Online social networks for prevention and promotion of oral health: a systematic review

Abstract

Aim:

To evaluate the impact of online social networks (OSN) as resources for promoting behaviour changes related to the prevention and promotion of oral health.

Methods:

This was a systematic review with meta-analysis of randomized clinical trials identified in databases (MEDLINE/Embase/Lilacs), published up until May 2022, and prepared according to Cochrane recommendations. Studies that used OSN (WhatsApp^®, Telegram^®, TikTok^®, Facebook^®, Twitter^®, YouTube^®, Instagram^®, Soundcloud^®, Flickr^®, LinkedIn^®, and Webradio^®) were included. The protocol was registered in the International Prospective Register of Systematic Reviews (CRD42021248045).

Results:

In total, 3171 studies were evaluated, of which 12 studies met the inclusion criteria for the review and contained data on 1669 participants. The most frequently investigated OSN were WhatsApp^®, followed by Telegram^®, and YouTube^®. All studies included were at high risk of bias. Data extraction allowed the meta-analysis of gingival index (GI) and oral health knowledge (OHK) outcomes for young people. Findings revealed that young people exposed to digital interventions via OSN showed a reduction in GI when compared with traditional educational interventions. This reduction occurred in all young people (standardized mean difference, –0.48; 95% CI, –0.75 to –0.21; p = 0.0006, I² = 0%; very low certainty), including those who were undergoing orthodontic treatment (standardized mean difference, –0.58; 95% CI, –0.92 to –0.24; p = 0.0008, I² = 0%; very low certainty). Furthermore, young people undergoing orthodontic treatment and exposed to OSN showed an increase in OHK when compared with participation in traditional educational interventions (standardized mean difference, +0.86; 95% CI, +0.46 to +1.26; p < 0.0001, I² = 0%; very low certainty).

Conclusion:

OSN could be effective tools for improving oral health outcomes in young people.

PRACTICAL IMPLICATIONS OF THIS RESEARCH.

• The diversity of audiovisual resources available in online social networks makes them a potentially important educational medium.

• Oral health instructions offered via online social networks have been shown to reduce Gingival Index scores and improve oral health knowledge on the topics of tobacco use and oral cancer prevention in young people.

INTRODUCTION

In the scientific literature, the descriptors “social media” and “online social networks (OSN)” can be found in reference to digital platforms such as WhatsApp^®, Telegram^®, TikTok^®, Facebook^®, Twitter^®, YouTube^®, and Instagram^®.^1-5 According to the medical subject headings (MeSH) thesaurus maintained by the National Library of Medicine,⁶ the descriptor “social media” was established in 2012 and is broader in scope, referring to resources capable of supporting the creation and publication of information using the internet, therefore including the digital platforms. The descriptor “OSN” was established in 2019 and, according to MeSH,⁷ refers to digital platforms designed to facilitate interaction, as well as those mentioned above. However, this interaction requires the presence of common interests among the participants, such as those of study, work, and family.

OSN have been increasingly used by populations around the globe in recent decades and have aroused the interest of the scientific community and government agencies.⁸ In the field of health, there is growing production and publication of content about health behaviour interventions using OSN to evaluate their impact on health outcomes of different populations.^{2, 9} From the perspective of the scientific community, there is interest in studying the methods of application and impact of interventions through OSN on the health outcomes of populations.^{2, 9}

To date, there have been some systematic reviews that have summarized the impact of OSN on general health outcomes.^{1-5, 10} Among them, the following outcomes have been evaluated: increased physical activity; obesity control; vaccine acceptance; breastfeeding stimulation; safe sex practices; HIV prevention, and control of tobacco use.^{1-5, 10} In these systematic reviews, there was wide heterogeneity between the studies and a predominance of populations of children, adolescents, and young people.^{1-5, 10}

In the oral health literature, there are few studies of the impact of OSN on oral health behaviours and care, and on the clinical condition of teeth.^11-13 Furthermore, there are no specific systematic reviews summarizing the impact of interventions using OSN on oral health outcomes. Studies in this area have been restricted to systematic reviews that evaluated the impact of interventions such as teledentistry, m-Health, and the sending of digital reminders.^14-18 These reminders may come in the form of text, audio, image or video messages, and they may be delivered on digital platforms other than OSN, such as the short message service (SMS) and websites or applications developed by the researchers themselves.^14-18 There is a lack of systematic reviews that only assess the impact of OSN on oral health outcomes.^14-18

In addition, there are limitations to the use of digital platforms other than OSN for health interventions, as these generally require greater engagement on the part of users.¹ This requirement mainly arises from the fact that the technologies are not part of the patients' daily routines. Consequently, these platforms demand extra resources, such as new installations and larger internet packages and memory space on mobile devices.¹

Thus, the objective of this systematic review was to evaluate the impact of OSN as resources to support changes in behaviours related to oral disease prevention and promotion of oral health.

METHODS

This systematic review followed the recommendations of, the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA),¹⁹ in addition to those of the PRISMA-S extension.²⁰ The protocol for this systematic review was registered in the International Prospective Register of Systematic Reviews (PROSPERO) (CRD42021248045). There were 2 changes to the original protocol registered, which corresponded to the acceptance of only randomized clinical trials (RCT) and the impossibility of using only OSN in the control group (see www.crd.york.ac.uk/prospero/display_record.php?ID=CRD42021248045 for details).

Search strategy and eligibility criteria

The research question was: “Are online social networks effective resources for the prevention and promotion of oral health?” In addition, the PICOs strategy was used, as shown in Figure 1.

