An evaluation of DeepSeek and healthcare professionals’ Q&A capabilities in improving patient-family satisfaction in the ICU
-
By
-
Min Li
-
Ruo-Yu Li
-
Wei-Jie Lai
-
Feng-Ling Yang
-
Cui-Bing Liu
-
June 8, 2026
-
Clinical Scorecard: Assessing the Impact of DeepSeek and Clinician-Family Communication on Patient and Family Satisfaction in the Intensive Care Unit
At a Glance
| Category | Detail |
|---|
| Condition | |
| Key Mechanisms | Use of DeepSeek, a large language model, to enhance clinician-family communication while ensuring clinician oversight. |
| Target Population | |
| Care Setting | |
Key Highlights
- DeepSeek outperformed clinicians in information completeness, clarity, and empathy (p < 0.05).
- 66.7% of family members reported satisfaction with DeepSeek responses (statistical significance needed).
- No significant differences in medical accuracy or safety assessment between DeepSeek and clinicians.
Guideline-Based Recommendations
Diagnosis
Management
- Establish protocols for integrating AI tools in clinical workflows.
Monitoring & Follow-up
- Regularly assess AI performance and its impact on communication outcomes.
Risks
Patient & Prescribing Data
Families of ICU patients seeking information and support.
AI-assisted communication may improve satisfaction but requires clinician oversight.
Clinical Best Practices
- Utilize AI tools like DeepSeek to enhance communication with patient families.
- Ensure clinician oversight in AI-generated communications to maintain safety and accuracy.
- Provide training for clinicians on effective use of AI tools.
Related Resources & Content
