Social Media Perceptions of Clubfoot Treatment: A Global Observational Study

Background: Clubfoot is a debilitating congenital anomaly that is often under treated in lower- and middle-income countries. There exists an active social media community with regards to clubfoot on both Instagram and Twitter. The purpose of this observational study was to investigate shared content related to clubfoot on social media. Methods: A search for public posts on social media regarding clubfoot was performed from October 31st, 2018 to October 31st, 2019. Three hashtags on Instagram and Twitter were analyzed: #clubfootsurgery, #clubfoot, and #clubfootmommas. Data was analyzed using Chi-squared tests with signicance set to p<0.05. Results: Overall, 2050 posts with clubfoot-related hashtags were identied as publicly shared on these social media platforms over the 12-month period. 1992 met the inclusion criteria. Pictures consisted of 93.9% of posts which overwhelmingly had a positive tone, while only 16 posts were explicitly negative. On Instagram, personal accounts authored 95.5% of posts and on Twitter, organizations shared 99.4% of posts. Content related to active treatment was shared 32.7% of the time - of these posts, 94.8% showed Ponseti casting or Ponseti bracing. Chi-squared analysis was performed comparing Instagram with Twitter for media format, timing, tone, author, content, visibility, and topic reference. Statistically signicant differences between the two (p<0.001) groups were found in each of these categories. Conclusions: Clubfoot treatment has a large social media presence with contributions from patients, treatment organizations, and physicians. Personal accounts on Instagram emphasize the patient experience, whereas organizations on Twitter focus primarily on education and increasing overall awareness.


Introduction
In recent years, the prevalence of clubfoot treatment has increased signi cantly as training programs around the globe have expanded, technology has improved, and available materials and providers have increased. Clubfoot, also known as congenital talipes equinovarus (CETV), has a global prevalence between 0.6 and 1.5 per 1,000 live births [1]. Of nearly 200,000 new cases per year, 90% occur in lowerand middle-income countries (LMIC) [2]. While widely regarded as a treatable congenital abnormality with the gold standard treatment the Ponseti method, clubfoot continues to leave many with lifelong de cits [3,4]. Contributing to this is the lack of early intervention, as only an estimated 15% of the diagnosed in LMIC start treatment with the Ponseti method [2]. De cits that persist later in life have been shown to lead to lifelong negative effects such as di culties nding a marital partner, experiencing limited educational opportunities, and increased poverty [5,6].
Social media provides a unique and powerful lens to examine the patient experience for those affected by clubfoot [7][8][9]. The use of social media is not limited to high-income countries (HIC) and continues to rise in LMIC [10]. The purpose of this observational study was to analyze publicly shared content on social media platforms, speci cally Instagram and Twitter, to gain an understanding of patient, physician, and organizations perspectives regarding clubfoot treatment. Speci cally, we evaluated posts for (1) media format (picture, text, or video); (2) tone (positive, negative, or neutral); (3) perspective (personal account, hospital, professional organization, physician, or other); (4) timing (active treatment, post-treatment, nontreatment); (5) content (education, advertising, research, news/media updates, or patient experience); (6) post visibility (number of hashtags); and (7) location. We hypothesized that Instagram would cultivate an environment for individuals to share daily updates with a relatively positive tone, and Twitter would focus more on scienti c advances and have a higher in uence of physicians.

Search
A search of public posts on Instagram and Twitter was performed on November 1, 2019 for posts from a 1-year time period: October 1, 2018, to October 1, 2019. Posts were identi ed by use of three hashtags: #clubfoot, #clubfootsurgery, and #clubfootmommas. A quantitative approach was taken to identify the hashtags that were most relevant to clubfoot and related treatment using the Symplur Healthcare Hashtag Project. [11] Other hashtags such as #clubfootsurgery, #clubfoot, #clubfootmommas, #equinovarus, #ponsetimethod, #ponseti, #talipesfoot, #talipes, #birthdefects, and #miraclefeet were considered; however, the three included hashtags had the most speci c and highest yield in regards to number of posts and relevance to the clubfoot treatment. The query of Instagram and Twitter resulted in a total of 1992 posts across both platforms including at least one of the three selected hashtags.

