Diagnosis and monitoring of bipolar disorder in general practice

In Australia, bipolar disorder affects 1.3% of the population at some stage of their life.1 General practitioners are often consulted for first presentations of this condition, and are frequently involved in ongoing monitoring and care of patients in conjunction with specialised mental health services. As with other persistent complex illnesses, GPs are well placed to coordinate the care of patients with bipolar disorder as they continue to provide other aspects of medical care and develop an understanding of the patient’s circumstances.

The peak onset of bipolar disorder is usually during the period from 15 to 30 years of age. The impact of bipolar disorder on patients, and individuals close to patients, is substantial.2 Disrupted relationships are common. Rates of unemployment are high — at the least, achievement of career aspirations may be significantly hampered. Many patients are forced onto government benefits. Furthermore, mortality rates are substantially increased, mainly due to higher rates of vascular disease, diabetes and obesity.3 At least a quarter of patients with bipolar disorder will attempt suicide on one or more occasions; at over 100 cases per 100 000 person-years (compared with the 2007 Australian general population suicide rate of 8.9 per 100 000 population per annum4), suicide rates are greatly increased.5

While hypomania and mania are the hallmark characteristics of the illness, depression is usually the predominant mood in those with bipolar disorder and has been associated with the greatest burden of disability.6 In a study on bipolar I disorder, patients experienced 32% of their weeks of follow-up in depression and 9% in mania or hypomania;7 in a study on bipolar II disorder, 50% of follow-up occurred in depression but only 1% in hypomania.8 Suboptimal function between discrete bipolar episodes, characterised by symptoms such as mild anxiety or depression, is common and tends to be under-recognised. Comorbid conditions including anxiety disorders (52% of patients in one Australian study) and substance misuse (39%) are highly prevalent.9

Patients with bipolar disorder report that delayed diagnosis and incorrect diagnosis are common. A study of participants in a United States bipolar disorder support group revealed that more than one-third sought professional help within a year of the onset of symptoms, but 69% were misdiagnosed — most frequently with unipolar depression — and consulted a mean of four doctors before receiving the correct diagnosis.10 Other frequently reported misdiagnoses included anxiety disorder, schizophrenia, borderline or antisocial personality disorder, alcohol or substance misuse and/or dependence, and schizoaffective disorder. Over one-third waited 10 years or more before receiving an accurate diagnosis.

The results of this study suggest two reasons for the common misdiagnosis of unipolar depression: first, many patients with bipolar disorder might not have experienced an episode of hypomania or mania by the time of misdiagnosis; second, for some patients with bipolar disorder, the occurrence of hypomanic or manic symptoms may not be elicited by clinicians or acknowledged by the patients. Mania and hypomania may have covert and subtle presentations, and some patients with hypomania may not recognise them as abnormal, rather perceiving them as normal or desirable (especially the feelings of being energised, confident and needing less sleep). Notably, participants in the US study rarely reported all their manic symptoms to a doctor — for example, fewer than one-third disclosed symptoms such as reckless behaviour, spending excessively and increased sexual interest or activity.10

The age of a patient influences the differential diagnosis. In younger patients, conditions such as attention deficit hyperactivity disorder and conduct disorder need to be considered. Conversely, when the initial onset occurs after the age of 40 years, organic causes such as medications (eg, steroids or dopaminergic agents for Parkinson’s disease), endocrine disorders (eg, hyperthyroidism) and neurological conditions (eg, frontal lobe tumours or cerebrovascular disease) should be excluded. In practice, however, such organic aetiologies are very uncommon.

On the other hand, there has been recent concern that the diagnostic pendulum may have swung too far, and that bipolar disorder (particularly bipolar II disorder) may be becoming overdiagnosed, especially in patients with unipolar depression or personality disorder.11 Some clinical authorities currently recommend diagnosing hypomania on the basis of brief periods, of only hours in duration, of elevated mood. Such a radical diagnostic shift risks labelling normal exuberance and enthusiasm as pathological mood disturbance. This is not merely an academic debate, as overdiagnosis could mean inappropriate use of mood-stabilising therapies and, consequently, insufficient attention paid to the psychological aspects of these conditions. Longitudinal monitoring is often necessary to make or refine the diagnosis of bipolar disorder.

