Feasibility and Validity Assessment of an AI-Generated Health Podcast: A Pilot Observational Study
试验速览
- 阶段
- 不适用
- 状态
- 招募中
- 入组人数
- 30
- 试验地点
- 1
- 主要终点
- Content Validity Index (CVI) of the Adapted Questionnaire
研究概览
简要总结
This study aims to explore a new way of delivering health information using an AI-generated podcast. The podcast, created with Google NotebookLM, uses verified content from the American Academy of Periodontology website to provide easy-to-understand information on gum health and prevention.
The goal is to determine whether this AI-generated podcast is a useful, engaging, and clear tool for educating the general public about health topics. Traditional health podcasts often feature expert interviews and can be lengthy, which sometimes limits their appeal and accessibility. By using AI to generate the podcast, investigator hope to offer a more standardized and concise presentation that avoids technical jargon.
To evaluate the podcast, investigator developed a questionnaire based on the Questionnaire for Assessing Educational Podcasts (QAEP). This questionnaire was adapted to better suit a non-specialist audience and covers four key areas: how easy the podcast is to access and use, the design and structure of the podcast, the clarity and completeness of the content, and the podcast's value as a learning tool.
Before using this questionnaire with the general public, investigator sent it to 10 experts in dentistry, public health, and communication for their review and feedback. Their input helped us make minor modifications to ensure the questionnaire is both clear and scientifically sound. After these revisions, investigator conducted a pilot study with 30 members of the general public who listened to the podcast and completed the questionnaire.
This study will assess the feasibility and validity of using an AI-generated podcast as a health education tool. The results will help determine if this approach can effectively improve public understanding of health information and may guide the future design of digital health communication strategies.
详细描述
Background and Rationale Recent advances in digital media have underscored the potential of podcasts as an innovative medium for disseminating healthcare information. Traditional health podcasts, while valuable, often suffer from limitations such as lengthy duration, inconsistent quality, and the use of complex medical jargon that may hinder public understanding. In contrast, artificial intelligence (AI)-driven content generation offers an opportunity to create concise, standardized, and accessible audio content. This study leverages Google NotebookLM to generate a podcast on gum health and prevention using publicly available information from the American Academy of Periodontology (AAP). By doing so, the research aims to explore whether an AI-generated health podcast can effectively enhance public health literacy.
Objectives
The primary objective of this pilot study is to evaluate the feasibility and validity of an AI-generated health podcast as an educational tool for the general public. Specific objectives include:
- Questionnaire Development: Develop and validate a structured questionnaire-adapted from the Questionnaire for Assessing Educational Podcasts (QAEP)-that captures public perceptions regarding podcast accessibility, design, content adequacy, and educational value.
- Reliability and Validity Testing: Assess the reliability and content validity of the newly developed questionnaire through expert evaluation and statistical analysis.
- Public Evaluation of Podcast Quality: Assess the general public's perceptions of the AI-generated health podcast, focusing on its accessibility, design, content quality, and overall educational value, using the validated questionnaire.
Study Design and Methods
研究设计
- 研究类型
- Observational
- 观察模型
- Other
- 时间视角
- Other
入排标准
- 年龄范围
- 18 Years 至 —（Adult, Older Adult）
- 性别
- All
- 接受健康志愿者
- 是
入选标准
- •Licensed professionals and recognized experts in relevant fields such as dentistry, public health, and health communication
- •Minimum of 5 years of professional experience in their respective fields
- •Prior involvement in healthcare education, research, or digital health communication
- •Willingness to review and provide detailed feedback on the adapted questionnaire
- •Ability to complete the survey in English
排除标准
- •Professionals without formal training or relevant expertise in the specified fields
- •Individuals with conflicts of interest that might compromise the objectivity of their evaluations
- •Experts who are unable to commit the necessary time to provide thorough feedback
- •General Public Group
- •Inclusion Criteria:
- •Adults aged 18 years or older
- •Individuals fluent in English
- •Access to a digital audio device and an internet connection to listen to the podcast and complete the survey
- •Consent to participate in the study
- •Exclusion Criteria:
- •Individuals with significant hearing impairments that might hinder the ability to comprehend the audio content
- •Healthcare professionals or individuals with advanced training in dentistry or health communication
- •Persons who have participated in similar educational podcast studies previously
结局指标
主要结局
Content Validity Index (CVI) of the Adapted Questionnaire
时间窗: From 7th February 2025 to 25th February 2025
Description: This outcome measure evaluates the content validity of the questionnaire adapted from QAEP, as determined by expert ratings. The Content Validity Index (CVI) quantifies the proportion of experts who rate questionnaire items as relevant or highly relevant. Unit of Measure: Score on a scale from 0 to 1, where scores above 0.78 indicate acceptable content validity
Internal Consistency Reliability of the Adapted Questionnaire
时间窗: From 7th February 2025 to 25th February 2025
Description: This outcome measure assesses the internal consistency reliability of the adapted questionnaire using Cronbach's Alpha coefficient, which measures how closely related the items are as a group. Unit of Measure: Cronbach's Alpha coefficient on a scale from 0 to 1, where values above 0.7 indicate acceptable reliability
Public Perception Score of the AI-Generated Health Podcast
时间窗: From 27th February 2025 to 15th April 2025
Description: This outcome measure assesses the general public's overall perception of the AI-generated health podcast using the validated questionnaire. Unit of Measure: Composite score on a 5-point Likert scale (1-5), where 1 represents "Strongly Disagree" and 5 represents "Strongly Agree"
Domain-Specific Evaluation Scores of the AI-Generated Health Podcast
时间窗: From 27th February 2025 to 15th April 2025
Description: This outcome measure assesses the general public's evaluation of specific domains of the AI-generated health podcast, including accessibility, design, content adequacy, and educational value. Unit of Measure: Mean scores for each domain on a 5-point Likert scale (1-5), where 1 represents "Strongly Disagree" and 5 represents "Strongly Agree"
次要结局
未报告次要终点
研究者
Dr Shiva Shankar Bugude
Assistant Professor
Qassim University