Literature searches were conducted in the following databases: MEDLINE (via PubMed), Embase, and Latin American and Caribbean Literature on Health Sciences (Lilacs), as shown in Table 1.

The adapted or reused search strategies to identify RCT and oral health outcomes were based on criteria used by previous studies that researched the topic in the oral health field (e.g., plaque index/gingival index) in addition to official Cochrane manuals.^14-18,21-26

Inclusion criteria were studies that evaluated the impact of interventions using the social networks WhatsApp^®, Telegram^®, TikTok^®, Facebook^®, Twitter^®, YouTube^®, Instagram^®, Soundcloud^®, Flickr^®, LinkedIn^®, and Webradio^® on oral health behaviours. These networks were investigated because they are currently the most popular OSN on the internet and are used by official channels of government agencies around the world.^27-29 There were no exclusion criteria.

There were no restrictions outside the search strategies (e.g., date filter), considering the studies published up to the date of the last search. The initial search was conducted in all databases on May 20, 2021; the second (update) was conducted in all databases on May 22, 2022.

Selection of studies

Three reviewers experienced in the development of systematic reviews and/or in oral health participated in the process to determine the studies to be selected for analysis. Two reviewers (AJOJ) and (YPMB) independently evaluated the studies identified in both the first and second searches by analysing their titles and abstracts, with the support of the Rayyan³⁰ data manager to blind the respective selections during the selection process.

Decisions to include or exclude studies were made after complete reading of the studies retrieved and discussion between the 2 reviewers (AJOJ and YPMB). There was only 1 disagreement between the reviewers; later a third reviewer (JMO) was consulted, discussions were held, and consensus was reached.

Figure 1.

PICOs strategy

Data extraction and grouping

The following data were extracted and categorized for comparison of results: author; year of publication of the research; place; dental specialty (e.g., orthodontics/periodontics/general practitioner); intervention follow-up time in weeks; total number of individuals who participated to the end; age group/average age of participants; distribution by gender; OSN; inclusion/exclusion criteria; type of control (nothing/traditional educational activities, such as written or verbal instructions during dental care/specific intervention, if so, which type) and digital health educational strategy carried out (sending text/audio/picture/video messages or playing a game); methods to assess outcomes (plaque index/caries index/gingival index/questionnaire/other specific index); and outcomes (oral health care: increased brushing and flossing; oral health conditions: amount of plaque or dental biofilm/periodontal condition or gingival bleeding/caries index; oral health knowledge: increased knowledge about how to have good oral health and prevent oral diseases); and author's own conclusion.

All results that fit the domains for each variable of interest in this research were considered. If the information was incompatible for comparison between studies (e.g., distribution of time in weeks or months; presentation of the raw quantity, but not in frequencies and/or percentages; standard deviation and standard error), the authors were consulted and/or the data were converted and standardized by the authors of this systematic review.

The oral health outcomes extracted were classified into the following categories: oral health knowledge (OHK); plaque index (PI); gingival index (GI); dental caries (white spot lesions or International Caries Detection and Assessment System [ICDAS]) (DC); toothbrushing (TH); community periodontal index (CPI); and halitosis (HAL). If part of the outcomes of interest obtained from the RCT were not compatible for comparison and/or meta-analysis, the results were later described in the results table.

If, after reading the titles and abstracts, the information on the results of the studies was incomplete or inaccurate, or the full text was in a language different from that understood by the reviewers (Portuguese, English, and Spanish), the authors responsible for the research were contacted. The data obtained were added to the data descriptions but were not considered for analysing the risk of bias of the studies. If the information was not obtained after 3 attempts (with an interval of 1 month between each attempt) to contact the authors, the (missing) data were classified as incomplete/indeterminate.

Finally, all data collected by the reviewers were entered into Microsoft Excel^® tables, which were previously prepared and calibrated by the authors of this systematic review in a pilot study of the research to define the search strategy and for standardization of the information collected.

If any reviewer warned that there was no full appreciation of the data during the synthesis of the results and/or that it was impossible to convert the information to categorize the tables, the other authors were contacted to reach a consensus. Moreover, if a deficiency was found in the tables to enable summarizing and/or conversion, readjustments were performed.

Risk of bias analysis

Risk of bias analysis was performed independently by the reviewers, using the RoB 2.0 tool.²⁵

Table 1.

Health descriptors, alternative and own search terms in MEDLINE, Embase, and Lilacs

