Introduction
Hallucinations in bipolar disorder have been the subject of little systematic research despite evidence that they are relatively common. In the most comprehensive study to date of prevalence of hallucinations in bipolar disorder (Baethge et al., Reference Baethge, Baldessarini, Freudenthal, Streeurwitz, Bauer and Bschor2005), records of all patients hospitalized at a Berlin teaching hospital between 1981 and 2000 were examined for evidence of hallucinations; in the bipolar disorder patients current hallucinations were reported as occurring in 11.8% of the sample. Observations of the phenomenology of bipolar disorder hallucinations have been summarized by Hammersley and Fox (Reference Hammersley, Fox, Larkins and Morrison2006).
Reports on clinicians' attributions for cause of hallucinations in bipolar disorder have a long history. Dr William Pargeter's classic 1782 study, Observations on Maniacal Disorders; contains case studies including a Mr Pascal who hallucinated a gulf of fire around him at all times, and two servant girls one of whom heard the voice of Christ, whilst the other “night and day saw the appearance of devils, sulphurous flames and other horrid images of the everlasting torture of the damned”. Pargeter attributed the cause of these hallucinations in every case to overwork and the nefarious presence of Methodism.
Method
Participants
Forty participants contributed to the pilot study; the participants were a subgroup drawn from a larger national randomized control trial (n-253) into cognitive behavioural therapy for bipolar disorder (Scott et al., Reference Scott, Paykel, Morriss, Bentall, Kinderman, Johnson, Abbott and Hazlehurst2006.), Participants were approached by their own trial therapists, given information about this secondary study and asked to consent to participate. The first 10 participants to give consent to each of the four trial therapists were included. Participants were a subgroup drawn from a large scale randomized control trial, and as such we can be confident that the participants were correctly diagnosed with bipolar disorder. All participants were either in remission, or experiencing depression or low level hypomania at the time of interview. No participants were manic at the time of interview.
Procedure
Assessment of hallucinations. As there is no existing reliable document for the purpose of gathering information on hallucinations in bipolar disorder, the authors chose to adapt Chadwick and Birchwood's (Reference Chadwick and Birchwood1995) Belief About Voices Questionnaire. Hallucinations were defined as true or pseudohallucinations according to the established criteria set by the KGV assessment (Krawiecka, Goldberg and Vaughn, Reference Krawiecka, Goldberg and Vaughn1977).
Detailed subjective descriptions of hallucinatory experiences were recorded during the interviews and allocated into one of four categories reflecting previous observations in the literature related to hallucinations in bipolar disorder. The categories were: hallucinations caused by illness; hallucinations caused by a magical/spiritual source, or caused by an omnipotent malevolent source; hallucinations caused by stress; hallucinations caused as a punishment and any other source. In order to ensure that the attributions had been categorized correctly, a researcher unconnected with the study and with extensive experience of bipolar disorder (KT) reclassified the attributions. Inter-rater reliability was 90% (Kappa = 8.75).
Results and discussion
Demographic information
Sixteen participants were men and 24 women, with a mean age of 42 (SD 10.02) and a mean age of illness onset of 26 (SD 8.35). Participants reported a mean of 6.85 (SD 3.39) episodes of depression and a mean of 4.79 (SD 3.54) episodes of mania/hypomania.
Prevalence, nature and appraisals of hallucinations
Nineteen of the 40 participants (47.5%) reported the experience of true hallucinations. Of these, visual hallucinations were most commonly reported; 17 out of 19 participants reported visual hallucinations, predominantly in manic or hypomanic states (N = 16). Conversely, of the 8 participants reporting auditory hallucinations, 7 reported that they did so when depressed.
All participants also reported pseudohallucinations (often pseudohallucinations taking a visual form in the manic prodrome), suggesting that rather than being discreet events, hallucinations may lie on a continuum with normal perception in a manner suggested by some cognitive models of psychosis (Bentall, Reference Bentall2003.) Frequencies are reported in Table 1.
Table 1. Subjective descriptions
Attributions
The most novel aspect of this pilot study concerns participants' attributions for hallucinations. First, the overwhelming majority of the participants held dual attributions for hallucinations; only one-fifth of participants reported that they attributed hallucinations to illness when in episode. However, all but two participants attributed hallucinations to illness when euthymic. A major difference between attributions for hallucinations in bipolar disorder and attributions for hallucinations in the schizophrenia spectrum concerns omnipotence/malevolence. In Chadwick and Birchwood's (Reference Chadwick and Birchwood1995) study of attributions for hallucinations in the schizophrenia spectrum, all participants reported their hallucinations to come from an omnipotent and malevolent source, and that this was a significant cause of distress. In this bipolar disorder sample only one-third of participants reported omnipotent/malevolent attributions in episode, and post episode only one participant continued to hold omnipotent/malevolent attributions. Interestingly, all but one of the participants who reported omnipotent/malevolent attributions also all reported auditory hallucinations, and reported hallucinating when depressed. This suggests that distress caused by omnipotent/malevolent attributions may be predominantly linked to the experience of hearing voices, or to the experience of depression or to both. Given the suggested association between auditory hallucinations and childhood trauma (Hammersley and Fox, Reference Hammersley, Fox, Larkins and Morrison2006), it would be interesting for future research to investigate any possible associations between omnipotent/malevolent attributions and trauma history. Similarly, of the participants who held magical/extrasensory attributions for their hallucinations, all reported visual hallucinations, and all but one reported hallucinating when manic or hypomanic on at least one occasion. Details of content and form of hallucinations, and mood state in which hallucinations occurred, and attributions in and out of episode can be seen in Table 1.
The main weaknesses of this study are the small sample size and the reliance on retrospective descriptions of attributions for hallucinations, which may have been subject to recall bias and the lack of a pre-existing valid and reliable data collection instrument. However, the study does offer new and potentially fruitful data in research that may inform the growing desire to incorporate psychological therapies into the medical treatment of bipolar affective disorder; in particular, concerning models that place emphasis on interpretations and attributions of internal experiences in bipolar affective disorder such as the SPAARS model (Jones, Sellwood and McGovern, Reference Jones, Sellwood and McGovern2005).
Acknowledgements
The authors would like to thank Professor Steven Jones and Dr Fiona Lobban from the Spectrum Centre at Lancaster University for their assistance in the preparation of this manuscript. A full version of this brief report is available on request from the first author. p.hammersley@lancaster.ac.uk
Comments
No Comments have been published for this article.