Ethical Considerations in Pediatricians’ Use of Social Media

Professional Education and Collaboration

Faced with multiple competing obligations and rigorous productivity expectations, physicians may need to consult with colleagues about challenging cases or research current evidence quickly. Social networking sites can help close this gap, as long as they are compliant with privacy requirements of the Health Insurance Portability and Accountability Act (HIPAA) (which would include identity verification and sufficient encryption). Physician-to-physician communication can also occur on public channels such as Twitter, through which groups of physicians congregate and “follow” each other for information sharing, thought leadership, and advocacy in pediatric public health. The American Academy of Pediatrics (AAP), for example, maintains a list of more than 700 tweeting AAP members, known as “Tweetiatricians.”¹⁷

Overall, 90% of physicians today use social media for professional purposes, including finding and reading relevant medical information.¹⁸ Social media are especially popular with early-career pediatricians who may be more familiar with digital technology, have received mentoring in its use during training, or attend conferences in which tweeting is encouraged to ask questions or provide feedback. Social media allow learners to express their opinions and ask questions without fear of embarrassment and has been observed to “flatten the hierarchy” of medicine by giving each person an equal voice.¹⁹

Patient Education and Advocacy

Studies have revealed that 8 in 10 Internet users go online for health information,²⁰ and more than half of these people get medical information specifically from social media.²¹ Social media tools can be used to share relevant and important public health topics and to guide the public, patients, and families who follow a physician’s channels to creditable resources that have been reviewed by physicians (such as online media articles, blogs, journal articles, or important recall or safety information). Communication may range from general advice, such as encouragement to get the seasonal influenza immunization, to broad-ranging notifications about public health emergencies.

Beyond mere ease and scope, education via social media may be preferable for patients who have moved beyond the model of acquiring knowledge from a medical professional in the clinical workspace. Some patients prefer “disintermediation,” or gathering information directly from online sources. Recognizing that not every such source is trustworthy, physicians can encourage a middle ground. The so-called “apomediary model” allows physicians to guide patients to reliable information that they can access independently before turning to their physician for consultation and clarification.²²

Social media can also be used to implement advocacy initiatives to enhance public health. Examples include appeals to contact elected officials to advocate for policies and legislation that improve the health of children and their families and the promotion of public and community health via immunization campaigns. Recent polls suggest that fully two-thirds of Americans believe that social media are important for getting elected officials to pay attention to issues and that social media platforms are “at least somewhat important for creating sustained movements for social change.”²³

These social media opportunities are widely accessible to, and can be used to influence, public opinion, as recent interference in US elections attests.²⁴ Opponents of certain medical recommendations or public policies can provide conflicting or inaccurate information that may trigger discussions and debate.²⁵ Although currently only 9% of physicians use social media to comment on posts or participate in group discussions or online chats,¹⁸ physicians participating in social media may help prevent misconceptions from dominating online discussions.

Social media can help patients and families understand the range of expertise of a given practice and learn of additional resources in their community. In some cases, with patient authorization, social media can permit the transmission of personalized information (such as appointment or medication refill reminders), which can help improve adherence to treatment regimens.²⁶

Social media can also benefit patients by increasing awareness of community health needs. For instance, appeals for voluntary blood donation on social media can help to mitigate acute shortages. Organ donation recruitment has also increased in response to Facebook appeals.²⁷

Patient Empowerment

Social media have made the world smaller and have provided a global community to support patients with special needs and chronic health conditions as well as individuals who may be experiencing social isolation or discrimination. Peer-to-peer health care affords patients the opportunities to develop a support network, share experiences, and learn about new opportunities for research or treatment.²⁸ Many private companies,^29,30 including some focused on pediatrics,³¹ have such online communities for patients to connect with each other and work to systematically improve their health outcomes. Facebook³² and Twitter³³ have both allowed the creation of “supportive disease subcultures.” As one patient noted, “The internet has made our small disease larger and we are able to educate many more people now.”³⁴

In addition to fostering community and providing education, social media also hold the promise of improved diagnostics. Patients (and their advocates) can now share reports online of unusual symptom complexes that have stumped local physicians. These disclosures may be similar to diagnostic dilemmas among other patients and provide physicians with diagnostic clues.²⁶ Identifying groups of patients with a common condition can also spur further research on that disease’s pathophysiology and treatment.^35,36 This process has been likened to crowdsourcing, or bringing together the collective wisdom of diverse groups of patients and physicians.²²

Increased Patient Access

In regions with inadequate infrastructure to support optimal telemedicine, social media sites, such as Facebook, have been used instead.³⁷ Ideally, social media should not replace face-to-face encounters with physicians, but social media may supplement those encounters by creating virtual clinics, thus allowing a more rapid response to an urgent health care situation.

