Anna Jijo, Maria Jijo, Swetha Manoj and Anitha Veerachamy
Aim This study aims to evaluate and compare the quality of educational content generated by two artificial intelligence tools— ChatGPT and DeepSeekAI—for educating parents about paediatric respiratory illnesses- croup, bronchiolitis, pneumonia, pertussis.
Methodology A cross-sectional analysis was conducted using AI-generated educational content. Each response was assessed for word and sentence count, average words per sentence, syllables per word, readability (Flesch Reading Ease Score and Grade Level), similarity percentage (Quillbot) and reliability (modified Discern Score). Statistical analysis was performed using independent sample t-tests, with p<0.05 considered significant.
Results DeepSeekAI responses were longer (mean word count: 422 vs. 333.75) and included more sentences. But no statistically significant differences were found in any variable, including readability (Grade Level: 9.22 vs. 9.30; Ease Score: 41.17 vs. 40.97) and reliability (2.25 vs. 2.00). Similarity scores were also comparable between the two tools (36.02 vs 32.1)
Conclusion With the increasing use of artificial intelligence (AI) in health communication, tools like ChatgPT and DeepSeekAI offer potential for generating accessible and reliable educational material. Both ChatGPT and DeepSeekAI generated parent education materials of similar quality, readability, and reliability. The findings of this study suggest that AI tools can be utilized for developing parental educational content for common paediatric respiratory conditions.
Key Message We conducted this study to assess and compare the quality of educational content created by ChatgPT and DeepSeekAI for parent education on common paediatric respiratory illnesses; croup, bronchiolitis, pneumonia, pertussis. Both ChatGPT and DeepSeekAI generated parent education materials of similar quality, readability, and reliability which suggests that they can be used for parental educational content generation.