JMIR Publications

ABSTRACT

Background:

Few studies have estimated the prevalence of schizophrenia in the Japanese population. Previous estimations often relied on self-reported physician diagnoses or typical schizophrenia symptoms, which may have underestimated the true prevalence owing to stigma, poor insight, and lack of access to healthcare among respondents. We developed an artificial neural network (ANN)-based schizophrenia classification model (SZ classifier) using data from a large-scale Japanese web-based survey. The SZ classifier incorporates demographic data, health-related backgrounds, physical comorbidities, psychiatric comorbidities, and social comorbidities as feature variables. The SZ classifier's performance was evaluated through internal and external validations, and promising results were demonstrated. Our findings suggest that the ANN-based model has significant potential as an effective tool for classifying schizophrenia cases and estimating prevalence.

Objective:

This study aimed to estimate the prevalence of SZ by applying an SZ classifier to random samples from the Japanese population.

Methods:

A total of 750 random samples with age, sex, and regional distributions similar to Japan's demographic structure were selected and tested using the SZ classifier to calculate the crude prevalence of schizophrenia. We further refined this estimate by excluding false-positive cases and including false-negative cases to determine the actual prevalence of schizophrenia.

Results:

The crude prevalence of schizophrenia in the general population of Japan, estimated using the SZ classifier, was 8.3% (95% Confidence Interval (CI): 6.6%, 10.1%). After adjustment, the actual prevalence of schizophrenia in the general population was estimated to be 1.6% (95% CI: 0.7%, 2.5%).

Conclusions:

This estimated prevalence was slightly higher than that reported in previous studies, possibly due to a more comprehensive disease classification methodology or, conversely, model limitations. This study demonstrates the capability of an ANN-based model to improve the estimation of schizophrenia prevalence in the general population, offering a novel approach to public health analysis.