Prescribing and borderline personality disorder

Accurate diagnosis is fundamental to effective management of borderline personality disorder, but many patients remain undetected.

The first-line management for borderline personality disorder is psychosocial treatment, not drugs. There are major prescribing hazards including polypharmacy, overdose and misuse.

Drug treatment might be warranted for patients who have a co-occurring mental disorder such as major depression.

If a drug is prescribed for borderline personality disorder, it should only be as an adjunct to psychosocial treatment. There should be clear and collaborative goals that are regularly reviewed with the patient.

Use single drugs prescribed in limited quantities for a limited time. Stop drugs that are ineffective.

Borderline personality disorder is a severe mental disorder that has its onset during adolescence and emerging adulthood.¹ It affects up to 3% of the population^2,3 and occurs almost equally among males and females.⁴

The disorder has a fourfold higher prevalence among primary care patients than among the general population.⁵ It affects around one-quarter of primary care patients with depression⁶ and one in five psychiatric outpatients.^7,8 In these settings females outnumber males by a ratio of up to 4:1.

Borderline personality disorder is a leading contributor to the burden of disease in our community. It is associated with adverse long-term outcomes that include severe and persistent functional disability,⁹ high family and carer burden,¹⁰ physical ill health,¹¹ and premature mortality,¹² including a suicide rate of 8%.¹³ People with the disorder use mental health services continuously for long periods of time.¹⁴ After schizophrenia, borderline personality disorder is the most costly psychiatric disorder to treat in Australia on a per case basis.¹⁵

Psychosocial treatment is the primary therapy but access to this is poor. Despite the effectiveness of treatment, persisting psychopathology (e.g. borderline personality disorder features, depressive and anxiety symptoms) and functional impairment remain clinically problematic.

While it is easier to provide prescriptions than psychosocial treatments, evidence does not support the use of drugs as first-line or sole treatment.¹⁶ Nevertheless, psychotropic drug use is common and needs careful management and review.

Borderline personality disorder is characterised by a pervasive pattern of instability in emotional regulation, interpersonal relationships and self-image, along with marked impulsivity.¹⁷ Clinically, this often manifests as recurrent self-harm and suicide attempts. The person frequently describes a chaotic lifestyle and relationships, reckless behaviours likely to harm the individual (e.g. impulsive substance use, unsafe sex), chronic dysphoria and anxiety, severe mood instability and reactive aggression.

People with borderline personality disorder typically present to health services during times of crisis, following self-harm, because of the consequences of impulsive and self-damaging behaviour, or because of poor physical, sexual and reproductive health. Interpersonal problems are at the very heart of borderline personality disorder and these are usually magnified during crises. For example an individual might have unrealistically high expectations or demands of care from health practitioners and when these are not met, he or she might respond aggressively. It is important to remember that clinicians are likely to see patients at their lowest ebb, but this is when they are in greatest need of timely and coherent assistance.

The DSM-5 criteria for borderline personality disorder are listed in Box 1.¹⁸ Although the DSM-5 requires any five of these nine criteria in order to make a diagnosis, even low levels of borderline pathology (e.g. one criterion) are associated with substantial increases in psychosocial impairment.¹⁹

Despite the evidence of the reliability and validity of the diagnosis, and the treatability of the condition,²⁰ many people with borderline personality disorder remain undiagnosed in clinical practice. This places them at risk of being given treatments that are ineffective or even harmful.²¹ The central task for diagnosing personality disorder is to separate ‘state’ (transient aberrations in mental state) from ‘trait’ (long-standing patterns of thinking, feeling, behaving, perceiving and relating). Many mental state disorders can present with features that are similar to borderline personality disorder. For example, affective dysregulation is characteristic of both bipolar disorder and borderline personality disorder. Also, the current definition of depression incorporates non-specific forms of dysphoria that overlap with borderline personality disorder. What distinguishes borderline personality from these other disorders is that the features are present most of the time and comprise part of the patient’s ‘usual self’. These patients will tell you that this is how they ‘usually are’. Although various tools are available to aid the diagnosis of borderline personality disorder, clinical application of the DSM-5 criteria is sufficient in a busy clinical practice. Each of the nine criteria should be enquired about and considered in turn.

A common cause of misdiagnosis of borderline personality disorder is to rely on ‘gut feeling’ when a patient presents as interpersonally abrasive, sullen or hostile, particularly if the individual also engages in self-harm. Such diagnoses are often unreliable because they do not assess each of the DSM-5 borderline personality disorder criteria and do not take into account other reasons for such presentations, such as temporary aberrations in mental state, depression or other disorders.

Another cause of diagnostic confusion is the high rate of comorbid conditions. Comorbidity with other personality disorders and with mental state disorders is the norm. At times, these other disorders (e.g. mood, anxiety, eating and substance use disorders) can overwhelm the clinical picture, but this does not indicate that the underlying personality pathology is unimportant or should be a secondary concern.²¹ Rather, there is evidence to suggest that personality disorder might be a key vulnerability factor for recurrent mental state disorders.^22,23 Patients with borderline personality disorder who have these co-occurring conditions should be treated for these conditions in accordance with best practice. However, there should not be a disproportionate emphasis given to the immediate relief of mental state pathology at the expense of managing the borderline personality disorder.

