Fast Track Fasenra

Validated a ML approval predictor concept for a complex biologic prescribing workflow using evidence-based methodologies
Industry
Life Sciences
Role
Sole UX Researcher and Designer
Client
AstraZeneca
Year
2019
Challenge
Clinicians needed confidence at point‑of‑care when biologic approvals hinge on many moving parts. I concepted and validated an ML approval predictor tool that gave biologic coordinators a critical piece of information: the likelihood of Fasenra being approved by insurance for a given patient.
Approach
I kept fidelity honest and the questions sharp. I framed 29 hypotheses across success metrics, built an interface that was realistic enough to test without leading, and did scrappy recruiting including cold calls and visits to physician offices to get real voices in the room.
Outcomes
Invalidated 6 out of 29 hypotheses and got signal on 3. Delivered a journey map, hypothesis tracker, and findings to guide what to build next—and what not to.
Hypotheses Created
29
Hypotheses Invalidated
6
Hypotheses w/ Signal
3.0
Key Activities

• Hypothesis framing (29 total)

• Journey mapping across stakeholders

• Concept + UI for predictor outputs

• Guerilla outreach (calls + office visits)

• Three physician interviews; synthesis + next steps

‍