AI for Healthcare Policy & PR
Proof-of-concept: reimagining healthcare workflows with generative AI innovation
AI Search Engine
AI Chatbot
*Note: to ensure client confidentiality, some details have been redacted.
Overview
A public healthcare network partnered with IBM as part of a provincial directive to modernize internal operations. We incubated two generative AI use cases and created functional MVPs to serve as proof-of-concept solutions.
Applicable Links
1) Identify existing data challenges and operational bottlenecks that hinder healthcare service delivery across the province.
2) Prioritize and scope GenAI use cases by balancing impact, feasibility, and the needs of diverse stakeholders.
3) Deliver high-value MVP solutions within a 2-month timeframe.
I led research and UX activities to scope and scale the use I worked closely with the team on research and UX activities to shape and grow our use cases, turning insights into actionable design requirements. into design requirements.
Nour Al-Safi, Leandra Keller, Stephanie Duguay, Maya Michniewicz
Prioritized use cases
Conversations with healthcare professionals highlighted key issues that resulted in two use cases:
Nurses needed a way to quickly search and understand existing policies. There were 4,000 internal policies across different webpages and documents — some of which were redundant or outdated.
The public relations team faced challenges developing content from scratch to issue public statements. Existing templates lacked the necessary nuance. Tight deadlines further complicated the process, leaving little time for final reviews.
"The policy search process can be time-consuming & inefficient."
"Maintaining policies is a highly manual process."
"Creating communications material is demanding. It is time-consuming because of approval wait times and it takes multiple iterations to create a draft!"
PROCESS
We adopted a human-first approach to a technical solution. I made listening to healthcare professionals central to each step:
Discovery
Content inventory & audit
We initiated discussions with healthcare leadership to gain a deeper understanding of core challenges, existing data, and underlying infrastructure. I co-facilitated a series of workshops to scope and prioritize two high-impact generative AI use cases.
Personas & User Flows
Content inventory & audit
My team members conducted semi-structured interviews with various healthcare professionals. After the interviews, I developed personas and user flows based on attitudinal and behavioural insights.
Co-design workshops
Content inventory & audit
I held design workshops sessions with frontline users—such as nurses and public relations staff—to clearly understand key challenges and validate our findings. I worked alongside business leaders and data scientists to develop capabilities that directly addressed user pain points.
SOLUTIONS
We fine-tuned the solution based on emerging research insights and collaboration, balancing function with user impact.
Policy search & maintenance tool
Healthcare staff are able to quickly find & verify policy & healthcare directives.
Policy managers can now use a natural-language search to find relevant information, as well as identify and flag outdated policies. We designed built-in reporting capabilities.
Contextual Communications Generator
PR teams now have a baseline for high-quality and context-aware communications.
PR can adapt tone and messaging to suit specific speakers and audiences—whether it's a politician, the general public, or the medical community—while staying mindful of current news, research, and global events.
Impact
The MVP solutions were deployed and were estimated to achieve the following outcomes:
On average, saving 100,000 hours per year, leading to 31% improved process efficiency
On average, saving 5.9 hours per PR request, leading to 83% improved response time
Buy-in and continued partnership
Our client was eager to scale the two MVP solutions, leading to ongoing collaboration and more innovation-based endeavours on the horizon.