Supporting Guidance on AI, and Fast

Organizations – both public and private – delivering services across various industries and domains are increasingly needing guidance on the use of Artificial Intelligence. The word “guidance” should not be taken lightly as it directly impacts an organization’s reputation and those receiving its services.

What is guidance?

Think of guidance as a set of guidelines or recommendations that should be followed to achieve a set of objectives. For many organizations, especially in healthcare, guidelines for the use of any technology or new process should consider quality principles such as safety and timeliness. For example, consider an organization’s objective of improving access to care by leveraging existing technologies – it will need to consider aspects such as: which technologies are available? Which ones are proven to be effective (or cost-effective) in improving access to care? How is it best deployed? Will the target users adopt and use this technology? Have there been studies showcasing their impact? Once they are used in our operations, what are some considerations we need to keep in mind? How will its use impact our processes and services? How will it impact our workforce? Will we need additional training?

Why is evidence-based guidance important?

Embedding the latest evidence into guidance ensures decisions are based on what is proven to work or has shown promise, leading to better patient outcomes and more consistent, high-quality care. It builds trust, promotes cost-effective practices, and allows for timely updates as new research emerges. Ultimately, it supports ethical, legally defensible, and efficient healthcare delivery.

How do researchers develop guidance?

Each researcher has one or a few areas of research areas they explore in depth. For example, a researcher may focus on the use of AI techniques in improving access to health services for adults with chronic diseases. Too specific – right? As a standalone study, it is not sufficient to provide guidance on using AI for improving access to care; however, collectively with other studies conducted on the same topic, it can provide a comprehensive picture.

This is where literature reviews including systematic reviews and meta-analyses play a key role. These reviews, which are conducted by researchers who understand the field, will follow guidelines such as the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) to ensure they consider all the studies relevant to informing guidance on a particular topic. 

The Wilson&Wilbur process for developing guidance on the use of AI in service delivery

1. Collaboratively working to gain a clear understanding of the organization’s objectives and aspirations for embedding AI.

2. Developing research questions to which answers will support the development of tailored guidance.

3. Developing a search strategy which includes all key terms that should be considered.

4. Searching the literature and applying PRISMA guidelines to our research process – this includes appraising the quality of published studies.

5. Synthesizing findings for guidance development.

6. Developing guidance using a grading system to establish the strength of the evidence supporting each recommendation or guideline.

7. Supporting organizations in implementing these guidelines and recommendations through tailored supports.

Importantly, to ensure we have the latest expertise involved in guidance development, we tap into our network of researchers who work in the specific area of interest. This allows us to gather data and evidence that may not be published yet.

Want to experience the evidence-based way of implementing AI but do not have the luxury of time? Let’s talk!

The W&W Team