Publication Details
Title:
Framing, public health, and machine learning: Ghana’s noncommunicable disease crisis through the lens of the news media
Details:
Noncommunicable diseases (NCDs) have become an urgent public health crisis and a significant cause of death, especially in developing countries like Ghana, where news media coverage influences public understanding of health issues, just as it does in other jurisdictions. This study employs an unsupervised machine learning method on 505 Ghanaian news articles published between 2014 and 2024 to analyze how the themes related to NCDs and the main attribution frames are presented. The results identified four primary thematic frames: stakeholder partnerships and crisis management efforts, Ghana’s NCD risk factors, systemic barriers to managing the crisis, and advances in healthcare technology. Additional attribution analysis revealed two key frames: a lifestyle frame, highlighting individual choices, and a socio-economic frame, connecting NCDs to poverty, weak health systems, urbanization, and environmental factors. Findings reflect global trends while emphasizing the influence of local structural factors in NCD crisis narratives. The study demonstrates the usefulness of computational methods for analyzing large news media text corpora, showing how Ghanaian news media mirror international framing patterns and also reveal unique local barriers and dependencies during public health emergencies.