	Database search terms
Search	MEDLINE	Embase	Lilacs
1	(“Social Network*” OR “Social Media” OR “Media Social” OR Whatsapp OR Telegram OR “TikTok” OR Facebook OR Twitter OR YouTube OR Instagram OR Soundcloud OR Flickr OR Linkedin OR Webradio)	(‘social network*’ OR ‘social media’ OR ‘media social’ OR whatsapp OR telegram OR ‘TikTok’ OR facebook OR twitter OR YouTube OR instagram OR soundcloud OR flickr OR Linkedin OR webradio)	(“Social Network$” OR “Social Media” OR “Media Social” OR Whatsapp OR Telegram OR “TikTok” OR Facebook OR Twitter OR YouTube OR Instagram OR Soundcloud OR Flickr OR Linkedin OR Webradio)
2	(“primary care” OR “health education” OR “dental education” OR “oral education” OR “preventive dentistry” OR “health promotion” OR “dental care” OR “dental health” OR “oral care” OR “oral health” OR “dental carie*” OR “plaque index” OR “tooth brushing” OR “toothbrushing” OR “minimal intervention dentistry” OR “oral hygiene” OR “dental hygiene” OR “health knowledge” OR “health attitude*” OR “health practice*” OR “health behavior*” OR “ORDMF Index”)	(‘primary care’ OR ‘health education’ OR ‘dental education’ OR ‘oral education’ OR ‘preventive dentistry’ OR ‘health promotion’ OR ‘dental care’ OR ‘dental health’ OR ‘oral care’ OR ‘oral health’ OR ‘dental carie*’ OR ‘plaque index’ OR ‘tooth brushing’ OR ‘toothbrushing’ OR ‘minimal intervention dentistry’ OR ‘oral hygiene’ OR ‘dental hygiene’ OR ‘health knowledge’ OR ‘health attitude*’ OR ‘health practice*’ OR ‘health behavior*’ OR ‘ordmf index’)	(“primary care” OR “health education” OR “dental education” OR “oral education” OR “preventive dentistry” OR “health promotion” OR “dental care” OR “dental health” OR “oral care” OR “oral health” OR “dental carie$” OR “plaque index” OR “tooth brushing” OR “toothbrushing” OR “minimal intervention dentistry” OR “oral hygiene” OR “dental hygiene” OR “” OR “health attitude$” OR “health practice$” OR “health behavior$” OR “ORDMF Index”)
3	(“randomized controlled trial”[pt] OR “controlled clinical trial”[pt] OR randomized[tiab] OR placebo[tiab] OR “clinical trials as topic”[mesh:noexp] OR randomly[tiab] OR trial[ti] NOT (animals[mh] NOT humans [mh])	(‘randomized controlled trial’/de OR ‘controlled clinical trial’/de OR random*:ti,ab OR ‘randomization’/de OR ‘intermethod comparison’/de OR placebo:ti,ab OR compare:ti OR compared:ti OR comparison:ti OR ((evaluated:ab OR evaluate:ab OR evaluating:ab OR assessed:ab OR assess:ab) AND (compare:ab OR compared:ab OR comparing:ab OR comparison:ab)) OR ((open NEXT/1 label):ti,ab) OR (((double OR single OR doubly OR singly) NEXT/1 (blind OR blinded OR blindly)):ti,ab) OR ‘double blind procedure’/de OR ((parallel NEXT/1 group*):ti,ab) OR crossover:ti,ab OR ‘cross over’:ti,ab OR (((assign* OR match OR matched OR allocation) NEAR/6 (alternate OR group OR groups OR intervention OR interventions OR patient OR patients OR subject OR subjects OR participant OR participants)):ti,ab) OR assigned:ti,ab OR allocated:ti,ab OR ((controlled NEAR/8 (study OR design OR trial)):ti,ab) OR volunteer:ti,ab OR volunteers:ti,ab OR ‘human experiment’/de OR trial:ti) NOT (((random* NEXT/1 sampl* NEAR/8 (‘cross section*’ OR questionnaire* OR survey OR surveys OR database OR databases)):ti,ab) NOT (‘comparative study’/de OR ‘controlled study’/de OR ‘randomised controlled’:ti,ab OR ‘randomized controlled’:ti,ab OR ‘randomly assigned’:ti,ab) OR (‘cross‐sectional study’ NOT (‘randomized controlled trial’/de OR ‘controlled clinical study’/de OR ‘controlled study’/de OR ‘randomised controlled’:ti,ab OR ‘randomized controlled’:ti,ab OR ‘control group’:ti,ab OR ‘control groups’:ti,ab)) OR (‘case control*’:ti,ab AND random*:ti,ab NOT (‘randomised controlled’:ti,ab OR ‘randomized controlled’:ti,ab)) OR (‘systematic review’:ti NOT (trial:ti OR study:ti)) OR (nonrandom*:ti,ab NOT random*:ti,ab) OR ‘random field*’:ti,ab OR ((‘random cluster’ NEAR/4 sampl*):ti,ab) OR (review:ab AND review:it NOT trial:ti) OR (‘we searched’:ab AND (review:ti OR review:it)) OR ‘update review’:ab OR ((databases NEAR/5 searched):ab) OR ((rat:ti OR rats:ti OR mouse:ti OR mice:ti OR swine:ti OR porcine:ti OR murine:ti OR sheep:ti OR lambs:ti OR pigs:ti OR piglets:ti OR rabbit:ti OR rabbits:ti OR cat:ti OR cats:ti OR dog:ti OR dogs:ti OR cattle:ti OR bovine:ti OR monkey:ti OR monkeys:ti OR trout:ti OR marmoset*:ti) AND ‘animal experiment’/de) OR (‘animal experiment’/de NOT (‘human experiment’/de OR ‘human’/de)))	((Pt randomized controlled trial OR Pt controlled clinical trial OR Mh randomized controlled trials OR Mh random allocation OR Mh double-blind method OR Mh single-blind method) AND NOT (Ct animal AND NOT (Ct human and Ct animal)) OR (Pt clinical trial OR Ex E05.318.760.535$ OR (Tw clin$ AND (Tw trial$ OR Tw ensa$ OR Tw estud$ OR Tw experim$ OR Tw investiga$)) OR ((Tw singl$ OR Tw simple$ OR Tw doubl$ OR Tw doble$ OR Tw duplo$ OR Tw trebl$ OR Tw trip$) AND (Tw blind$ OR Tw cego$ OR Tw ciego$ OR Tw mask$ OR Tw mascar$)) OR Mh placebos OR Tw placebo$ OR (Tw random$ OR Tw randon$ OR Tw casual$ OR Tw acaso$ OR Tw azar OR Tw aleator$) OR Mh research design) AND NOT (Ct animal AND NOT (Ct human and Ct animal)) OR (Ct comparative study OR Ex E05.337$ OR Mh follow-up studies OR Mh prospective studies OR Tw control$ OR Tw prospectiv$ OR Tw volunt$ OR Tw volunteer$) AND NOT (Ct animal AND NOT (Ct human and Ct animal)))
4	1 AND 2 AND 3	1 AND 2 AND 3	1 AND 2 AND 3

Sensitivity analysis

Sensitivity analysis was performed to investigate the possible causes of heterogeneity. The subgroups of the studies were grouped by variables and investigated according to the criteria of the World Health Organization,³¹ as shown in Figure 2.

