A scoping review of crowdsourcing and collective intelligence in oral healthcare

Abstract

Objective

Diagnostic and treatment errors are detrimental yet relatively common in primary healthcare settings. Crowdsourcing and collective intelligence (CI) in the form of peers’ opinion and group decision-making can lead to modifications in diagnosis and treatment planning by dentists. In this review, it was explored their use in oral healthcare delivery, investigating specific applications and evaluating their merits and drawbacks in oral healthcare.

Methods

A scoping review approach was chosen as the optimal study design, due to the complexity of the topic. A literature search from 1980 to December 2024 using selected search terms and a search strategy was implemented across six databases: Medline, Scopus, Web of Science, SciELO, LILACS, and CINAHL. Inclusion criteria specified studies published in English, Spanish, and Portuguese, focusing on CI applications in oral healthcare settings.

Results

Of the 265 identified studies, 63 abstracts and 20 full-text studies were screened, and 7 studies were included in the qualitative synthesis. The findings suggest that CI may significantly enhance patient management in primary healthcare settings by employing various group decision-making models such as multidisciplinary teams (MDTs) and the Delphi method. MDTs, comprising professionals from different disciplines, can enhance patient care through collaborative problem-solving and holistic approaches. The Delphi method allows for reliable, expert-driven decisions through anonymous feedback rounds. CI not only addresses diagnosis and treatment planning challenges but also improves resource allocation. While these approaches can reduce individual biases and systematic errors, challenges such as conformity bias and groupthink were also acknowledged.

Conclusion

This review highlights the significance of diverse expert opinions and structured consensus-building in advancing collective intelligence in oral healthcare. It emphasizes CI’s potential to improve patient management through collaborative decision-making models. However, further research is necessary to explore CI’s specific applications and effectiveness in oral healthcare.

Introduction

In many healthcare settings, diagnosis and treatment planning are the responsibilities of a single clinician’s judgement to decide the diagnosis and treatment of a patient [1]. Whilst individual judgements rely on common and widely accepted cognitive strategies, which are useful in making efficient decisions, these can also lead to severe and systematic errors [2]. For example, the strategies used to solve uncertain outcomes may be misled by limitations in recalled information, extrapolation, and inaccurate estimations [2]. An over-reliance upon individual judgements, makes the diagnostic process vulnerable to inherent biases and cognitive strategies the may not be optimal. Typically, only complex cases are discussed amongst peers and appropriate specialists.

Diagnostic errors are among the most common, costly, and catastrophic medical errors [3]. The rate of diagnostic error in outpatient care amongst the US adult population has been reported are approximately 5% with around half of these are harmful [4]. A similar proportion has been reported in the UK [5]. A failure of cognitive reasoning is common experience by those who think intuitively rather than critically [6]. When applied to healthcare settings, the propensity for individual systematic and predictable errors is of great concern to patient treatment outcomes.

To reduce the limitations of individual diagnosis, combining expert judgements into a group decision can lead to improved diagnostic accuracy in disease detection, identification, and prediction [7]. A consensus framework is constructed by harnessing the experiences of many minds and may function as a way to mitigate the limitations of individual judgements. Groups can make better decisions because of having more brain power, more information, and diverse perspectives [8]. Therefore, in situations where uncertainty prevails, it may be worthwhile for decisions to be founded upon consensus frameworks. Theoretically, collective intelligence (CI) should show superiority as individual practitioners aggregate their clinical expertise [9]. Research indicates that collaborative approaches outperform individual diagnoses in primary healthcare settings [10–12]. CI can be achieved by expert interactions in multidisciplinary team (MDT) meetings [9]; or by aggregating independent decisions made by individual experts [9], although multiple iterations may be required to reach consensus [13, 14].

Teamwork is essential in healthcare, particularly in medicine, where collaboration commonly occurs during ward rounds and MDT meetings [13–17]. Group consensus methods utilising anonymous communication and independent judgments, like the Delphi method, offer predictability but can complicate the process and extend timeframes [18]. Additionally, the emergence of interactive online platforms and large language models (LLMs) present innovative opportunities to use CI in healthcare [19].

In this study, collective intelligence was defined as the pooling of decisions among practitioners with expertise, to form a consensus [15]. Surowiecki describes four criteria for effective CI [15]. The first condition is the independence of individuals within the group, whereby the initial opinion of each participant is not influenced by others within the same group. The second condition is diversity of opinion, where everyone hold unique knowledge of the matter at hand. The third condition is decentralisation, where opinions have equal weight. Finally, the consensus requires a method to aggregate these multiple opinions to create a collective decision [15].

