Clinical Guide

How to Follow Up AI Screening for Suspected Tardive Dyskinesia

What should clinicians do after a remote AI screen is positive or unusable for suspected tardive dyskinesia?

A remote AI screen is only useful if clinicians have a clear response pathway for positive alerts and failed video capture. This guide summarizes the follow-up actions directly described in the article for patients taking antipsychotics after suspected TD is flagged or the video cannot be analyzed.

  1. Treat a positive screen as a prompt for clinician evaluation

    When the algorithm identifies suspected TD, move promptly to physician assessment rather than accepting the AI output as definitive. The article states that when suspected TD is identified, a psychiatrist is justified in evaluating the patient and determining the diagnosis and actions to follow. It also states that a health care professional's evaluation is essential to confirm the diagnosis required for prescribing treatment.

  2. Prioritize a formal AIMS assessment

    Arrange an AIMS evaluation by a trained care team professional when the alert is reported. The discussion states that an AIMS can be completed with greater urgency after an alert and that the tool may help clinicians provide monitoring at the standard pace or with increased frequency. The article repeatedly positions AIMS-based clinician assessment as the reference standard for follow-up.

  3. Escalate to in-person or telehealth AIMS when video capture is inadequate

    If the video cannot be analyzed because of network, camera, environmental, or user-following issues, do not rely on the remote screen. The limitations section states that in cases with unresolvable issues, the person can be referred for an in-person or telehealth AIMS with the provider. This is especially important because poor-quality video excluded 17% of participants in Study 3.

  4. Maintain clinical suspicion despite a low-risk or incomplete remote result

    Do not let a negative or low-risk remote result override concern when the clinical picture still suggests TD. The article acknowledges that the method can miss movement, especially if involvement is limited to toes or feet, because those regions are not directly assessed. Continued evaluation is warranted when symptom pattern or examination history raises concern outside the protocol's visual field.

Clinical Considerations

  • The article does not provide a full downstream treatment algorithm after diagnosis, only that clinicians should determine diagnosis and next actions after a positive screen.
  • Because the system is a screening tool, positive results should not be used alone to prescribe treatment.
  • Video-quality failure is common enough to require an alternate pathway, with 17% of participants excluded in the home-setting study.
  • Remote screening may miss isolated foot or toe movements because those areas are not directly captured.

Bottom Line

A positive or unusable remote TD screen should trigger prompt clinician AIMS-based follow-up, not independent treatment decisions or dismissal of concern.

Read full article
Physicians Postgraduate Press, Inc. (PPP) makes no warranties about the accuracy or completeness of any information published in The Journal of Clinical Psychiatry or other PPP materials, and disclaims liability for any use or non-use of that information. Clinicians should not rely solely on these materials and should exercise their own professional judgment when making patient care decisions on an individualized basis.