Inclusion Criteria
Only posts relating to human participants were included. All posts referencing clubfoot were included, and those discussing other procedures were excluded. Only posts in English were included to avoid misinterpretation and the lack of funds to pay for an interpreter. Data was collected and analyzed by two independent reviewers (H.S. and L.Y) familiar with the Ponseti method and related clubfoot treatment. Interrater variability was resolved by review of original media and discussion with the rst author (E.K) to achieve agreement. Data analysis was performed in Microsoft Excel. A binary categorical scoring scheme was used for media format, perspective, timing, tone, content, post visibility, and geographic location.
Tone was determined by overall explicit positive, negative, or neutral expression in the text accompanying and media. In the majority of cases this referred to positive connotations to tackling the global burden of CTEV with positive accurate information on management and early detection. No posts were excluded for interrater variability.
Anonymity and Risk of Harm: Much ethical consideration was given to use of the data of a potentially vulnerable population found within this study. The data aggregated was determined to be public and thus reasonable to be accessed as our research group only accessed public accounts that may be viewed by any user worldwide without requesting speci c access to the content. A thorough review of the terms and conditions of both Twitter and Instagram was performed to ensure the aforementioned methods aligned with site policies. Twitter explicitly states that any posted information is "viewable and searchable by anyone around the world" and Instagram states that posted information is "considered public information." [12,13] In addition, only posts including the aforementioned hashtags were collected for analysis, indicating that these users desired their content to reach a larger population that is similarly interested in clubfoot related happenings; and thus, reasonably expected their content to be accessed by strangers. No private account information was accessed in any manner.
Given the potential sensitive nature of the subject matter, there is risk of harm if identi able data were to be ampli ed more broadly than originally intended. As such, we sought to preserve the anonymity of the authors with the following methods. No direct quotes or account names were collected or are shared throughout this manuscript. Similarly, all data presented is deidenti ed and unable to traced back to the original authors without complete replication of the study methods. Due to the sensitivity of the data, no public data sets will be made available. Upon analysis, no data was suspected to be shared by participants under the age of 13, however if encountered, any such data would have been eliminated from analysis. The data presented addresses key themes within each of the platform's clubfoot related communities and does not intend to infer judgement on the study populations. Recommendations from the Social Media Research: A Guide to Ethics were closely followed. [14] Results Instagram Table 1 (Table 1) Twitter Similarly, a summary of the content analyzed from Twitter is available in Table 1. A total of 493 tweets were included in the study. These tweets were almost exclusively (490/493, 99.4%) authored by organizations, with only three total posts authored by either a personal account or a hospital. A majority of the posts had a positive tone (87.42%), 11.76% had neutral tones, and only 0.81% posts having a negative tone. As for timing, 11.2% of posts showed active treatment, 4.26% showed obvious posttreatment content, and 84.8% of posts shared content unrelated to treatment. In terms of content, 75.9% of posts contained an advertisement, 28.2% referenced media or news, and patient experience was only represented in 16.9%. Each post had an average of 1.99 hashtags. The location of all posts was determined in the same manner as posts from Instagram. Posts originated predominantly from the United States (354/493, 71.8%). In total, 44 different countries from across the globe were represented.
Comparison: Chi-squared analysis was performed comparing Instagram with Twitter for media format, timing, tone, author, content, visibility, and topic reference. Statistically signi cant differences between the two (p<0.001) groups were found in each of these categories, as depicted in Table 1.   Discussion CTEV is considered the most common serious musculoskeletal birth defect across the globe [5,15].
Considering the high prevalence of the disease, the low level of treatment within LMIC, and ability of the disease to harm in ways beyond the physical level, there is great opportunity for improvement in current management [6,16]. Multiple studies have been done comparing the Ponseti method to other treatment types such as the French method; however, the Ponseti method continues to produce the best outcomes and satisfaction [17][18][19][20][21][22]. Despite an adequate method to treat clubfoot, patients and families continue to face uncertainty and fear when receiving this diagnosis [23,24]. One author described their experience as a "stressful, complicated journey." Online social communities have the ability to provide an avenue of relief and connection with those across the globe [23]. The online community with respect to clubfoot has a presence in over 50 countries, spans multiple platforms, and unites the perspectives of organizations, families, and patients.
Instagram has an active social media clubfoot community. The authors of the posts predominantly (1,320/1,440, 92%) post from the United States, the United Kingdom, and Australia (Fig 3). Despite this, over 40 different countries were represented and thus giving this community a far-reaching in uence.
These posts were shared predominantly (1432/1499, 95.5%) from personal accounts and contained a positive tone (1157/1499, 77.2%). Given the positive nature of these posts and focus on the patient experience (893/1499, 59.6%), the stories shared on Instagram indicate the platform has provided individuals a forum to share the daily successes, updates, and obstacles. Posts are rarely shared by physicians and health professionals, which perhaps, has allowed those within the community to connect with others traversing similar hardships and allowed for a sense of vulnerability to develop within the Instagram community. One individual with a child diagnosed with clubfoot commented that Instagram has allowed her to see adults with clubfoot do "amazing things" and has given her con dence that one day her child will also be able to do those things (Anonymous, unpublished).
Twitter similarly has an active clubfoot community but with a different demographic. Many of the posts were made from within the United States, but as compared to Instagram, were not shared from personal accounts but primarily from organizations that are based within the U.S and have a large global reach. One public account for a large, international organization authored the highest number of tweets (168/493) and is based in the United States, however, also conducts work in 26 other countries across the globe thus showing that location data may be skewed towards higher income countries [25]. The included tweets were most frequently advertisements (374/493, 75.8%), many of which were focused on fundraising. With a much smaller in uence from personal accounts and posts focused on patient experience, the Twitter community is one of education, awareness, and fundraising. Physicians had an underwhelming presence on Twitter but may be more active on other social media sites not included in this study [26]. It is also plausible that the content shared by the organizations on Twitter were authored by physicians and other providers working within the organizations and thus may represent a similar viewpoint. The demographics and topics represented within this platform form a pillar of the clubfoot social media community, one that emphasizes the scienti c progress in the treatment and care for those with clubfoot [27].
Active treatment posts from both Instagram and Twitter were analyzed (Fig 2) for the type of content and treatment shown. Ponseti casting and bracing posts made up over 45% of active treatment posts on Instagram and over 50% of active treatment posts on Twitter. Post-treatment posts frequently showed children riding bikes, running, standing, and displaying their feet with a smile on their face. The community often highlights positive content and is described by multiple individuals as an outlet to express thoughts and feelings related to present struggles. In a private conversation, one individual said "I nd it so refreshing that I feel so connected to each of the [other] mom's journeys" in regard to how unifying the community can be in hard times (Anonymous, unpublished).
While the combination of these two platforms create the image of a holistic community with collaboration between physicians, patients, researchers, and families -the reality is that there is little communication between them. The differences in content posted about and the demographics of the posters highlight this point. With little representation of organizations on Instagram and a smaller portion of posts highlighting the patient experience on Twitter, a perception of divide has been created. Are these communities then at a loss for the lack of collaboration between one another? And can steps be taken to remediate this disunity?
Worth considering rst is if this virtual distance between the two groups has been bene cial for those directly affected by clubfoot by creating an open space for vulnerability and community. Without the in uence of healthcare providers and researchers this community may have been given the room it needed to develop. However, given the positive tone of many of these posts and numerous photos of patients and families smiling, it is reasonable to hypothesize that many have found their clubfoot management to be an overall positive experience. This in addition to literature demonstrating that patients using social media for health-related topics are often driven by a search for social support, health information, as well as exchanging advice supports the notion that collaboration between the two groups may be vital [28][29][30]. Unity between these two communities may be further explored through studies analyzing potential methods for connecting with patients through social media while maintaining patient anonymity in addition eliciting the desires from patients and their families. Physicians and health professionals ought not enter blindly into these communities with the intention of immediately connecting with patients and their families but may begin considering ways to incorporate social media connections into their practice to improve care for those with clubfoot.
Geographic representation: Given the distribution of both of the online social media communities represented in Figs 3A, 3B, and 3C, it is clear that the expanse of this clubfoot community is far-reaching. In an era where collaboration across the globe can be instantaneous, representation from individuals from different countries and their unique viewpoints could not be of higher importance. Within the clubfoot community, 53 countries had one or many individuals author at least one post. These 53 countries, however, only represent where the author is based in and does not include the numerous countries that individual or organization represent. If united perhaps through an online forum or another platform, the wide expanse of clubfoot in social media could be used to deepen the sense of community it already has started to develop [8].

