood of participants holding at least one misconception. Poisson regression is used to report the mean ratio and adjusted mean ratio (AMR) of oral health misconceptions (outcome) by the engagement with oral health information and self-reported oral health (exposure variables). Poisson regression was chosen because it is suitable for modeling count data, particularly when the outcome variable represents a frequency count of misconceptions per participant. We confirmed the appropriateness of the Poisson model by assessing equidispersion using the Pearson chi-square statistic, which indicated no overdispersion. Sex, education (as a measure of socioeconomic status), and nationality were selected and adjusted as potential confounders based on their theoretical relevance [41] and observed associations with both the exposure and the outcome in exploratory analyses.

A sensitivity analysis was performed for both regression models to assess the potential impact on the results if the response “don’t know” was reclassified as “agree” in the 14 statements addressing oral health misconceptions. We reclassified “don’t know” responses based on the assumption that these individuals might be more susceptible to misconceptions due to a lack of knowledge or confidence. This provided a more conservative, robust estimate of misconception prevalence, allowing us to examine the influence of uncertainty on our findings across different analyses. A 2-tailed α level of .05 was used to test relationships in both directions, and all statistical analyses were conducted using Stata/BE (version 17.0; StataCorp).

This study was conducted in compliance with the ethical standards set by the institutional review board of King Saud University. Research ethical approval was obtained, and the study was registered with the Institutional Review Board from the Committee of Human and Social Research Ethics, King Saud University (registration KSU-HE-23848). Participants were informed about the study objectives, procedures, and their rights before participating in the survey. Informed consent was obtained electronically via a mandatory “Agree” button at the start of the questionnaire. Only individuals who provided consent were able to proceed with the survey. The survey content was specifically designed to be low-risk without including sensitive or intrusive questions. Data were collected anonymously, with no personally identifiable information linked to the responses. The platform (SurveyMonkey) complied with strict data security and privacy protocols to ensure the protection of participants’ responses. IP addresses were monitored to identify duplicate entries within a 24-hour period and then omitted before analysis. Participation was voluntary, and no monetary or material compensation was provided to participants in this study. This study did not involve the use of images or media containing identifiable participants.

Using targeted social media campaigns across Instagram, X (Twitter), TikTok, and Snapchat, we observed notable differences in reach and interaction across platforms. TikTok achieved the highest reach (836,477) and link clicks (67,678) during its 30-day campaign, followed by Instagram, which reached (53,341) users and generated (1007) link clicks over 60 days. X (Twitter) and Snapchat had comparatively lower interactions. Across all platform advertisements, the overall view rate was 6.6%, and the participation rate was 0.7%.

Out of 640 individuals who initially responded to the questionnaire, 387 (76%) individuals completed it fully; the remaining (n=253) either provided incomplete responses or exited the survey after clicking “disagree” to participate in the questionnaire. These partial responses were excluded from the analysis to ensure data completeness.

The final sample size consisted of 248 (64%) female and 139 (35.9%) male participants (Table 1). The largest age group was 15-24 years (n=221, 57.1%), followed by 25-34 years (n=77, 19.9%). Most participants held a bachelor’s degree (n=184, 47.5%) or a high school diploma or less (n=138, 35.6%). Regarding occupation, more than half of the participants were students (n=199, 51.4%), while 97 (25%) participants were unemployed. In terms of marital status, 299 (77.2%) of participants were single, while 82 (21.1%) of participants were married. Saudi participants accounted for 72.3% (n=280) of the sample, and most participants resided in Riyadh (n=258, 66.6%). Significant sex-based differences were observed in occupation (P<.001) and marital status (P=.02), with male participants more likely to be employed and married compared to female participants, who were predominantly unemployed and single (Table 1).

Female participants were significantly more likely than male participants to use social media for oral health information (n=156, 68.7% compared to n=71, 31.2%; P=.02) and follow dental-specific accounts (n=126, 72% compared to n=49, 28%; P=.003; Table 2). For self-reported oral health, female participants were significantly more likely to rate their oral health as good (n=98, 73.1%) or average (n=125, 60.1%), while poor oral health was significantly slightly more common among male participants (n=20, 44.4%; P=.02).

The patterns of social media use are summarized in Table 3. Of the total sample, the most frequently used platform on a daily basis was WhatsApp (n=344, 89.8%), followed by Instagram (n=304, 78.9%), Snapchat (n=239, 62%), and YouTube (n=196, 50.7%). The majority of our sample reported not using Line (n=374, 97.1%), Facebook (n=295, 76.4%), and LinkedIn (n=271, 70%). Telegram had a proportion of weekly users (n=134, 35.1%), while TikTok showed mixed use with 49.7% (n=192) reporting daily use and 31% (n=120) reporting no use.

