Review of Contemporary Risk Management Challenges in the Treatment of Patients With Borderline Personality Disorder

The term “risk management” serves as a shorthand to describe the potential for injury to patients in the course of medical care, as well as the vulnerability of clinicians to liability litigation or complaints to administrative agencies (e.g., medical boards) related to these injuries. Risk management topics are generally included in psychiatry residency training and continuing medical education as an afterthought; nevertheless, clinicians in practice often devote a significant amount of time and energy to addressing risk management concerns (1). These considerations are particularly relevant for clinicians working with patients who have primary or co-occurring personality disorder pathology; certain aspects of borderline personality disorder pathology will often make consideration of risk a fraught and confusing process. Fear of litigation and complaints to administrative organizations from patients and families can heighten clinician anxiety in decision making, and, at times, can dominate clinician concerns. Increased risk management concerns may be a key factor in the antipathy many clinicians have toward working with patients diagnosed as having borderline personality disorder (2). Although this aspect of clinicians’ attitudes about working with this patient population is not routinely described, it stands to reason that concerns about liability or complaints may contribute to avoidance of these patients, a pattern that has been well documented (3).

Risks Inherent in Treating Patients Who Have Borderline Personality Disorder

Inherent risks specifically associated with the treatment of patients with borderline personality disorder were detailed in the early years following the introduction of the diagnosis in the 1980 DSM-III. Since then, the following developments have had an impact on these risks: changes in general treatment patterns in psychiatry, including an exponential increase in the use of psychotropic medications; marked reduction in the average length of stay of psychiatric hospitalizations; changes in training opportunities; developments in personality disorder diagnostic classification; emergence of empirically validated treatments for the disorder; and an explosion of choices in ways to communicate with patients through various technologies. These phenomena have contributed to extra layers of complexity in these matters.

The dilemmas inherent in the clinician’s attempt to balance professional competence with the genuine legal concerns associated with the practice of psychiatry may be magnified in the treatment of patients with borderline personality disorder (4). The psychiatry risk management literature generally stresses the necessity of weighing informed management of medical-legal concerns with avoidance of burdensome defensive practice. Although most psychiatric risk management writing underscores certain prototypical predicaments, including evaluating suicidality or the need for psychiatric hospitalization, a survey of psychiatric risk management didactics will reveal the routine omission of material explicitly related to the management of patients with borderline personality disorder pathology as an area of focus.

Two reviews of legal difficulties encountered by psychiatrists underscore the pattern of complaints related to a number of key concerns: Simon and Shuman (5) reviewed data from psychiatric claims (from the Psychiatrists’ Program managed by Professional Risk Management Services, Inc.), by loss, for all states from 1998 to 2008. These data (5) listed the following causes of loss: incorrect treatment (38%), suicide or attempted suicide (17%), drug reaction (14%), incorrect diagnosis (7%), and unnecessary commitment (4%). Reich and Schatzberg (6) reviewed data related to malpractice claims and actions of regulatory agencies (e.g., state medical boards) from 1990 to 2009. They examined publicly available insurance data (malpractice frequency and type) and data from the National Practitioner Data Bank (from required reports of malpractice settlements and hospital discipline). They found that, among physicians, psychiatrists may be at lower risk than other medical specialties for malpractice litigation, but at higher risk for complaints at the state board level. Among the areas of most frequent clinical difficulty associated with regulatory agency involvement were accusations of negligence and incompetence, inappropriate prescribing, and inappropriate contact with patients. Severity of injury was associated with risk for malpractice suits, with suicide attempts and suicide as the leading causes, and incorrect diagnosis, incorrect or ineffective treatment (including medication errors), improper detention, and inappropriate sexual behavior as other high frequency causes.

The areas of concern in psychiatric malpractice litigation and complaints to medical boards outlined previously (e.g., suicide attempts and suicide, medication errors or adverse reactions, and complaints related to incorrect detention) will be familiar topics for clinicians working with patients diagnosed as having borderline personality disorder. Suicidality has been described as the behavioral specialty of patients with this disorder; research (7–9) has underscored this observation, as 70% percent of patients with borderline personality disorder have a lifetime history of attempted suicide, with an estimation of up to 10% for death by suicide. Although no medications have been approved for the treatment of symptoms, multiple studies (10) have reinforced that patients with borderline personality disorder are high users of almost all categories of psychiatric medications. Complicating this situation, clinical experience suggests that many treaters, including psychiatrists, are both first reluctant to make a diagnosis of borderline personality disorder and then to share this diagnosis with patients and families (11). Nevertheless, the aforementioned areas of risk apply in the treatment of patients with borderline pathology, both when these patients are first diagnosed with the disorder and subsequently when this diagnosis is shared with the patient and family, and in those situations when the borderline pathology is linked to increased risk management concerns but is not diagnosed or is diagnosed but not disclosed.

