Sex differences in schizophrenia spectrum disorders: insights from the DiAPAson study using a data-driven approach

Area of Science:

• Neuroscience
• Psychiatry
• Machine Learning

Background:

• Schizophrenia Spectrum Disorders (SSD) exhibit known sex differences in onset and symptom presentation.
• Areas like time perception and positivity in SSD remain underexplored.
• Machine learning applications in understanding these sex differences are limited.

Purpose of the Study:

• To investigate sex differences in Italian patients with SSD.
• To utilize a data-driven approach, including machine learning, to identify patient clusters.
• To explore underexplored aspects such as time perception and positivity in relation to sex.

Main Methods:

• Assessed 619 Italian patients with SSD (198 females, 421 males) using standardized clinical tools.
• Collected data on demographics, clinical characteristics, symptoms, functioning, positivity, quality of life, and time perspective.
• Employed Principal Component Analysis (PCA) and Gaussian Mixture Model (GMM) for data-driven clustering and validation.

Main Results:

• Males were more likely to be single and less educated; females smoked more.
• Males exhibited more severe psychiatric and negative symptoms, with a less negative past perception.
• Females showed better interpersonal functioning; PCA and GMM identified two main sex-differentiated clusters.

Conclusions:

• Identified distinct sex differences in SSD, supporting tailored treatments for males and females.
• Machine learning highlighted unique SSD phenotypes, emphasizing the need for sex-specific interventions.
• Advocates for a multifaceted, interdisciplinary approach to address sex-based disparities in SSD and improve quality of life.