Using Smartphones to Measure Symptoms of Schizophrenia in Patients Enrolled in a Clinical Trial
In this poster you will learn:
- Traditionally, assessments of symptom and disease severity in individuals with schizophrenia have been highly subjective, limiting validity as well as intra- and inter-rater reliability. This diminishes sensitivity to change, increases time and costs associated with data collection, decreases power and increases the sample size required to detect treatment effects.
- The requirement for a live rater who completes a lengthy interview limits the ability to conduct assessments frequently during a clinical trial as well the breadth of ecological sampling that is possible, and places significant burden on both study participants and clinicians.
- Novel, low-burden methods capable of sensitively measuring disease severity and treatment efficacy with greater ecological validity are needed.
- Negative symptoms of schizophrenia represent a cluster of debilitating symptoms for which treatment remains inadequate. They include diminished normal patient behaviors with decreased emotional, facial and verbal expressivity, decreased amount and content of speech, reduction in interests, social interactions, movement. Together these symptoms profoundly impact overall functioning.
- Advances in artificial intelligence (AI) and machine learning (ML) when applied to smartphone technology allow these behaviors to be measured with high accuracy and sensitivity. We outline a protocol design that leverages this technology to begin to establish the validity and reliability of this methodology for measuring clinically relevant patient behaviors in a clinical trial involving individuals with schizophrenia.