Medical social media (SoMe) has continued to grow in popularity and complexity with an estimated 90% of clinicians using it personally and 65% professionally. 1 Likewise, over half of clinicians report using SoMe for educational purposes, 2 and 85% of the general public use SoMe to connect with and seek healthcare information. 3 These platforms have numerous benefits, facilitating networking, education, advocacy, and outreach. It is not uncommon for conferences to now have sessions highlighting SoMe as a powerful tool for practice building and advertisement. Crowdsourcing can also help both patients and clinicians navigate challenging clinical decisions by seeking advice from experts across practices and continents, which would be difficult to access or facilitate efficiently through other forums.
However, these numerous benefits come with an equally diverse set of risks including concerns regarding patient privacy and dignity, conflicts of interest, interprofessional respect, information accuracy/misuse, and equity and justice. 1 4 5 6 7 Even online activity occurring on personal accounts can have professional implications, affecting enrollment, employment, and litigation. A growing number of organizations, professional groups, and literature are calling attention to these issues, but, in the authors' opinion, these risks are generally underdiscussed and appreciated. This article uses a series of anonymized case examples to highlight some of these challenges and offers suggestions on how to avoid potential pitfalls while taking advantage of the many beneficial aspects of medical SoMe.
Patient Privacy and Dignity
An interventional radiologist regularly shares Instagram stories under the title “A Day in the Life of IR.” He shares a case of “the largest pseudoaneurysm I've ever seen” for which he performed embolization and stenting. “If you have pulsating grapefruit in your groin, that's not normal folks!” Later he's called in for active extravasation. He shares a picture of the pelvic angiogram captioned “Joyride + shots = don't drink and drive.” The patient died a few days later from traumatic injuries.
The most widely discussed issues related to medical SoMe tend to involve patient privacy and confidentiality. Both legally and ethically, clinicians are required to maintain confidentiality by not sharing patients' identifiable information. 8 Sharing de-identified information for educational purposes is permissible, but the line between identifiable and de-identified is not always as clear as it may seem. For example, sharing patient identifiers such as a name, date of birth, or medical record number is clearly in violation of HIPAA, leaving the clinician at risk for potential criminal and civil charges. However, HIPAA can also apply to information such as the disease entity, treating facility, and date if it could be used to identify the patient. 9 10 In the authors' opinion, these less obvious violations tend to be more common where one shares sensational or unique cases soon after they occurred, making them potentially identifiable and in violation of HIPAA without documented patient consent. 11 Even something seemingly innocuous like sharing a challenging #FilterOUTFriday case the same day is likely identifiable by someone familiar with the patient.
Beyond privacy concerns, medical SoMe posts can also undermine patient dignity. Ethically clinicians have an obligation to protect and support the dignity of their patients. We are taught to treat patients and their loved ones with respect, upholding their autonomy, needs, and preferences regardless of their backgrounds. 8 12 However, we are also human with our own dignity, autonomy, and opinions, which can be threatened in clinical interactions. Healthcare can be tragic and draining, so it is not uncommon for clinicians to share their frustrations with their coworkers or even make light of the macabre as a means of coping. This social phenomenon is not unique to medicine and has been called “gallows humor.” 13 Gallows humor is not inherently wrong but can have negative effects, particularly in the public light of SoMe. Sharing a picture of the “grossest foot” one has ever seen or making fun of a seemingly de-identified patient can degrade a person, loved one, or tragic event to a punch line or notch in someone else's belt. Beyond compromising that clinician's relationship with current and future patients, it can undermine viewers' trust in healthcare in general.
To mitigate these risks, clinicians should ensure the clinical stories and images they share are truly de-identified, not only by removing patient identifiers but also considering the context and timing and what is shared. For example, it can be helpful to delay posting about an episode of care until there has been sufficient time to not identify the patient. If one regularly shares patients' stories and images, it would be prudent to obtain and document consent for the specific use. For example, a prior survey showed that 90% of patients found it acceptable to reuse images of them for individual educational purposes, but only 42% considered reuse on SoMe acceptable. 14 Clinicians should also consider whether their post upholds the dignity of the patient. A useful rule of thumb is to ask oneself, “if this was me or my loved one, would I be okay with my doctor or nurse posting this?” Table 1 provides an overview of additional suggestions.
