The Role of Theory in Translating Clinical AI Research into Practice

by Jon Scaccia February 16, 2024

In the realm of healthcare, the implementation of Artificial Intelligence (AI) has been met with equal parts enthusiasm and skepticism. As we stride forward, the need to narrow the chasm between potential and practice in clinical AI has never been more critical. A recent bibliometric study, Evaluating the translation of implementation science to clinical artificial intelligence: a bibliometric study of qualitative research (read the full article here), sheds light on how the application of Theories, Models, or Frameworks (TMFs) in qualitative clinical AI research could be the bridge we desperately need.

The Role of TMFs in Clinical AI

Implementation Science, a field dedicated to the study of methods to promote the integration of research findings into healthcare practice, is still in its formative years. A key aspect of this science is the application of TMFs, which are essentially various theoretical approaches, models, or conceptual frameworks developed to understand and guide the process of implementing changes in healthcare settings. Despite their proven benefit in other healthcare interventions, the study finds a striking underutilization and inconsistent application of TMFs in clinical AI research.

Insights from the Study

The study spans research published from 2014 to 2022, analyzing 202 eligible studies of which only 34.7% applied a TMF. Interestingly, while the number of publications increased eightfold during the study period, the proportion applying TMFs didn’t show a significant increase. The majority of TMFs were only applied once, indicating a lack of consensus or standardization in their application. Furthermore, the study highlights a lack of clear rationale for TMF selection in most cases, suggesting that the current application of TMFs in clinical AI research may not fully utilize their potential to guide and improve implementation.

Implications for Public Health Practitioners

For public health practitioners, understanding and applying the right TMFs can significantly enhance the adoption and integration of AI in clinical settings. It provides a structured way to analyze and address the complexities involved in implementing AI systems, considering the unique demands of healthcare environments. However, the study’s findings indicate a need for a more strategic and informed approach to the selection and application of TMFs. Increasing the accessibility and engagement with theory-informed practices among the implementation science community is essential.

The Path Forward: Increasing Rigor and Frequency of TMF Application

To truly harness the potential of clinical AI, the study recommends a concerted effort to increase the rigor and frequency of TMF application in research. This involves developing clear guidelines for TMF selection based on the specific demands of the research question and the intervention being implemented. Building a repository of well-characterized, popular TMFs, accompanied by tools and guidelines for their application, can help streamline their adoption in future research. Moreover, training and capacity-building initiatives can empower more researchers and practitioners to engage with TMFs effectively.

Conclusion

The study presents a compelling case for the more robust application of TMFs in clinical AI research. By doing so, we can accelerate the translation of AI innovations into real-world clinical practice, ensuring that the promising potential of AI in healthcare is fully realized. As researchers, practitioners, and stakeholders in public health, it’s crucial to advocate for and adopt a more theory-informed approach to clinical AI implementation, bridging the gap between theory and practice.

Be the Health Change-Maker – Get Informed Weekly!

Step into the role of a public health change-maker with ‘This Week in Public Health.’ Each issue brings you closer to the heartbeat of community health, innovative research, and advocacy. It’s more than news; it’s a platform for transformation. Subscribe for free and join a community of informed individuals driving positive change in public health every week!