The clinical features of mania include elevated, expansive or irritable mood, accelerated speech, racing thoughts with flight of ideas, increased activity and reduced need for sleep. Patients may develop grandiose ideas, act recklessly (including increased spending) and show increased sexual drive and activity. Symptoms often appear abruptly. Where symptoms are less severe and of shorter duration, the term hypomania is used.

An episode of acute mania is a medical emergency, so early recognition and diagnosis is critical. Patients have the capacity to destroy their reputations, relationships and finances within a short period. Insight and judgement are usually impaired early, even in the absence of delusions, and involuntary hospitalisation is frequently required to protect the patient. If outpatient treatment occurs, it is essential to monitor risky behaviour such as financial indiscretion and potential harm to others (eg, hazardous driving).

A useful clinical adage is that patients with acute mania are “always worse than they seem”. An apparently reasonable presentation during a brief assessment may mask more serious dysfunction. Reports from family and friends should be taken seriously, but interpreted with an understanding of the patient’s normal function and the nature of these relationships. Some clues to the presence of hypomania and mania are summarised in Box 1.

Manic relapses are often due to poor medication adherence. Other common causes include substance misuse (particularly cannabis, cocaine or amphetamines), antidepressants and stressful life events.

While the chaos and disruption of mania leads to major concern for friends and relatives, it is the depressive episodes that are associated with the marked disability and the high suicide rates linked to bipolar disorder. Risk management, particularly the recognition of suicidality, is a crucial responsibility for the GP.

As discussed above, depression is more likely than mania to be the first presentation of bipolar disorder, leading to diagnostic uncertainty between unipolar depression and bipolar disorder (yet to manifest with mania) in those with an initial depressive episode. It is apparent that there are no pathognomonic characteristics of bipolar depression compared with unipolar depression. There are, however, replicated findings of clinical characteristics that are more common in patients with bipolar I depression than in those with unipolar depression, and vice versa, or that are observed in patients with the initial presentation of unipolar depression who “convert” to bipolar I disorder over time.

Common clinical features of patients with bipolar I depression are:6

• course of illness with earlier age of onset, shorter duration of episodes and more prior episodes

• symptoms such as worthlessness, low self-esteem, social withdrawal, hypersomnia, hyperphagia or weight gain, atypical features (eg, “leaden paralysis”), lability of mood and psychotic features

• mental state signs such as psychomotor retardation (lower activity levels)

• family history positive for bipolar disorder.

Some other features of depression said to be indicative of the future development of bipolar disorder include induction of hypomania or mixed states by antidepressants, abrupt onset of depression, seasonal pattern, mixed presentations with hypomanic symptoms, and postnatal onset.

Rather than proposing a categorical diagnostic distinction between bipolar depression and unipolar depression, we have recommended a probabilistic (or likelihood) approach to the distinction13 — in other words, a differential likelihood of experiencing the symptoms and signs of depression listed above. The suggested criteria for this probabilistic approach are detailed in Box 2.

The probabilistic approach is premised on a current major depressive episode in patients with no clear prior episodes of hypomania or mania and specifies features that indicate the greater likelihood of bipolar I depression or unipolar depression. These features are useful to take into account when making treatment decisions for: depressed patients in whom past history of hypomanic or manic episodes is ambiguous; “unipolar depressed” patients with a family history of bipolar disorder; and young patients presenting with recurrent depressive episodes only (where it is unclear whether this represents a first presentation of bipolar disorder or unipolar depression). This approach has been developed to help clinicians identify depressed patients who have an increased likelihood of going on to develop bipolar disorder.

To our knowledge, no formal treatment studies using this approach have been carried out to date. At present, we do not recommend that clinicians immediately diagnose and treat depressed patients with these features as definitely having bipolar disorder, but commend practitioners to seriously consider this possibility for such individuals as treatment progresses.

