PhyPal pilot will track four workflow measures against clinic's pre-launch records
Founder Ranjeet Randhawa told RuntimeWire the California workers' compensation pilot will compare four measures with the practice's own records; about 15,000 patients are expected to go through PhyPal over three months.
By Emerson Redding · Published
Why it matters
The pilot has a defined operational scorecard, but Randhawa said its before-and-after design cannot rule out other practice changes and will produce directional results rather than establish that PhyPal caused any changes.
PhyPal's California workers' compensation pilot will track four workflow measures and compare them with the practice's own pre-launch records, founder Ranjeet Randhawa told RuntimeWire in an exclusive interview.
The measures are “time from physician recommendation to authorization submission,” “the share of requests complete on first submission,” “staff time per authorization” and “intake completion rate.” “Baselines come from the practice’s own records before go-live,” Randhawa said.
The pilot is with a multi-site California practice with more than 80 providers, Randhawa said. About 15,000 patients are expected to go through PhyPal over three months. Active capabilities include conversational pre-visit intake, assembling clinical evidence from patient conversations and prior records, and preparing and tracking authorization requests.
“It is an operational pilot, not a randomized trial. The comparison is the practice’s own workflow before PhyPal: we measure the same steps before and after go-live,” Randhawa said.
Randhawa said the before-and-after design cannot rule out other changes at the practice, so the results will be treated as directional. The company will not report clinical outcomes from this pilot because the design cannot attribute changes to the platform.
Clinicians review the system's output and retain decisions about diagnosis, treatment and whether to request treatment, Randhawa said. “The system prepares; physicians decide.” Authorization requests cannot be submitted without the treating physician's signature, and staff handle administrative steps, he said.
Randhawa said PhyPal is building a multi-layered documentation architecture to separate and classify records, extract facts and check them against rules before information reaches a physician. RuntimeWire previously reported on PhyPal's conversational AI for clinical workflows in workers' compensation.
“The system prepares; physicians decide.”
— Ranjeet Randhawa, PhyPal founder
Emerson Redding is a synthetic AI reporter for RuntimeWire. Accountable editor: Ryan Merket.