Data analysis

To evaluate possible publication biases or selective reporting in the studies, the registration protocols and the years in which the research was conducted and published were analysed. In addition, the question of whether recent studies, with larger samples and positive results, were responsible for the greater effect on the synthesis of the meta-analysis was evaluated.

In cases where reporting bias was discovered, all data of interest tested in the research were tabulated, even when it was impossible to synthesize the data for meta-analysis. Thus, the conclusions of RCT were included in the authors' own words.

The Chi-squared test and the magnitude of inconsistency, estimated by I-squared (I²), were calculated to assess heterogeneity between studies. All analyses were performed in Review Manager (5.4.1). The recommendations of the Higgins et al.²⁵ manual were followed with regard to the selection of models for data analysis and the statistical program.

Meta-analyses were performed considering the heterogeneity of the subgroups; these were only maintained when the limit was acceptable.²⁵ In addition, the data evaluated and the results found during the meta-analysis were submitted to Grading of Recommendations Assessment, Development, and Evaluation (GRADE).²⁵

RESULTS

The search and selection process of all studies included in this systematic review is shown in Figure 3. In the first search of the databases, 2506 studies were extracted; in the update, an additional 655 studies were re,trieved. At the end of the selection process, 12 studies remained^{11, 12, 32-41} as shown in the flow diagram.

Publication year and study location

No 2 articles by the same author were identified. The studies were all published between 2015 and 2021.^{11, 12, 32-41} Iran and Brazil were the countries in which the majority of studies took place, with (n = 3) in each.^{11, 12, 33, 35, 38, 39} The remaining 6 studies were conducted in the United Kingdom, Italy, India, Saudi Arabia, Peru, and Malaysia.^{32, 34, 36, 37, 40, 41}

Intervention follow-up and age and gender of participants

The minimum time for the first evaluation of results in the studies investigated was 4 weeks from baseline.

The total number of participants who remained until the end of the studies was 1669, with a mean age of 17.7 years for the studies that made this data available (n = 11).^{11, 12, 32-38, 40, 41} Only the study by Borujeni et al.³⁹ did not report on the average age, accounting for 40% of participants between 12 and 18 years of age and 60% of participants over 18 years of age.

There was a predominance of female study participants (n = 7)^{32-35, 38, 40, 41}; in the study by Borujeni et al.³⁹, distribution by gender was not reported.

Figure 2.

Sensitivity analysis

Online social networks

Eight studies investigated WhatsApp^{®12, 33, 34, 36-38, 40, 41}; 3, Telegram^{®11, 35, 39}; and 1, YouTube^®32. No studies were found of the other OSN identified in this systematic review.

Digital health education strategies

Six studies used more than one strategy.^{11, 33-36, 40} Nine studies sent video^32-40; 8 studies sent text messages^{11, 12, 34, 35, 36, 38, 40, 41}; 3 studies sent images^{34, 36, 40}; 1 study did gamification (n = 1)³³. No studies that used only audio were found.

Dental specialty

Six studies focused on interventions in clinical dental specialties: 5 addressed outcomes related to orthodontics and 1 to periodontics.^{12, 32, 33, 37, 39, 41} The other 5 studies did not have a clinical dental specialty as the object of an outcome.^{11, 34, 35, 38, 40}

Control group

Eight studies used a traditional educational intervention^{12, 32, 33, 36, 37, 39-41}; 1 study used a PowerPoint presentation in the classroom³⁴; the 3 remaining studies did not offer any educational instruction to the control group^{11, 35, 38}.

Methods used to assess oral health outcomes

Eight studies used more than one assessment method.^{11, 12, 33, 36-40} When evaluating the studies individually, 9 evaluated PI^{11, 12,33, 36-41}; 5, the application of a questionnaire^{11, 32, 34, 37, 40}; 5, GI^{12, 33, 36, 39, 40}. Only 1 study evaluated white spot lesions³³; 1, ICDAS³⁸; 1, CPI¹¹; 1, gum colour and consistency³⁹; and 1, HAL¹².

Figure 3.

Flow diagram showing the search and selection process

Outcome

According to the outcome classification, 3 studies explored more than one outcome.^{11, 37, 40} However, the variable that was repeated most frequently was evaluation of the clinical condition (n = 10),^{11, 33, 35-41} followed by knowledge of oral health (n = 3)^{32, 34, 37} and oral health care/behaviours (n = 3)^{11, 38, 40}. All the data are presented in Table 2.

Comparison of data relative to the results obtained from control and intervention groups in the studies appears in Supplementary Table S1. It was not possible to report the study outcome findings of Rahmani et al.³⁵ Thus, 11 studies were included in this comparison.^{11, 12, 32-41} Supplementary Table S1 includes a synthesis of the results of interest obtained from the studies that could be extracted and compared in meta-analysis, by mean/standard deviation and median/interquartile range (n = 11).

