Back

Leveraging artificial Intelligence and online psychotherapy to achieve efficient and coordinated services within a healthcare setting: A quality improvement initiative

Stephenson, C.; Eadie, J.; Holmes, C.; Asadpour, K.; Gutierrez, G.; Kumar, A.; Jagayat, J.; Patel, C.; Sajid, S.; Knyahnytskyi, O.; Yang, M.; Reshetukha, T.; Moi, C.; Barrett, T.; Shirazi, A.; Verter, V.; Soares, C.; Omrani, M.; Alavi, N.

2025-02-25 psychiatry and clinical psychology
10.1101/2025.02.22.25321970 medRxiv
Show abstract

ObjectivesThis study aimed to implement an artificial intelligence-assisted psychiatric triage program, assessing its impact on efficiency and resource optimization. MethodsThis quality improvement initiative recruited patients on the waitlist for psychiatric evaluation at an outpatient hospital. Participants (n=101) completed a digital triage module that used natural language processing and machine learning to recommend a care intensity level and a disorder-specific digital psychotherapy program. A psychiatrist also assessed the same information and the decisions for care intensity and psychotherapy programs were compared with the artificial intelligence recommendations. ResultsThe overall wait time to receive care decreased by 71.43% due to this initiative. Additionally, participants received psychological care within three weeks after completing the triage module. In 71.29% of the cases, the artificial intelligence-assisted triage program and the psychiatrist suggested the same treatment intensity and psychotherapy program. Additionally, 63.29% of participants allocated to lower-intensity treatment plans by the AI-assisted triage program did not require psychiatric consultation later. ConclusionsUsing artificial intelligence to expedite psychiatric triaging is a promising solution to address long wait times for mental health care. With future accuracy refinements, this could be a valuable tool to implement in hospital settings to assist care teams and improve mental health care. This could result in increased care capacity and improved workflow and decision-making.

Matching journals

The top 7 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.