Reproducibility of Cognitive Profiles in Psychosis Using Cluster Analysis

Abstract

We found evidence of four cognitive subgroups of patients with psychosis, with cognitive profiles that map closely to those produced in our previous work. Clusters were associated with clinical and community variables and a measure of premorbid functioning, suggesting that they reflect meaningful groupings: replicable, and related to clinical presentation and functional outcomes. (JINS, 2018, 24, 382-390).

Implications of data-driven analyses for personalized therapy in psychosis: a systematic review of cluster- and trajectory-based modelling studies

“…In addition, two studies were conducted on negative symptoms 29,97 , one study on positive symptom 98 , three studies on positive and negative symptoms 35,99,100 , and three studies on positive and negative schizotypy. 67,101,102 The reported clustering methods were K-means or non-hierarchical clustering analysis 35 32,85,86,91,93 reported four clusters and one study 94 discovered five clusters of patients. One study found three clusters in unaffected siblings based on neurocognitive function.…”

“…101,102 In addition, poor cognitive impairment cluster(s) were predicted by age, gender, non-Caucasian ethnicity, low socioeconomic and educational status, poor premorbid adjustment, low premorbid and current IQ, early age of illness onset, long duration of illness, severe positive and negative symptoms, poor social cognition, high antipsychotics dosage, use of second generation antipsychotics, and poor functioning and poor quality of life. 32,37,65,66,[79][80][81][82][83][84][85][86][87][88][89][90][91][92][93][94][95][96] Siblings subgroups with impaired neurocognitive function were predicted by young age, low educational status, low IQ, poor premorbid adjustment, and severe positive schizotypy (Figure 4). 66 Figure 4: Schizophrenia spectrum circle illustrating predictors of symptomatic clusters.…”

“…Therefore, the models can be unstable and parameter estimates of clinical symptoms may not converge to a consistent set of subgroups and lack a direct relationship to clinical reality. 73,91,105 For example, intermediate clusters and trajectories substantially vary between studies. 91 We advocate that study results should be applicable, comparable, generalizable and interpretable into clinical practice.…”

“…73,91,105 For example, intermediate clusters and trajectories substantially vary between studies. 91 We advocate that study results should be applicable, comparable, generalizable and interpretable into clinical practice. We also propose to validate models using additional comparable statistical methods, combine statistical methods of subtyping with empirical methods, and work together with clinicians to create a common understanding and clinically relevant clustering or trajectories nomenclature.…”

Cognitive and clinical predictors of community functioning across the psychoses

“…Within groups, heterogeneity of cognitive performance is high, suggesting that even within diagnosis not all patients exhibit the same pattern or degree of impairment. Empirical grouping methods, such as cluster analysis, have been used to identify groups of patients with similar cognitive profiles across the psychoses, and have found distinct cognitive profiles, including global impairment, moderate or mixed impairment, and cognitively “normal” functioning, which are detected across diagnostic groups and in transdiagnostic samples (Burdick et al, ; Lewandowski, Baker, McCarthy, Norris, & Öngür, ; Lewandowski, Sperry, Cohen, & Ongür, ; Van Rheenen et al, ). However, in these studies patients with BD with psychosis tend to be overrepresented in cognitively intact subgroups, whereas patients with SZ are overrepresented in globally impaired subgroups (Lewandowski et al, ).…”