Sutter Health & AstraZeneca Partnership

Interactive Presentation
Live PreviewNavigate through the presentation below. Use arrow keys or click through slides. Initial loading may take a few seconds.
Primary care physicians spent 7-10 minutes of every 45-minute diabetes appointment clicking through multiple EHR screens to find critical patient data. This left minimal time for meaningful patient interaction and clinical decision-making. The fragmented data view also led to missed care opportunities, therapeutic inertia, and suboptimal diabetes management outcomes.
Collaborated with Sutter Health and AstraZeneca to design CM-SHARE, focusing on user-centered design and evidence-based clinical workflows. Assumed ownership of the entire application during PM transition. Implemented algorithm-driven medication recommendations addressing both non-adherence and therapeutic inertia (JMIR 2022). Developed validated efficiency metrics using EHR audit log data (Nature 2022). Created color-coded critical data visualization and automated aggregation from multiple sources.
Transformed diabetes care delivery across 15+ providers managing ~800 monthly visits. Published research in JMIR Medical Informatics (Sept 2022) demonstrating how the tool addresses medication non-adherence and therapeutic inertia. Second publication in Nature Scientific Reports (Oct 2022) validated clinical efficiency improvements through EHR audit log analysis. Physicians reported meaningful time savings and improved patient interactions, with feedback highlighting how the tool restored time for patient care.
Baseline: EHR audit log data from 15 primary care providers managing ~800 monthly diabetes visits. Tracked 6 months pre-implementation showing 7-10 minutes per visit spent navigating EHR screens. Data source: Epic Systems audit logs, validated by clinical workflow observations.
Why EHR-integrated web app vs standalone tool: Higher physician adoption despite Epic technical constraints. Why diabetes-first vs multi-condition: Ensured depth over breadth for clinical validity. Chose color-coded visualization over text-heavy displays based on cognitive load research and physician shadowing (n=20 sessions).
Time savings validated through Nature Scientific Reports peer-reviewed study (Oct 2022) using EHR audit log analysis. Clinical impact measured via JMIR Medical Informatics publication (Sept 2022) on medication adherence and therapeutic inertia. Provider satisfaction via quarterly surveys (n=15 providers, 89% satisfaction). Patient outcomes tracked through HbA1c improvements over 12 months.
2-second page load requirement limited data richness (chose 8 key metrics vs comprehensive display). Epic integration constraints vs custom-built flexibility. Diabetes-only focus vs expanding to HTN/hyperlipidemia (deferred to maintain quality). Required consistent EHR data quality (not all clinics eligible). Manual data validation required initially (automated later).
JMIR Medical Informatics • September 2022
Nature Scientific Reports • October 2022
Led collaboration between Sutter Health clinicians, AstraZeneca research team, Epic integration specialists, and clinical informatics team. Conducted 20+ physician shadowing sessions and workflow mapping exercises.
Balanced comprehensive data display with cognitive overload - limited to 8 key metrics on main screen. Chose Epic-integrated web app over standalone tool for better adoption despite technical constraints. Prioritized diabetes over expanding to other chronic conditions to ensure depth over breadth.
Worked within Epic's strict integration requirements and HIPAA compliance. Limited by 2-second page load requirement and existing EHR data quality. Had to accommodate varying technical literacy levels among providers.
Unlike generic EHR plugins, CM-SHARE was purpose-built for diabetes workflows with evidence-based algorithms. Differentiated from Epic's native tools through superior UX design and validated time savings published in peer-reviewed journals.
This project taught me the critical importance of clinical shadowing and understanding actual workflows versus reported workflows. The 3-year journey from analytics contributor to product owner showed me how persistence and taking ownership can transform career trajectory. Publishing peer-reviewed research validated our evidence-based approach and opened doors for speaking at medical conferences.
Product Owner & Clinical Workflow Designer
Domain: Digital Health / Clinical Decision Support