Assessing self-harm: challenging.
I met with the patient and the best friend in the ED, and they hastened to reassure me that this was all a big misunderstanding, and could they please go home? This seemed not unreasonable to me - wasn't it plausible for a the patient to have reasoned that 80 mg of Paxil should be more helpful than 20 mg at a moment of acute distress? I did my best to assume the role of a reassuring physician, stating that everything would be all right.
Then, I presented the case to the attending psychiatrist, whom I later came to know as a thoughtful, dedicated, and compassionate physician. I proposed that contacting the outpatient psychiatrist and planning for close outpatient follow-up seemed like a reasonable plan. The attending psychiatrist, however, disagreed completely. He concluded, with quiet and unyielding conviction, that the circumstances were murky, that the patient's impulse control was questionable, and that, all things considered, a brief psychiatric admission would be the safer course of action than a discharge home. In the end, the patient was admitted on a psychiatric hold, over her objections, and I did not hear any more about her treatment course or outcome. Was the patient intending to cause serious harm to herself by overdose? Was she seeking transient relief from psychic pain, without intent to cause harm? I never found out.
Such cases are hardly unusual. But they highlight one of the psychiatrist's most important tasks: namely, the assessment of the potential for self-harm and suicidality in the face of murky and ambiguous evidence.
There is research evidence that as a profession, we could be doing a better job of assessing and documenting the risk of self-harm. In a much-noted study from Columbia University, Dr. Mary Bongiovi-Garcia and colleagues found that almost 19% of suicide attempters and nearly 30% of those with suicidal ideation were not identified as such in the chart at the time of psychiatric admission (J. Affect. Disord. 2009;115:183-8).
In the course of its deliberations, the DSM-5 Task Force has been considering the inclusion of two new diagnostic entities to clarify the assessment of self-harm, and suicidal ideation and intent. The first is nonsuicidal self-injury (NSSI), which has been proposed to describe individuals who have "engaged in intentional self-inflicted damage to the surface of his or her body ... with the expectation that the injury will lead to only minor or moderate physical harm."
The second is suicidal behavior disorder (SBD), which is intended to describe individuals who took action on at least one occasion within the previous 24 months with the expectation of ending their lives.
The final period for public comment on the DSM-5 draft proposals closed on June 15, 2012. Subsequently, it was announced on dsm5.org that the field trials for NSSI had not produced reliable results, but that both NSSI and SBD were still being considered and revised for inclusion in the manual. At press time, the American Psychiatric Association indicated that either or both diagnostic proposals could be included in the DSM-5.
Including SBD and/or NSSI on Axis I of the DSM-5 should compel clinicians to assess the lethal intent behind acts of self-harm in a more consistent and standardized manner. The two diagnoses would foreground key information about suicidal intentions in the diagnostic formulation. And their inclusion would likely spur clinicians to think carefully about safety issues as they develop their treatment plans. From the standpoint of diagnostic precision, the inclusion of SBD and / or NSSI might reduce the overdiagnosis of borderline personality disorder based solely on a history of past episodes of self-harm. As the authors of the SBD proposal point t out, those who have made serious suicide attempts car-F ry a surprisingly wide variety of Axis I and Axis II diagnoses (Am. J. Public Health 2010;100:2473-80), not just major depression and borderline personality.
The inclusion of these entities would carry with it some uncertainties and risks. For example, would a prior diagnosis of SBD or of NSSI affect insurability? Would SBD and NSSI, by virtue of their very definitions, constitute preexisting conditions? Would a past diagnosis of SBD dissuade clinicians from discharging patients from the emergency department even when that could be done safely?
It is difficult to predict the answers to these questions in advance of the publication of DSM-5, just as it was difficult to know how best to assess the self-injurious intent of that patient I saw in the ED. All things considered, though, Axis I entities that mandate careful evaluation of the intentions underlying self-harm would appear to advance the field of psychiatric nosology. I hope the diagnostic proposals are incorporated into the DSM-5.
MICHAEL BRODSKY, M.D.
DR. BRODSICY is the medical director of Bridges to Recovery in Pacific Palisades, Calif, as well as the director of psychiatric services at the Venice (Calif) Family Clinic.