In addition to being a health insurance company, Oscar has a virtual health offering where members can see virtual providers (often for free). To support those visits, Oscar also has a home grown Electronic Medical Record and Clinical Documentation system. Though the visits were being documented thoroughly, documentation effort was eating into provider capacity. This research aimed to explore where providers were spending the most time documenting and recommend ways we could reduce the time spent.

• Observational time study shadowing 4 hours of appointments per participant across both phone and video visits

• Culminated to 20+ hours of observation

Sample: 10 providers, a mix of those in primary care and urgent care as well as representation of new vs. existing patient visits, reasons for visit, and tenure of provider

• Time work vs. Mind work: Providers were dedicating substantial time work for documentation - generating note content, waiting for the patient to join the call, tool loading times, etc. But there was also extensive mind work - forming diagnosis and orders, crafting patient-friendly instructions. Any effort to reduce documentation time should shave off time work but not get in the way of mind work.

• The tool’s existing layout required significant scrolling up and down to document a range of issues patients would surface across structured and unstructured fields. All of these would need to be re-editing if the patient added more context or detail.

• The existing tool created redundant charting because it required the provider to document an argument in plain text and translate that into structured fields like orders, diagnosis codes, etc.

Following this study, the Lead Designer for this tool executed a strong Idea Gen workshop to define the near-term tactics and long-term strategy to reduce documentation time and increase provider’s ability to be present during patient visits.

I continued to consult on early concepts. This work has led to an impressive AI scribing technology that has reduced appointment documentation time by 25%.

Additionally, as of July 2025, 8,400 primary care visits (97% of all primary care video visits) have been conducted using AI scribing

91% of providers report an increased ability to focus on the patient during the visit! Read more about this technology here.

More than anything, this study speaks to the importance of a baseline. Because of the variation in provider documentation styles and lacking tracking, our system could not generate a strong documentation time baseline. That’s where UX Research brought value. This time study uncovered the gravity of documentation burden and helped generate ways to reduce that burden.