Several landmark articles (12–16) about risk and management in the treatment of patients with borderline personality disorder have been authored by Thomas Gutheil, a key figure in American forensic psychiatry. Gutheil’s articles have described risks, including challenges in the assessment of suicidality, boundary violations, and abandonment by the clinician.

Gutheil’s articles on risk in work with this patient population reflect their time of publication; in that era, many psychiatrists focused on psychotherapy, and prescription of medication did not dominate their work as it does now. In addition, psychoanalytic theory and practice still greatly influenced psychiatrists’ work with patients diagnosed as having borderline personality disorder, at a time predating the introduction of evidence-based treatments for the disorder. Gutheil described the frequent patterns related to transference and countertransference currents that often fueled risk management concerns. Challenges in the assessment of suicidality were described as reflecting the frequently observed mix in these patients of communicative suicidality, often marked by nonsuicidal self-injurious behavior or suicide gestures, and aspects of borderline pathology, such as impulsivity or jointly occurring substance use disorder, that predisposed these patients to risk of death by suicide. Boundary violations were explored as end results of the therapist’s insufficient appreciation of frequently observed patterns of alternating idealization and devaluation experienced by some patients, as well as associated eroticized transference, meaning superficially loving sentiments observed in patients that were paradoxically infused with aggression and destructiveness. Risk of patient abandonment by the clinician was understood as a possible manifestation of underappreciated negative countertransference reactions.

Contemporary Trends Contributing to Risk

Changes in General Treatment Patterns in Psychiatry, Including High Rates of Medication Use

Gutheil’s seminal articles (12–16) on borderline personality disorder and risk management were written at a time when most practicing psychiatrists still offered psychotherapy as part of their treatment repertoire. Some of Gutheil’s articles preceded the explosive growth of pharmacotherapy, well documented beginning in the 1990s, with the introduction of fluoxetine. From the early 1970s on, there have been many concerted efforts to identify medications useful in the treatment of patients diagnosed with this disorder (17). Over the years, mood stabilizers, benzodiazepines, antidepressants, and antipsychotic agents, among other choices, have all been studied, and results have consistently suggested a limited role for pharmacotherapy in the treatment of patients with the condition. That said, patients with this disorder are known as frequent recipients of complex and burdensome polypharmacy, despite essentially no data indicating that these medication regimens are helpful. Prescribers are frequently faced with a patient population in marked distress, often with highly concerned families involved, disposed to take medications, sometimes at higher than usual doses, with limited efficacy both for the primary symptoms, and very often, for the putative jointly occurring conditions that the medications might be targeting. With the introduction of each new medication subcategory, new risks for the prescriber emerge. These risks vary but are consistently concerning. Lithium is toxic in overdose and requires assiduous monitoring of thyroid and renal functioning. Benzodiazepines are sometimes disinhibiting and potentially habit forming and toxic when combined with alcohol. Older antidepressants were toxic in overdose or associated with dangerous adverse effects for patients unable to manage strict dietary restrictions. Newer antidepressant medications can be associated with weight gain or iatrogenic hypomanic or manic states. Older antipsychotic agents are associated with development of tardive dyskinesia, and newer second-generation antipsychotic agents are linked to metabolic syndrome, including obesity, hypertriglyceridemia, and hypercholesterolemia. Risks associated with pharmacotherapy are additionally complicated by widening use of off-label, experimental interventions, such as ketamine infusions and microdosing of hallucinogens.

The near ubiquity of split treatment in current practice makes the prescribing psychiatrist’s obligation to obtain permission to contact other clinicians involved in a patient’s care ever more important (18). Although the pertinent case law did not involve the treatment of a patient with borderline personality disorder, appropriate communication with a prior clinician has been established as standard of care (19). Frequently observed polypharmacy and split treatment complicate the risks of drug reactions, incorrect treatments, and medical errors.

