Bipolar disorder (BD) is a complex potentially devastating illness, associated with losing 14 years of effective activity and dying 9 years early. There are substantially elevated risks of substance abuse and suicidal behaviour; a recent study indicated a 16% additional risk for suicide attempts for BD and comorbid substance abuse. Studies of the course of BD indicate that, if anything, the pattern of relapse and recurrence worsens as the individual ages. Pharmacotherapy of BD presents complex challenges, because of the need to treat effectively the different phases (e.g. manic, depressed and subsyndromal) of the disorder. A number of recent guidelines have drawn together evidence from available research. The British Association for Psychopharmacology (BAP) guidelines recognized the shortage of high quality studies and the need to combine scientific data with clinical experience in their recommendations. BAP, World Federation of Societies of Biological Psychiatry (WFSBP) and American Psychiatric Association (APA) guidelines all recognize the importance of lithium, valproate, oral antipsychotics and benzodiazpenes as treatment options in mania and mixed states. None of the guidelines recommend antidepressant monotherapy for bipolar depression, but rather the use of antidepressant and antimanic agents in combination. Both APA and BAP guidelines recommend lithium or valproate as front-line medications for long-term treatment. However, although both medications are effective in randomized controlled trials (RCT) relapse rates of 37–40% are reported.
Furthermore, partial or non-adherence to longterm treatment is reported in up to 50% of BD patients. Important factors in adherence include management of medication side-effects and also the beliefs that patients have about their illness and medication. If more effective treatments (including better adherence to available pharmacological treatment) are to be delivered, it is necessary to understand how psychological and social factors impact on the course of BD. Patelis-Siotis concluded that researchers should _pursue the understanding of cognitive processes in BD which would allow us to refine and develop cognitive behaviour therapy (CBT) interventions unique to this disorder. This review will consider the role of these factors before reviewing current individual and family approaches to BD. This information will then be discussed in relation to future psychological intervention research.
Individual psychological factors
Neurocognitive deficits have been reported in BD patients, including in selective attention, working memory, backward masking and controlled visual information processing. Birth cohort studies have indicated that individuals destined to develop BD and severe affective disorder show mild cognitive deficits in childhood and adolescence prior to the onset of illness; suggesting such factors are not restricted to illness episodes alone. These factors are relevant to understanding the aetiology of BD and also to the effective delivery of therapy.
An important issue is whether particular neurocognitive deficits extend into euthymic periods of BD. Quraishi and Frangou’s systematic review highlighted the presence of deficits in sustained attention, inhibitory control and verbal memory in remitted BD. However, they suggested executive functioning deficits (planning, concept formation, shifting of set), may not be abnormal in fully remitted BD. More recently, Martinez-Aran et al. identified verbal memory and frontal executive deficits across manic, depressed and euthymic BD in comparison with healthy controls. Significant deficits in verbal memory and some measures of executive functioning in euthymic BD persisted after controlling for level of subsyndromal symptoms. These authors suggested that these trait deficits might have an important negative impact on illness course. Furthermore, because number and duration of previous episodes were associated with severity of verbal memory impairment it was proposed that neurotoxic effects of stress-related hypercortisolaemia might be responsible for these cognitive deficits through triggering hippocampal and prefrontal damage. Sobczak et al. reported executive functioning and verbal memory deficits in first-degree relatives of BD 1 patients, which worsened after acute reduction in central 5-HT activity. Sobczak et al. suggested 5-HT mediated frontal lobe dysfunction as an important marker for BD 1 disorder. Lopez-Figueroa reported serotonergic dysregulation consistent with decreased 5HT levels in the dorsolateral prefrontal cortex of BD patients.
Taveres et al. have argued that the performance of depressed and manic patients is most clearly differentiated on _hot_ cognitive tasks (which include affective material). Thus, manic and depressed patients show biases on an affective _go/no-go_ task compared with controls; manic patients to positively valenced material and depressed to negative targets. Furthermore, the performance of depressed subjects was impaired by negative feedback, but not if this feedback had high information content. Tavares et al. have argued that performance on tasks of this type are also subserved by the prefrontal cortex.
Studies of social cognition have reported dysfunctional assumptions and attributional biases in BD. Scott found higher levels of interpersonal dependency (sociotropy) and stronger beliefs in needs for social approval and perfectionism in euthymic BD than controls. Scott and Pope reported that within a BD group, dysfunctional attitudes were most apparent in depression, but scores were also elevated in hypomania. It also appears that BD patients make attributions for negative events which are as depressogenic as those of unipolar depressives and that individuals with a tendency towards BD have particular coping styles with respect to mood change.
