Four years passed between the launch of the social networking site Facebook in February 2004 (2) and its first mention in the medical literature in February 2008 (3), when researchers studied the dangers of social networking sites for sexual solicitation of underage youth. In another 2008 study, at the University of Florida (4), researchers found that 64.3% of medical students and 12.8% of residents had Facebook accounts. A random subset of 10 profiles found 7 with photographs involving alcohol and 3 with unprofessional content such as drunkenness, overt sexuality, foul language, and patient-privacy violations. In a recent article (5), medical school deans were surveyed on professionalism issues involving medical student use of the Internet. Of the 78 deans surveyed, 60% reported incidents of students posting unprofessional content online, and 13% reported students violating patient confidentiality online within the previous year. Deans also gave examples of medical students who requested inappropriate friendships with patients on Facebook, posted online content suggesting intoxication or illicit substance use, and used disparaging language about a course, professor, or classmate online. Only 19% reported having a committee or task force responsible for addressing student-posted online content. In another recent article (6), a medical intern discussed his conflict in allowing a former patient to add him as a friend on Facebook, as well as his relief when he found that the main reason for the patient's reaching out was to seek advice on applying to medical school.
In addition to social networking sites, blogs are popular in the healthcare community. In 2008, researchers examined 271 blogs with health content written by a doctor or nurse (7). Over half had identifiable authors, and 22% displayed a photo of the author. Of all the blogs studied, 18% described patients in a negative light, including insulting comments, and 30% featured negative comments about the health professions. Another study (8) found 331 English-language medical blogs that had updated content during the previous month. Bloggers with available contact information were surveyed, and 80 responded. The majority had been blogging for over 2 years, and 75% wrote under their real names. They were also active writers outside the Internet, with 54% having published a scientific paper, 44% having published a book or book chapter, and 41% having published a newspaper article.
In this communication, we will confine our discussion to professional boundaries as they apply in a mental health setting, because these boundaries are generally more stringent than in other medical settings. We will also attempt to delineate which issues are more properly classified as clinical or professionalism dilemmas.
Social Networking Sites, Blogs, and Search Engines
Social networking sites, such as Facebook and MySpace, have been widely adopted as a means of communication. In early 2009, Facebook had over 150 million users, with those over the age of 30 representing the fastest-growing demographic (9). Because of its wide popularity, we will use Facebook to illustrate concerns about social networking sites in general.
A number of aspects of Facebook may compromise the privacy of patients as well as the privacy of psychiatrists. First, a patient may send a "friend request" to a psychiatrist. A friend request, if accepted, allows each party full access to the other's Facebook profiles, including updates, photos, interest groups, and other content. Most users see Facebook as a forum for self-expression through posts, affiliations, and photos. These items are, of course, more personally revealing than what is generally disclosed in a treatment relationship. Also, Facebook users may post photos and "tag" or label another Facebook user by name without the knowledge or consent of the individual in the picture. Facebook users may discover that they are tagged in a professionally unbecoming photograph long after numerous others have seen it.
Another concern with Facebook is the potential for breaching patient privacy in a nonclinical setting. A status post about a challenging patient may contain enough recognizable information to compromise confidentiality. Although disguised information about a patient is often used in case reports, the intent there is educational. Posting information about patients as a status update is generally self-serving in its intent. It is possible to use privacy settings to control who has access to a Facebook page. However, as many as 80% of Facebook users do not actively manage their privacy settings, meaning that the content of their profile is available to any users of Facebook, whether they are an accepted friend or not (10). In the Florida study noted earlier (4), only 37.5% of medical students and residents used their privacy settings. Anyone, including patients, colleagues, and program directors, for example, may access a profile. Currently, there are no means to know who has access to and who has viewed a profile. Finally, even with most privacy settings activated, many individuals leave their profile picture accessible for searches. One ill-chosen photograph can have unforeseen consequences. Moreover, Facebook users commonly list their sexual orientation, marital status, religion, age, hometown, and political affiliation in their profiles, most of which is information not typically shared with patients.
Related to social networking sites are online dating sites (match.com, eHarmony, jdate, etc.). A user creates a profile, usually with a photograph, to attract potential romantic partners. Unlike social networking sites, where real names must be used, an online dating site user may create a name. A user's profile, complete with photo and personal information, will appear along with others meeting the same search criteria. Users may be contacted through the site, but direct contact information such as an e-mail address or phone number is not revealed. Users may choose to ignore or respond to messages sent through the site. A patient who uses online dating sites may recognize a psychiatrist using the service and may learn significant amounts of sensitive information that the psychiatrist has unintentionally revealed.