Reduction in the Average Length of Stay of Psychiatric Hospitalizations

The marked reduction in the average length of stay of psychiatric hospitalizations since the 1980s, and the emerging crisis of the overall reduction of psychiatric inpatient services available, together have added challenges for psychiatrists working with patients diagnosed as having borderline personality disorder (20). Although these pressures may be somewhat beneficial, by limiting the ease of unhelpful, derailing psychiatric hospitalizations—as has been a well-described phenomenon of the past—for those situations in which a psychiatric hospitalization would be necessary or beneficial, these pressures can contribute to increased risk as psychiatrists are tasked with treating these patients with high-acuity as outpatients, with limited support or social service assistance.

Long emergency department waits and assignment to inconvenient facilities may make clinicians less likely to recommend or require hospitalization, potentially increasing risk to the clinician. In clinical situations that may necessitate hospitalization for patients with borderline personality disorder, essentially meeting standard of care (e.g., observation of developing persistent psychotic thinking, concerning co-occurring active substance use, or suicidality with active planning and intent), may be delayed because of these real-world inconveniences. These realities can complicate the psychiatrist’s management of suicide assessment, sometimes predisposing clinicians to overinvolvement or “heroic” efforts, which then present their own risks. In the same vein, limited access to treatment for conditions commonly co-occurring with borderline personality disorder, such as substance use and eating disorders, can add to clinician risk when providers are obligated to treat conditions themselves that might be best addressed through specialized services.

Changes in Training Opportunities

Changing health care economics have led to a drastic reduction in inpatient length of stay, leading to the demise of many facilities that offered long-term inpatient treatment options for patients with borderline personality disorder (20). These units were often fabled training facilities for psychiatry residents and fellows motivated to learn to work with patients with moderate-to-severe personality disorder pathology. The disappearance of these units as teaching opportunities was not countered with a robust increase in training didactics about work with this patient population. One survey of residency directors, for example, noted that only about one-half of the residency directors who responded indicated that their training programs offered any didactics related to diagnosis and treatment of borderline personality disorder (21). The result of these changes may be a generation of psychiatrists who have not had sufficient supervision in treating the disorder and may not be familiar with emerging, empirically validated interventions, but who nevertheless have obligations to care for the most symptomatic and impaired patients diagnosed as having this disorder.

Developments in Diagnostic Classification

Since its inception, borderline personality disorder has been an unusually controversial diagnosis. Despite repeated studies supporting its validity, the diagnosis has been subject to persistent scrutiny and dismissal. Researcher George Valliant expressed in the title of an article on this subject that “the beginning of wisdom is never calling a patient borderline” (22). Other expressions of doubt about the diagnosis have come from feminist critics (23) who have maintained that aspects of the diagnosis are inherently misogynistic. In the 1990s, the emergence of the concept of complex posttraumatic stress disorder further complicated diagnosing borderline personality disorder, as many influential thinkers suggested that that those diagnosed as having the disorder may have had insufficient attention paid to aspects of their trauma history (24).

Since 1980, many stakeholders have suggested renaming borderline personality disorder, reflecting concerns that the name is confusing, pejorative, or both. Other names proposed have included “emotional regulation disorder,” in an attempt to reduce stigma and promote acceptance (25). Most recently, the DSM-5 appendix introduced a hybrid model of personality disorder diagnosis, which may well become part of the official DSM in its next iteration. This hybrid model (26) moves away from the familiar categorical model on which much of the borderline personality disorder research is based.

Risk management calculations in work with patients with borderline personality disorder will likely be affected by the clinician’s acceptance of the diagnostic category and comfort in sharing the diagnosis with the patient, family, and other providers involved. Misgivings about the nosology, and preferences for other ways of communicating certain core diagnostic elements, will presumably affect the clinician’s attention to risk of injury for the patient and to legal vulnerability for the provider. If the hybrid categorical and dimensional approach to assessing personality pathology becomes the accepted approach, then too there will be new challenges, because clinicians may not recognize or be familiar with a new name for the disorder.

Emergence of Empirically Validated Treatments for Borderline Personality Disorder

The introduction of Linehan’s dialectical behavior therapy (DBT) (27) marked a new era in the treatment of patients with borderline personality disorder. DBT’s impressive evidence base, multiple rigorously conducted treatment trials, and marked positive reception among patients and treaters, have contributed to a sense of optimism about work with these patients. Over the years, DBT has been widely disseminated and used not only for patients with borderline personality disorder, but also for patients with mood, substance, eating, and anxiety disorders, among other conditions (28). In addition, research (29) that has examined dismantled DBT interventions, as in those situations in which a DBT skills group alone might be offered, has supported clinician interest in offering only one part of a standard, full DBT package.