The aim of this scoping review was to explore how crowdsourcing and collective intelligence are used in oral healthcare delivery. The review aims to answer the question: In the context of oral healthcare, what specific applications of crowdsourcing and collective intelligence approaches can be identified, and what are their respective merits and drawbacks in improving diagnosis, treatment planning, and patient care? Additionally, it evaluated the merits and drawbacks of these models in diagnosing and treatment planning in oral healthcare. In so doing, the review highlights the implications of CI in practice and identifies areas for future research aimed at leveraging peer validation for enhancing patient care.

Methods

Design and eligibility criteria

A scoping review approach was chosen because it serves as a preliminary exploration of the available literature to identify the evidence and assess the scope of literature on this topic [20–22], and allowed for synthesizing information from diverse sources, including qualitative as well as quantitative research. This scoping review was reported according to Preferred Reporting Items for Systematic reviews and Meta-Analyses Extension for Scoping Reviews (PRISMA-ScR) guidelines [23].

The review included primary studies and conference papers focussed on the use and outcome of aggregated opinions of multiple practitioners in healthcare, published in English, Spanish or Portuguese. Study participants could be either or the same or mixed disciplines, and with the same of different levels of expertise; making decisions based on real or simulated cases; making decisions related to diagnosis, treatment, patient management or achieving consensus for sets of statements. Studies that investigated the implementation of collective decisions were also included. There was no restriction on study design, but a publication date was set from 1980 to November 2025.

Secondary studies, or those published solely as abstracts or posters, as well as studies lacking full texts, published results, were excluded. Additionally, studies that focused exclusively on the technical aspects of crowdsourcing or collective intelligence in dentistry, letters to the editor, and conference abstracts were also omitted from consideration.

Sources of information and search strategy

A systematic search of the literature was conducted on MEDLINE, SciELO, LILACS, Web of Science, CINAHL and Scopus. Reference listings of selected articles were hand searched to identify other possible studies. Articles published in the grey literature [information not published through traditional academic channels], were searched ad-hoc. The details of the search strategy used, and results are given in Table 1.

Table 1.

Searched strategies and sources of information

Database	Search Strategies	Results
Scopus	(TITLE-ABS-KEY(“decision-making” OR “Collective intelligence” OR “Interdisciplinary” OR “Interdisciplinary treatment”)) AND (TITLE-ABS-KEY(“clinical application” OR “Diagnosis” OR “Dentistry” OR “endodontic”)) AND (TITLE-ABS-KEY(“periapical lesions” OR “Deep bite” OR “Wisdom of crowds”))	93
Medline	((“decision-making”) OR (“Collective intelligence”) OR (“Interdisciplinary “) OR (“Interdisciplinary treatment”)) AND ((“clinical application”) OR (“Diagnosis”) OR (“Dentistry”) OR (“endodontic”)) AND ((“periapical lesions”) OR (“Deep bite”) OR (“Wisdom of crowds”))	71
Web of science	((“decision-making”) OR (“Collective intelligence”) OR (“Interdisciplinary”) OR (“Interdisciplinary treatment”)) AND ((“Diagnosis”) OR (“Dentistry”) OR (“endodontic”))	49
CINAHL	(“decision-making” OR “Collective intelligence” OR “Interdisciplinary” OR “Interdisciplinary treatment” OR “clinical application” OR “Diagnosis” OR “Dentistry” OR “endodontic” OR “periapical lesions” OR “Deep bite” OR “Wisdom of crowds”) AND (dental OR dentistry OR “oral health”)	10
LILACS	((“decision-making”) OR (“Collective intelligence”) OR (“Interdisciplinary “) OR (“Interdisciplinary treatment”)) AND ((“clinical application”) OR (“Diagnosis”) OR (“Dentistry”) OR (“endodontic”)) AND ((“periapical lesions”) OR (“Deep bite”) OR (“Wisdom of crowds”))	11
SciELO	((“decision-making”) OR (“Collective intelligence”) OR (“Interdisciplinary”) OR (“Interdisciplinary treatment”)) AND ((“Diagnosis”) OR (“Dentistry”) OR (“endodontic”))	31
Total		265

Study selection and data extraction

All references identified were exported into Rayyan online software (https://rayyan.qcri.org) for selection. The selection process to identify studies that potentially meet the inclusion criteria involved independent screening of title and abstract of studies retrieved by two reviewers (RM and RvM). Any disagreement between the reviewers over the eligibility of studies, was resolved through discussion until consensus was reached. The reviewers were not blinded to the authors or journals. The reasons for exclusions were recorded. On a second stage the full text of all relevant and potentially relevant studies was obtained.

