Analysis of the Brazilian-Portuguese Content on Autism Spectrum Disorder Available on YouTube Videos

Abstract Aims Information about Autism Spectrum Disorder (ASD) is available in different videos on YouTube. However, these videos might present inaccurate or outdated information. The aims of this study were to 1) identify characteristics of Brazilian-Portuguese videos about ASD (i.e., type of content, number of views, likes, and dislikes); 2) analyze the content of informative videos about ASD, by assessing their trustworthiness and overall quality; and 3) investigate the most popular topics of informative videos about ASD over time. Methods This is a cross-sectional study that investigated Brazilian-Portuguese ASD content on YouTube. Videos were selected by two examiners and categorized as experiential or informative. The Discern checklist and the Global Quality Score (GQS) were used to analyze the trustworthiness and quality of informative videos. Results In total, 216 videos were analyzed, of which 19.5% and 85% were classified as experiential and informative, respectively. The majority of informative videos presented moderate trustworthiness and quality. Videos about ASD clinical aspects were the most popular. Conclusion YouTube offers a large number of experiential and informative videos on ASD. However, some of these videos fail to present reliable and additional sources of information for stakeholders. Efforts to promote knowledge translation about ASD on YouTube are necessary.

disorders, epilepsy, intestinal and urinary disorders, genetic syndromes, and psychiatric disorders (Bauman, 2010;Oliveira et al., 2023).There is evidence that most cases of ASD result from a combination of genetic predisposition and environmental factors during brain development (Kreiman, & Boles, 2020).ASD is usually diagnosed between four and five years of age, considering the patient's social characteristics and stereotypes (Fountain et al., 2011;Maenner et al., 2021;Mandell et al., 2010;Zwaigenbaum et al., 2015).In low-and middle-income countries like Brazil, ASD diagnosis can occur at around six years of age or more (Sociedade Brasileira de Pediatria, 2019).
Given the high prevalence of ASD, there has been exponential growth in the number of studies involving individuals with this condition.Despite this, in clinical practice, health professionals have difficulties in implementing evidence-based information (Longo et al., 2017;Morris et al., 2011).Furthermore, the scenario of available information for caregivers of individuals with disabilities is complex, disconnected, difficult to access, and often contains misconceptions about ASD.This might be associated with reduced well-being and quality of life for children and their families (Gibson et al., 2017).Scientific information produced in research typically takes more than 10 years to be implemented in practice, causing a significant impact on the quality of health services (Lupatini et al., 2022;Morris et al., 2011).Hence, strategies are needed to facilitate and promote trusted knowledge among health professionals, individuals with ASD, and their families.
Knowledge translation (KT) is the integration of synthesis, dissemination, and knowledge application for stakeholders, to improve the quality of health services (Canadian Institutes of Health Research, 2015).The Knowledge-to-action framework is a theoretical model that presents a conceptual map with dynamic stages, from knowledge creation to its implementation in practice (Graham et al., 2006).Among these dynamic stages, the knowledge products or tools are crucial to facilitate dissemination in a clear and easy-to-understand format (Graham et al., 2006).Internet-based information is one type of knowledge tool that disseminates health content, even in low-and middleincome countries (Associac ¸ão Brasileira de Internet, 2022; Chag & Donghwi, 2021;Onder & Orhan, 2021).In these countries, information regarding ASD and other disabilities is available on many social media platforms, such as YouTube (Bellon-Harn et al., 2020;Leite et al., 2021).
YouTube is a popular source of non-peer-reviewed videos, uploaded by individuals.It is one of the most popular social media, with one billion hours of viewing per day (Mangan et al., 2020;Onder & Orhan, 2021).Ninety-six percent of Brazilian adults use this social medium at least once a day (Youtube Insights, 2017).The millions of videos available on YouTube represent a wide spectrum of interests of different users, including educators, academics, and researchers (Snelson, 2011).Thus, within the Knowledgeto-action KT framework, YouTube might be a promising tool that can be used by stakeholders for ASD knowledge dissemination.Despite the benefits of accessing health information on YouTube, the information presented on this social medium might not always be reliable (Furtado et al., 2022).Misleading information in YouTube videos about ASD can present undated and inaccurate data that might not be helpful for patients, families, and healthcare providers (Fern andez-Llatas et al., 2017).
Many studies have analyzed the informative content on YouTube in different health conditions, such as diabetes, hypertension, and cerebral palsy (Baytaroglu & Emrah, 2021;Cakmak, 2021;Furtado et al., 2022;Kumar et al., 2014;Silva et al., 2020).Some of these studies showed that informative videos on YouTube presented low to moderate trustworthiness and quality.Regarding ASD content, Kollia et al. (2017), Azer et al. (2018), andLacruz-P erez et al. (2021) analyzed the characteristics and the informative content of the most widely accessed YouTube videos on ASD.These studies identified popular informative topics (e.g., multidisciplinary treatment, social and emotional behaviors, and support services), and critically appraised the content of videos in English and Spanish.The authors reported that misleading information might be presented in YouTube ASD informative videos (Kollia et al., 2017;Lacruz-P erez et al., 2021).To the best of our knowledge, the appraisal of Brazilian-Portuguese ASD YouTube content has not yet been conducted.An analysis of informative Brazilian-Portuguese content on YouTube about ASD would clarify whether this social medium can be used as a reliable information source for Brazilian patients, families, and health professionals.
The objectives of this study were: 1) to identify the characteristics of Brazilian-Portuguese videos about ASD (i.e., type of content, number of views, likes, and dislikes); 2) to analyze the content of informative videos about ASD, by assessing their trustworthiness and overall quality; and 3) investigate the most popular topics of informative videos about ASD over time.

