This site is intended for health professionals only

At the heart of general practice since 1960

Borderline personality disorder

Consultant psychiatrist Dr John Wilkins answers questions posed by GP Dr Stefan Cembrowicz on a challenging mental health condition

Consultant psychiatrist Dr John Wilkins answers questions posed by GP Dr Stefan Cembrowicz on a challenging mental health condition

1 How common is a diagnosis of borderline personality disorder (BPD)?

The incidence and prevalence of BPD (also known as emotionally unstable personality disorder, borderline type) is affected to some extent by the diagnostic criteria used. DSM IV, favoured by the American Psychiatric Association, is more specific than the impressionistic description in ICD 10, the system favoured in the UK. However, the prevalence is generally thought to be 2% of the population, 10% of psychiatric outpatients and 30-60% of cases in a specialised personality disorder clinic. These are American figures and the numbers in the UK are probably lower because of our reluctance to make a formal diagnosis of personality disorder. (In the US, a diagnosis is needed for insurance purposes.) Nevertheless, it is an important problem for psychiatrists, community mental health teams (CMHTs), GPs and A&E departments.

2 What characteristics do typical patients with this condition exhibit?

There is no such thing as a typical patient in cases of personality disorder (PD). However, typical symptoms or behaviours associated with BPD include:
• patterns of intense but unstable relationships
• instability of mood, often with changes that occur over very short time spans (minutes or hours), rather than the longer time spans (weeks and months) associated with bipolar disorder
• impulsive behaviour, including drug and alcohol misuse, sexually indiscriminate behaviour and eating disorders, particularly bulimia
• a tendency to catastrophise
• poor self image
• in some cases, paranoid symptoms and hallucinations or pseudo-hallucinations
• self-harming that relieves symptoms, albeit temporarily.

3 How do you distinguish adolescent turbulence from borderline personality disorder? Lifestyle chaos and poor anger management are common in the population – how can you make the diagnosis in the GP setting? Do questionnaires help?

I consider it unwise to diagnose personality disorders before the age of 18 and preferably not 21. Before this, the personality is still developing. However, persistence of symptoms tends to distinguish personality disorder from adolescent turmoil. Symptoms of more than one year's duration are unlikely to be associated with adolescent problems.

Because of the pejorative nature of the diagnosis, I would recommend that it only be made by an experienced consultant psychiatrist. However, instability of relationships, occupational instability, mood swings and self-harm are usually good indicators that the diagnosis of BPD might be appropriate. In my view, questionnaires don't help.

They assess the situation at a particular point and, bearing in mind the instability that is central to the diagnosis, there is a tendency to over-diagnose using questionnaires. This would apply to screening instruments used in primary care and assessment instruments used by psychiatrists. There is always a tendency by patients to assess their symptoms as extreme.

4 The US (DSM-1V) and UK (ICD 10) criteria for diagnosing borderline personality disorder differ. Which is more appropriate?

There is little difference between the criteria in DSM IV and the description in ICD 10 for emotionally unstable personality disorder, borderline type. The NHS will always use ICD 10, while research projects tend to use DSM IV. The results can usually be applied to ICD 10 cases.

5 Patients with a label of borderline personality disorder present a large caseload to psychiatrists and GPs – they may be frequent attenders at GP surgeries and out-of-hours services too. But there is often a negativity towards treating them. How should GPs tell a patient they have this condition and how can they advise them to manage themselves?

BPD can be very difficult to manage. I wouldn't recommend that GPs make a diagnosis for the reasons set out above. A referral to the CMHT would be appropriate, where they should be assessed by an experienced mental health professional, preferably a consultant. A telephone discussion beforehand helps so that the CMHT know what the expectations of primary care are.

Clear roles and responsibilities need to be defined in these cases more than in any other. Who will prescribe? What will be prescribed? Even more important, what will NOT be prescribed? Contingency plans need to be made for crisis management. Easy access to appointments with the person with responsibility for the case in secondary care is important.