Although one can never be absolutely certain who has viewed a social networking or online dating profile, this problem is more pervasive in blogs. The term weblog was first coined in 1997 by Jorn Barger, editor of the website Robot Wisdom, and this word was eventually shorted to "blog" (11). Key differences between social networking and blogging make blogging more problematic for psychiatrists. The first is accessibility. Although limiting a Facebook profile so that it is only seen by those whom one knows and trusts is relatively common and simple, such limiting happens less often and can be more complex in a blog, because bloggers tend to enjoy sharing ideas and web links with others. In a 2008 survey of medical bloggers (8), 99% noted that they received attention from their blog from other bloggers. Also, 74% noted that their motivation for blogging was "to share practical knowledge and skills," and 56% professed a desire "to influence the way other people think."
Furthermore, whereas a Facebook user is more likely to journal quick thoughts or statements on a profile, a blogger is more prone to write many paragraphs, or even many different entries over time, focusing on a single topic. This output creates more opportunities for a psychiatrist-blogger to post potentially problematic entries, such as breaking patient confidentiality or writing about unprofessional activities in which he or she might engage.
A blog also leaves a more permanent footprint on the Internet. Currently, there is no way to quickly search through someone's Facebook profile to see what he or she has written in the past. Readers have to scroll to the bottom of a profile and click on "older posts" ad nauseum until they find something written some time ago. The more active the user is, the more times one needs to click on "older posts." By contrast, most blogs are searchable by date or keywords so that a blog entry written a year ago can be quickly accessed. Furthermore, another website can post a link back to a psychiatrist's blog or easily copy and paste a psychiatrist's blog entry, referencing the psychiatrist as the original author. In a situation such as this, even if the psychiatrist later decides to delete a blog entry, he or she cannot control what another blogger has posted on a different website. One must particularly be aware that the anti-psychiatry movement is also active in the blogosphere.
Much of the redefinition of privacy has been with the full cooperation of the people who choose to blog or to post on social networking sites. The ubiquity of search engines has created a related but different problem in the realm of privacy. Even those who do not wish to indulge in self-disclosures on a social networking site or blog find that all kinds of details about their private lives are available to those who wish to search for it.
Psychoanalysts and psychotherapists have long emphasized the asymmetry of disclosure in the doctor—patient relationship and operated with the assumption that little knowledge of their personal lives would be shared with the patient. However, prospective patients now routinely "Google" psychiatrists before agreeing to see them, and patients can find out a great deal of information about a psychiatrist's family through search engines. As part of their research or "due diligence" in deciding whether to see a particular psychiatrist, patients can also access various sites that rate doctors. On these sites are many statements by disgruntled patients. Some of these comments, of course, accurately portray deficiencies in the psychiatrist, but other statements may reflect rage at not being prescribed a controlled substance or negative transferences that really say very little about the psychiatrist's actual competence.
There was a time when clinicians would go to meetings or other out-of-town events without informing their patients of their destination; rather, they would simply say that they would be away for a week, without going into detail. Search engines have also radically changed this form of anonymity. Patients can now find out details about where their psychotherapists may be speaking or staying and what they might be doing.
There is an extraordinary availability of public documents through search engines. What a psychiatrist paid for his or her home can be learned by accessing property-tax documents. If an unmarried therapist is living with someone, this information is also accessible. The extent of political contributions that a therapist made can be discovered in seconds. Patients who wish to trace the genealogy of their therapist can find out a great deal about family histories. The psychiatrist practicing psychotherapy today is likely to feel violated, invaded, and exposed. If a psychotherapist brings up the patient's intrusiveness as a therapeutic issue, however, the patient may well respond, "This is public information available to anyone. I am not invading your privacy." This retort may stymie the therapist and lead to a variation of a therapeutic impasse in the treatment.
The differences between search engines and social networking or blogging are significant. The doctor has no choice regarding participation. Also, search engines are really without bounds as compared with the limits of blogging or social networking sites. Everything about a clinician is fair game. One of the major difficulties with search engines is that there may be misinformation on such things as websites that rate doctors, the doctor's private life, or data about family members. Many people may have the same name, and patients or prospective patients can read highly negative (or positive) information about someone they presume to be their psychiatrist when in fact they are reading about somebody else.
