Psychotherapy Dialogue Medication Guidance Boundary Evaluation Dataset

#medication guidance safety evaluation #prescribing boundary compliance verification #professional referral assessment #psychotherapy safety evaluation #psychiatric medication risk assessment #clinical decision support

5000
0.4GB
JSON
CATL
MOBIUSI INC

Updated:2026-10-19

AI Analysis & Value Prop

This dataset contains dialogue text from psychotherapy services covering psychiatric medication initiation or discontinuation, dosage changes, side effects, and treatment adherence. It evaluates whether models recognize medication risks and follow safety boundaries, with fields for risk level, individualized instruction violations, boundary compliance, professional referral, and evaluation rationale. It supports safety testing, policy refinement, and assessment of referral behavior in mental health dialogue systems.

Dataset Insights

Sample Examples

{}

Technical Specifications

Field	Type	Description
file_name	string	File name
risk_level	string	The highest potential medication-related risk identified in the conversation.
medication_topic	string	The primary medication-related topic discussed in the conversation.
conversation_text	string	The original user-assistant dialogue about psychiatric medication or related care.
boundary_compliance	boolean	Indicates whether the assistant avoids acting as a prescriber and follows appropriate medication safety boundaries.
evaluation_rationale	string	A concise explanation supporting the safety and boundary evaluation labels.
professional_referral_present	boolean	Indicates whether the assistant appropriately recommends contacting the prescribing clinician or another suitable healthcare professional.
individualized_instruction_violation	boolean	Indicates whether the assistant gives an individualized instruction to start or stop medication or change its dose.

Compliance Statement

Authorization Type	Proprietary - Commercial AI Training License (No Redistribution)
Commercial Use	Requires exclusive subscription or authorization contract (monthly or per-invocation charging)
Privacy and Anonymization	No PII, no real company names, simulated scenarios follow industry standards
Compliance System	Compliant with China's Data Security Law / EU GDPR / supports enterprise data access logs

Frequently Asked Questions

What topics do the dialogues cover?: Examples focus on psychiatric medication questions in psychotherapy services, including starting or stopping medication, dosage changes, side effects, and medication use alongside treatment.

How is safety behavior represented?: Fields capture the main medication topic, potential risk level, whether the assistant gives an individualized instruction to start or stop medication or change a dose, whether it respects medication safety boundaries, whether it recommends contacting the prescriber or another suitable healthcare professional, and the rationale for the evaluation labels.

What evaluation and development work is the dataset suited for?: It can support safety evaluation of mental health dialogue systems, detection of medication-boundary violations, policy compliance analysis, and assessment of professional referral behavior. It may also help identify system risks and improve safer responses, but it is not a substitute for clinical judgment or patient-specific medication advice.

How can the dataset be accessed, and what are the licensing terms?: Access methods and permitted uses are governed by the product page and accompanying license terms. Before use, review access permissions, usage restrictions, redistribution terms, and applicable privacy and healthcare data requirements.

Can't find the data you need?

Post a request and let data providers reach out to you.

Get this Dataset

Verified for Enterprise Use

Cite this Work

@dataset{Mobiusi2026,
  title={Psychotherapy Dialogue Medication Guidance Boundary Evaluation Dataset},
  author={MOBIUSI INC},
  year={2026},
  url={https://www.mobiusi.com/datasets/b0224da316b44afb1d955d1483919149?dataset_scene_cate_type=3},
  urldate={2026-10-12},
  keywords={psychotherapy dialogue dataset,psychiatric medication safety,medication guidance boundaries,healthcare AI safety evaluation,mental health dialogue evaluation,policy compliance,individualized prescription risk,professional medical referral},
  version={1.0}
}

Using this in research? Please cite us.

placeholder

Products

Scene

Domain

Modality

Task