EPS: Early Psychosis Screener
Predict conversion to psychosis at 12 months
Schizophrenia affects about 1% of the world’s population, and there has been little progress on effective treatment. There is growing evidence that if individuals with attenuated psychotic syndrome (APS) or first episode psychosis (FEP) can be identified and treated early, their prognosis might be greatly improved.
TeleSage developed the EPSI (Early Psychosis Screener for Internet) to predict the likelihood of an individual’s conversion to psychosis within 12 months. The paper version of the screener (EPS-26) is available for free download here. The EPS-9 is a newer and free online assessment which uses the nine most informative self-report questions from the EPSI and machine learning to assess whether the individual might be advised to see a clinician or retake the assessment in the near future.
EPS-9: BRIEF ONLINE SCREENING
The EPS-9 is a brief survey that is available now and can be taken for free. It includes the most informative items from the EPSI. After completing the assessment, individuals may be advised that their results are most similar to those of people who were NOT at increased risk of developing a psychotic disorder. Since people’s results may change over time, individuals may be advised to seek rescreening in about six weeks. Some individuals may be advised that their results are most similar to those of people who might benefit from an in-person evaluation from a licensed clinician.
EPSI: ONLINE ADMINISTRATION
The EPSI estimates 12-month psychosis risk using 64 simple self-report survey items and machine learning. After completing the assessment, individuals may be advised that their results are most similar to those of people who were NOT at increased risk of developing a psychotic disorder. Since people’s results may change over time, individuals may be advised to seek rescreening in about six weeks. Some individuals may be advised that their results are most similar to those of people who might benefit from an in-person evaluation from a licensed clinician. The results of the screening can be brought into a clinician’s office for further assessment.
EPS-26: PAPER ADMINISTRATION
The paper version of the Early Psychosis Screener consists of a series of simple self-report screening questions. If the individual scores above a certain threshold, they are considered to be at greater risk for psychosis.
The EPS-26 is available for free download.
Regardless of the results of any of these self-report surveys, if an individual believes he/she is suffering from a mental illness, he/she should seek an in-person evaluation from a licensed clinician.
“In their paper “The Early Psychosis Screener (EPS): Quantitative Validation against the SIPS Using “Machine Learning”, Brodey et al. (2018) share compelling preliminary evidence in support of their newly created Early Psychosis Screener (EPS). Creation of the EPS differed relative to other more established measures, as an expert panel was employed for item development, modification, and selection, and cognitive interviewing techniques were implemented before field testing to promote stronger construct validity, with the ultimate goal of improving accuracy of true risk prediction. The EPS also benefited from having North America’s largest CHR research network in NAPLS, as well as COPE, as its validation sample. Additionally, a machine learning analytic approach leveraged cutting-edge statistics to optimize the number and type of items included in the measure to improve efficiency. An important set of questions for psychosis-risk screening research is to determine what screener to use, when, with whom, and how. The EPS should now be added to the conversation.”
UNIVERSITY OF MARYLAND, DEPARTMENT OF PSYCHIATRY
Initial Item Development
The core principles for developing items included:
- Use short sentences e.g. “My thoughts were being controlled.”
- Use standard response sets e.g. Never, Rarely, Sometimes, Often, Always
- Use single 30-day timeframe (symptoms can fluctuate)
- Limit to fifth-grade reading level
- Avoid idiomatic language
- Avoid double negatives e.g. ‘not’, ‘never’
- Avoid words with multiple meanings
- Request direct experiential evidence
- Avoid judgmental language
To create the EPSI, we reviewed the assessment literature on early psychosis as well as psychosis and authored 490 items that might be predictive of psychotic conversion. To assure breadth of coverage, we fit these items into 25 general bins:
- Thought Broadcasting and Telepathy
- Ideas of Reference
- Persecutory and Paranoid
- Thought Insertion, Deletion, and Guilt
- De-realization and Erotomania
- Mild Auditory Hallucinations
- Prominent Auditory Hallucinations
- Visual and Tactile Hallucinations
- Affective Flattening
- Avolition and ADLs
- Concentration and Confusion
- Drug Use
- Alcohol Use
- Social Functioning
- Work and School Functioning
- Social Functioning Relative to Period 12 Months Prior
Expert Panel Review
Reviewed by eight members, including Michael First, MD
- Rate items 1-3 on clarity and centrality
- Conduct three rounds of review; items deleted, revised, added
- Develop instructions
- Reduce 490 items to 301 items that scored 2.5+
- 7 NAPLS (Calgary, Emory, UCLA, UCSD, UNC, Yale, Zucker Hillside)
- Columbia University
- One non-clinical control site (9 IRBs)
All clinical participants received SIPS screening:
- SIPS 0,1,2 = Clinically Low Risk (CLR)
- SIPS 3,4,5 = Clinically High Risk (CHR)
- SIPS 6 = First Episode Psychosis (FEP)
Attempted to follow all CHR patients 12 to 24 months or until conversion if sooner
A priori analytic plan:
Spectral Clustering Analysis plus Non-linear Support Vector Machine (SVM)