AI in Psychiatry

Digital health policy, telehealth oversight, and the future of AI-assisted mental health care — research from Columbia University and Stanford Medicine

The rapid expansion of telepsychiatry companies has outpaced regulatory oversight. In my work published in JAMA Psychiatry — co-authored with colleagues at Stanford Medicine — I documented how some digital platforms prioritize visit volume over patient safety, prescribing stimulants and other controlled substances with minimal clinical evaluation.

As AI tools increasingly enter psychiatric practice, the questions my research addresses become more urgent: How do we integrate technology without compromising clinical standards? What regulatory frameworks do we need? And how do we ensure AI serves patients, not just profits?

Published Research

Integrating Telehealth into the Mental Health Ecosystem

JAMA Psychiatry • Viewpoint • Sultan RS, Zhang AW, Singh MK

This paper examines the intersection of rapid telehealth expansion and psychiatric care quality. My co-authors at Stanford Medicine and I analyzed how digital health companies like Cerebral and Done scaled quickly by prescribing controlled substances — including stimulants for ADHD — with evaluations that fell far below clinical standards. We proposed a framework for integrating telehealth responsibly, including minimum evaluation standards, outcome monitoring requirements, and regulatory recommendations.

Read the full analysis

Key Findings & Recommendations

The Problem

  • Some telehealth platforms prescribe controlled substances after evaluations lasting under 10 minutes
  • Regulatory oversight has not matched the speed of digital health expansion
  • Patient safety concerns include inadequate monitoring of side effects and drug interactions
  • Volume-based business models incentivize rapid prescribing over thorough evaluation

The Framework

  • Minimum evaluation standards before prescribing controlled substances via telehealth
  • Mandatory outcome monitoring and follow-up protocols
  • Regulatory requirements for digital health platforms treating psychiatric conditions
  • Quality metrics that prioritize patient outcomes over visit volume
  • Transparent reporting of prescribing practices and patient outcomes

Research Directions

As AI rapidly transforms healthcare, my work bridges the gap between technological innovation and responsible clinical implementation in psychiatry. Current areas of investigation include:

  • AI-assisted diagnosis — Can algorithms improve ADHD detection accuracy, particularly in underdiagnosed populations like women and adults?
  • Digital phenotyping — Using smartphone and wearable data for continuous mental health monitoring without replacing clinical assessment
  • LLMs in clinical decision support — The opportunities and necessary guardrails for using large language models in psychiatric practice
  • Ethical frameworks — Ensuring AI tools in psychiatry serve patients and maintain clinical standards, not just reduce costs
  • Regulatory policy — What oversight structures are needed as AI becomes embedded in mental health delivery

Frequently Asked Questions

What has Dr. Sultan published about AI in psychiatry?

I published a viewpoint in JAMA Psychiatry titled "Integrating Telehealth into the Mental Health Ecosystem," co-authored with researchers from Stanford Medicine. The paper examines how some digital psychiatry companies prioritize volume over safety, and proposes a framework for safe integration of AI and telehealth into psychiatric care.

What are the risks of telepsychiatry that Dr. Sultan's research identifies?

My JAMA Psychiatry research identified several risks: some digital platforms prescribe stimulants and other controlled substances with minimal evaluation (sometimes under 10 minutes), regulatory oversight hasn't kept pace with telehealth expansion, patient safety is compromised when volume is prioritized over quality, and many platforms lack adequate monitoring systems for side effects, drug interactions, and treatment outcomes.

How can AI be safely integrated into mental health care?

My framework includes: maintaining clinical oversight of AI-assisted decisions, establishing regulatory standards for digital health platforms, requiring adequate evaluation before prescribing controlled substances via telehealth, implementing quality metrics that prioritize patient outcomes over visit volume, and ensuring AI tools augment rather than replace clinical judgment. The goal is harnessing technology's potential while protecting patients.

Explore More Research

ADHD Research & Resources

411-cited JAMA study, comprehensive knowledge base, 70+ pages of clinical resources for patients and clinicians.

Cannabis & Mental Health

NIH NIDA-funded research on adolescent cannabis use, psychosis risk, and evidence-based resources.
All Publications