• We're very committed to research, and see how Wysa can work for different user groups and conditions.

     

    We need help. We're looking for partnerships with research and academic institutions around the world to help us create studies to extend Wysa's reach and impact.

     

    Please reach out to us at hello@wysa.ai.

     

    We'd love to hear from you!

    Read about Wysa Studies below..

  • Wysa Effectiveness and Engagement Studies

    At Wysa , we continuously evaluate Wysa performance and continue to test Wysa effectiveness, usability, and engagement through our research collaborations.

    Study Summary

    Along with external researchers, we conducted a real-world mixed methods study on the effectiveness and engagement of Wysa on anonymous global users.

    Key findings include:

     

    1. Wysa high usera group had significantly higher improvement (40% higher) compared to the low userb group. With a Mann-Whitney P=.03 and a moderate effect size of 0.63. This is comparable with effect sizes for online human therapy for depression.
    2. 79% of users in the high user group improved on their symptoms of depression.
    3. 68% of the user feedback found Wysa experience helpful and encouraging.

    a High users of Wysa: Those who engaged with Wysa on two screening days (separated by 14 or more days) as well as at least once in-between those days (n= 108).

    b Low users of Wysa: Those who only engaged with Wysa only on the two screening days but never engaged in-between (n= 27).

     

    Read the peer-reviewed study published in the open access journal JMIR mHealth & uHealth1.

    Study Results

    A within-group analysis showed that users who engaged more with Wysa showed a higher average reduction in depression symptoms (reduction of 31% in pre scores) as compared with those who engaged less with Wysa (reduction of 18% in pre scores).

     

    A between-group analysis revealed that the average impact or improvement (pre- minus post- depression scores) among the high user group was significantly higher (mean 5.84 [SD 6.66]) compared with the low user group (mean 3.52 [SD 6.15]); with a Mann-Whitney P=.03 and with a moderate effect size of 0.63.

     

    1 Inkster, B; Sarda, S; Subramanian, V (2018). "An Empathy-Driven, Conversational Artificial Intelligence Agent (Wysa) for Digital Mental Well-Being: Real-World Data Evaluation Mixed-Methods Study". JMIR Mhealth Uhealth. 6 (11): e12106. PMID:30470676.

    URL: https://mhealth.jmir.org/2018/11/e12106/

All Posts
×