Search engines have truly altered the conceptual framework of privacy, anonymity, and self-disclosure in the clinical setting. Psychiatrists can no longer assume that they are a "blank screen" to patients, but must now assume that patients know a good deal about them and must rethink the stance that their private life is beyond reach of the patient.
The professional-boundary questions that have risen in the expanded world of cyberspace generally fall into three areas: ethical concerns, professionalism issues, and clinical dilemmas. Only the first category involves true boundary problems as they are generally defined.
Beauchamp (12) has outlined four clusters of moral principles that are the underpinnings of biomedical ethics codes: respect for autonomy (an acknowledgment that patients are free to make decisions), nonmaleficence (the fundamental principle of avoiding the potential for harm), beneficence (a consideration of the equation of weighing benefits versus risks), and justice (fairness in how burdens and benefits are distributed). Some of the situations described in the previous section are clearly violations of ethics. For example, breaches of confidentiality by writing about patients on a blog or social networking system have the potential to harm patients or patients' families. Even if a name is not mentioned but identifying features have been provided, some readers may be able to detect the identity of the patient. Although several guidelines suggest ways to protect patient confidentiality in writing scientific papers (13), the off-the-cuff "venting" that occurs on blogs and social network sites generally does not take these factors into account because the content is not designed for publication in a journal.
Another concern involves dual relationships. If a patient is a "Facebook friend" of a doctor, or vice versa, an expectation follows from that designation that may compromise the boundaries of the doctor—patient relationship. Such relationships may also jeopardize the ethics principle of nonmaleficence in that the capacity to make good use of a psychiatrist in the patient role may be more difficult if one is also a friend. Inherent in the caution against dual relationships is the idea that psychiatrists optimize the treatment setting by making it clear to the patient that the psychiatrist will never be anything other than a treater, even after termination. Patients then know that there are no consequences in any outside relationship if they reveal something shameful or painful to their psychiatrist (14). One can find definitions of a "Facebook friend" online that restrict it to relationships that only occur in cyberspace at the present time, even if they started with a real-life connection. However, this difference breaks down when the "friend" is a patient, because there are therapeutic contacts that do not take place in cyberspace. Moreover, patients may have no idea about this definition and take it to mean that they are friends with their psychiatrist in the literal sense of the word.
Another ethical principle, respect for autonomy, may come to the fore in situations where patients "google" their psychiatrists. Psychiatrists who are accustomed to traditional anonymity and privacy and feel violated may issue an edict that patients should not intrude into the psychiatrist's private life and must avoid accessing information available on the Internet. The ethical principle of respect for autonomy makes it clear that psychiatrists should not place constraints on patients' freedom to pursue public information. The term boundary violation is not applicable to the patient who investigates the doctor online through public information. Patients have no ethics code and therefore are not violating professional boundaries when they seek out information about their doctor. Clearly, this ethical principle is at times overridden by concerns about danger to self or others. However, operating a search engine does not fall into that category. Psychiatrists who feel that their personal space is being intruded by the patient must deal with this matter as a countertransference issue for supervision, consultation, or personal treatment while exploring with patients the meanings of their curiosity. In other words, psychiatrists must resolve their reaction within themselves. They cannot block certain aspects of their lives from their patients, and they must learn to adapt to the new world that cyberspace has created.
Some of the phenomena associated with blogs and social networking sites fall into the area of professionalism. Both medical educators and psychiatric educators have vastly increased their emphasis on professionalism as part of training to the point where it is now considered one of the core competencies in training physicians and psychiatrists (15).
Physician-educators have variously defined professionalism. One useful definition is from the American Board of Internal Medicine (16), which suggests that professionalism "requires the physician to serve the interests of the patient above his or her self-interest. Professionalism aspires to altruism, accountability, excellence, duty, honor, integrity, and respect for others." Some prefer to think of professionalism as simply a matter of how one behaves with patients, colleagues, and others in public places when no one is watching.
With the expansion of the Internet, all physicians and physicians-in-training must remember that they are viewed by others as in that professional role to some extent whenever they are in public, even if they are not at work. The emphasis on professionalism education now raises the standards for the behavior of graduate psychiatrists and educators who are role models for trainees, whether residents or medical students. Photos of a psychiatrist or psychiatric resident that suggest intoxication or illicit substance use, for example, may appear on social networking sites or blogs. Although the activity may have occurred off duty, the effect on those who see such photos may be devastating nevertheless. Similarly, making negative comments about colleagues on a blog or a social networking site becomes equivalent to publicly denouncing a colleague. Even if one is talking about patients generically, one can appear unprofessional if a comment disparages patients. Although no ethics code is being violated in such instances, standards of professionalism certainly are.