Table 1. Suggested best practices for medical social media.
| Patient privacy and dignity |
| • Consider routinely obtaining and documenting consent to use a patient's de-identified clinical story or image online. This should be specific about intended use, i.e., posting on SoMe rather than just for “educational purposes.” This should not be buried within a standard consent form that the patient is unlikely to see |
| • Remove identifiable information including patient identifiers. Delay posting about an episode of care until there has been sufficient time to not identify the patient. Particularly rare conditions may not be able to be shared anonymously |
| • Consider the intent of the post and whether the patient is portrayed in a manner that would undermine their dignity. Does the post make light of a patient, their body, or their condition? |
| • Document consent in posts when possible |
| Information accuracy and misuse |
| • Use disclaimers in one's profile and/or posts to note that the information shared is not meant to be medical advice |
| • Share references/evidence to support educational content shared |
| • Be wary of using advice from SoMe to make treatment decision without additional evidence/experiences to support the decision |
| • Consider seeking and offering advice that will affect treatment decisions via less public forums such as SIR Connect or direct communication offline |
| • Avoid offering medical advice to patients or their loved one via SoMe. Instead direct them to a secure messaging system through one's practice or means of making an appointment. Clinicians can also encourage the person to seek emergency services if they feel the condition is emergent |
| Conflicts of interest |
| • Disclose relevant relationships associated with information shared on medical SoMe, e.g., “I'm a paid consultant of X” when posting about a product from X |
| • Be wary of less obvious COIs that may influence behavior and potential effect on those who will read the post |
| • Respect copyright laws that may restrict reuse of an image on SoMe |
| Interprofessional respect |
| • Avoid attacking or disparaging colleague or other specialties online, particularly behaviors that may be construed as cyber bullying, defamation, or stalking. Respectful debate and inquiry is acceptable |
| Justice and equity |
| • Consider potential biases in the content shared and whether it may perpetuate stereotypes, biases, and inequality in healthcare |
Information Accuracy/Misuse
A newly graduated IR is referred a patient with severe peripheral artery disease and non-healing foot ulcers. The patient has extensive, complex disease burden. The IR posts angiogram images on Twitter asking for advice about how best to treat his patient. An experienced IR in another state shares her experience with similar patients. The less experienced IR attempts her suggested approach and has major complications ultimately resulting in amputation. The patient's family sues and during the discovery period, the more experienced IR is subpoenaed and later accused of practicing in state in which she is not licensed.
Although medical SoMe can be a powerful source for sharing ideas and education, there is no peer review process for what we tweet. The most popular and widely shared information on SoMe is often the most sensational and not necessarily the most accurate. 15 16 As such, clinicians should be wary of the unfiltered nature of this content, particularly when using this information to make treatment decisions. For information consumers, clinicians should ensure that they have additional evidence for their treatment decisions—saying that one did something because they heard it on Twitter likely will not go well in a deposition, even if the source was someone they trust.
Likewise, we should be wary of the accuracy of what we post and whether it could be construed as medical advice. The SIR code of ethics notes that IRs should not “use any forum or medium of public communication to advertise themselves through deceptive, misleading, or untruthful information.” 12 Disclaimers can be helpful here to a degree, noting that something is simply your opinion or that your posts are not meant to represent medical advice or the views of your institution/practice. SoMe content is generally permanent and discoverable, so clinicians can be subpoenaed if another clinician views your advice as expert opinion and uses it to guide a decision that leads to a lawsuit. Clinicians can also run into issues when patients or family members become their friends on SoMe and ask for medical advice through these forums. For example, we are aware of situations in which physicians have joined disease-specific patient forums and have acted as advisors/thought leaders for the group including treating some of them as patients. The general advice across prior literature and guidelines is to direct requests for medical advice to a secure standard messaging or email system through your practice and/or to direct them to make an appointment. 3 6 7 17 This is because offering advice online can lead to privacy and licensure issues, leaving the clinician and their practice at risk.
While it is not wrong for the less experienced IR in the case to ask for help, assuming that the images and clinical information shared was truly de-identified, using a public forum for professional advice is, in general, fraught with ethical and privacy issues. It would be important for him to consider his comfort/experience with what was suggested and whether it is a well-accepted approach, which could be supported by literature or expert opinion in court. Ideally, he would have disclosed the risky nature of the procedure and documented the patient's informed consent to proceed. Although it is also not necessarily wrong for the more experienced IR to offer her experience, she should provide a reference or note that it is simply her experience that took years to be comfortable performing. It would be preferable, in our opinion, to pursue and offer this advice via less public forums such as privately in a direct conversation offline.
Conflicts of Interest
An IR regularly posts videos of IR procedures on TikTok with a large following. She's asked by a device representative to post a video of her using their new cryoablation system. They will provide the system for a discounted rate if she continues posting videos when she uses it. She captions the video, “tired of those old ablation systems? HuNamics' new EzICE made my day easy and efficient!” Shortly after the post, HuNamics invites her to give a paid talk at an upcoming conference.