Clinical Research and Recruitment

Social media can expand research opportunities and connect researchers to peers, novel ideas, and potential human subjects or study participants. Social media tools are not only used to broadcast ideas; they can also be used to search for information, opinions, and ideas or simply to listen. Such tools allow researchers to contact patients with rare conditions and locate research subjects lost to follow-up as well as reach out to younger patients who might otherwise not learn about or become interested in participating in research.³⁸ Social media can inform researchers about patient responses to treatment and adverse drug effects and trigger modifications in treatment plans.³⁹

Use of social media in research is ethically complex, however. Whereas most research studies are developed by experienced researchers and approved by their organization’s institutional review board or human subjects committee, studies involving social media may bypass this review process because social media use is seen as public, thus calling into question whether it is necessary to obtain the participants’ informed consent.⁴⁰ Protection of human subjects is critical, however, and researchers who use social media in their studies should continue to follow the policies of their institution’s ethics committee or institutional review board. Investigators should disclose that a research study is being considered, be clear about the risks and benefits of the research, and, if possible, obtain consent from potential study subjects before proceeding with the study rather than merely lurking to obtain data.³⁸ Without such transparency, potential subjects might post personal items on social media that they would have kept private if they had been aware of the research taking place (from a practical viewpoint, there should be no expectation of privacy for information voluntarily posted online⁴¹).

Identification of and Counseling for High-risk Behaviors

As digital natives, most pediatric patients expect to communicate through social media. Adolescents, in particular, spend a great deal of time online⁴² and may feel more at ease sharing personal information in that context rather than in face-to-face conversation. Teenagers may also be more open to health-related messages and advice communicated through social media.⁴³ Social media can, therefore, be used to identify self-disclosed high-risk behaviors and explore ways of mitigating risk and accessing appropriate resources.^44,45 Such conversations are more straightforward when the pediatric patient has specifically granted the physician access to material posted on social media, although such communication itself raises issues related to blurring of appropriate boundaries (as discussed below). Greater discretion is required regarding social media information that is in the public domain.

Acquisition of Important Information

In rare cases, social media have been used to obtain specific and important information relevant to patient care. For instance, the authors of one case study describe how a Facebook search allowed a medical team to locate the family of a patient with amnesia.⁴⁶ Taken too far, however, indiscriminate queries (sometimes referred to as “patient-targeted googling”) present significant ethical complexity and concerns, which will be addressed below.

Nurturing Compassion Through Narrative

Medical practice today can be challenging and exhausting. Sharing inspirational anonymized patient stories can nurture compassion among health professionals, reminding them about why they chose the profession in the first place.⁴⁷ Humor has also been shown to prevent burnout,⁴⁸ and so-called “collective venting” can be cathartic.⁴⁹

By the same token, because social media are publicly accessible, insufficiently anonymized patient narratives (and even sufficiently anonymized ones that are written in a negative venting tone) may reflect negatively on the authoring physician. Even deidentified patient information holds the potential to be hurtful, both to that specific patient and to other patients who might fear that their private information could be shared in a similar manner. Denigrating language is inherently personal and reflects poorly on the physician and the profession of medicine itself.⁵⁰ There is evidence that sharing dehumanizing narratives and language begins early in medical training, with medical students shown to frequently express themselves in this way online.⁵¹ A majority of medical schools now report unprofessional online conduct.⁴⁸ For precisely this reason, schools are starting to put forth policies regarding social media expression.⁵²

Even when details are deidentified and a patient narrative is expressed in respectful terms, a majority of medical educators still believe that discussing it on social media requires explicit patient consent.^53,54 Given the abundance of readily accessible information, nondisclosure of names, addresses, or other traditionally verifiable information is not a guarantee that one’s identity will remain anonymous.