Study Design
This is a cross-sectional study.No Institutional Ethics Committee approval was required because patient data were not used.

Video Searching
Video searching followed the steps reported by Furtado et al. (2022).Only one examiner was designated to search YouTube videos in order to provide one consistent list of videos and avoid the presence of algorithms.This search was conducted in April 2021 on the YouTube website (www.youtube.com)without date restriction.The keywords "autism spectrum disorder" or "Autism" in Brazilian Portuguese were used.In addition, the examiner took the following steps before performing the search, to minimize the influence of algorithms: 1.The examiner disconnected the YouTube personal account; 2. Adblock software was installed to avoid advertisements; 3. The examiner searched in the anonymous mode.
The examiner searched for the 200 most viewed videos on the website (for each keyword used).This number of videos was recommended by other YouTube content analysis studies (Furtado et al., 2022;Garg et al., 2015), considering that YouTube users usually do not look for more than this number of videos when searching content (Garg et al., 2015).The examiner listed the titles and hyperlinks of these videos.
After the search process, two independent examiners, familiar with the ASD body of evidence conducted the video selection, data extraction and categorization, and content analysis of the eligible videos.These examiners were a physical therapist with seven years of experience in pediatric rehabilitation and a physical therapy undergraduate-level final-year student.Each step (i.e., selection, extraction, categorization, and analysis) was considered complete after the consensus of both examiners.In case of disagreement in any step, a third examiner (also a final year physical therapy undergraduate level student) was requested.

Video Selection
Before the selection process, the examiners gathered to discuss the study eligibility criteria.Videos were included if they were in the Brazilian-Portuguese language and presented any content related to ASD.Videos were excluded if they were duplicates, live videos, advertising, or videos with different content other than ASD.

Data extraction and categorization
The examiners independently extracted descriptive information about the selected videos (title, duration, year of uploading) and the video's interactive parameters (number of likes and dislikes; number of comments; and number of views).After data extraction, examiners independently categorized the videos into two groups: 1. Informative videos: Videos that presented ASD educational content for individuals with ASD, families, or health professionals.To be classified as informative, videos needed to present at least one of the following topics: etiology, diagnosis, classifications, clinical characteristics, epidemiology, assessment tools, multidisciplinary treatment, and/or social aspects.These topics were based on the Autism Spectrum Disorder Guidance Manual, from the Brazilian Pediatrics Society (Sociedade Brasileira de Pediatria, 2019).2. Experiential videos: Videos that presented narratives and experiences of individuals with ASD and families.Experiential videos presented content related to biographies, testimonies, interviews, documentaries, reports, or fictional stories representing individuals with ASD, and/or videos showing their ' daily routines (therapy sessions, school, or home environment).
Interviews, documentaries, or reports that presented more than 10% of the video duration of informative content (e.g., etiology data or ASD characteristics) were classified as informative.This criterion was established by the authors since no standardized criteria to discriminate between experiential and informative videos are described in YouTube analysis studies.