Much like the discussion about information accuracy, conflicts of interest (COI) disclosures are not required for SoMe posts. It is not uncommon for influencers to get paid to promote products on their accounts or “thought leaders” in healthcare to be paid to give a talk or seminar about their experience with a given product. The potential legal issues here include antikickback and self-referral (Stark) laws. 18 19 Technically, the device company in the case could not directly offer a discount of other monetary gift in exchange for the IR using their system without being in violation of antikickback laws. However, there are many legal ways in which companies can reward clinicians for promoting their products.
Ethically, COIs are not inherently wrong but have the potential to be. We all have conflicting interests, financial or otherwise. 20 The potential ethical issue in healthcare is whether a COI biases behavior, causing a clinician to behave in a manner not in a patient's best interests—will the IR in the case keep using EzICE to keep getting compensated by HuNamics despite inferior patient outcomes, whether consciously or subconsciously? Previous work suggests even small rewards can bias behavior, 21 22 and people tend to be limited in their abilities to recognize their own biases. 23 This has been previously described as “bounded ethicality,” that people are more likely to justify their own behavior even if they would condemn the same behavior of someone else. 23 Additionally, the COI issues that are publicized and taught tend to be egregious, making it easier to distance ourselves from them and overlook less obvious biases. 24 25 Even prompted disclosures for conferences and publications are often incomplete 24 ; so, it is likely that SoMe is particularly vulnerable to bias from COIs that is seldom discussed or acknowledged.
To mitigate these issues, clinicians should appreciate the complexity and power of COIs on clinical behaviors, which extends beyond financial quid pro quos. 26 Medical SoMe influencers and prominent members of specialty communities have substantial power to influence the behaviors of those who look up to them. What may seem like an innocent post about an amazing thrombectomy session after your favorite device representative stopped by with their newest product can have meaningful ethical implications. The authors encourage clinicians to consider these often subtle biases, and if a clear relationship exists, disclose it in the post so that others can weigh this when considering what is shared. This also aligns with the SIR code of ethics, which suggests minimizing and disclosing COIs. 12 Table 1 provides additional suggestions.
The examples above illustrate three potential sets of ethicolegal issues related to medical SoMe: patient privacy and dignity, information accuracy/misuse, and COI. Another very relevant issue is that of interprofessional respect and potential cyber bullying. Medical specialties and practices will always be competitive. 27 This is not inherently wrong but when competition for patients bleeds over into frank online disparagement of our colleagues, we believe that an ethical line has been crossed. This can threaten the ethos of teamwork in medicine by fracturing cross-specialty relationships and fostering harmful stereotypes about other colleagues and/or specialties. Likewise, cyber bullying can lead to litigation for defamation or can spiral into stalking. These additional issues are less commonly discussed in the literature on medical SoMe ethics and should be explored further by physicians and their professional societies. One example of thoughtful deliberation in this area is the recently published recommendations from the Society for Vascular Surgery, 6 which IR societies could consider modeling a similar set of recommendation upon to provide a citable resource for our specialty.
In summary, medical SoMe is a powerful tool for education, community/practice building, and advocacy that can also be used in ways that raise serious ethical and legal concerns. IRs should familiarize themselves with their practices'/institutions' policies regarding SoMe use if they exist. Either way, the authors hope this piece can serve as a resource to take advantage of the many benefits of medical SoMe while avoiding potential pitfalls. #PauseBeforeYouPost
Acknowledgments
This work was supported by the SIR and SIO Applied Ethics Working Group.
Funding Statement
Funding None.
Footnotes
Conflict of Interest The authors have none to disclose.
