A predictor model of treatment resistance in schizophrenia using data from electronic health records.

ABSTRACT:

Objectives

To develop a prognostic tool of treatment resistant schizophrenia (TRS) in a large and diverse clinical cohort, with comprehensive coverage of patients using mental health services in four London boroughs.

Methods

We used the Least Absolute Shrinkage and Selection Operator (LASSO) for time-to-event data, to develop a risk prediction model from the first antipsychotic prescription to the development of TRS, using data from electronic health records.

Results

We reviewed the clinical records of 1,515 patients with a schizophrenia spectrum disorder and observed that 253 (17%) developed TRS. The Cox LASSO survival model produced an internally validated Harrel's C index of 0.60. A Kaplan-Meier curve indicated that the hazard of developing TRS remained constant ov

SUBMITTER: Kadra-Scalzo G 

PROVIDER: S-EPMC9484642 | biostudies-literature | 2022

REPOSITORIES: biostudies-literature