Back to the library

Key question New

AI-processed · Author-approved

Can simulated consultations with conversational chatbots increase medical students' and physicians' willingness to prescribe digital health applications?

“Do you have a Digital Therapeutic for that?” Design and Evaluation of Conversational Virtual Patients for DTx Prescription Training in Medical Education
Laura Ebers, Mustapha Addam, Marc Roger Herzog, et al. · International Conference on Information Systems (ICIS) · 2026
Illustration for: Can simulated consultations with conversational chatbots increase medical students' and physicians' willingness to prescribe digital health applications?

Loads Spotify's player; Spotify may set cookies. This device remembers your choice.

Key findings

  • Participants reported satisfaction significantly above the scale midpoint with the conversational simulation format, rating it as engaging and motivating.
  • Interacting with the virtual patient produced statistically significant increases in participants' willingness to engage with digital therapeutics and their willingness to prescribe them; these were self-reported shifts, measured with a retrospective pre-assessment, among the 19 participants whose consultation reached the DTx episode.
  • Hypotheses regarding perceived authenticity, perceived skill development, and overall digital therapeutic knowledge gain were not supported: limited conversational naturalness explains the authenticity result, while the paper attributes the non-supported results for skill development and knowledge gain to the brevity of a single session.
  • The observed shifts reflect increased awareness rather than comprehensive knowledge acquisition within a preliminary sample of 29 participants.

What it means for you

CIO / IT Executive

Procure and pilot a conversational virtual patient platform to simulate digital therapeutics consultations for clinical staff.

IT Manager

Configure chatbot persona dialogue trees to specifically handle migraine consultation scenarios and guide users toward digital therapeutic episodes.

Business Strategist

Develop a scalable training product targeting medical students and physicians to address low statutory prescription rates for digital therapeutics.

Researcher

Design longitudinal multi-session studies to measure whether extended interaction with conversational virtual patients yields actual skill development and knowledge gain beyond initial awareness.

Policymaker

Integrate conversational simulation tools into medical education curricula to boost healthcare providers' willingness to prescribe statutory digital therapeutics.

Summary

This study evaluates whether interactive conversational virtual patients can prepare healthcare providers to evaluate and prescribe digital therapeutics. The researchers designed four chatbot patient personas simulating migraine consultations and tested them in an online experiment with 29 medical students and physicians using a multi-level training evaluation framework.

Problem

In Germany, DTx (DiGA) can be prescribed and are reimbursed by statutory health insurance, yet prescription rates remain low. Digital therapeutics (DTx) remain underused because clinicians lack familiarity, confidence, and formal training regarding their clinical indication and prescription processes. At the same time, medical education rarely covers digital applications, and traditional training with human simulated patients is expensive and hard to scale.

Transcript
Host: Welcome to A.I.S. Insights — powered by Living Knowledge. I’m Anna Ivy Summers. Today, we are discussing a study titled “Do you have a Digital Therapeutic for that?” Design and Evaluation of Conversational Virtual Patients for DTx Prescription Training in Medical Education, authored by Laura Ebers, Mustapha Addam, Marc Roger Herzog, and Till J. Winkler. Alex, could you share what this study examines?

Expert: Hello Anna. This study investigates whether interactive, conversational virtual patients can prepare healthcare providers to evaluate and prescribe digital therapeutics, commonly referred to as DTx. The researchers developed four chatbot patient personas simulating migraine consultations and tested them in an online experiment with twenty-nine medical students and physicians.

Host: What is the primary problem this study addresses in healthcare?

Expert: In Germany, digital health applications can be prescribed and are reimbursed by statutory health insurance, yet prescription rates remain low. Clinicians frequently lack familiarity, confidence, and formal training regarding clinical indications and prescription workflows. At the same time, medical education rarely incorporates digital therapeutics, and traditional training with standardized patients, that is, trained individuals who portray patients, is expensive and hard to scale.

Host: How did the researchers approach this challenge methodologically?

Expert: They designed four virtual patient personas with varying migraine histories and technological familiarity. The twenty-nine participants conducted text-based medical consultations with these personas, followed by an evaluation based on a multi-level training framework to assess user reactions, learning outcomes, and willingness to prescribe.

Host: What were the key findings from this evaluation?

Expert: Satisfaction with the training format was rated significantly above the scale midpoint, with participants describing the simulation as engaging and motivating. In addition, among the nineteen participants whose consultation reached the DTx episode, the study recorded statistically significant increases in participants' willingness to engage with digital therapeutics and their willingness to prescribe them. It is important to note that these increases were self-reported, using a retrospective pre-assessment.

Host: Were there hypotheses that were not supported by the data?

Expert: Yes. Hypotheses regarding perceived authenticity, perceived skill development, and overall digital therapeutic knowledge gain were not supported. Within this preliminary sample of twenty-nine participants, the observed shifts reflect increased awareness rather than comprehensive knowledge acquisition.

Host: Looking at the business and practical implications, why does this matter for the digital health sector?

Expert: As an implication for digital health companies, conversational virtual patients represent a promising, scalable complement to standardized patients. Rather than relying entirely on conventional continuing medical education or static informational materials, organizations can use simulated interactions to introduce clinicians to digital health applications and their specific prescription pathways.

Host: What does the study suggest for future iterations of these simulation systems?

Expert: The study provides two distinct explanations for the limited outcomes. First, improving dialogue naturalness is needed for perceived authenticity. Second, building clinical confidence likely requires repeated or longer training rather than a single session.

Host: Conversational virtual patients demonstrate clear utility in raising awareness and openness toward digital therapeutics, while further refinement in conversational depth and multi-session design will be essential for deeper skill development. Alex, thank you for sharing these insights, and thank you to our listeners for joining us on A.I.S. Insights — powered by Living Knowledge.
Artificial Intelligence & Automation Information Systems & Management Human-Computer Interaction Data Analytics & Big Data Digital Transformation & Strategy Cybersecurity & Privacy Cloud & Infrastructure E-Commerce & Digital Platforms Operations & Supply Chain Sustainability & Green IS

Digital Therapeutics, Virtual Patients, Medical Education, Conversational Agents, Prescription Behavior, Digital Health