Table 4 highlights the prevalence of oral health misconceptions among participants. Overall, 96.9% (n=375) of participants reported having at least one or more of the identified misconceptions. When stratified by sex, 93.5% (n=130) of male and 98.8% (n=245) of female participants were found to hold at least one or more misconceptions. Common misconceptions included “During pregnancy, the baby absorbs calcium from the mother’s teeth and bones” (n=337, 87%) and “Pregnant women should only undergo dental treatment after childbirth” (n=245, 63.3%). Additionally, 59.9% (n=232) of participants believed that “You should not eat anything when going for tooth extraction.” Sex-based differences were observed for the misconception that “Using a hard-bristled toothbrush makes teeth whiter,” with male participants more likely to agree (n=38, 57.5%) than female participants (n=28, 43.4%; Table 4).

Logistic and Poisson regression analyses (Table 5) revealed significant associations between following dental-specific accounts on social media with lower odds of having any misconceptions (AOR 0.41, 95% CI 0.22-0.78) and a lower count of misconceptions (AMR 0.87, 95% CI 0.77-0.98). Trust in social media for oral health information was significantly associated with a higher count of misconceptions (AMR 1.16, 95% CI 1.02-1.31). Self-reported oral health showed that participants with poor oral health had a higher count of misconceptions (AMR 1.23, 95% CI 0.99-1.50), although the association was not statistically significant. After conducting the sensitivity analysis by combining “I don’t know” with “agree” as a single variable, there was no difference in most of the observations, except that participants who perceived social media as a good source of oral health information had a statistically significant higher count of misconceptions (AMR 1.18, 95% CI 1.02-1.37). This reclassification suggests that uncertainty about oral health information may reflect susceptibility to misconception.

This study investigated the patterns of social media use and the prevalence of oral health misconceptions among Saudi adults, focusing on the relationship between engagement with oral health information on social media and self-reported oral health with oral health misconceptions. The study highlighted significant associations between following dental-specific accounts and lower oral health–related misconceptions, while higher trust in social media information was associated with a higher count of oral health–related misconceptions. Additionally, self-reported poor oral health was associated with higher counts of oral health–related misconceptions.

A significant portion of our sample consisted of students and women. The high proportion of students in our sample reflects that students are known to be highly active on social media and rely on these platforms for information seeking, including health-related information. This aligns with findings from a scoping review indicating that social media plays a dominant role in health education in Saudi Arabia and that younger individuals and women frequently seek medical and health-related advice through digital platforms [42].

The results of our targeted social media campaigns reveal distinct differences in audience reach and interaction across various platforms. TikTok, in particular, demonstrated a significant advantage in terms of reach. This aligns with previous studies that highlight TikTok’s rapid rise as a dominant platform for user interaction, especially among younger audiences [43,44]. Instagram, known for its visual-centric content, also proved effective in reaching a wide audience and generating meaningful interactions in this study. Instagram has been shown to be effective for health communication because of its ability to attract users through images, videos, and infographics, which enhance user interaction and retention [45]. In contrast, X (Twitter) and Snapchat showed lower interaction rates, which can be explained by their platform-specific characteristics. In contrast, both X (Twitter), with its focus on quick updates and real-time communication, and Snapchat, with its temporary content format appealing to younger users seeking informal and immediate interactions, limit their potential for sustained and broad interaction [46].

TikTok’s effectiveness in disseminating health information can be attributed to its short-form video format, which caters to modern users’ preference for quick, easily digestible content [47]. This format has not only captured the attention of younger audiences but also influenced other social media platforms to adopt similar features. The platform’s popularity among younger demographics further enhances its role in health communication, as it serves as a primary source of information for this age group. Moreover, TikTok’s participatory features, such as duets and challenges, encourage user participation and the viral spread of health-related content [48].

Our methodology assumed that using social media advertisement campaigns would effectively reach our target population, with advertisement reach aligning with the demographic distribution in Saudi Arabia, where men represent a larger proportion (61.2%) [49]. The term “ad recipient” refers specifically to individuals who were exposed to the advertisement (ie, saw the advertisement) without necessarily engaging with it by clicking. Men comprised the majority of advertisement recipients across all platforms: Snapchat (n=28,235, 56%), TikTok (n=625,665, 74.32%), Instagram (n=24,625, 57.2%), and X (Twitter; n=107,116, 61%), similar to the male demographic distribution in Saudi Arabia [49]. However, despite greater advertisement exposure among men, women represented the majority of questionnaire respondents. This difference in interaction may be explained by behavioral differences, as previous studies have found that women are more likely to interact with health-related advertisements and content on social media than men [50-52]. This trend may be driven by greater health awareness among women, a higher likelihood of managing family health matters, and cultural norms that encourage women to be more involved in health-related decision-making [53,54].

This study identified several common oral health–related misconceptions. The most prevalent misconceptions include the belief that pregnant women lose calcium during pregnancy, dental procedures should be postponed during pregnancy, eating before dental extractions is harmful, and scaling weakens tooth structure. These findings are consistent with earlier studies conducted in Saudi Arabia, which also documented widespread myths about oral health [26-28].

