Early results within the EP and Cath Lab show meaningful gains driven by more complete and accurate capture:
When Allina Health and AssistIQ announced their partnership last fall, the goal was simple: Establish a consistent, reliable way to capture supplies, implants, and tissue used in procedures while reducing steps for clinicians.
The underlying belief was that if Allina Health’s clinicians were given the technology that made product capture and data quality dependable, everything downstream would change; Allina could consistently bill for what’s chargeable, more accurately order what teams will need, and use reliable data to make more informed decisions.
Now, with more than five months of use across thousands of cases, Allina Health is seeing what dependable AI-powered technology enables in practice.
It wasn’t for a lack of trying; clinical teams had multiple methods in place to document products, including barcode scanners and paper implant sheets. But these methods were time-consuming and took them away from their focus on patient care.
Even with strong teams in place, those methods introduced gaps that were hard to reconcile. Leaders lacked a reliable source of truth to learn from. While waste was suspected, it could not be measured in a way that enabled confident, decisive action.
For Tom Lubotsky, Chief Supply Chain Officer at Allina Health, this went deeper than revenue.
Without confidence in the data, conversations about cost, variation, or improvement stall. Accurate product capture is not just a technical requirement, but a prerequisite for trust.
AssistIQ introduced computer vision–based capture that records supply, implant, and tissue usage as it happens, without relying on barcodes or QR codes that may not exist. Products are identified visually, expiration dates, lot numbers, and serial numbers are captured automatically, and the data flows directly into Epic and Workday.
For Chris Schultz, Director of Supply Chain at Abbott Northwestern Hospital, that shift addressed a very practical problem.
Over the first 5.5 months, across more than 3,000 cases, capture accuracy exceeded 99%. Just as importantly, the system continued to learn. When new or unfamiliar items appeared, they were identified and added, improving accuracy over time.
With accurate capture in place, analytics stopped being a reconciliation exercise and became a reliable source of insight across teams.
Once capture became reliable, previously hidden patterns surfaced, particularly in complex procedural areas like EP and CV care.
In EP, average monthly billed amounts increased by 32%, alongside a 201% increase in captured chargeable products. Cath Lab saw even larger gains, with a 171% increase in average monthly billed amounts and a 364% increase in chargeable products captured.
These gains were not driven by increased case volume, but instead reflected the impact of how providing clinicians with the right technology leads to more complete documentation of what was already being used.
Another meaningful shift has been Allina Health’s ability to quantify waste for the first time.
Historically, waste in procedural environments was suspected, but rarely measured in a credible way. Without case-level visibility, it was difficult to distinguish between unavoidable loss, workflow issues, or product failures.
With AssistIQ’s analytics, that changed.
For Allina Health, the longer-term value of accurate capture extends well beyond today’s metrics.
With reliable, case-level data in place, teams can begin to move from hindsight to foresight. Patterns in supply and implant use that were previously hidden can now be analyzed across locations, service lines, and physicians. Over time, that visibility can inform how cases are prepared, how inventory is staged, and how teams ensure the right products are available when and where they are needed.
Over time, Allina sees this as a foundation for more predictive use of data, from anticipating supply needs to streamlining case preparation and reducing friction for clinical teams. That progress depends on one thing above all else: data that people trust enough to act on.