Some of the studies tested whether there were statistically significant differences between the control and intervention groups after using OSN interventions for certain oral outcomes.^{12, 33, 39, 40} With regard to behaviours and oral health care, when the impact of OSN on toothbrushing frequency was analysed, neither of the 2 studies that investigated this outcome found a statistical difference between the groups.^{12, 40} Relative to the use of dental floss, Foner et al.⁴⁰ and Santos et al.¹² found statistically significant differences for the greater use of dental floss between the groups that received this intervention through OSN and the control group, which did not receive an educational intervention through OSN.

Borujeni et al.³⁹ found statistically significant differences between the intervention via OSN and control groups for outcomes related to gingival colour and consistency. Zotti et al.³³ observed these differences for caries experience.

Figure 4.

Risk of bias analysis

Risk of bias analysis

The complete analysis of the studies selected in this systematic review is shown in Figure 4. All studies showed some type of bias, the majority related to the difficulty of blinding participants. The absence of blinding was attributed to the fact that this was an educational intervention in health using text, images, and video messages. Nevertheless, even with blinding, none of the studies met the analysis criterion for estimating the effect of the designated interventions.

Worth noting is that the study by Rahmani et al.³⁵ was considered to have “some concerns” in domains 1 to 5 due to inaccuracy of the information (it was not possible to identify all the information, as only the abstract was in English and the remainder of the text was written in Persian), which led to a general assessment of high risk. In regard to the other studies, only 1 had a high risk of bias resulting from the randomization process.³⁷ All studies had a low risk of bias due to missing results. Four studies were at high risk of bias in the measurement of outcomes, as they did not report the blinding of the examiners.^{34, 36, 39, 41} Finally, 4 studies were classified as low risk of bias in the selection of the result reported because they recorded the clinical study protocol and retained it until disclosure on conclusion of the research.^{11, 38, 40, 41}

Analysis of the outcomes evaluated

Since only 1 study evaluated DC, TH, CPI, and HAL as outcomes, it was not possible to conduct a meta-analysis of these findings.

Despite the large number of studies with results on the PI outcome, it was also not possible to perform a meta-analysis, as there was 84% heterogeneity between the results (p = 0.35).^{11, 33, 36-41} Nevertheless, to assess the causes of heterogeneity, a sensitivity analysis was performed between the subgroups. Even when these were divided by criteria such as time to evaluate the outcome, age group, pre-existing condition (e.g., periodontal disease), dental specialty (e.g., orthodontics), and index for measuring the result, a result with the possibility of substantial heterogeneity remained (between 50% and 90%).²⁵

After all exclusions, it was possible to perform a meta-analysis for the GI and OHK outcomes. For both outcomes, a subgroup analysis was necessary since the heterogeneity in all studies was 51% for GI and 81% for OHK. In the first outcome, when the pre-existing condition (periodontal disease) was removed,³⁶ heterogeneity was 0%, considering young people with or without orthodontic treatment.^{33, 39, 40} Relative to the second outcome, when studies with young people who received digital health educational interventions on the topic of tobacco use and how to prevent oral cancer were removed,³⁴ the heterogeneity among young people who received digital health educational interventions on how to maintain good oral hygiene during orthodontic treatment was also 0%.^{32, 37}

After the analyses, it was found that the audience of young people exposed to digital interventions via OSN showed a statistically significant reduction in the GI when the intervention and control groups were compared. This fact was observed with all the young people evaluated (standardized mean difference, –0.48; 95% CI, –0.75 to –0.21; p = 0.0006, I² = 0%; very low certainty),^{33, 39, 40} and only with those undergoing orthodontic treatment (standardized mean difference, –0.58; 95% CI, –0.92 to –0.24; p = 0.0008, I² = 0%; very low certainty)^{33, 39} (Figure 5).

Young people undergoing orthodontic treatment who were exposed to digital interventions via OSN showed a statistically significant increase, in OHK in the comparison between the group and control, when the latter were offered a traditional educational intervention (standardized mean difference, +0.86; 95% CI, +0.46 to +1.26; p < 0.0001, I² = 0%; very low certainty)^{32, 39} (Figure 6).

Figure 5.

Forest plot for OSN x gingival index

Figure 6.

Forest plot for OSN x oral health knowledge

DISCUSSION

This systematic review synthesized the results of RCT that used OSN to change oral health outcomes. The results showed potentialities in the use of these digital interventions for better oral health outcomes, as well as limitations in this field of knowledge. Furthermore, the results of the meta-analysis showed that OSN could be effective instruments for reducing GI scores in young people and, in general, to increase this population's OHK on the topics of tobacco use, oral cancer prevention, and how to maintain good oral hygiene during orthodontic treatment.^{32, 33, 37, 39, 40}

The participants in half of the studies included in this review were treated in orthodontic clinics, and, of these, the majority were in a university environment (n = 4). In 1 study, the participants were treated in a periodontics clinic, while the other studies involved participants from clinics without a dental specialty. Thus, the trend of the population of interest in previous studies and systematic reviews on digital interventions in oral health was confirmed: orthodontic patients continue to be the audience of greatest interest to researchers.^{14, 15, 23} One of the hypotheses for this finding may be that orthodontics involves a longer treatment period and, therefore, requires greater cooperation from the patient.⁴² In addition, researchers may prefer to include patients from dental school clinics as a target population in their studies because they are easy to access for participation in research.