References
- 1.Ventola C L. Social media and health care professionals: benefits, risks, and best practices. P&T. 2014;39(07):491–520. [PMC free article] [PubMed] [Google Scholar]
- 2.Bowser D. Clinical Care Options; 2017. Social media behavior and attitudes of us physicians: implications for continuing education providers - a clinical care options (CCO) white paper; pp. 1–7. [Google Scholar]
- 3.Bennett K G, Berlin N L, MacEachern M P, Buchman S R, Preminger B A, Vercler C J. The ethical and professional use of social media in surgery—a systematic review of the literature. Plast Reconstr Surg. 2018;142(03):388e–398e. doi: 10.1097/PRS.0000000000004692. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Denecke K, Bamidis P, Bond C. Ethical issues of social media usage in healthcare. Yearb Med Inform. 2015;10(01):137–147. doi: 10.15265/IY-2015-001. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Chretien K C, Kind T. Social media and clinical care: ethical, professional, and social implications. Circulation. 2013;127(13):1413–1421. doi: 10.1161/CIRCULATIONAHA.112.128017. [DOI] [PubMed] [Google Scholar]
- 6.Gifford E D, Mouawad N J, Bowser K E. Society for Vascular Surgery best practice recommendations for use of social media. J Vasc Surg. 2021;74(06):1783–17910. doi: 10.1016/j.jvs.2021.08.073. [DOI] [PubMed] [Google Scholar]
- 7.Ethics, Law, and Humanities Committee, a Joint Committee of the American Academy of Neurology, American Neurological Association, and Child Neurology Society . Busl K M, Rubin M A, Tolchin B D. Use of social media in health care-opportunities, challenges, and ethical considerations: a position statement of the American Academy of Neurology. Neurology. 2021;97(12):585–594. doi: 10.1212/WNL.0000000000012557. [DOI] [PubMed] [Google Scholar]
- 8.Beauchamp T L, Childress J F. New York: Oxford University Press; 2019. Principles of Biomedical Ethics. 8th ed. [Google Scholar]
- 9.45 C.F.R. § 160103
- 10.45 C.F.R. §§ 164.502(d)(2), 164.514(a) and (b)
- 11.Cifu A S, Vandross A L, Prasad V.Case reports in the age of Twitter Am J Med 201913210e725–e726 [DOI] [PubMed] [Google Scholar]
- 12.SIR Code of Ethics. Society of Interventional RadiologyAccessed February 25, 2022 at:https://www.sirweb.org/globalassets/aasociety-of-interventional-radiology-home-page/about-us/governance/sir-code-of-ethics_approved-november-2018.pdf
- 13.Watson K. Gallows humor in medicine. Hastings Cent Rep. 2011;41(05):37–45. doi: 10.1002/j.1552-146x.2011.tb00139.x. [DOI] [PubMed] [Google Scholar]
- 14.Wyatt K D, Finley A, Uribe R. Patients' experiences and attitudes of using a secure mobile phone app for medical photography: qualitative survey study. J Med Internet Res. 2020;22(05):e14412. doi: 10.2196/14412. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Pershad Y, Hangge P T, Albadawi H, Oklu R. Social medicine: Twitter in healthcare. J Clin Med. 2018;7(06):E121. doi: 10.3390/jcm7060121. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Lee J Y, Sundar S S. To tweet or to retweet? That is the question for health professionals on twitter. Health Commun. 2013;28(05):509–524. doi: 10.1080/10410236.2012.700391. [DOI] [PubMed] [Google Scholar]
- 17.Collaboration for Outcomes on Social Media in Oncology (COSMO) . Attai D J, Anderson P F, Fisch M J. Risks and benefits of Twitter use by hematologists/oncologists in the era of digital medicine. Semin Hematol. 2017;54(04):198–204. doi: 10.1053/j.seminhematol.2017.08.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.42 U.S.C. § 1320a-7b
- 19.42 U.S.C. § 1395nn
- 20.Lo B, Ott C. What is the enemy in CME, conflicts of interest or bias? JAMA. 2013;310(10):1019–1020. doi: 10.1001/jama.2013.221227. [DOI] [PubMed] [Google Scholar]
- 21.Lundh A, Lexchin J, Mintzes B, Schroll J B, Bero L. Industry sponsorship and research outcome. Cochrane Database Syst Rev. 2017;2:MR000033. doi: 10.1002/14651858.MR000033.pub3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.DeJong C, Aguilar T, Tseng C W, Lin G A, Boscardin W J, Dudley R A. Pharmaceutical industry-sponsored meals and physician prescribing patterns for Medicare beneficiaries. JAMA Intern Med. 2016;176(08):1114–1122. doi: 10.1001/jamainternmed.2016.2765. [DOI] [PubMed] [Google Scholar]
- 23.Chugh D, Bazerman M H, Banaji M R. Cambridge University Press; 2005. Bounded ethicality as a psychological barrier to recognizing conflicts of interest; pp. 74–95. [Google Scholar]
- 24.Hsieh L J, Madadi S R, Shore K T, Keller E J, Makary M S. Potential bias in image-guided procedure research: a retrospective analysis of disclosed conflicts of interest and open payment records. J Vasc Interv Radiol. 2022;33(02):141–147. doi: 10.1016/j.jvir.2021.08.026. [DOI] [PubMed] [Google Scholar]
- 25.Chambers T S. No Nazis, no space aliens, no slippery slopes and other rules of thumb for clinical ethics teaching. J Med Humanit. 1995;16(03):189–200. doi: 10.1007/BF02274154. [DOI] [PubMed] [Google Scholar]
- 26.Wiersma M, Kerridge I, Lipworth W. Dangers of neglecting non-financial conflicts of interest in health and medicine. J Med Ethics. 2018;44(05):319–322. doi: 10.1136/medethics-2017-104530. [DOI] [PubMed] [Google Scholar]
- 27.Kreindler S A, Dowd D A, Dana Star N, Gottschalk T. Silos and social identity: the social identity approach as a framework for understanding and overcoming divisions in health care. Milbank Q. 2012;90(02):347–374. doi: 10.1111/j.1468-0009.2012.00666.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