Inappropriate Self-Disclosure

Despite the potential benefits of social media, significant risks remain. One important risk is inappropriate personal self-disclosure by physicians online, which can negatively impact a physician-patient relationship or one’s employment. Even if an inappropriate posting on social media is entirely unrelated to one’s medical practice, it nevertheless reflects on one’s professionalism.⁵⁵ As the American Medical Association Code of Ethics states, “The ethical obligations of physicians are not suspended when a physician assumes a position that does not directly involve patient care.”⁵⁶ This includes online disclosures, and all professionals, not just physicians, are judged by how they comport themselves online.⁵⁷

By its very nature, social media invite inappropriate posting because “anonymity can breed disinhibition.”⁵⁸ Some posts are clearly inappropriate, such as selfies of grinning clinicians posing with weapons during a humanitarian mission⁵⁹ or uncivil microblog posts (eg, tweets).⁶⁰ Some posts may be more ambiguous but in the absence of nonverbal cues can easily be misinterpreted. Comments addressing patients, as well as those regarding one’s employer or clinical setting, reflect negatively on both the subject, the poster, and the professional, potentially leading to disciplinary action or even termination. Tweeting or posting from work, especially if excessive, may be perceived as not paying attention to clinical duties and may violate institutional policy.⁶¹

There are permanent implications to such disclosure, too, because despite subsequent attempts to purge one’s online profile, what goes online stays online. Unflattering social media content may affect a clinician’s future employment opportunities.⁶² Despite this risk, one study revealed that only one-third of medical students set their Facebook pages to private.⁶³ This may be significant given that some institutions may consider social media and the content of postings in promotion and tenure.⁶⁴

From a specifically pediatric perspective, poor digital citizenship undermines a pediatrician’s important duty to serve as a role model for patients and families using social media. The AAP has previously encouraged pediatricians “to increase their knowledge of digital technology so that they can have a more educated frame of reference for the tools their patients and families are using, which will aid in providing timely anticipatory media guidance as well as diagnosing media-related issues should they arise.”⁴²

To prevent inappropriate disclosure, it is best to pause before posting by taking a moment to reflect on the potential impact of a post should it be seen by one’s colleagues, patients, or families. It is in a physician’s best interest to consider separating their personal and professional social media presence and efforts. The professional platforms can have public settings and content for education and advocacy, and the personal platforms can be set to private. Restricting access to one’s private social media accounts may be beneficial in light of reports of a few patients using social media to stalk their physicians.⁶⁵ It is also wise to remember that online postings are discoverable in legal proceedings. Facebook’s privacy policy clearly states that they may “access, preserve and share your information…in response to a legal request (like a search warrant, court order, or subpoena) if [they] have a good faith belief that the law requires [them] to do so.”⁶⁶

Simply having a presence on Facebook could permit a process called tagging. Tagging involves a facial recognition algorithm that identifies an individual from photographs posted online. An unprofessional pose from a friend’s party could end up appearing on one’s own Facebook feed. To prevent being tagged, Facebook settings can be modified.

Unfortunately, tagging may also occur from many social media platforms through comments and tweets. Even if a pediatrician were to elect not to have a social media presence at all, others are able to post information and photographs about the physician, including videos recorded and photographs taken without the physician’s knowledge or consent. For this reason, routinely monitoring and curating one’s online presence is recommended. Pediatricians can get a sense of their online “footprint,” as well as their online “fingerprint” generated by their online work and advocacy, by regularly searching their online presence.

Blurring Relationship Boundaries

As stated above, a recent poll found that most millennials and nearly half of adults want to follow or connect with a health care provider on social media.¹¹ In one study, nearly 20% of adult patients attempted to communicate with their physician through Facebook.⁶⁷ Accepting this type of relationship can be problematic, however. “Friending” blurs the boundaries of the professional relationship, not only heightening the possibility of inappropriate physician self-disclosure⁶⁸ but also introducing a level of mutuality that can undermine a patient’s privacy and patient-physician interactions. For example, if a physician discovers something on a patient’s Facebook page that is not consistent with the patient’s self-report, should the physician raise this issue with the patient directly? How might that discussion impact the patient’s perception of privacy? Accepting a friend request may also raise expectations of off-duty availability or that clinical questions will be answered via less secure social media.

Recognizing this ethical challenge, 75% of physicians decline friend requests from patients¹⁸ and only 5% have ever initiated one.⁷ More than 80% of medical educators believe that it is never or rarely acceptable to become social networking friends with patients.^53,54 The American College of Physicians,⁶⁹ American Medical Association,²⁶ British Medical Association,⁷⁰ and Federation of State Medical Boards¹⁵ all discourage accepting (and certainly initiating) friend requests with current or former patients, although the recent American College of Obstetrics and Gynecology statement permits some measure of discretion in this area.⁷¹ Additionally, some employers prohibit friending patients.

