A case series exploring possible predictors and mechanisms of change in hearing voices groups



Objectives. Hearing voices groups (HVGs) are increasingly common in National Health Services and are often preferred to individual therapy by both service users and providers. Whilst a range of approaches exist, mixed results have been found and only cognitive behaviour therapy (CBT) was supported by well-controlled trials. This study aimed to explore possible predictors and mechanisms of change in a seven-session CBT group for voice hearers.

Design. An exploratory case series design was used.

Method. Fifteen outpatients with a diagnosis of schizophrenia or schizoaffective disorder completed a HVG and carried out weekly measures of distress, negative beliefs about voices, self-esteem, effective coping strategies, and activity levels. Visual inspection and quantitative rules were used to group participants with similar results and cross-correlations and t tests were used to verify key findings.

Results. Several pathways emerged across therapy. Despite measurement frequency, changes on different outcomes tended to occur simultaneously, making conclusions about mechanisms difficult. However, changes in beliefs about voice malevolence and omnipotence correlated most frequently with changes in distress. Visual analysis indicated 53% of participants improved on a measured outcome but satisfaction scores were higher, with 93% feeling the group helped them deal with their problems more effectively. Clients especially valued the chance to meet similar others.

Conclusions. The results suggest HVGs are valued by clients, regardless of their background or symptoms. HVGs should emphasize testing negative beliefs about voices and allow space for supportive discussions between clients. Recommendations for future research are discussed, including consideration of benefits not detected by outcome measures.

Practitioner Points

  • • The evidence for Hearing voices groups (HVGs) is mixed, despite their popularity. There is therefore a need for process research to enhance efficacy.
  • • Within our case series (N= 15) employing weekly measurement, changes in negative beliefs about voices correlated more closely with distress than changes in coping strategies, activity levels, or self-esteem.
  • • Baseline symptom scores could not predict dropouts from therapy, indicating clients should not be excluded based on demographic or symptom scores.
  • • Satisfaction data indicated outcome measures did not capture all the benefits of HVGs and most clients valued normalization more than any change in outcome.