Data charting process

One reviewer (RM) extracted relevant data from eligible studies, and an additional reviewer checked the information extracted for accuracy (RvM) (non-independent verification of data extraction). The reviewer extracted the following information from each article using a standardized, predefined data collection form: author, year, aims of studies reviewed, study design, makeup of the expert panel, type of patient case (real or simulated), type of aggregated opinion or decision, form of consensus, results, authors’ conclusion and details of the relevance of each study to CI. This is summarised in Table 2, using a format adapted from Radcliffe and collaborators [9].

Table 2.

Summary of characteristics of included studies

Author (Year)	Type and number of experts	Study description	Type of data (Real/Simulated)	Type of opinions aggregated	Collective intelligence framework	Relevance to collective intelligence
Krischer and Dixon (1982)	8 Speech pathologists, 5 General medical practitioners, 6 dental specialists, 1 nurse	Analytical study of how different specialty groups assess probability outcomes for treatment modalities.	Simulated (scenario)	Treatment alternatives	Multidisciplinary team	Treatment preferences changed after decision analysis, indicating the influence of structured decision-making processes on individual preferences.
Reit and Gröndahl (1984)	35 Chief dental officers from the Public Dental Health Organization in Sweden	The study presents a decision tree for clinical management of endodontically treated teeth with periapical lesions.	Simulated cases (clinical records radiographs)	Treatment alternatives	Group decision	Highlights the variability in expert opinions and decision-making process, underscoring the collaborative nature of decision-making.
Reit, et al. (1985)	13 General dentists and 8 endodontists	The study investigates decision analysis regarding to management of periapical lesions in endodontically treated teeth.	Simulated cases (clinical records radiographs)	Treatment alternatives	Multidisciplinary team	The study indicates that for treatment decisions practitioners rely more on simpler judgments than on subjective assessments.
Reit and Gröndahl (1987)	13 General dentists and 8 endodontists	Analytical study of Delphi method to compare individual decisions to group consensus.	Simulated cases (clinical records)	Probability of success/failure of treatment options	Delphi method	Investigated the influence of Delphi as a framework for group consensus to improve decision-making under uncertainty.
Perasso et al. (2018)	3: Orthodontist, prosthetist, general dentist	Case report of collaboration over a patient with aesthetic complaints	Real case	Treatment plan	Multidisciplinary team	The treatment plan achieved significant improvements clinically and patient satisfaction.
Bhatt et al. (2021)	3: Postgraduate endodontic students, endodontist	Comparison of CBCT findings with conventional radiographs in endodontic cases	Simulated cases (clinical records radiographs)	Diagnosis and treatment plan	Multidisciplinary team	Additional information from CBCT scans resulted in the alteration of the initial diagnoses as well as subsequent treatment plans.
Ganhewa et al. (2023)	13 General dentists	The study explored the implementation of CI to improve diagnostic accuracy and treatment planning.	Simulated cases	Diagnosis and treatment plan	Delphi method, Interactive Platform	The study suggests that CI can enhance diagnostic accuracy and treatment planning, improving confidence and collaboration.

CI Collective intelligence, CBCT Cone beam computed tomography

Results

The search identified 265 citations, with two citations identified manually. After duplicates were excluded, 63 were screened by title and abstract, and 20 full-text articles were selected. Of the full-text articles reviewed, 13 were excluded. Finally, data from seven articles were extracted. A PRISMA chart (Fig. 1) shows the details of each step of the selection process. All studies included in this review were observational studies. Most of the studies were released in the 1980 s [24–27], while three were published from 2018 to 2023 [28–30].

Fig. 1.

PRISMA flowchart demonstrating the selection process of articles retrieved from different web sources

Included studies varied in the diversity of expert groups participating. Except for one study focusing solely on general dental practitioners (GDP) [30], most studies included GDP alongside different dental specialists, such as endodontists [24–27, 29], another study included GDP and different dental specialists (i.e., orthodontist and prosthodontists) to improve diagnosis to treatment efficiency to address patients’ outcomes concerns [28]. Additionally, one study explored how different health professionals (i.e., general medical practitioners, speech pathologists, and nurses) assessed probability outcomes referred to cleft palate treatment modalities [27].