Since the 1990s there has been growing optimism and enthusiasm for the treatment of borderline personality disorder. There is now a variety of effective evidence-based psychosocial treatments (Box 2). Referral for one of these first-line treatments is recommended.^16,24 No single treatment is recommended over another and they have common core features (Box 2).²⁵ Treatment guidelines include one published by the Australian National Health and Medical Research Council (NHMRC).¹⁶

Despite advances in psychosocial treatment for borderline personality disorder, improvements remain suboptimal.^24,25 Access to treatment is limited, and dropout rates are high (15–77%).²⁶ The NHMRC guidelines provide advice that can be implemented when referral to specialist services for borderline personality disorder is unavailable.

Pharmacotherapy has been investigated as a stand-alone or adjunctive treatment option for borderline personality disorder. There are many high-quality reviews^16,27,28 including a Cochrane review.^27,29 The Cochrane group found 33 randomised controlled trials in adults with borderline personality disorder,^27,29 but commented that the overall evidence base for prescribing is unsatisfactory. The literature is hampered by small trials (less than 50 patients) of numerous drugs, short treatment periods (mean duration of 12 weeks), diverse outcome measures, infrequent replication of findings, and lack of independence from the pharmaceutical industry.^30,31 In addition, controversy has surrounded the methods of one research group and the integrity of the findings.²⁹

The available evidence^16,27-29 does not support a prominent role for selective serotonin reuptake inhibitors in the treatment of borderline personality disorder, despite their widespread use. Other drugs require further investigation. Mood stabilisers (topiramate, sodium valproate, lamotrigine) have shown some effect in reducing affective dysregulation and impulsive aggression. Antipsychotics such as aripiprazole, olanzapine and quetiapine have shown some effect in reducing cognitive–perceptual symptoms and affective dysregulation. Omega-3 polyunsaturated fatty acids might reduce the overall severity of borderline personality disorder.²⁷

Although the evidence for drug therapy is less conclusive than for the psychosocial interventions, US data indicate that prescribing rates for borderline personality disorder are paradoxically high. Drugs are prescribed for 78% of patients for more than 75% of the time over a six–year period and polypharmacy occurs in 37% of patients,³² perhaps reflecting clinical needs and pressures. To date, there has been no study of Australian prescribing practices, but clinical experience suggests that the situation might be similar.

All prescribing occurs in the context of the doctor–patient relationship. However, this takes on particular significance when the condition being treated is fundamentally a relational disorder. Interpersonal difficulties and challenges, particularly maladaptive help-seeking, are part of the everyday work of caring for people with borderline personality disorder. This need not be a deterrent to becoming involved in their care, but it does require a degree of self-reflection and self-monitoring.

In caring for patients with borderline personality disorder, strong thoughts or feelings on the part of the doctor can be used as a reminder to reflect on what is happening in the doctor–patient relationship: Am I feeling ‘pulled’ or ‘pushed’ to respond in particular ways? Why won’t the patient do what they need to? Who is responsible for change? Am I doing all the work? Am I responding as I might usually do, or am I treating this patient differently? Some patients will invite clinicians to behave differently with them and we can be aware of these invitations without automatically feeling ‘cheated’ or ‘manipulated’ and becoming more rigid and inflexible.

Some strategies to assist in this task can be drawn from the common characteristics of evidence-based treatments in Box 2, e.g. encouraging patients to develop a sense of self-control, and helping patients to connect feelings to events and actions. Collaborative goal setting is fundamental to successful prescribing. This involves being clear about what it is that the patient is seeking, and agreeing to common goals. A key role for the clinician is to keep one’s eye on these goals, assisting the patient to persist with the chosen strategy, without becoming too rigid.

Clinicians need to be aware of their own limits and needs and should be willing to communicate these to the patient, rather than dress them up as the patient’s needs. For example, the doctor’s responsibility to prescribe safely, legally and ethically might be at odds with the patient’s demands for immediate relief from distressing symptoms.

Clinicians also need to admit and be willing to openly discuss difficulties. It is often much easier to start prescribing than to stop it. Honest and open discussion about whether a trial of a drug is warranted or has been effective is fundamental.

Borderline personality disorder is a leading cause of disability and mortality and is common in clinical practice. Although it was once considered ‘untreatable’, the outlook for patients with borderline personality disorder is much improved, with a range of effective psychosocial treatments available for the disorder. However, access to these treatments is limited in the Australian healthcare system. Despite this, drugs should not be used as primary therapy for borderline personality disorder because they have only modest and inconsistent effects. They do not change the nature or course of the disorder. A time-limited trial of a drug can be considered as an adjunct to psychological therapy, to manage specific symptoms, but caution is needed.

Conflict of interest: Andrew Chanen was a member of the development group for the National Health and Medical Research Council’s Clinical Practice Guideline for the Management of Borderline Personality Disorder.

Acknowledgement: The authors would like to thank Dr Louise McCutcheon for comments on an earlier version of this manuscript.