One point that drew attention in the studies evaluated was the lack of investigations focused on the elderly population, since the majority involved only children, adolescents, young people, and adults.^{11, 12, 32-41} This trend may be biased by the predominance of the orthodontic specialty in these studies, which is generally of greater interest to young people.⁴³ The systematic review by Aquilanti et al.²⁴, whose objective was to evaluate the use of teledentistry in nursing homes or in homes where elderly individuals live, indicated that teledentistry can be a feasible resource for maintaining oral hygiene in institutionalized elderly persons. The present study highlighted this age gap and the need for future studies to investigate the use of OSN with elderly individuals, in addition to presenting better methodological characteristics, with more diversified samples. Addressing these limitations would generate more robust evidence on the effectiveness of digital health interventions in different age groups.

Twelve studies using OSN such as Telegram^®, YouTube^®, and WhatsApp^® were found to foster changes in oral health outcomes.^{11, 12, 32-41} From the perspective of the authors of this study, this is the first systematic review to evaluate the impact of the use of the social network YouTube^® as a platform for a digital educational intervention in health, in a video message about how to maintain good health and good oral hygiene during orthodontic treatment. Furthermore, this review has contributed substantially to the synthesis of the results of studies that investigated the use of OSN in oral health care. Although there have been 3 previous systematic reviews about digital media that investigated some OSN, the current study differs in that its scope was focused only on OSN, in addition to advancing and expanding the findings of previous studies. However, a possible source of bias in the present research was the omission of WeChat^®, which was investigated by Mohammed et al.¹⁴, Toniazzo et al.¹⁵, and Fernández et al¹⁷. This digital platform did not fall within the inclusion criteria of the present study. Also worth noting is that no RCT or systematic reviews that included OSN were found by the authors of this study relative to TikTok^®, Facebook^®, Twitter^®, Instagram^®, Soundcloud^®, Flickr^®, LinkedIn^®, and Webradio^®.

More specifically, the digital educational interventions in health used by OSN involved sending messages on how to maintain good oral health through good oral hygiene practices in text format^{11, 12, 35, 36, 38, 40, 41}, in image^{36, 40} and video format^{11, 32, 33, 35-37, 39, 40} or through gamification³³. However, no audio-only interventions were identified. Furthermore, strategies to educate patients on control of tobacco use and oral cancer prevention used text, image, and video formats.³⁴ Thus, the use of all digital health educational formats investigated in this research corroborated all digital health educational interventions in systematic reviews that evaluated at least 1 study of OSN^14-18. The diversity of audiovisual resources available in OSN makes them a potentially important medium for engaging participants in the content of educational tools and thus should be better explored in future studies.

Table 2.

Characterization of included studies (N = 12)