Limitations
This study is not without limitations. Due to the nature of the social media platforms used, only public posts with the selected hashtags were analyzed. Thus, a large number of posts may have been excluded if they were made private or did not include a relevant hashtag. Yet, the search across the two platforms included a total of 1992 posts over an extended study period of one year, generating a large sample for analysis. In general, social media users tend to highlight positive aspects of their subject, potentially creating a bias toward positive outcomes while failing to capture the negative experiences that patients may face [31]. Although other social media sites such as Facebook and Snapchat are popular among social media users, Instagram and Twitter were chosen for their ability to provide a large data set of public posts with relevant hashtags that could be objectively analyzed for content.

Conclusion
In conclusion, the social media community with regards to clubfoot represents numerous countries and individuals from diverse backgrounds. Social media provides unique insight into the patient experience that ought to be utilized [32]. With input from physicians, researchers, parents, patients, and organizations, the community represents many perspectives and backgrounds. The content shared is overwhelmingly positive and frequently focused on the patient experience. There does, however, exist a divide between the patient and family focused Instagram community and the education, awareness and fundraising focused Twitter community. Physicians, health professionals, and organizations ought to enter into the patient centric online communities with humility and a desire to learn. True connections and collaborations may be formed through intentional effort and evaluation of the unique speci cs of their patient base and local in uence. With such expansive reach and representation, these divided communities ought to collaborate to form a higher sense of unity for the continued improvement in the treatment and care for those with clubfoot.

Declarations
Ethics approval and consent to participate: The authors recognize that social media analyses require careful review of ethical principles. In order to uphold the highest of ethical standards, the Ethical Guidelines for Internet Research from the British Psychological Society as well as the Economic and Social Research Council were carefully reviewed. Four main principals include: Respect for the Autonomy, Privacy and Dignity of Individuals and Communities, Scienti c Integrity, Social Responsibility, and Maximizing Bene ts and Minimizing Harm. In regard to these principals, no identi able information or quotations were used throughout the study, when quotations were used from private discussion, they were identi ed as such throughout the text. No formal ethical approval was sought prior to this study. As stated in the ESRC's guidelines, we assessed the risk of making a discovery of illegal images or activities as low -as the main predicted authors of tweets and Instagram posts (inferred from previous studies) were likely to be professional organizations, healthcare providers, and patients who were sharing updates about treatment and unlikely would include any illicit information. With the addition of no speci c patient information being shared, the authors felt it reasonable to proceed without formal ethical approval. Recommendations from the Rocial Media Research: A Guide to Ethics by Townsend and Wallace were also followed.