Some patients switch to mania rapidly and without warning. A considerable proportion, however, have a transitional phase with early warning signs that begin hours, days or weeks before the onset of frank manic symptoms. Interestingly, early warning signs before the development of a bipolar depressive episode are less commonly observed. With training, both the patient and his or her family can identify the behaviour changes in this transitional phase. If such signs can be reliably identified, countervailing strategies can be developed. It may be possible to develop an early warning relapse profile for individual patients.

Possible early warning signs for hypomanic or manic episodes include increased activity and busyness; reduced need for sleep; impulsive behaviour; speaking in a caustic manner; and telephoning friends indiscriminately. Possible early warning signs for depressive episodes are feeling tearful, moody, withdrawn, snappy, slowed down, negative, stubborn, pessimistic, hopeless or excessively self-doubting. Therapeutic strategies to prevent the development of full manic or depressive relapses are discussed in this Supplement in the articles on psychotherapy and pharmacological treatments (see Lauder et al; Malhi et al).14,15

While there are a number of formal clinician-administered assessment tools and self-rating tools for patients that can be used to screen for and monitor bipolar disorder,16,17 these have been predominantly developed for use in research settings. In general, they are not widely used in clinical settings — neither in psychiatric nor general practice. One of the more widely used screening tools is the Mood Disorder Questionnaire (MDQ),18 which screens for a lifetime history of mania or hypomania, although there has recently been concern about the potential of this instrument to over-diagnose bipolar disorder.19 Examples of scales used commonly in research settings include the Bipolar Depression Rating Scale, developed in Australia,20 and the Young Mania Rating Scale.21

Of more clinical pertinence is the use of daily mood charts,22-24 which are widely used in clinical settings. These charts allow patients (and thereby their clinicians) to monitor daily changes in mood, specific target symptoms, stressors, activity levels and daily routines. The charts provide patients with an overview of their progress and are increasingly used as a component of psychosocial treatment programs (see Lauder et al, );14 for example, daily mood charts that focus on identification and management of early warning signs are useful in psychotherapy.

Bipolar disorder can be extremely taxing for the families and carers of patients. Families and carers need and want information and education about the illness, as well as continuing support during times of crisis. They can become active participants in elements of treatment such as encouraging appropriate use of medication, regulating lifestyle and monitoring for relapse.

Loss of insight during acute manic episodes poses particular problems for families and carers, as patients may deny their illness and resist treatment. Treating doctors must balance the need to respect patients’ privacy with fulfilment of their obligation to provide adequate care. Decisions may need to be made about the legal competence of patients, and the point at which action should be taken to deem patients incompetent and a danger to themselves and others. Dangers are not always physical and can include damage to reputation, income and relationships. Dependents normally under the care of the patient may also be at risk, and treating doctors should be aware of their legal and ethical responsibilities to protect children from harm. For those prone to frequent relapses, it is often useful — while the patient is well — to agree on a preferred management strategy for future recurrences, although it must be emphasised that such advance care directives cannot be legally binding.

The GP must consider the ongoing welfare of the patient as paramount, as behaviour may be markedly disturbed during episodes of mania. Patients may engage in behaviour with severe future ramifications, such as embarking on unwise business schemes, spending large amounts of money, or becoming aggressive with a potential of physical harm to others. Despite the patient’s protestations at the time, the clinician has an imperative to intervene in such situations, although this may arouse considerable patient anger or irritation.

For patients who are prone to spending or borrowing large amounts of money during manic episodes, serious consideration should be given to strategies such as:

• arranging for the patient to provide a family member with power of attorney

• using relevant state bodies to take over the patient’s financial affairs

• reducing the patient’s capacity to spend large amounts of money (eg, by encouraging the patient to forgo credit cards).

GPs have a central role in facilitating diagnosis, accessing specialist care, and providing continuing monitoring and support for patients with bipolar disorder. The GP can play a pivotal role in early identification of this serious and common condition.