This question becomes more complex in a pediatric context given the reliance of pediatric patients on social media as a forum for expression and communication. It is generally inadvisable for pediatricians to accept friend requests from current patients (and certainly to initiate them). However, declining a friend request might seem like it could compromise communication or even give offense to the requestor. If so, the physician should meet face to face and talk with the requestor and discuss with them why a dual relationship is not wise or in the patient’s best interest. An alternative way to maintain communication would be to redirect the requestor to the pediatrician’s professional site or a separate platform on which no other personal or professional posts are made.

Unlike in other specialties, pediatric patients “age out” of the therapeutic relationship with their physicians. Social media present a way for pediatricians to stay in contact with former patients. When not doing so is felt to represent a significant mutual loss, it is up to the pediatrician’s discretion whether to remain in touch with former patients via social media.

Becoming social media friends with the parent of one’s patient is also not without ethical complexity given that that the parent is usually tasked with making decisions for the patient and may be unduly influenced in the decision-making by the friend relationship. Such friending may also create an avenue of communication that could lead to an unintended disclosure of identifiable health information to the patient and family. Ethical issues in this friendship may become even more complicated if the social media relationship continues once the pediatric patient reaches the age of majority and the parent is no longer responsible for making decisions for the patient. It is, therefore, problematic to engage in social media contact with parents of patients if social media are the only basis of the relationship.

In some instances, however (such as small communities or areas in which the physician and the parent have other mutual interests), such contact can be a normal expression of friendship. Once again, it is imperative to maintain proper boundaries by reserving clinical discussions for separate forums.

Up to this point the discussion has been focused on established patients attempting to become friends on Facebook, but the reverse process (ie, friends on Facebook becoming patients) has also occurred. Generally speaking, a physician-patient relationship consists of a patient seeking assistance in a health-related matter and the physician agreeing to undertake diagnosis and treatment.⁷² This does not require an in-person meeting and could conceivably occur online. Although specific requirements vary from state to state, it is likely that, regarding social media, the broadest notion of what constitutes a physician-patient relationship could be adopted.⁷¹ Clinical questions posed through social media could, therefore, potentially create a professional obligation with documentation and follow-up requirements similar to those of an office visit as well as a potential liability risk, which may not be covered by malpractice insurance. What began as a kind-spirited attempt to offer general assistance could ultimately result in medicolegal obligations (and liability), underscoring the need for physicians to be extremely cautious about health care communications with personal friends and about posting anything that could be construed as representing a diagnosis or treatment.

Conflict of Interest

As noted above, social media provides valuable opportunities for health advocacy and patient education but can also raise the possibility of conflict of interest. For example, 11% of physician-written health care blogs include named products.⁷³ Tweets may involve “suspect promotions.”⁷⁴ Because using social media is relatively easy, many more clinicians may be tempted to consider a paid promotion without understanding the ethical and legal implications of such conflicts of interest if promulgated without appropriate disclosure. Such arrangements have also occurred in traditional media and violate journalistic and educational ethics (not to mention Federal Trade Commission regulations⁷⁵) if not publicly disclosed. Even an individual message on Twitter (a tweet) allows pediatricians to include a conflict of interest statement despite the 280-character limit. Employment relationships should be revealed in online profiles and biographies whenever possible.

Confidentiality

The Hippocratic Oath states, “I will respect the privacy of my patients, for their problems are not disclosed to me that the world may know.” The HIPAA Privacy Rule, which covers “individually identifiable health information” based on 18 possible identifiers, including “full-face photographic images (and any comparable images),”⁷⁶ also applies to social media, as do state privacy laws.⁵⁰ Physicians have an ethical and legal obligation to keep protected health information confidential. Despite this duty, there have been reports of physicians sharing details about patients or inadequately deidentifying facts and photographs^77,78 because even blocked-out faces in photographs may still be identifiable. Social media postings that reference a medical encounter with a specific time and date could be seen by patients and damage trust in the physician-patient relationship and likely violate HIPAA.

To facilitate appropriate use of such digital communication, several legal issues need to be addressed. These include licensure across state lines or international borders,⁷⁹ adequate malpractice insurance coverage, reimbursement for telecare (if available), and maintaining confidentiality (because HIPAA also applies to social media communications). When a physician is communicating about health care or rendering advice directly to patients or families, HIPAA compliance requires the use of a secure site with encryption. Removing any identifiable patient details preserves the patient’s anonymity and underscores the practice’s compliance with state and federal law. The option of 24/7 patient communication also demands appropriate attention to the physician’s own work-life balance.

