The authors proposed catatonia diagnostic criteria that require the presence of three neuropsychiatric symptom clusters, rated over 24 hours; this system differs from other symptom clustering proposals and is intended to increase diagnostic rigor over Bush-Francis Catatonia Rating Scale (BFCRS) or DSM-5 criteria. By applying new BFCRS item score thresholds, symptoms were clustered into three categories to comprise the Research Diagnostic Criteria for Catatonia (RDCC): akinesia (criterion A), unusual motor signs (criterion B), and behavioral signs (criterion C). RDCC symptom clusters were analyzed in four prospectively evaluated patient groups (delirium, medical, affective, and psychosis) (N=341). Use of the RDCC, compared with the DSM-5-TR and BFCRS, resulted in far fewer diagnoses of catatonia in the four patient groups: medical, N=1 out of 42 (2%); affective, N=1 out of 45 (2%); psychosis, N=3 out of 53 (6%); and delirium, N=0 out of 201. Permutations of the RDCC with more relaxed criteria were assessed, requiring either symptom thresholds or numbers of symptoms to meet criteria, resulting in catatonia rate gradations between those obtained with the RDCC and those obtained with current systems. The Cochrane Q test found that the DSM-5-TR was not dissimilar to the RDCC, if fulfilling numerical thresholds for criteria A-C, although any level of symptom severity was allowed. Confirmatory factor analysis with three goodness-of-fit indexes validated the RDCC. The RDCC requires akinetic symptoms on the basis of literature demonstrating their high BFCRS prevalence and exploratory factor analysis co-loadings, plus symptoms from unusual motor and behavioral signs. Compared with current lenient diagnostic approaches, having the symptoms required by the RDCC produced lower catatonia rates in the psychosis, affective, and medical groups and revealed no patients with catatonia in the delirium group. Subdividing DSM-5-TR symptoms into several different criteria may improve diagnosis. RDCC symptom clusters are both research data-based and amenable to further research for validation.
Read full abstract