Regarding the decision-making design, one study looked at real cases – this refers to the diagnosis, management, or decision making of a patient where the group decision influences the patient treatment [28]. The other six studies used simulated cases where decisions were generated from de-identified data such as images and radiographs [24–27, 29, 30].

By type of decisions collected, five studies assessed decisions regarding patient clinical management [24–28]. The other two studies looked at both, diagnostic and patient management decisions [29, 30]. In the studies examining diagnostic and treatment decision accuracy, the main types of information evaluated by participants were radiographic images alone or in conjunction with comprehensive descriptions of the clinical history.

Consensus was derived through collaborative group discussions in the form of panels or formal MDT meetings. One study used the Delphi method within an interactive platform enabling the rapid sharing of cases and harnessing of collective intelligence [30]. The remaining studies had both independent and collaborative phases [25–29], one of which employed the Delphi method [26]. Specific comparisons between individual and group performance were not made in any of the studies included.

Outcomes

The studies reviewed investigated whether diagnosis, or treatment planning, would be enhanced by collective expertise. Two studies illustrated how the quality and validity of the consensus obtained with each round of Delphi was also investigated [26, 30]. Results showed there was a trend, indicating an increase in accuracy of group opinion and consensus after each round. Reit and Gröndahl indicated that increasing consensus after each round may present problems [26]. However, they also concluded that despite the general increase in agreement, the individual clinician’s confidence levels did not change [26]. One study explored the influence of the Delphi method as a framework for achieving group consensus through an interactive platform [30]. The use of the platform demonstrated superiority of CI, showing improved diagnostic accuracy with the use of this platform [30].

This review identified studies indicating that the use of multidisciplinary team (MDT) discussion methodologies facilitated comprehensive assessment and diagnosis, thereby improving treatment efficiency and outcomes [24, 27–29]. Krischer and Dixon [27] noted that different professions appear to have biases towards the preferred treatment modality choice. However, participants changed their treatment choice after decision analysis, indicating the influence of structured decision-making processes on individual preferences, or on simpler judgments [25].

Contributions from peers [28] and additional information [29] were also found to be useful in supplementing the individual’s decision-making process, especially when complex cases were being considered [28]. The review underscores that MDT approaches not only enhance patient care but also play a protective role in ensuring accurate assessments and optimal treatment strategies, ultimately improving overall outcomes for patients.

Discussion

This scoping review identified seven studies describing ways in which Clinical Intelligence (CI) has been utilized in oral healthcare settings. The studies included highlight the potential of CI to enhance decision-making processes within the field of oral healthcare. Findings suggest that there is a relative increase in the number of research publications focusing on CI in recent years, indicating that this area may be gaining interest among oral health researchers and practitioners. However, despite this growing interest, the applications of CI in oral health are not yet comparable to its implementation in other areas of healthcare, such as medicine, where it has been more widely adopted [9]. This disparity underscores the need for further exploration and development of CI strategies specifically tailored for oral healthcare settings, ultimately benefiting patient care and outcomes.

In the following paragraphs, findings will be discussed following four elements of Collective Intelligence, as outlined by Surowiecki [15]. This framework not only allows for a comprehensive exploration of the review findings but also contributes to guiding future studies and practical applications of CI in oral healthcare.

Independence of opinion

One characteristic of achieving high quality CI is ensuring that all opinions, at least in the first instance, are independent of one another. A deliberative environment such as an MDT is vulnerable to conformity bias, as interacting individuals are prone to being influenced by others. Participants within the group might also succumb to groupthink, informational influences, and social pressure, whereby, in an effort to reach a consensus, they may either hinder contributions or not critically evaluate other alternatives [31, 32]. This has implications for the final decisions as the meeting might not have effectively drawn out the expertise and opinions of all participants [18, 33]. Deliberative groups are also likely to get anchored [2] to the first opinion they hear.

Collective intelligence relying on anonymity in group processes shows potential for improving diagnostic accuracy. The studies showed that the Delphi method can be used to compensate for some of the limitations of deliberative environments such as MDTs and interactive platforms. as the Delphi method reduced social pressures and conformity bias. This is because Delphi enables independent clinicians to make initial diagnostic inputs anonymously. The Delphi method is widely used in dentistry for establishing consensus [34–36].