Impact on Professional Reputation

A significant number of patients use online rating sites to choose their physician. Given the impact of one’s online presence on professional reputation, it is advisable for physicians to monitor the status of their online identity.⁸⁰ Although most physician ratings are positive,^71,81,82 there have been reports of ratings sabotage (ie, fabricated negative reviews) by competing professionals or disgruntled patients.⁸¹ Given the anonymity of many reviews, such a situation may be difficult to remedy.⁸³

Responding thoughtfully and positively to critical reviews can be helpful. Physicians must remember that patients can post whatever they want about a situation, but the physician remains bound by confidentiality obligations. Responses must always be generalized, not specific to an individual. It is important to recall that HIPAA regulations prevent physicians from disclosing any protected health information about the patient, including acknowledging that the person is or was a physician’s patient.

Unfair or potentially fabricated reviews can be reported to the rating Web site or, depending on one’s work context, to the public relations division of one’s employer. Another solution can be to optimize search engine results so that one’s own practice site comes up first, before any potentially negative reviews. Encouraging more patients to review the practice online will likely provide a balanced and generally positive view.

The American Medical Association Code of Ethics not only mandates ethical behavior on the part of physicians but also requires physicians to report unprofessional behavior on the part of their peers.⁵⁶ This mandate extends to the unprofessional use of social media by other physicians. Difficulties inherent in such reporting, which include having to determine what content crosses the line and not wanting a colleague to get in trouble, may explain why few other codes include this requirement, although it is frequently included in social media guidelines written by hospitals, academic centers, and other health care employers.

Given these concerns, pediatric practices and other health care organizations may wish to formulate social media policies for their employees, which can clarify expectations, provide valuable tools and information, and ultimately protect patient confidentiality and privacy. These policies may also include expectations and standards for communicating with patients through social media, such as response time and documentation.⁸⁴

Inappropriate Acquisition of Information via Social Media

As noted above, there may be instances in which social media can be used to obtain specific and relevant information about patients (such as the example of locating the family of a patient with amnesia through Facebook). This practice has become increasingly common. In one study, 1 of 6 pediatric trainees had conducted Internet or social media searches for more information about a patient, and a similar percentage of faculty believed they would do so if that might help in patient care.⁸⁵

This practice could, however, expand to indiscriminate searches for information about patients.⁸⁶ So-called patient-targeted googling can be motivated by a genuine desire to understand more about one’s patients and perhaps gauge their adherence to treatment plans. But it can also stem from “curiosity, voyeurism, and habit.”¹⁵

Patient-targeted googling can generate 2 types of information, the first coming from the patient directly. Physicians may understandably be unsure how to use information gathered from social media sources.⁵⁴ For example, if an adolescent patient denies drinking alcohol but has posted on social media about drinking to excess, should the pediatrician confront the patient with this newfound information? To do so might compromise trust, but not doing so might preclude thoughtful intervention.

The other type of information patient-targeted googling may generate comes from third-party sources, ranging from news articles to posts from someone other than the patient. Such non–user-generated Internet content presents a different set of challenges. First, there is no guarantee that the information is accurate. Even if it is, acquiring knowledge about a patient that the patient did not directly provide (and may not be aware that the physician is in possession of) threatens to compromise trust. Uncertainty remains as to whether to reveal one’s knowledge of this information to the patient, either to confirm its veracity or engage in discussion about its content.

Before pediatricians engage in patient-targeted googling, it is important to identify the information they are seeking to acquire and determine if it is of sufficient importance to justify such a search. Clinton et al⁸⁷ suggest 6 questions to consider before engaging in patient-targeted googling:

• “Why do I want to conduct this search?”

• “Would my search advance or compromise the treatment?”

• “Should I obtain informed consent from the patient prior to searching?”

• “Should I share the results of the search with the patient?”

• “Should I document the findings of the search in the medical record?”

• “How do I monitor my motivations and the ongoing risk-benefit profile for searching?”⁸⁷

It is generally advisable to disclose any relevant patient information discovered through the Internet or social media to the patient so that everyone is working with the same set of facts. This disclosure is especially important before entering any information gleaned through that route into the patient record.⁸⁸