68/100

Moderate evidence

Cite
More
EmailLinkedInX

AI agreed with psychiatrists on workers' mental health-related presenteeism 72.2% of the time in one Japanese firm

Study title: Development of an artificial intelligence model to estimate psychiatrist-assessed mental health-related presenteeism

Evidence type
Controlled or validated comparison
Sources
1 sourcePeer-reviewed
Published
1 min read

Study title: Development of an artificial intelligence model to estimate psychiatrist-assessed mental health-related presenteeism

Objectives This study aimed to contribute to the development of an AI-based system that supports worker health and productivity by enabling early detection of presenteeism. We tested whether an AI model could assess mental health-related presenteeism with accuracy comparable to that of psychiatrists and whether the frequency of application use was comparable between avatar-based and real-person interfaces.

Read the study abstract

Objectives This study aimed to contribute to the development of an AI-based system that supports worker health and productivity by enabling early detection of presenteeism. We tested whether an AI model could assess mental health-related presenteeism with accuracy comparable to that of psychiatrists and whether the frequency of application use was comparable between avatar-based and real-person interfaces. Methods This study aimed to design a multivariable prediction model among white-collar employees in a Japan…

Development of an artificial intelligence model to estimate psychiatrist-assessed mental health-related presenteeism
Photo by Resume Genius on Unsplash · https://unsplash.com/@resumegenius?utm_source=truvace&utm_medium=referral

The gain

In one Japanese company, an AI model matched psychiatrists' ratings of mental health-related presenteeism with 72.2% accuracy across 1,631 video responses from 117 white-collar workers.

In brief

The primary outcome measure was the accuracy of the AI model in estimating workers' mental health-related presenteeism. Results The AI model achieved an overall accuracy of 72.2%, a macro F1 score of 0.64, and a weighted F1 score of 0.73 compared with psychiatrists' ratings.

Main points

  1. Objectives This study aimed to contribute to the development of an AI-based system that supports worker health and productivity by enabling early detection of presenteeism.
  2. We tested whether an AI model could assess mental health-related presenteeism with accuracy comparable to that of psychiatrists and whether the frequency of application use was comparable between avatar-based and real-person interfaces.
  3. Methods This study aimed to design a multivariable prediction model among white-collar employees in a Japanese company.

The full story

Objectives This study aimed to contribute to the development of an AI-based system that supports worker health and productivity by enabling early detection of presenteeism. We tested whether an AI model could assess mental health-related presenteeism with accuracy comparable to that of psychiatrists and whether the frequency of application use was comparable between avatar-based and real-person interfaces.

Methods This study aimed to design a multivariable prediction model among white-collar employees in a Japanese company. The participants comprised 117 white-collar workers who provided a total of 1,631 video responses to a standardized health-status question over a period of 10 working days.

Sources

1 on file
  1. Journal of Occupational Healthdoi:10.1093/joccuh/uiag055
    Peer-reviewed2026-09-21

Talk it over

Three questions to start on, two more if you have time
The AI model matched psychiatrists' ratings 72.2% of the time across 1,631 video responses from 117 white-collar workers. How would you weigh that level of agreement against the aim of early detection?

Start with a question

AI agreed with psychiatrists on workers' mental health-related presenteeism 72.2% of the time in one Japanese firmhttps://truvace.com/item/development-of-an-artificial-intelligence-model-to-estimate-psychiatrist-assesse

  1. The AI model matched psychiatrists' ratings 72.2% of the time across 1,631 video responses from 117 white-collar workers. How would you weigh that level of agreement against the aim of early detection?

    Answer in the thread
  2. All 117 workers came from one Japanese company. Who might be missing from that sample, and how could that change how well the model agrees with psychiatrists?

    Answer in the thread
  3. If you were one of the workers giving a daily video response over 10 working days, what would you want to know about who sees the AI's estimate?

    Answer in the thread
If you have more time
  • The researchers also compared avatar-based and real-person interfaces for how often workers used the application. What did they find, and how might the type of interface affect how honestly workers answer?

    Answer in the thread
  • In a workplace you know, who should decide whether a tool like this is used, and what evidence beyond agreement with psychiatrists would you ask for?

    Answer in the thread

What readers say