Breast cancer cases in Pakistan are reported to be among the highest in Asia, highlighting the need for effective media coverage. This study explores the informative and educative roles of 14 Pakistani English newspapers in breast cancer news stories ( n = 5783) over two decades. The data was systematically organized into two separate datasets, the Informative and the Educative, using keywords related to both categories. Latent Dirichlet Allocation (LDA) was used to identify latent thematic structures, which were subsequently refined through a transformer-based sentence embedding approach using Sentence-BERT to assign texts to semantically coherent themes. The analysis revealed distinct patterns, with informative content emphasizing scientific advancements, patient care, and healthcare infrastructure, while educative content highlighted awareness campaigns, prevention, and community engagement. Article length analysis showed minimal variation between the two types, reflecting balanced editorial attention. The study’s methodological framework and focus on South Asia provide important insights into media functions in health communication, with broader implications for global research.
Awais et al. (Fri,) studied this question.