The intersection of artificial intelligence with qualitative or mixed methods for communicable disease research: a...

Artificial Intelligence (AI) and machine learning (ML) have been embraced by the global health community as an innovative and transformative tool for communicable disease research. Concurrently, social science research is critical to understanding complex human behaviors related to disease prevention and control. Given that best practices on the use of AI in support of qualitative or mixed methods research are still being established, we sought to describe the current landscape of AI/ML use in qualitative and mixed methods research on communicable disease topics.

Scoping review.

We conducted a scoping review of studies incorporating AI/ML methods with qualitative or mixed methods for communicable disease research published before or on February 12, 2024, using PubMed, Scopus, Web of Science, and Embase platforms. Publications meeting our inclusion criteria were charted and analyzed to classify methodologies and assess primary reasons for using AI methods with qualitative or mixed methods.

We identified 1342 records, of which, 29 met our full inclusion criteria. AI and ML techniques were primarily used for: AI-assisted qualitative or mixed methods analysis; integration of AI with human-led qualitative research; and AI-assisted tool development validated through qualitative or mixed methods. Most studies examined data generated during public health emergencies, with social media data analysis as the predominant application. Findings simultaneously highlight AI's ability to rapidly process vast quantities of qualitative data, while also demonstrating the necessary role of human oversight to ensure contextual accuracy.

The advent of advanced AI and ML technologies enhances the efficiency and expands the breadth of qualitative and mixed methods for communicable disease research. However, methodological best practices should emphasize AI integration with human-led qualitative analysis to capture nuanced findings and ensure AI use is ethical and equitable. As such, these methods must be accessible in diverse global research settings, including countries with low- and middle-income economies where the burden of communicable diseases may be disproportionate.