The Clinical Outcomes in Measurement-based care Assessment of Symptoms and Severity (COMPASS) Scale (COMPASS)
Description
The COMPASS scale is a psychopathology rating instrument, developed to assess individuals with first episode psychotic disorders in community settings. The COMPASS scale is available in different lengths, derived from the original 12-item version using Mokken scale analysis:
1. 12-Item Version: The original assessment used in the RAISE-ETP study.
2. 10-Item Version: This version was derived by removing two items—Elevated/Expansive Mood and Impulsive Behavior—due to their low statistical scalability coefficients in the RAISE-ETP data.
3. 5-Item Version: A shorter version focused solely on positive and negative symptoms (Unusual Thought Content, Hallucinations, Conceptual Disorganization, Apathy, and Asociality) was also derived and met unidimensionality criteria. This version is considered useful in settings where minimizing assessment time is the primary concern.
Test Features
Administration details
Suggested in the original paper to use the time-frame ‘since the last treatment visit’.
Approx time: 05-15 minutesAssessment Reference (for this version):
- Robinson, D. G., John, M., Miller, A. L., Schooler, N. R., & Kane, J. M. (2024). The COMPASS scale for the assessment of individuals with first episode psychotic disorders. Schizophrenia research, 274, 307–314. https://www.sciencedirect.com/science/article/pii/S0920996424004511?via=ihub
Accessing the assessment and manual information
License details
The 12-Item COMPASS is in the public domain and free to use. It can be accessed in the Supplementary Materials of the original article.
Reliability and Validity
- As summarised from development paper: The preferred 10-item version demonstrated improved unidimensionality (high scalability coefficients, H>0.3) and successfully met the requirement for monotonicity, indicating it is statistically robust for ordering patients based on their sum scores. Furthermore, items assessing core psychotic symptoms (e.g., Suspiciousness, Hallucinations, Unusual Thought Content) showed high test-retest reliability (correlations >0.7). This information was based on data collected from community clinics, in first-episode psychosis patients in the US.