Given the limitations of MDTs, it can be argued that the use of a CI interactive platform can be advantageous within primary healthcare settings. Such a platform facilitates collective insights to reach a consensus, regardless of geographically location and can potentially improve confidence and accuracy of diagnosis. Furthermore, interactive platforms will likely become more prominent with technological developments. In this review, only one study used CI for diagnosis through an interactive platform [30]. Nonetheless, interactive platforms will have the very same pitfalls unless they are structured to allow independence of opinion where the participants are unable to view the preceding opinions. Social media platforms and deliberative forums do not meet this criterion as users will always read comments prior to contributing.

Whilst these studies indicated that the Delphi method can potentially establish group consensus, none of the studies reviewed described the application of CI in improving patient outcomes. Additional study is warranted to assess whether the Delphi method can be beneficial in oral healthcare beyond improving diagnostic accuracy, for example, as a method to improve patient care. Despite the Delphi method having some strengths, like MDTs, the Delphi method is ultimately vulnerable to problems of human cognition, motivation, inefficiency and practical limitation, which may affect the eventual consensus.

Diversity of opinions

It is suggested that diversity of opinions accounts for a wide variety of perspectives, enabling the patient’s condition to be scrutinised to a higher degree from multiple viewpoints [27, 28]. On the other hand, if the team members are too similar and therefore their collective skill base is narrow, different perspectives may not be considered even if a deliberative approach is undertaken [4]. This drawback warrants further research into the effects of a clinician’s demographical background in providing a different approach. Additional study is also needed on the role of students who at the time, were not fully qualified within their respective disciplines, in improving accuracy of the group diagnosis [40]. Based on current literature, we postulate that students and clinicians in training may be valuable team members when harnessing CI to improve diagnostic accuracy [19, 29].

Decentralisation of opinions

The decentralization of opinions aims to ensure no single voice has more power over others. However, deliberative groups may may fail in this regard as dominant and authoritative personalities, or those with impressive credentials may be given extra consideration [18, 32, 33, 37].

The inherent structure of the Delphi method allows for better decentralisation than other deliberative groups such as MDT. It is important to note that Delphi groups still make use of a moderator or communicator in between anonymous rounds to communicate the groups opinions to individual participants, which may lead to less-than-ideal conditions for satisfying decentralisation.

From this review, one study used collective intelligence for diagnosis through an interactive platform in a structured way [30]. This highlights a significant gap regarding clinical evidence for collective intelligence diagnostic systems. Diagnostic imaging represents an ideal medium through which the benefits of collective intelligence can be harnessed through digital platforms [10–12, 30, 38]. However, interactive platforms are by no means immune to this phenomenon. Although interaction is usually anonymous or pseudonymous, the participants are able to recognise ‘admin’ status or see the numbers of posts by, or various badges that have been awarded to certain participants. These platforms have grown in prominence, studies have demonstrated their potential for diagnosis through an interactive platform in a structured way [39].

Reliable method to aggregate opinions to form consensus

Group discussions, like multidisciplinary teams (MDTs), often achieve consensus, though the accuracy and implementation of that consensus is sometimes uncertain [40–43]. Social pressures like conformity bias and groupthink might aid in reaching agreement but can also compromise accuracy. While the Delphi workflow provides a structured approach to achieving consensus, one study [26], which evaluated two patient cases, ultimately could not reach a definitive conclusion. This inability was reported in a medical setting, which suggests that the processes of aggregation and consensus-building may still present challenges [44].

Another consideration is the group size, which has been reported to enhance accuracy through majority rule. In the present review, five of the included studies consisted of large group sizes (i.e., n ≥ 9) [24–27, 30]. However, some authors have concluded that groups as small as three [10, 31] can also improve diagnostic accuracy. This accuracy can be maximised in several ways. First, by increasing group sizes, and secondly, by applying different CI rules: majority vote, confidence rules, and quorum rules [10–12, 18, 38]. However, minority opinions may be overlooked when participants conform to those of the majority [13, 25, 44]. Thus, MDTs and the Delphi method, are susceptible to cognitive and motivational limitations, which may impact consensus.

This highlights the necessity for innovative approaches in these environments. For example, artificial intelligence (AI) is an area that is emerging to support decision-making in dentistry. None of the studies selected in this review evaluated AI. Although the accuracy is still under scrutiny [45], there is increasing interest in the application of AI in dentistry, warranting further exploration [46]. A review of AI-generated treatment recommendations indicates that AI can complement traditional methods and improve clinicians’ decision-making when refining treatment plans [47], suggesting that AI could serve as a valuable tool for enhancing treatment outcomes. Additionally, the FDI World Dental Federation asserts that AI can significantly improve diagnostics and treatment planning, advocating for its integration in oral healthcare to develop more efficient and higher-quality treatment plans, ultimately benefiting patient outcomes [48].