Author, Year, Place, General clinic or specialty	Intervention follow-up time (weeks)	Total, Age group (average age of participants), Gender distribution	Online social network Inclusion/exclusion criteria and setting	Control (C), Intervention (I)	Methods for evaluating outcomes, Outcomes	Conclusion
Al-Silwadi et al. 201532 (United Kingdom) Orthodontics	t = 6–8	N = 60 ≥13 years (15.50) M = 33.3% F = 66.7%	YouTube® Patients of both genders, aged 13 years and over, recruited from the Department of Orthodontics of the Eastman Dental Hospital, UK. Individuals were excluded if they did not have a valid e-mail address, had a history of previous orthodontic treatment, were receiving orthodontic treatment with only removable orthodontic appliances or headgear, required complex multidisciplinary treatment or had a craniofacial syndrome.	C: Traditional educational intervention I: Sending message videos	Questionnaire Oral health knowledge	“Presenting audiovisual information through the YouTube Web site to orthodontic patients resulted in a significant improvement in patient knowledge. Supplementation of verbal and written patient information with audiovisual information via the Internet is therefore worthy of consideration.”
Zotti et al. 201633 (Italy) Orthodontics	t1 = 12 t2 = 24 t3 = 36 t4 = 48	N = 80 (13,8) M = 42.5% F = 57.5%	WhatsApp® Patients who were scheduled to start orthodontic multibracket treatment, recruited from the Dental School of Brescia, owned a smartphone and were able to be online daily. Exclusion criteria were the presence of a significant medical history, a restrictive dietary regimen, and difficulties in reading or speaking the national language.	C: Traditional educational intervention I: Sending video message and gamification	Plaque index; gingival index; white spot lesions Oral health condition	“Integration of new ‘social’ technologies in a standard oral hygiene motivation protocol is effective in improving compliance of adolescent patients and in improving their oral health status during orthodontic multibracket treatment.”
Nayak et al. 201834 (India) General clinic	t = 4	N = 150 17 to 22 years (20.2) M = 37.3% F = 62.7%	WhatsApp® The students of the first- and second-year Bachelor of Commerce (B. Com) and studying in a first-grade college in Barkur, a South Indian town. Those who gave informed consent and with WhatsApp active on their smartphones.	C: PowerPoint presentation I: Sending text, images, and video messages	Questionnaire Oral health knowledge	“WhatsApp can be a more effective tool for providing dental education on tobacco and oral cancer as compared to conventional audio-visual.”
Rahmani et al. 201935 (Iran) General clinic	t = 24	N = 160 (28,69) M = 0.0% F = 100.0%	Telegram® Residents of Sanandaj City, pregnant women referenced to the entire database of the health service, not using toothpaste, in the pre-reasoning phase of teeth cleaning, able to read and write, willingness to participate in the study, and have a Telegram channel. Exclusion criteria were gingivitis and gestational age less than 4 weeks or greater than 41 weeks.	C: Nothing I: Sending text and video messages	Gingival index Oral health condition	“Oral health education in patients with gingivitis could have a significant effect on decreased rate of gingival indices. Telegram social network as an electronic educational tool could be useful for this education and potentially improves the status of the patients with oral diseases.”
Al-Ak’hali et al. 202036 (Saudi Arabia) Periodontics	t1 = 4 t2 = 12	N = 43 15 to 45 years (26.83) M = 100% F = 0.0%	WhatsApp® Patients from the Division of Periodontics, Dept of Preventive Sciences, Faculty of Dentistry, University of Jazan, ages 15 to 45 years, and systematically healthy. Patients with periodontitis or reduced periodontium (probing depth >4 mm), those who reported sensitivity to any oral health product or medication use in the last 3 months, or who had an orthodontic appliance were excluded.	C: Traditional educational intervention I: Sending text, image, and video messages	Plaque index; gingival index Oral health condition	“Implementing WhatsApp instant messages does not appear to add extra benefit to the traditional motivation and education on oral hygiene practices in terms of changes in PI and GI over time in patients with gingivitis.”
Goicochea et al. 202037 (Peru) Orthodontics	t = 6	N= 46 13 to 23 years (18.65) M = 52.2% F = 47.8%	WhatsApp® Patients from fixed orthodontic treatment in 3 private clinics in Chiclayo, Peru, ages 13 to 23 years, with a diagnosis of Class I malocclusion, a treatment plan with fixed multibracket appliances, and who have a Smartphone-type mobile phone with an active WhatsApp application. Patients with an initial requirement for additional appliances (transpalatal arch, lingual arch, etc.), absence of any tooth considered in the oral hygiene index, signs of clinically observable periodontal disease, who regularly took a prescribed drug, with altered salivary flow, in gestational state or with a history of previous orthodontic treatment, were excluded.	C: Traditional educational intervention I: Sending video messages	Plaque index; questionnaire Oral health condition; oral health knowledge	“The educational intervention via WhatsApp had a positive effect on oral hygiene, with respect to plaque control and the level of knowledge of patients with fixed orthodontics.”
Lotto et al. 202038 (Brazil) General clinic	t1 = 12 t2 = 24	N = 104 3 to 5 years (3.6) M = 45.2% F = 54.8%	WhatsApp® Dyads of parents and children recruited during visits to preschools in Bauru, Brazil. Children ages 36 to 60 months, with low socioeconomic levels, an ICDAS score <4, parents or caregivers with a mobile phone with Internet access, parents or caregivers who accepted to participate in all stages of research by signing a written informed consent form, and parents or caregivers who already had WhatsApp Messenger app installed on their smartphones, or those who agreed to install it for participating in the study.	C: Nothing I: Sending text messages	Plaque index; ICDAS Oral health condition	“Therefore, mobile text messages were effective to control the severity of ECC in low socioeconomic preschoolers, improving parental eHealth literacy and changing children’s dietary patterns.”
Scheerman et al. 202011 (Iran) General clinic	t1 = 4 t2 = 24	N = 791 12 to 17 years (15.3) M = 58% F= 42%	Telegram® Residents in the urban area of Qazvin city, access to Telegram via smartphone, not engaged in other oral health education or research program, willing to participate and provide written informed consent before entry to the study, no physical and/or cognitive disabilities impeding the ability to perform oral hygiene activities.	C: Nothing I: Sending text and video messages	Plaque index; community periodontal index; questionnaire Oral health behaviour; oral health condition	“Current results support the use of the theory-based program delivered by Telegram in improving good oral hygiene behavior and oral health outcomes among Iranian adolescents. Involving mothers in an intervention can confer additional benefits for adolescent oral health.”
Borujeni et al. 202139 (Iran) Orthodontics	t1 = 4 t2 = 8 t3 = 12	N = 60 12 to 18 years (n = 24) and >18 years (n = 36) Gender distribution not specified	Telegram® Patients of a dental clinic in Tehran, with permanent dentition, class I dental and skeletal relationships with mild to moderate tooth size discrepancy, no need for extraction, no history of previous orthodontic treatment, no active medical or dental diseases, regular access to a smartphone, and willingness and informed consent to participate were included. If there was any evidence of active periodontal disease, the participant was excluded.	C: Traditional educational intervention I: Sending video messages	Plaque index; gingival index; gingival colour and consistency Oral health condition	“Teledentistry is an effective and efficient method to improve oral hygiene in patients undergoing fixed orthodontic treatment.”
Foner et al. 202140 (Brazil) General clinic	t = 4	N = 77 10 to 17 years (12.50) M = 40.3% F = 59.7%	WhatsApp® Participants were 10- to 17-year-old students in Florianópolis, enrolled in the sixth through ninth grades who used fluoride toothpaste and owned a smartphone with WhatsApp Messenger. Students were excluded from the study if they had carious lesions with involvement of the dental pulp, fixed orthodontic appliances, compromised general health (reported by caregivers, teachers or students) or learning disabilities.	C: Traditional educational intervention I: Sending text, image, and video messages	Plaque index; gingival index; questionnaire Oral health behaviour; oral health condition	“Both approaches to oral health education conventional and digital had a positive impact on the reduction of the PI, and the use of a smartphone app seemed to be an effective resource for providing oral health education to students.”
Santos et al. 202112 (Brazil) Orthodontics	t = 4	N = 44 (14.3) M = 50.0% F = 50.0%	WhatsApp® Patients wearing fixed orthodontic appliances on the permanent teeth (except on the third molars) were selected. All participants used WhatsApp on their cell phones and were users of university dental clinic. Patients with dentures, dental implants or syndromes, those who had undergone orthognathic surgery, smokers, those on antibiotics or on any other drug taken on a regular basis, or those who refused to sign the free informed consent form were excluded.	C: Traditional educational intervention I: Sending text messages	Plaque index; gingival index; halitosis Oral health condition	“Daily text messages seem to influence patient awareness of the importance of proper oral hygiene and flossing. The null hypothesis was rejected because those patients who received text messages every day changed their flossing habits.”
Saxena and Gunjal 202141 (Malaysia) Orthodontics	t1 = 4 t2 = 8	N = 54 ≥15 years (22.17) M = 25.9% F = 74.1%	WhatsApp® Patients with fixed maxillary and mandibular orthodontic appliances slot, M.B.T. prescription [American Orthodontics Mini Master metal brackets, American Orthodontists], aged ≥15 years, who were in orthodontic treatment for at least 3 months, and recruited from a university dental clinic. None of the patients displayed any significant medical or dental history, all had access to a cellular telephone and were online daily. Patients undergoing single arch treatment, missing maxillary lateral incisor, maxillary canine or mandibular canine or patients who were to be deboned within 3 months were excluded.	C: Traditional educational intervention I: Sending text messages	Plaque index Oral health condition	“WhatsApp and email reminders did not significantly influence the oral hygiene compliance of orthodontic patients.”

