Reliability and validity of the Korean version of the comprehensive assessment of at‐risk mental states

Abstract

Aim Although the comprehensive assessment of at‐risk mental states (CAARMS) is one of the most widely used instruments for identifying individuals meeting the criteria for ultra‐high risk (UHR) of developing psychosis, the psychometric properties of the Korean version of the CAARMS (CAARMS‐K) have not been studied to date. Thus, we tested the reliability, validity, and factor structure of the CAARMS‐K in a Korean population. Methods The CAARMS‐K was administered to 96 UHR individuals. The inter‐rater reliability and internal consistency of the CAARMS‐K were evaluated. In addition, its factor structure was investigated using principal‐axis factor analysis. Concurrent validity was examined by comparing the extracted CAARMS‐K factors with the five factors of the positive and negative syndrome scale (PANSS) using Spearman's correlation tests. The percentage of transition to psychosis was examined at 1‐ and 2‐year. Results The inter‐rater reliability was good (intraclass correlation = .89), and the internal consistency was acceptable (Cronbach's alpha = .812). A five‐factor solution demonstrated the best simple structure and accounted for 43.1% of all‐item variance. The five factors extracted from the factor analysis were similar to the PANSS five factors and significantly correlated with the factors of the PANSS. Among the 96 UHR individuals, 11 (11.5%) and 17 (17.7%) developed psychotic disorders during the 1‐ and 2‐year follow‐up, respectively. Conclusions The CAARMS‐K is a reliable and valid instrument for the assessment of UHR individuals in Korea.

Factorial structure of the Comprehensive Assessment of At-Risk Mental States in help-seeking individuals: mapping the structure and the prediction of subsequent transition to psychosis

“…In the CAARMS, Cognitive Change is a subscale comprising impaired attention, thought block, and racing thoughts, whereas Disorganized Speech includes two additional aspects of formal thought disorder: circumstantiality and tangentiality. Thus, our "cognitive-disorganized" factor manifests some resemblance to the "communication-attention and concentration problems" and "disorganization/cognitive" dimension found in one of the SOPS studies (32), in the other CAARMS studies (18,20), and also the disorganization dimension identified in studies of patients with established psychosis, which typically comprises formal thought disorder and attentional impairment (43). Indeed, the two symptoms most consistently related to disorganization in established schizophrenia -"odd or bizarre behavior" and ''conceptual disorganization'' (thought disorder)-load with the cognitive symptoms (subjective and objective cognitive change) in this sample, supporting the notion of a unique factor, i.e., the cognitive/disorganized factor.…”

“…Third, factor analysis of the CAARMS extracted six factors in our sample. Previously, factor analytic studies on the CAARMS revealed three underlying factors (negative-interpersonal, communicational-cognitive-behavioral disorganization, and perceptual-affective instability component) (17) and five factors (negative, anxiety, disorganization/cognitive, self-harm, and manic dimension) (18,20). Other studies conducted on a conceptually similar instrument, the Scale of Prodromal Symptoms (SOPS), have found a three-model solution (31,32).…”

“…To our knowledge, only three studies have explored the dimensional structure of CAARMS to better understand the UHR subjects' psychopathology, suggesting the presence of three (17) or five (18,20) factors.…”

Development of the PSYCHS: Positive SYmptoms and Diagnostic Criteria for the CAARMS Harmonized with the SIPS

“…Interrater reliability (IRR) for CHR‐P ascertainment has also been excellent, both for the SIPS (median kappa across 16 published samples 0.89) (Woods et al, 2019) and the CAARMS (median across three studies 0.845) (Fusar‐Poli et al, 2012; Miyakoshi et al, 2009; Paterlini et al, 2019). IRR for attenuated positive symptoms has also been excellent for both SIPS (median ICC across 21 published samples 0.88) (Woods et al, 2019) and CAARMS (median ICC or Pearson r across eight studies 0.89) (Braham et al, 2014; Fusar‐Poli et al, 2012; Lho et al, 2021; Miyakoshi et al, 2009; Paterlini et al, 2019; Wang et al, 2022; Yokusoglu et al, 2021; Yung et al, 2005).…”

Development of the PSYCHS: Positive SYmptoms and Diagnostic Criteria for the CAARMS Harmonized with the SIPS

“…Interrater reliability (IRR) for CHR-P ascertainment has also been excellent, both for the SIPS (median kappa across 16 published samples 0.89) (Woods et al, 2019) and the CAARMS (median across three studies 0.845) (Fusar-Poli et al, 2012;Miyakoshi et al, 2009;Paterlini et al, 2019). IRR for attenuated positive symptoms has also been excellent for the SIPS (median ICC across 21 published samples 0.88) (Woods et al, 2019) and CAARMS (median ICC or Pearson r across eight studies 0.89) (Braham et al, 2014;Fusar-Poli et al, 2012;Lho et al, 2021;Miyakoshi et al, 2009;Paterlini et al, 2019;Wang et al, 2022;Yokusoglu et al, 2021;Yung et al, 2005).…”