What is being measured by formal thought disorder scales? An item-level content analysis.

Clinical Snapshot

Key Findings

• P Value: Not applicable for this type of analysis

• Effect Size: Weak content overlap across all scale pairs (specific JSI values not provided in abstract)

• Primary Outcome: Content overlap between FTD scales measured by Jaccard Similarity Index

• Nnt Or Sensitivity: 56 discrete FTD phenomena identified; 20% of features were idiosyncratic to single scales

• Confidence Interval: Not reported in abstract

Clinical Application

High feasibility - provides practical framework for standardizing FTD assessment in clinical practice Highly relevant to Australian mental health services; aligns with RANZCP guidelines emphasizing standardized assessment tools; could inform TGA considerations for digital mental health tools measuring FTD Clinicians and researchers working with patients with psychotic disorders, particularly schizophrenia and related conditions

Abstract

BACKGROUND: Formal thought disorder (FTD), characterized by disruptions in the flow and form of thought, is a core feature of psychosis. But its measurement is fragmented across numerous rating scales, leading to its continued neglect in both research and clinical practice. To determine if different FTD scales measure the same underlying construct, we need to assess the degree to which the content of commonly used FTD scales overlaps with each other. METHODS: We conducted a systematic review to identify all standardized, clinician-rated scales used to measure FTD in psychotic disorders. From this set, we extracted individual items and derived a consensus list of 56 discrete FTD phenomena. Two independent clinical experts conducted item-to-item mapping for every scale item onto these FTD phenomena. Content overlap between scales was quantified using Jaccard Similarity Index (JSI). We determined the overall coverage achieved via several combinations of FTD scales. RESULTS: The 15 scales, comprising 207 items, showed weak content overlap. The mean JSI across all scale pairs was low, and no single phenomenon featured across all scales. While some core FTD phenomena (e.g. 'incoherence', 'poor speech content', 'drifting-off') were represented in many scales, 20% of all identified features were idiosyncratic, appearing in only one scale. CONCLUSIONS: Existing FTD rating scales capture a wide but heterogeneous array of symptoms with poor content overlap. This lack of harmonization challenges the comparability of mechanistic and interventional studies. We highlight the need for a consensus-based, standardized measurement of FTD and provide a comprehensive checklist to advance the research and clinical practice.

References

1. 1.Sreeraj, V. S., Voppel, A., Venkatasubramanian, G., & Palaniyappan, L. (2026). What is being measured by formal thought disorder scales? An item-level content analysis. Psychological Medicine. https://doi.org/10.1017/S0033291726104152