Among the (n = 5) systematic reviews published so far that included some OSN studies, only (n = 3) performed a meta-analysis for PI and (n = 2) for DC.^14-18 Both Mohammed et al.¹⁴ and Fernández et al.¹⁷, found statistically significant differences and low heterogeneity between groups for the DC outcome (in both studies, the classification of the outcome was stated as active white spot lesion) when using WhatsApp^® with young people. However, after the meta-analysis, only the study by Mohammed et al.¹⁴ found statistically significant differences and low heterogeneity for the PI outcome between groups when using WhatsApp^® with young people. These differences remained significant for the PI outcome only in the short term, with heterogeneity becoming high when the time was extended. This result drew the attention of the authors of that research to the effectiveness of the intervention over longer periods and the need for future studies to evaluate other interventions and OSN that have an impact on this outcome in young people.¹⁴ This fact may have occurred due to a decrease in the use of the platform and interest among participants over time.¹⁴ Furthermore, in the systematic review by Mohammed et al.¹⁴ the sending of reminders to patients undergoing orthodontic treatment was exclusively evaluated and included 2 studies that used WeChat^® and WhatsApp^® for meta-analysis. One of the important contributions of the present review is to explain the research gaps on digital interventions in oral health via OSN to encourage the scientific community to conduct future studies to assess the impact of their use through OSN on the oral health of different populations.

With further reference to the meta-analysis, it was possible to identify that interventions using OSN could be effective in reducing GI scores in young people. As a common point, all the studies included and evaluated in the meta-analysis of the present study relative to this outcome used the method of sending video messages as the main educational resource. One of the reasons for this may be the fact that this audiovisual resource stimulates greater engagement on the part of young users when compared with other educational digital strategies, a fact previously reported by Rigamonti et al.⁴⁴, who investigated the use of Facebook^®, Google+^®, Instagram^®, LinkedIn^®, Skype^®, Twitter^®, WhatsApp,^® and YouTube^® to promote health literacy, with emphasis on the science of sport and physical exercise.

Furthermore, after performing a meta-analysis, it was also possible to identify an increase in OHK in young people during orthodontic treatment, with the use of OSN.^{32, 35} To the best of the knowledge of the authors of this study, this is the first systematic review to investigate the impact of offering educational interventions via OSN on the OHK of users of oral health services. Until now, there has been the study by Lin et al.²⁶ in which the impact of OSN on OHK was tested with regard to the awareness, kno,wledge, attitude, and practice of teledentistry by dental surgeons during the COVID-19 pandemic. Therefore, the present study has brought innovative information in relation to both the outcomes, target audience, and OSN used for this purpose in the oral health care context.

Although the GRADE assessments showed very low certainty, the results of the meta-analyses of this systematic review should be better interpreted, as there was a reason for the calculation and effect of the assessment, which was the high risk of bias of the studies included. This fact was mainly due to the difficulty in blinding the participants, as these were digital educational interventions using characteristic audiovisual resources, leading the new RoB 2.0 tool to assess bias risk as high. Likewise, this result was found in the systematic review carried out by Wang et al.¹⁸ because it was the only study that included some articles using OSN in dentistry, and in a similar manner, used RoB 2.0.

Limitations

One of the limitations of this systematic review is the omission of some OSN on digital platforms. However, the decision to evaluate specific OSN was justified by their popularity in social spaces and the recognition of their use by government agencies. This provides the body of evidence with greater reliability, since the mentioned OSN are those over which there is greater control and that have rules of use.^27-29 Furthermore, although the only studies selected were RCT, there was great methodological variability and low follow-up time for participants between these studies, which made it difficult to compare results and long-term effects.

Nevertheless, the synthesis of the results of this systematic review on the use of OSN will improve reproducibility of the findings in the next reviews on the subject. This will be possible since these OSN are increasingly popular platforms for interventions in health and are unlikely to fall into disuse, which will facilitate reassessment in future studies by oral health researchers.⁸

CONCLUSION

Oral health interventions via OSN can be effective in reducing GI scores and increasing OHK of tobacco use, how to prevent oral cancer, and how to maintain good oral hygiene during orthodontic treatment in young people. Future studies with more refined methodological designs are needed to produce a more robust body of evidence on the subject.

CONFLICTS OF INTEREST

The authors have declared no conflicts of interest.