Many of these oral health misconceptions can negatively impact oral health by discouraging preventive behaviors and delaying necessary treatment. For example, myths about pregnancy and oral health, such as the belief that “Pregnancy causes calcium loss in teeth” or that “Dental treatment should be avoided during pregnancy,” may lead to untreated dental issues, increasing the risk of adverse pregnancy outcomes like preterm birth and low birth weight [55]. Similarly, misconceptions about primary teeth being unimportant contribute to early childhood caries, which affects children’s overall well-being and increases the likelihood of future dental problems [56]. Additionally, beliefs in ineffective home remedies, such as using salt for whitening or avoiding professional scaling, can lead to poor oral hygiene and a higher risk of periodontal disease [57,58]. Based on these observations, addressing these common oral health–related misconceptions through targeted public health campaigns and education is essential. Engaging the community to participate in the educational content, primarily through social media platforms, would be an active way of enhancing people’s awareness.

The high prevalence of misconceptions underscores the challenges in addressing oral health information in the digital age. While social media platforms provide an avenue for disseminating information, their effectiveness depends on the quality and credibility of the content shared.

This study observed associations between trust in social media for oral health information and higher misconceptions are consistent with broader concerns about the role of social media in propagating misconceptions in health communication [59]. The user-driven nature of social media echo chambers, combined with algorithm-driven content personalization, can reinforce existing beliefs, making users more susceptible to content that aligns with their biases and less likely to evaluate or seek out alternative perspectives critically. Social media algorithms on platforms such as TikTok, Instagram, and Facebook select content based on users’ past interactions, prioritizing posts that align with their preference history. As a result, users who have previously interacted with nonevidence-based health claims are more likely to be exposed to similar misconceptions [60,61]​. Conversely, the protective effect of following dental-specific accounts aligns with prior studies that emphasize the importance of verified content from credible sources in mitigating health misconceptions [62-65]. On social media, accounts maintained by professionals or organizations with expertise in a given field act as anchors of credibility, reducing the likelihood that users will engage with inaccurate information [59].

Our research specifically investigates oral health–related misconceptions in Saudi Arabia, an area understudied despite social media’s growing influence in shaping health beliefs. This study’s multiplatform social media approach enabled a comprehensive analysis of social media’s role in oral health information. The use of targeted advertisements across multiple platforms, including Snapchat, TikTok, Instagram, and X (Twitter), allowed for the random selection of participants, enhancing the representativeness of the sample. We provided robust insights into the associations between engagement behaviors and self-reported oral health with oral health misconceptions by using advanced statistical modeling, including logistic and Poisson regression analyses. However, the cross-sectional design limits the ability to establish temporal relationships between the variables. Moreover, the reliance on self-reported data introduces potential response and recall biases, as participants may have overstated their engagement with oral health information or underreported misconceptions due to social desirability. Additionally, recall bias could have affected their ability to recall their pattern of social media use accurately. Furthermore, while we adjusted for key confounders, unmeasured variables may have influenced both the exposures and outcomes, potentially affecting the observed associations.

Our findings align with previous research that highlights social media’s role as a valuable channel for disseminating health information and as a potential source of misconceptions [19,66]. Therefore, the highly used platforms like WhatsApp, Instagram, Snapchat, and YouTube, as observed in this study, highlight their potential as valuable tools for web-based dental public health campaigns in Saudi Arabia. These platforms are widely used daily, reflecting their integration into the community’s digital habits and their capacity to reach broad and diverse audiences. To address oral health–related misconceptions, health authorities, dental professionals, and regulatory bodies should consider the following:

While this study provides valuable insights, future research is needed to investigate how different content formats (eg, videos, photos, and participatory posts) and features (eg, likes, comments, and shares) influence the dissemination and retention of oral health information on social media. Experimental studies can assess which formats are most effective in improving knowledge and correcting oral health–related misconceptions, while longitudinal research can explore how repeated exposure to accurate content impacts behavioral change over time.

The study underscores the role of social media in shaping oral health perceptions among adults in Saudi Arabia. Engagement with dental-specific accounts was associated with fewer misconceptions, while trust in oral health information in social media was associated with higher misconceptions. Platforms like WhatsApp, Instagram, TikTok, and Snapchat are widely used in Saudi Arabia and offer the potential for targeted dental public health campaigns. Strategic collaborations with credible dental professional creators to disseminate accurate and engaging oral health information. Creating Instagram reels for quick educational content or using TikTok’s targeting features could enhance information retention and user interaction. Although this study focuses on Saudi Arabia, its findings have broader relevance, particularly in regions with high social media penetration and similar digital health-seeking behaviors, such as the Gulf countries. From a policy perspective, legalizing and monitoring oral health–related advertisements and information in social media should be implemented by regulatory bodies to protect public health.

Future research should build on these findings to investigate the effectiveness of varying social media platforms’ content formats and features in improving oral health knowledge and correcting misconceptions through experimental studies. Longitudinal research should also explore the long-term impact of repeated exposure to accurate information on behavioral change. These insights provide a basis for stakeholders to formulate evidence-based, innovative strategies to improve public health in the context of the digital age.