Assessing the Impact of AI Education on Hispanic Healthcare Professionals’ Perceptions and Knowledge

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

General Impression

The authors present the result of a brief survey on perceptions of artificial intelligence conducted at the Puerto Rico Medical School pre- and post an AI course. The survey shows different levels of familiarity with the technology among participants and demonstrates that knowledge about AI increases after completion of the course. This being a one-time survey at one school, the results are not generalizable and do not add much to the fast-evolving public debate on artificial intelligence. Methodological flaws  further diminish the value of the manuscript and its limitation on subjective data without a suitable control cannot be overcome by a simple rewrite. Publication is thus not recommended.

Detailed critique

Introduction: The authors stress the present study is the first of its kind conducted at a Hispanic institution, which is an important feature. However, there is neither a mention of the demographics of the study population (i.e., do the participants actually identify as Hispanic?) nor a discussion of differences between the outcomes of the present study and the numerous reports of similar surveys in different populations.

Materials and Methods:

Participant selection, line 62. The background of study participants is unclear. The methods section says that participants were recruited from a rather large pool of administrator, IT professionals and students, with priority given to technical staff and investigators. It is thus not clear why the discussion focuses on Medical Students. Furthermore, the section also mentions that non-Spanish speaking individuals were included. No mention is made anywhere in the paper about the actual composition of the survey participant population in terms of age, background, sex, and self-reported ethnicity.

Description of intervention: There are no details on the intervention, other than the somewhat cryptic acronym for the title (AIML+HDR v2). How long was it, how were the learning outcomes assessed, what were the objectives?

Regulatory approval. Documentation of regulatory approval of the study is missing. What were the incentives for completing the research survey?

Results:

Participant demographics. Despite the title of the section, no description of the participant demographics is given. The included survey questions (Table 1) give reason to believe that the information was not collected.

Results in general: The description of the results is very extensive and repeated in the discussion section. There are a lot of data displayed, but there is not much effort in interpreting the findings. Factor analysis, effect size determination and an estimate of statistical power would have been far more useful than the numerous graphical representations of questions.

Learning outcomes: While it is not surprising that learning occurs when a student takes a course, it is not clear how effective the course was. Was there a control group? Were there objective measures of learning (i.e., quiz scores) that could have been more informative?  

Author Response

"Please see the attachment" 

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

I would like to thank the authors for allowing me to review this manuscript. I'm personally interested in the role of AI in Education, and this was an interesting manuscript to read. I just have a few suggestions to make, which I think will improve the overall reporting of the manuscript.

1. If you emphasize the type of institution in the abstract and the keywords (Hispanic), could you consider including it in the title of the manuscript? Do authors consider this relevant to undestand the soundness of the manuscript?

2. The introduction is clear and concise, it presents the justification of the study. However, if the authors want to include the specific demographic Hispanic population, they should at least mention why they believe this is a relevant factor.

3. You could re-structure the Methods section into less suybsections. Design-Participants-Outcomes-Statistical analysis would be more appropriate in my opinion. It will facilitate reading the manuscript.

4. Lines 62-64. If the sample is only composed by medical students, then this should be mentioned instead of "healthcare professionals" in the Introduction, to clarify to whom is the assessment conducted.

5. Lines 67-68. Remember this self-selected or convenient sample might be a strong limitation to draw conclusions in this study.

6. Line 71. What do you actually mean by "specialized subtitles"? How was the translation process conducted?

7. Line 74. Please inlude the full name of the acronym the first time it is mentioned

8. Lines 144-145. This sub-classification must be explained prior to the data analysis section.

9. Sample size calculation to reach sufficient statistical power is missing. Please detail how did authors calculate the minimal sample size required in this study. If this was not done, it is a limitation.

10. Lines 165-167. This should be described in the Methods section

11. Discussion. Do not divide this section. Usually the Discussion does not have subsections

12. Line 305. Do not repeat Results in the Discussion section, this is unnecessary.

13. Conclusions. Conclusions must be simple statements that address every objective. Write a simplified version of this section please.

Author Response

"Please see the attachment" 

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

Thank you for addressing my concerns, the article has improved.