Utility of large language models as information tools for nursing care in gout: a comparative study of DeepSeek and ChatGPT - Summary - MDSpire

Utility of large language models as information tools for nursing care in gout: a comparative study of DeepSeek and ChatGPT

  • By

  • Xia Pan

  • Yali Wang

  • QiaoLan Yang

  • Jing Wang

  • Yun Tong

  • Duanfeng Zhang

  • Xiaofeng Lv

  • Chun Zheng

  • Miaoyin Wu

  • Tianwang Li

  • Li Tang

  • Zhengping Huang

  • May 7, 2026

  • 0 min

Share

Objective:

To assess and contrast the efficacy of DeepSeek-R1 and ChatGPT-4.0 as digital information resources for nursing practitioners in managing patients with gout.

Key Findings:
  • DeepSeek-R1 had a higher FKGL (13.04) compared to ChatGPT-4.0 (11.41), indicating lower readability.
  • ChatGPT-4.0 had a higher mDISCERN quality score (4.30) than DeepSeek-R1 (3.98), though not statistically significant.
  • DeepSeek-R1 provided more current citations (average age 3.57 years) compared to ChatGPT-4.0 (5.42 years).
  • DeepSeek-R1 had 4 invalid reference links, while both models cited clinical guidelines predominantly.
Interpretation:

Both LLMs provided high-quality, evidence-based responses, but ChatGPT-4.0 was more coherent in language, despite DeepSeek-R1 offering more recent references.

Limitations:
  • The study focused on a limited number of inquiries regarding gout.
  • The statistical significance of some findings was not established.
Conclusion:

While both models are effective in providing nursing information for gout management, ChatGPT-4.0's clarity and coherence make it a preferable resource despite DeepSeek-R1's more current references.

Original Source(s)

Related Content