Disentangling compliance with command hallucinations: Heterogeneity of voice intents and their clinical correlates.

Background: Earlier studies suggested that perceptions of voice intents (benevolence, malevolence) are associated with different psychological and behavioral responses including compliance with command hallucinations (CH). However, to our knowledge, no studies have examined the clinical differences between subgroups of clients with different perceptions of the intents of their CH. In order to better understand the risk for compliance with CH, our objectives were 1) to compare sociodemographic and clinical profiles of subgroups of clients (based on perceptions of CH intents); and 2) to investigate their specific associated risk factors for compliance with CH.

Method: We analyzed the MacArthur Violence Risk Assessment Study, focusing on 181 participants with psychosis reporting CH. Group comparisons and within-group ordinal logistic regression analyses were performed using sociodemographic and clinical measures such as the BPRS, BIS-11 and NAS-PI.

Results: Of the 181 participants, 102 (56.4%) reported having only malevolent voices, 14 (7.7%) rated them as benevolent only, 58 (32.03%) as benevolent and malevolent, and only 7 (3.86%) as neutral only. Results showed that individuals with malevolent voices had more emotional disturbance while those with benevolent CH had more severe positive psychotic symptoms and were more certain that they would comply in the future. Moreover, childhood physical abuse, belief about having to obey as well as psychotic symptoms significantly predict compliance with malevolent CH in a multivariate model.

Conclusions: Our results suggest that researchers and clinicians should consider perceptions of voice intents when both assessing risk of compliance with CH and developing relevant intervention targets.