Content analysis of informative videos
To analyze the content of the informative videos, the following information was extracted: 1. Content domain; 2. Information provider; 3. Type of presentation; 4. Funding information.
The content of the videos was analyzed according to their trustworthiness and general quality, using the modified Discern checklist and the Global Quality Score (GQS).Both instruments present established psychometric properties (Bernard et al., 2007;Charnock et al., 1999) and have been used to analyze YouTube content (Garg et al., 2015;Kumar et al., 2014;Li et al., 2020;Strychowsky et al., 2013).
The Discern checklist is a five-point checklist that evaluates the accuracy of health information presented in videos (Charnock et al., 1999).Discern total scores vary from zero to five, considering five different topics (one point each) (Supplemental material 1).Higher scores on this scale represent better trustworthiness of the video.The GQS evaluates quality, flow, and information usefulness.According to this scale, the informative videos were classified into five categories (Singh et al., 2012): 1. Poor quality, poor flow of the video, a lot of information missing, not at all useful for patients or professionals; 2. Generally poor quality and poor flow, some information listed but many important topics missing, of very limited use to patients or professionals; 3. Moderate quality, suboptimal flow, some important information is adequately discussed but other information is poorly discussed, somewhat useful for patients or professionals; 4. Good quality and generally good flow.Most of the relevant information is listed, but some topics are not covered, that are useful for patients or professionals; 5. Excellent quality and flow, very useful for patients or professionals.
Video quality, flow, and usefulness, evaluated by the GQS, were based on DISCERN scores; language and terminology used by the presenters; and video relevancy for the stakeholders (patients, parents, or health professionals).The examiners established appropriate criteria to consistently assess the video's trustworthiness and quality (Supplemental Material 1).Considering the broad body of evidence of ASD that could be included in YouTube videos, the examiners conducted group discussions about popular topics of informative videos, before formally assessing trustworthiness and quality.In addition, to ensure consistency among examiners, before the formal evaluations an inter-examiner reliability analysis was conducted with 10 videos, using the interclass correlation coefficient (ICC) and the Kappa index.The analysis results showed appropriate inter-examiner reliability for the Discern checklist (ICC ¼ 0.70) and GQS (Kappa index ¼ 0.80), according to Bruton et al. (2000).

Data Analysis
Descriptive statistics were conducted using frequency, for categorical variables, and mean and standard deviation (or median and interquartile intervals for non-normal distribution), for continuous variables.

Results
From 400 videos searched, 216 videos were eligible for this study, of which 42 (19.5%)videos were categorized as experiential, and 174 (80.5%) as informative.The detailed flow chart with the search and selection process is shown in Figure 1.Table 1 presents the interaction parameters of all selected ASD YouTube videos.Experiential videos presented a higher number of interaction parameters (views, likes, and dislikes) in comparison with informative videos.Interaction parameters were also analyzed separately by content topics of the informative videos: etiology, diagnosis, ASD classifications, ASD clinical characteristics, epidemiology, assessment tools, multidisciplinary treatment, and social aspects.Videos related to assessment tools presented a higher number of views and likes.
Table 2 details the analysis of all informative videos.ASD clinical characteristics, diagnosis, and multidisciplinary treatment were the main topics covered in the YouTube videos.Most informative videos were presented in lecture format, by different healthcare professionals, mostly psychologists, and physicians.There was a small contribution from rehabilitation professionals in this group of videos.
A timeline with all informative ASD YouTube content topics is presented in Figure 2. The most relevant informative YouTube content started to be published in 2014.It is possible to visualize that content related to clinical characteristics is the most popular   Most of the informative videos (51.1%) scored three points on the DISCERN checklist, representing moderate trustworthiness.Detailed information on DISCERN and GQS is also presented in Table 1.The least scored topics on the Discern checklist were "4.Are additional sources of information listed for patient reference?"(19.5%) (e.g., videos that mention that there are many articles and texts about a treatment option without referring to or providing them to the consumers).The second and third least covered topics (71.8% and 76.4%) were "3.Are reliable sources of information used?" (e.g., videos that did not inform who the presenter was and their background) and "5.Are areas of uncertainty mentioned?"(e.g., videos that present the benefits of a treatment option but do not present any study about it).In consequence, most informative videos (89%) showed moderate quality and flow, considering the GQS criteria.All content topics presented moderate trustworthiness (median 3 out of 5 in the DISCERN) and moderate quality in the GQS, with the exception of assessment tools that presented good quality and flow.

