Determine if work environment requires adaptation

Why does this matter?

Health AI tools and work environments function dialectically – changes in the work environment can affect use of the AI product and changes in an AI product can affect the work environment. 

  • There are times when the best way to improve use of a health AI product is to intervene in the work environment. Many activities within the work environment are required to support the AI product, facilitate downstream actions prompted by an AI output, empower front-line workers to intervene and to improve patient care.
  • When intervening in the work environment is required, organizations with effective change management programs may be able to rely on existing structures and programs.

How to do this?

Step 1: Ensure that prerequisites are met to appropriately intervene in the work environment

  • Monitor the work environment affected by the AI product post-rollout to ensure that relevant information is available to inform any adaptations.
  • Develop strong communication channels and feedback loops with front-line workers who directly use the AI product as well as with managers who are responsible for the quality of care.
  • Ensure clear understanding of incentive structures and priorities within the work environment that influence use of the AI product. 

Step 2:  Monitor performance targets to confirm that a problem has emerged within the work environment

Step 3:  Specify the problem to be addressed to determine appropriate actions 

  • See the table for the full list of types of shifts and associated actions, which include:
    • Training front-line workers on optimal use of the product.
    • Re-initiating information dissemination about the AI product.
    • Shifting promotion ladders, performance evaluations and job titles to support and incentivize the use of a new AI product.

Step 4:  Communicate changes 

  • Communicate changes to the work environment that alter the patient care experience with affected clinicians in the implementation context.

“it really helps to talk to providers, or whoever is going to benefit from this tool. So often they’re not aware. So we rolled something out. We think that it was all this. There was a lot of communication. But then when we start monitoring the usage of those tools, people are not using them. They then say, “I forgot. Somehow I didn’t even know that it was there, and that it would be useful.” A lot of this is communication and re-training, and making sure that users are aware of the benefits of this tool and the long term impact.”

Technical Leader

adopt health ai