Regular care plan reviews are important and the attendance of the GP is desirable. With these clear boundaries patients can then begin to learn how to manage their symptoms and what strategies will help. Remember, always deliver what you promise and don't promise too much.

6 How should GPs deal with these patients in a crisis? When may drugs be useful?

Patients may turn to their GP in a crisis. A contingency plan is important.

Some 50% of patients have depression and may warrant treatment with antidepressants, although there is little research on the efficacy of medical treatment. Anti-psychotics are helpful if there are psychotic or pseudo-psychotic symptoms. Depot injections of flupentixol have been shown to be helpful in extreme cases, usually at a low dose. Avoid benzodiazepines if at all possible; if they seem necessary in a crisis, keep prescriptions to a minimum and do not repeat them without review either by yourself or secondary care.

7 What are the forensic implications of this condition?

The most likely association is with sexual abuse. Individuals who suffer from this condition have usually been traumatised and sexual trauma is commonly associated. People with BPD will probably be involved in abusive relationships and may themselves be perpetrators of abuse. There is an association with criminality generally, and with inter-personal violence and arson in particular.

8 What is the connection between borderline personality disorder and psychotic illnesses? Is there a genetic link?

BPD is five times more common in first-degree relatives and is also associated with other psychiatric disorders in the family. Whether this is a genetic link is unknown. Psychotic or pseudo-psychotic symptoms are a feature of the condition. It is not generally associated with major psychoses, although both schizophrenia and bipolar disorder are usually on the differential diagnosis.

9 Are there any specialised units for managing patients with borderline personality disorder? What is the long-term outlook for patients with this condition?

Specialised units are few and far between and tend to be tier four services. Examples in the South East are the Crisis Recovery Unit at the Bethlehem Royal Hospital in Beckenham, Kent; the Winterbourne Day Unit in Reading; and the Henderson Hospital in Sutton, Surrey.

The natural history of the condition is that it tends to dissipate during the third and fourth decades of life. Studies have shown that after 10 years, 50% no longer satisfy the diagnostic criteria.

10 Are there any self-help groups or resources for these patients – or their families?

There are no organised self-help groups that I am aware of, although there are sometimes groups set up in an ad hoc fashion for sufferers and their families. Sometimes an addiction model is used, such as a 12-step programme much in the same way as for addicts more generally. There is no data about the efficacy of such approaches.

Take-home points

• Borderline personality disorder is thought to affect 2% of the population and 10% of psychiatric outpatients.
• It is unwise to diagnose personality disorders before age 18 and preferably not before 21. Symptoms of more than one year's duration are unlikely to be associated with adolescent turmoil.
• There is no typical patient but typical symptoms or behaviours are associated.
• Instability of relationships, occupational instability, mood swings and self-harm are good indicators of the condition.
• Questionnaires tend to overdiagnose.
• Half of patients have depression and may warrant treatment with antidepressants, although there is little research on efficacy.
• Make crisis plans.
• Antipsychotics are helpful if there are psychotic or pseudo-psychotic symptoms.
• Individuals with this condition have usually been traumatised and sexual trauma is commonly associated.

What I will do now

Dr Cembrowicz comments on the answers to his questions

• Remember the addresses of the specialised units
• Hope that at least some of my patients will grow out of this condition by the age of 40
• Remember the importance of clear boundaries and limits for these patients
• Try to help them appreciate the nature of their own personality – that they are people with rapidly volatile moods who think little of themselves and may be impulsive
• Where possible ask for a secondary care diagnosis – especially if psychotic features may be present
• Make a care plan for them, and avoid benzodiazepines

Dr Stefan Cembrowicz is a GP in Bristol

Rate this article  (1 average user rating)

Click to rate

  • 1 star out of 5
  • 2 stars out of 5
  • 3 stars out of 5
  • 4 stars out of 5
  • 5 stars out of 5

0 out of 5 stars

Have your say