Thomas and Bentall reported hypomania scores in an undergraduate sample were predicted by rumination, distraction and engagement in dangerous activities [using an expanded version of the Nolen-Hoeksema Coping Styles Questionnaire]. Self-reported depression scores were associated with ruminative style only, consistent with previous reports. This pattern of results has recently been replicated in a further behavioural high-risk sample.
Other studies have indicated a role for personality factors in BD, including neuroticism, interpersonal sensitivity and conscientiousness. These trait factors indicate the importance of approaches that address long-term vulnerabilities, in addition to efforts to achieve initial symptomatic improvements. Psychosocial approaches therefore need to both equip the individual with strategies over the long-term and facilitate environmental changes that may have protective effects for the individual.
Family and social factors
Johnson et al. his review concluded that life events tend to precede illness episodes. Ellicott et al. found this pattern was independent of changes in medication compliance. However, the impact of life events and psychosocial stress may vary during the course of illness. This issue has been considered in relation to Post’s behavioural sensitization and kindling model in which substantial disrupting events are required to trigger affective episodes in the early stages of BD, but as the illness course continues the individual becomes sensitized to triggers, and less significant life events are necessary. Research is inconclusive on this point, with some studies reporting higher sensitivity to life events later in the illness course, whilst others suggest such sensitivity is limited to earlier illness episodes. A recent study found no relationship between number of episodes and either kindling or sensitization; but did suggest age may moderate reactions to stress. These results indicate there are likely to be complex interactions between events and their consequences for different individuals at different points in their illness course. Greater clarity might be achieved by studies that are more specific about the nature of particular events for the individual. Thus, Johnson et al. found manic, but not depressive, symptoms increased in BD following a significant goal attainment event. Malkoff-Schwartz et al. reported that social rhythm disrupting life events preceded manic, but not depressive, episodes in BD.
As with depression and schizophrenia, another relevant aspect of psychosocial stress is family atmosphere. Dore and Romans found 92% of partners struggled to maintain their relationship and 62% felt they would not have begun the
relationship had they known about the illness. Carer burden has been related to frequency of emotion-focused coping strategies (attempts to manage own emotions by avoidance or resignation); which indicates the importance of evaluating carer’s own response to BD. Butzlaff and Hooley’s review of expressed emotion (EE) found associations between high EE (critical comments and emotional over-involvement of carers) and relapse in BD. Recently, Wendel et al. confirmed that high EE relatives of BD patients viewed symptom behaviours as being more under the patient’s control than did those with lower EE. In addition, patients exhibiting higher levels of subsyndromal symptoms during meetings with relatives were at higher risk of relapses and more likely to experience high EE comments from relatives. These findings indicate that high EE responses of relatives are associated with both their interpretations of the behaviours of BD patient, and with the behaviours themselves.
Efficacy of individual CBT and family work A number of studies have explored the effectiveness of psychological and family interventions. This evidence has been recognized in BAP, APA and WFBP guidelines which all identify a role for structured psychological therapy in the adjunctive treatment of BD.
Colom et al._s recent RCT study of group psychoeducation indicated significant impact on relapse rates and plasma results indicative of higher lithium medication compliance at 2 years follow-up. Five reports of controlled studies of
individual CBT indicate reductions in both relapse risk and subsyndromal symptomatology in BD. However, although Lam et al. reported reduced risk of relapse, reduced rates of admission and shorter time in episodes for BD individuals who received CBT, 44% of those receiving CBT relapsed within 12 months. Although much superior to the comparable rate of 75% in controls, it indicates the need for interventions which can further improve such figures.
There are two recently published reports of RCTs which have used family focused treatment (FFT) a psychoeducational approach which also trains patients and relatives in family communication and problem solving. These studies by Miklowitz et al. and Rea et al. reported fewer relapses in FFT patients compared with controls. Again, although effective, these studies still reported substantial numbers of patients relapsing in the FFT condition [35% in Miklowitz et al.].
- Jones SH, Bipolar Disorders, 2005 Feb; Vol. 7
The article above contains foundational information. Articles below contain optional updates.
Reflection Exercise #9
The preceding section contained information about psychological therapies for bipolar disorder. Write three case study examples
regarding how you might use the content of this section in your practice.
Partial or non-adherence to longterm treatment is reported in what percentage of bipolar disorder patients? Record the letter of the correct answer the