Christopher R. Thomas, M.D.
Chair, AACAP Rights & Legal Matters Committee

The physician’s responsibility for keeping patient information confidential as described in the Hippocratic Oath has been considered essential over the ages for the therapeutic alliance between patient and physician. States recognize and define only certain limited exceptions, such as suicidal risk, when confidentiality can be breached by the physician without patient consent, usually in the interests of public safety. Recent legislation in the United States (Health Insurance Portability and Accountability Act, 1996), Canada (Personal Information Protection and Electronic Documents Act 2000), and the United Kingdom (Data Protection Act 1998) in response to electronic media and medical databases has sought to reinforce the privilege of patients in controlling their personal information and its disclosure.

A similar trend has occurred in expecting consent for information involving patients in academic and professional publications. In 1977, the New York Supreme Court found in favor of a patient suing a psychiatrist for publishing a disguised account of the patient’s treatment without consent, stopping the book from further distribution, and awarding damages for breach of fiduciary obligation (Doe v. Roe 1977). A separate complaint to the General Medical Council in the United Kingdom (Court 1995), when a psychiatric patient thought her identity had not been sufficiently disguised in a journal article, along with other concerns, prompted the International Committee of Medical Journal Editors (ICMJE) to issue revised guidelines for manuscripts (ICMJE 1997). The ICMJE Uniform Requirements for Manuscripts Submitted to Biomedical Journals: Writing and Editing for Biomedical Publication, revised again in 2007, state clearly:

Patients have a right to privacy that should not be infringed without informed consent.... Identifying details should be omitted if they are not essential. Complete anonymity is difficult to achieve, however, and informed consent should be obtained if there is any doubt....

Over 500 medical journals and publications, including major psychiatric publications, have signed onto these guidelines and established procedures to protect patient privilege and confidentiality in published reports. The AACAP News’ guidelines (Bernet 2007) state:

The author should make a statement at the end of the submission that (1) identifying information regarding patients has been changed or omitted and (2) every parent and patient (if able) have given permission for the publication of information about their case.”

There are exceptions to these standards, such as when the case represents a composite of two or more patients, or the information is already in the public domain. It would appear then that the requirement for consent in most patient publications, including case reports, is a settled matter in our evolving medical ethics until one looks at the psychoanalytic press. An ongoing debate persists among psychoanalysts as to how to properly handle ethical concerns for patient privacy in academic publications (Furlong 2006). The International Journal of Psycho-Analysis rejected any consent requirements (Gabbard 2000), opting instead that authors offer description of the means of disguise in efforts to protect patient identity.

The reluctance of psychoanalysts to join other medical writers in adopting these standards is the result of the centrality of case reports in psychoanalytic publications and psychoanalytic beliefs about obtaining patient consent. Unlike other studies, case reports cannot easily protect patient identity by averaging details over a group of individuals and psychiatric description typically requires accurate details that increase the likelihood of identification. Freud grappled with this problem at the conception of psychoanalysis, reporting in his preface to the Dora case (Freud 1905):

It is certain that the patients would never have spoken if it had occurred to them that their admissions might possibly be put to scientific uses….But in my opinion the physician has taken upon himself duties towards science as well… Thus it becomes the physician’s duty to publish what he believes he knows of the causes and structure of hysteria… so long as he can avoid causing direct personal injury to the single patient concerned.

Freud argued that patients would edit their thoughts in analysis if they knew what they revealed would potentially be published, undermining their treatment. Freud therefore published without patient consent deciding that the obligation to society and treatment of future patients outweighed the potential risk of harm to the individual patient. Given his eventual fame and awareness of his publications, it is hard to imagine that any later patient entering treatment with him did not expect to appear in a case report. Freud’s argument is essentially the same put forward by Levine and Stagno (2001), that request for consent to publish in some cases can be harmful to patients and would erode the valuable contribution of case reports in psychiatry. The first part of the argument, that asking consent would somehow be potentially harmful to the patient’s treatment by its effect on the patient-physician relationship (Howe 2003), rests on the principle of therapeutic privilege. It is a rarely used and controversial legal doctrine that permits a physician to withhold information if the physician determines that it would be harmful to the patient or interfere with the patient being able to decide about a recommended treatment (Canterbury v. Spence1972; Somerville 1984).

Therapeutic privilege is usually invoked only when the potential benefit obtained by withholding information is for the individual patient, not others. If asking for consent to publish is determined by the therapist to be harmful for an individual patient, then the ethical solution is clear – Don’t Publish. That is why the argument is usually coupled with the prospect of potential benefit to society, including future patients and trainees, from publishing case reports (Draper & Rogers 2005). This position holds that the ethical principle of potential benefit to others outweighs respect for the autonomy of the individual patient. It is carefully dissected by Kochiras (2006), who shows that neither of the ethical conditions necessary to support this argument is suitably met in this case, that there is a greater balance of benefit than harm or that there is no morally acceptable alternative.

Some have held the position that disguise alone satisfies the need to protect the patient’s privacy in case reports (Dugan 2008; Gabbard 2000, Draper & Rogers 2005). The problem is the information thought to be necessary in accurately describing psychiatric cases conflicts with removing enough detail to make the patient truly anonymous. There are no accepted guidelines on how to approach this task, and the therapist is left alone to decide if masking is sufficient (Kantrowitz 2004). Editors are in the position of trying to decide if the author’s masking is adequate, unaware as to what details might trigger recognition for a patient, the patient’s family, and acquaintances. The potential harm to patients if they or others discover and identify them in publications is substantial. The risk of patients coming across publications mentioning them has dramatically increased with the advent of electronic media. Just ask your patients if they have ever Googled your name or searched PubMed.

There are no uniform policies in place to educate psychoanalysts on the ethics of confidentiality (O’Neil 2007), with only 25 percent of societies and institutes providing copies of their code of ethics to candidates in a survey of United States, Canadian, and United Kingdom psychoanalytic organizations. The same survey found that 30 percent did not offer an ethics course that addressed the issue of confidentiality. Public surveys have found there is great concern about the use of patient information in medical research without consent (Robling et al. 2004), although the information from analytic patients on this issue is limited (Kantrowitz 2005). Masking or removing unnecessary information is still required even after obtaining a patient’s consent to publish in order to protect privacy to the greatest possible extent. That is why the AACAP News uses the conjunction “and” instead of “or” in the instructions for authors regarding consent and removing identifying information.

Other arguments raised by psychoanalysts are that patients have been upset even after giving consent for publication, and therefore obtaining consent does not prevent potential harm (Levine and Stagno 2001). That misses the point that patients’ rights should be respected especially in those situations that present potential harm. Just because a patient might still develop tardive dyskinesia does not mean that therapists should not bother to obtain informed consent prior to starting a neuroleptic medication.

Case reports remain an important means of communication and teaching about clinical issues in psychiatry. Current legal and professional standards expect that we protect patient privacy and privilege with respect to their medical information. Publication guidelines instructing authors to obtain consent from patients serve to protect patients and authors. The current AACAP News “Instructions to Authors” fulfill those standards and should not be weakened. To publish without seeking patient consent is deception on the part of the therapist and undermines the therapeutic relationship whether the patient discovers it or not. Legal precedent is clear, and authors are liable if they publish information about patients without consent. Interestingly, Helene Deutsch reported that Dora did learn that Freud had written about her and was proud of it (Lipton 1991). One wonders what would have happened to the course of psychoanalysis and obtaining consent to publish case reports if she had been upset and decide to sue Freud instead.