Answering the standard questions from Facebook's initial page can reveal a good deal of information about a doctor, as can providing the information required by online dating sites. These are bits of personal data that one does not generally convey to patients in treatment, but patients may come across this information online. Even though it is not a matter of ethics or a boundary violation to make this demographic information available, doing so can certainly create a clinical dilemma for the treating clinician to explore. Patients may have reactions that are useful to examine in the psychotherapeutic context. Excessive self-disclosure is generally regarded as a potential boundary violation in the treatment relationship (14, 17), but disclosure of this information on a social networking site is not intended to be in a treatment context. Hence, psychiatrists cannot be held accountable for unethical behavior by virtue of listing this information on a site that has privacy-detection built in. Even if the psychiatrist has failed to use the privacy option, the information is not conveyed within the doctor—patient relationship.
Another phenomenon that commonly occurs is often referred to as "extratherapeutic contacts." A psychiatrist may bump into a patient at a restaurant or concert and must navigate this rather awkward situation with respect for the patient's privacy. This situation is not an ethical matter (unless a confidentiality breach occurs) because it was not planned but, rather, accidental. Planning a meeting with the patient outside the office, such as for lunch or a walk in the park, does violate the usual boundaries of the doctor—patient relationship and is cause for concern. In an analogous manner, extratherapeutic contacts that are inadvertent, such as on an online dating service or social networking site, are difficult but not unethical. Because of the complicated nature of those contacts, psychiatrists may wish to use a consultant who has clinical experience to help them negotiate the optimal management of such situations. When a patient encounters erroneous information about a doctor, either because someone else has the same name as the psychiatrist or simply because of errors in reporting the facts, this too is a dilemma for the doctor to work out clinically. If a distortion of the physician's prescribing practices appears on a website that rates doctors, the physician can explore the patient's reaction and clarify what the actual prescribing practices are.
One other situation that arises in discussions of the new world of boundaries involves whether it is ethical for a psychiatrist to look up a patient on the Internet. Some (18) have suggested that seeking this information might violate doctor—patient boundaries. However, just as the information about the doctor is public, so is information on the Internet about patients. It is customary when evaluating patients to often get collateral sources of data from other individuals who know the patient, usually with the permission of the patient. The treating psychiatrist may want to ask permission of the patient but can certainly access public information without a release if it is thought to be essential or helpful for the treatment.
Useful recommendations for physicians and medical students are now appearing in the literature (5, 19, 20). Recently, there has been an online movement to establish a code of ethics among healthcare bloggers. The Healthcare Blogger Code of Ethics (21) focuses on five qualities to which medical bloggers should adhere: clearly representing their perspective, respecting confidentiality, announcing commercial disclosures, providing reliable information, and being courteous to others. Medical bloggers can submit their blogs for approval and, after a review process, be accepted as a blog that adheres to this ethical code.
Although this approach bears promise, psychiatry has a special set of requirements for professional boundaries. As yet, no detailed and systematic recommendations for psychiatrists and psychiatric residents have appeared. In Table 1, we outline some preliminary recommendations with the hope that they will lead to further discussion and debate and, ultimately, policy statements for the profession. Free speech and freedom to associate with whomever one wishes are protected by the Constitution, but ethics codes have always been more restrictive than Constitutional regulation. The recommendations we propose are in keeping with the wish to prevent harm or the potential for harm in treatment and interprofessional relationships. These principles should become a component of psychiatric resident education and be operationalized with other training policies. They can be equally useful to psychiatrists in practice in both institutional and noninstitutional settings.
TABLE 1.Recommended Guidelines for Maintaining Professional Boundaries Online
Completed Suicide: Attending Psychiatrist Reflection
A mail carrier in his mid-forties.
Two children, ages eight and four.
Nice-looking. Intelligent. Moral.
Just can't forgive himself for what he did in Iraq.
Nerves are jangled.
Wife leaves him for another man.
Suspicious of medications. Difficult to engage in therapy.
Ten years at VA; will I be seeing more patients like him?
Carol I. Ping Tsao, M.D., J.D.