Design, implementation, and evaluation of a local podcast for community health messaging

Thesis Abstract

This study investigates how a locally produced podcast can enhance community health messaging by addressing information gaps, trust, and engagement among diverse residents in a mid-sized urban setting. The problem addressed is the persistent gap between health information dissemination and residents’ awareness, comprehension, and behavioral uptake, exacerbated by limited access to traditional health campaigns and varying trust in health authorities. The aim is to design, implement, and evaluate a sustained local podcast that conveys evidence-based health information in an accessible, culturally resonant, and action-oriented format. Specific objectives are (1) to co-create a podcast series with community stakeholders and health professionals, (2) to implement a 12-episode pilot highlighting preventive health practices and service navigation, (3) to assess changes in health information literacy, trust in health messaging, and intended preventive behaviors among listeners, (4) to analyze listener engagement and reach across demographic segments, and (5) to develop a scalable model for replicating the approach in similar urban communities. Methodologically, the study adopts a mixed-methods design underpinned by the Health Communication Theory and the Technology Acceptance Model to examine both diffusion of innovations and user acceptance. The population comprises adults aged 18 and above residing in the target city. A stratified random sample of 400 residents will be recruited for quantitative survey pre- and post-intervention, with a 6-month follow-up of 200 respondents to assess sustained impact. Additionally, purposive sampling will select 20 community stakeholders and 12 health professionals for qualitative interviews to provide contextual insights. Data collection instruments include a structured questionnaire measuring health information literacy, trust in health messaging, perceived usefulness, and self-reported preventive behaviors, as well as in-depth interview guides and podcast episode analytics. Podcast episodes will be produced in collaboration with local health departments, community organizations, and multilingual voice talent, ensuring accessibility across English, Spanish, and Mandarin speakers. Data analysis employs a sequential explanatory approach. Descriptive statistics, t-tests, and multiple regression analyses will assess changes between pre- and post-tests and identify predictors of information uptake and behavior change. A repeated-measures ANOVA will evaluate sustained effects at the 6-month follow-up. Qualitative data from interviews will be analyzed thematically using NVivo, with triangulation against podcast analytics (download counts, listening duration, episode completion rates) to illuminate how content, format, and delivery influence engagement. A content analysis of episode transcripts will examine frequency and framing of health messages. Model specification includes regression equations linking information literacy and trust to reported preventive behaviors, controlling for age, education, and prior health status. Key expected findings include (a) increased health information literacy scores and higher trust in health messaging among listeners, (b) statistically significant upticks in intended and self-reported preventive behaviors such as routine screenings and vaccination intent, (c) higher engagement metrics for episodes featuring community voices and practical demonstrations, and (d) identification of demographic subgroups with differential impacts, informing targeted tailoring. The study anticipates that co-creation and local credibility will be critical drivers of message acceptance, with multilingual episodes expanding reach and comprehension. The contribution to knowledge lies in demonstrating a rigorous, scalable design, implementation, and evaluation framework for community-based health podcasts, bridging theory and practice by integrating diffusion and user-acceptance perspectives with empirical performance data. It also provides actionable guidance on governance, sustainability, and replication in similar urban settings. The main conclusion is that locally co-produced podcasts can meaningfully strengthen health information literacy, trust, and preventive behaviors when designed with community input, multilingual access, transparent sourcing, and measurable feedback loops. Practical recommendations include establishing formal partnerships with health agencies, embedding episodes within existing community programs, adopting ongoing listener analytics dashboards, and scaling the model to other health topics and languages, with emphasis on iterative refinement informed by sentinel listener feedback.