As with all other empirically validated treatments for borderline personality disorder, DBT explicitly describes the clinician’s process of first making the diagnosis and then sharing it with the patient (30). However, because DBT has increasingly been used for patients with a variety of psychiatric diagnoses, this diagnosing and sharing is no longer considered standard in the offering of DBT. Clinicians who may not want to raise the subject of the diagnosis may, therefore, have greater discretion to recommend DBT for, as an example, a comorbid condition, such as a mood disorder. In this process, some of the key risk management protections that might come from first diagnosing borderline personality disorder, sharing the diagnosis with patients and families, and then documenting the likely manifestations of the disorder, may well be lost.

Given the robust development of evidence-based treatments for patients with the disorder, it is interesting to consider the 1975 landmark lawsuit Osheroff vs. Chestnut Lodge. These proceedings involved a suit by a patient, Dr. Osheroff, against the storied psychiatric facility, Chestnut Lodge, in Maryland, for failure to treat his mood disorder symptoms with appropriate medication (31, 32). The case, which was settled out of court and therefore never became case law, has regardless long been used as a distilled communication for the risk clinicians would incur by offering only psychotherapy instead of pharmacotherapy for certain patients. Up until now, there has not been a widely publicized legal action about the corollary, meaning the failure to diagnose a personality disorder, and to then offer evidence-based treatment, if appropriate. Such a suit might address the widespread phenomenon of patients with the disorder, who are incorrectly diagnosed over many years, and therefore denied appropriate treatment, while vulnerable to multiple, repeated, and damaging interventions of little value.

Explosion of Choices in Patient Communication Technologies

The explosion of options for communication with patients adds an extra dimension to provider risk management. Communication through e-mail, texting, and social media outlets, now routine for many clinicians, can add complications related to confidentiality, patients’ perceptions of clinician availability, and the possibility of a gradual erosion of boundaries. The rapid move to virtual sessions during the COVID-19 pandemic brought to the fore the possible boundary concerns associated with conducting treatment with patients who were often in their own homes, sometimes in their bedrooms, and not always dressed and groomed as they might be for an in-person office visit.

Overinvolvement that can portend boundary violations may be facilitated by communication channels, such as texting and e-mail. Familiarity through social media exposure may be another element in the erosion of boundaries, as patients may be exposed to personal information about clinicians in a way that had not been possible in the past.

The passage of the 2020 21st Century Cures Act requires health care providers to give patients access to all the health information in their electronic medical record “without delay” (33). This federal regulation adds yet another component to risk management challenges with patients with borderline personality disorder. It has been a widespread practice for some clinicians to in effect “keep two sets of books,” by documenting borderline personality disorder symptoms and diagnoses without necessarily describing them as such to patients. Appropriate documentation about risk, for example, as might relate to a seemingly paradoxical decision to recommend a lower level of care for a patient with borderline personality disorder who conveys chronic daily suicidality, is made more complicated if the diagnosis has not been discussed with the patient and family. Clinicians may be influenced by knowing that patients are legally entitled to access their medical records. Complaints or suits related to incorrect diagnoses may increase as patients more routinely review their medical records and, perhaps, find surprising information about the psychiatric diagnoses documented.

Conclusions

Psychiatrists will treat patients with primary or comorbid borderline personality disorder during the course of their careers; patients with borderline personality disorder are significantly overrepresented in most outpatient and inpatient settings. Failure to diagnose the condition or withholding information about a reliably made diagnosis from the patient and family, will not mitigate clinician liability. John Gunderson, a pioneer in research on borderline personality disorder, noted that the liability in treatment of this patient population “largely derives from countertransference enactments—excessive availability, punitive hostility, personal involvement, and illusions of omniscience and omnipotence” (8). The risk management challenges described by Gunderson have been further complicated in recent years by the changes outlined previously. Nevertheless, it should be possible in psychiatric practice to treat these patients safely and without undue apprehension. Clear, consistent guidance about risk management in the treatment of patients with the condition may, over time, contribute to diminished stigma and increased access to care for a patient population that has historically been marginalized.