Discussion
YouTube is an important source for disseminating information about health conditions, diseases, or procedures (Madathil et al., 2015).The current study analyzed the characteristics of the most commonly accessed videos on YouTube about ASD in Brazilian-Portuguese and provided a detailed analysis of the content, trustworthiness, and quality of informative videos.In general, more than 80% of the videos published on YouTube are informative and mainly address information related to the diagnosis and characteristics of individuals with ASD.Most informative videos presented moderate trustworthiness and quality and were created by physicians and psychologists.
YouTube is a valuable source of information on ASD, featuring several videos with informative or experiential content (Azer et al., 2018;Kollia et al., 2017).Despite the low number of experiential videos in Brazilian-Portuguese found in the present study, they showed a greater number of interactive parameters (views, likes, and dislikes).It has been speculated that experiential content on YouTube might facilitate networking and opportunities for advocating and sharing experiences (Mazanderani et al., 2013).Considering the knowledge-to-action framework, experiential videos can play an important role in the dynamic stages of the action cycle by identifying problems; adapting knowledge to real contexts; and exposing, monitoring, and evaluating barriers and facilitators to knowledge use (Graham et al., 2006).This finding suggests that Brazilian ASD patients and their families might use YouTube as a source to share experiences with others.
The contents related to clinical characteristics and a diagnosis of ASD were the most popular in YouTube Brazilian videos, followed by multidisciplinary treatment.We believe that Brazilian families might seek information about clinical features and diagnosis due to difficulties in ASD early diagnosis in Brazil (Sociedade Brasileira de Pediatria, 2019).Epidemiology and, in particular, assessment tools were the least discussed topics in the ASD videos on YouTube.Even though there are many assessment tools for healthcare professionals to assess different clinical aspects in individuals with ASD (Hanratty et al., 2015;Wigham et al., 2019), this content is not commonly presented on free platforms.The same result was found by Furtado et al. (2022) who analyzed informative YouTube videos about cerebral palsy.The small number of videos that provide information on the epidemiology of ASD can be explained by the scarcity of epidemiological data on ASD in Brazil (Paula et al., 2011).In the Autism Spectrum Disorder Guidance Manual, from the Brazilian Pediatrics Society (Sociedade Brasileira de Pediatria, 2019) there is no information on Brazilian epidemiological data.Data provided in the guideline only refer to United States epidemiological information (Sociedade Brasileira de Pediatria, 2019).Therefore, Brazilian YouTube video producers have little information to provide about ASD epidemiology.
In general, the results of the present study show that informative videos about ASD present moderate trustworthiness and quality.The scarce information about the video's presenter, as well as the lack of availability of additional sources of information frequently decreased Discern checklist scores.In addition, scientific evidence was moderately included in the Brazilian videos, also preventing better trustworthiness and quality.Similar results were found by Azer et al. (2018) andLacruz-P erez et al. (2021), when analyzing informative ASD content on YouTube videos in English and Spanish, respectively.Lacruz-P erez et al. (2021) stressed the importance of Spanish videos about ASD to include additional credible sources to facilitate access to high-quality information for stakeholders.Therefore, for this social medium to also better present informative content about ASD in Brazilian-Portuguese, video producers must provide appropriate evidence-based reliable sources of information for consumers.
In the current study, the informative videos about ASD available on YouTube were produced mostly by physicians or psychologists.The absence of rehabilitation professionals, such as occupational therapists and speech-language pathologists was unexpected.Both professions are well established in Brazil and their role in the multidisciplinary treatment of individuals with ASD is widely known (Case-Smith & Arbesman, 2008).One of the reasons for the little involvement of these professionals in the production of videos may be the unavailability of time and motivation.In Brazil, rehabilitation professionals face exhausting working hours, precarious working conditions, and low wages (Almeida, 2013).Another surprising finding of the present study was the low participation of parents and caregivers in the production of informative videos (23% of the videos).Given the importance of informative YouTube videos in KT, the participation of parents and caregivers in the production of videos of this nature should be encouraged.
Informative content on YouTube began to be published in 2014, with a significant increase between 2016 and 2018.Clinical aspects were the most popular content found.The higher number of videos published on ASD from 2016 to 2018 might be associated with advances in screening and early diagnosis criteria for ASD (Maenner et al., 2021).Furthermore, this increase in ASD videos on YouTube may be related to the more frequent use of YouTube as a source of information sharing by healthcare professionals (Li et al., 2020).The reduction in the production of informative content about ASD on YouTube occurred in 2020 when the COVID-19 pandemic took hold.This finding might be justified by the increasing visibility of other social media during the pandemic, such as Instagram.
The current study has some limitations that should be mentioned.Even though we provide a content evaluation of the informative videos, detailed specific informative video analysis (i.e., the target public and comments content) was not conducted.In addition, content analysis of the experiential videos was not performed.Considering the importance of these videos, future netnographic and qualitative studies that perform an in-depth analysis of Brazilian YouTube videos could provide rich information regarding ASD.Future studies can also investigate if the Brazilian informative videos about ASD are useful and relevant, considering the perception of individuals with ASD and their families.Furthermore, due to the extension and unstructured presentation of saved live videos (lives) on YouTube, it was not feasible to include this format in our content analysis.Since these videos are becoming popular on social media, future studies might seek strategies to critically analyze their content.

Conclusion
YouTube includes large content about ASD in Brazilian-Portuguese. Informative videos about ASD include several topics, mostly presented by health professionals.The majority of these videos presented moderate trustworthiness and quality since they failed to present reliable and extra sources of information.Efforts to better promote knowledge translation about ASD on this type of platform are extremely necessary.

Figure 1 .
Figure 1.Flow chart diagram of video searching and selection.ASD: autism spectrum disorder.
an increase in publications from 2014 to 2018.In addition, videos about the diagnosis started to increase in 2017 and reached their peak in 2018.There was an overall decrease in the number of videos published in 2019 in comparison with 2014 to 2018.

Table 1 .
Interactive parameters.trustworthiness and quality of informational videos content.

Table 2 .
Content analysis of informative videos.
Figure 2. Timeline of YouTube informative ASD topics.