Although we were methodical in our review, it had some limitations. Firstly, while our search strategy included studies written in English, Spanish, or Portuguese, we may have missed relevant articles published in other languages. Secondly, as our initial search was based on published studies, unpublished studies might have been missed. The initial screening process was based on key phrases within the title so it is likely that we might have excluded relevant studies. the combination of terms that yielded the highest number of studies was ultimately selected as the final strategy. This was followed by a more specific search. A hand-search for relevant literatures was also conducted. Therefore, any effect, if present, would have been minimal due to the inclusive nature of our main search in the six databases.

None of the studies included examined the challenges associated with CI, specifically time constraints, complexity, and the phenomenon of non-implementation. Non-implementation of CI decisions has been explored across various medical settings [40–43]. These studies found that the most common factor for non-implementation included failure to consider patient preferences. The presence of comorbidities and the availability of further clinical information were also reported as a significant reasons for non-implementation [42, 43], which may suggest that patients with multiple health conditions may affect adherence to clinical guidelines. Additionally, there are other factors described as influencing both the implementation of CI decisions and clinical decision-making. For example, the final treatment decision rests on the patient themselves, often based on personal preferences [45]. Whatever the case may be, factors contributing to the non-implementation of CI decisions in oral healthcare have not been investigated, indicating a need for further research in this area.

Furthermore, the studies reviewed, with one exception [28], presented results obtained via simulated or past clinical cases and thus should be interpreted with caution, as the diagnostic tasks undertaken for these studies are not true reflections of real clinical conditions. Furthermore, none of the studies included focused on the unique challenges faced by public oral healthcare services. These gaps emphasize the need for further research into how CI approaches can be effectively implemented within public oral healthcare settings. Thus, while studies in our review generally demonstrated positive participation rates, it is important to acknowledge that awareness of participation in a research setting may have played a role (Hawthorne effect), which may not be replicated in clinical settings [49]. Therefore, this scoping review must not be considered as final, but as a step forward to identify the nature and extent of the current literature and research evidence in this topic.

Conclusion

This scoping review provides an overview of the current literature on the use of collective intelligence in oral healthcare, with several studies showing that collective intelligence, either via MDTs, the anonymous Delphi method, or other interactive platforms, improves diagnostic accuracy and leads to better treatment outcomes. Based on the present findings, the broader utility of harnessing collective intelligence using interactive platforms is just beginning to be recognized within oral healthcare. The present scoping review reveals the need for more collective intelligence research targeted toward oral healthcare. Given the large amount of research indicating the benefits of collective intelligence in medical and healthcare settings, it is hoped that this review will encourage further research within the oral healthcare professions.

Abbreviations

CBCT

Cone beam computed tomography

CI

Collective intelligence

CINAHL

Cumulative Index of Nursing and Allied Health Literature

LILACS

Latin American and Caribbean Literature on Health Sciences

LLMs

Large language models LLM

MDT

Multidisciplinary team

PRISMA-ScR

Preferred Reporting Items for Systematic reviews and Meta-Analyses Extension for Scoping Reviews

SciELO

Scientific Electronic Library Online

Authors’ contributions

RM: Participated in the conception and design of the study, acquisition of data, analysis, and interpretation of data; as well as drafting of the manuscript and its critical revision, and approval of the final version.NC: Participated in the conception and design of the study, acquisition of data, analysis, and interpretation of data; as well as drafting of the manuscript and its critical revision, and approval of the final version.MG: Participated in the conception and design of the study, and interpretation of data; as well as drafting of the manuscript and its critical revision, and approval of the final version.RvM: Participated acquisition of data, analysis, and interpretation of data; as well as drafting of the manuscript and its critical revision, and approval of the final version.AL: Participated in the conception and design of the study, acquisition of data, analysis, and interpretation of data; as well as drafting of the manuscript and its critical revision, and approval of the final version.

Funding

Not applicable.

Data availability

All data generated or analysed during this study are included in this published article (Please see Table 2).

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

Rodrigo Mariño is a Senior Editor for BMC Oral Health. The other authors declare no conflict of interest.