Self-Funded

A New Way To Predict (And Price) Healthcare Risk

Mar 31, 2026
Guest
Ali Panjwani
Ali Panjwani, Founder and CEO of Merit Medicine, builds ML models on massive claims data to forecast and price employer healthcare risk. He explains predicting spend from simple census files. Conversation covers how better risk selection improved a carrier’s loss ratio, avoiding biased underwriting, data tokenization and privacy, and financing high-cost cell and gene therapies.
Ask episode
AI Snips
Chapters
Transcript
Episode notes
INSIGHT

Models Must Evolve With New Therapies

  • Merit Medicine trains models continuously to detect evolving risks like new rare-disease therapies and changing standards of care.
  • Ali explains they update models as treatments and diagnostics change so underwriting reflects current clinical realities, not stale rules.
ANECDOTE

From Biotech Pricing To Employer Risk

  • Ali left biotech public policy because he wanted to follow dollars to the payers and solve employer cost problems.
  • His family exposure to chronic conditions and a public policy degree pushed him from pricing drugs to predicting employer risk.
INSIGHT

Predict Risk Using Only A Census File

  • Merit can acquire first-dollar pharmacy and medical claims using only a census file (name, DOB, gender, zip) by partnering with large data aggregators covering ~95% of commercial lives.
  • They tokenize and de-identify members, enabling member-level insights without violating HIPAA.
Get the Snipd Podcast app to discover more snips from this episode
Get the app