Reducing variation remains one of healthcare’s most persistent challenges. While most health systems can identify differences in physician preference, product utilization and procedural costs, understanding those differences requires trusted, case-level procedural data. Without it, conversations about standardization often rely on anecdotal evidence, manual reporting and individual perspectives rather than objective insights.
Northwell Health’s Hospitals in Connecticut and the Hudson Valley sought to increase visibility into procedural data to reduce variation and improve clinical, operational and financial performance. Rather than relying on disparate data sources, the organization needed a trusted view of procedural performance to support physician engagement, collaboration and better decisions.
Northwell Health’s experience highlights three strategies that can help health systems move beyond identifying variation to understanding its causes and driving meaningful improvement.
1. Build a trusted foundation for decision-making.
Northwell realized that variation alone wasn’t the problem — different patient populations and clinical circumstances often warrant different care. The challenge was distinguishing care approach variation from variation driven by habit, preference or inconsistent practice patterns.
To gain visibility, Northwell centralized procedural data across multiple facilities and workflows, creating a more consistent view of physician utilization patterns.
“We had the data in multiple places,” said Deanna Marchiselli-Flynn, director of Clinical Optimization at Northwell. “Procedural Analytics pulled it together in one location.”
Like many health systems, Northwell had realized savings opportunities through contract negotiations and commodity purchasing. The next opportunity required a deeper understanding of physician preference items, where variation often has a greater impact on both clinical standardization and cost stewardship.
One review found a physician using demineralized bone matrix (DBM) in nearly 90% of applicable cases, compared with about 10% among peers. Rather than make assumptions, the team asked: What can explain this difference, and is it clinically justified?
The challenge was no longer data collection. It was transforming fragmented information into trusted insights that physicians, value analysis leaders and operational teams could use to make better decisions together.
2. Connect stakeholders around a single source of truth.
To sustain variation reduction, the team needed physicians, value analysis, quality and supply chain teams working together with connected data and shared organizational goals.
By combining case-level procedural data with pricing benchmarks, quality measures and operational insights into one platform Northwell, gained insight into who is using what products, how often they’re using them, and where variation happens.
Instead of debating whether variation existed, Northwell could focus on engaging the right organizational stakeholders early and build more informed strategies around product evaluations, contract negations, conversions and standardization opportunities.
3. Empower physicians with objective evidence.
Before engaging physicians, Northwell ensured its data could withstand scrutiny and support evidence-based conversations.
Value analysis coordinators used the platform to evaluate utilization patterns and support category reviews. By combining utilization and supply chain data, they gained visibility into what was purchased but how it was used in patient care, transforming reporting into decision-support.
“The data often reveals things people didn’t realize were happening,” said Brooke Palanzo, value analysis coordinator, surgical services, Northwell. “That’s where the opportunities for improvement start.”
Northwell wasn’t looking to eliminate physician choice. Rather than relying on generalized observations, they brought physician-level utilization data, procedural trends and benchmark information into discussions, shifting focus from debating variation’s existence to understanding practice patterns.
In one example, surgeons believed they were receiving favorable pricing because of longstanding vendor relationships. However, benchmark comparisons revealed they were paying significantly more than peer organizations for the same product.
“Surgeon’s time is valuable,” shared Amanda Saviano, value analysis coordinator, surgical services at Northwell. “If you can bring all the information from the Clinical Database and Procedural Analytics together in one place and present a complete picture, that’s key.”
Today, decisions are evaluated through the lens of quality, outcomes and total cost of care — not cost alone. By combining utilization data, benchmark pricing and clinical performance metrics, teams can determine whether products improve patient outcomes and identify opportunities to reduce variation without compromising quality.
From Visibility to Lasting Change
Northwell’s experience shows that reducing unwanted variation begins with a shared view of procedural performance built from clinical, operational and supply chain data. The goal isn’t to limit physician autonomy — it’s about giving leaders a shared understanding of where variation exists, why it exists and where greater consistency can improve quality, efficiency and value.
As healthcare grows more complex, those that succeed will move beyond reporting variation to understanding it, engaging clinicians around it and using it to drive continuous improvement.
To learn more, visit https://www.vizient.com/procedural-analytics.
At the Becker's 11th Annual IT + Revenue Cycle Conference: The Future of AI & Digital Health, taking place September 14–17 in Chicago, healthcare executives and digital leaders from across the country will come together to explore how AI, interoperability, cybersecurity, and revenue cycle innovation are transforming care delivery, strengthening financial performance, and driving the next era of digital health. Apply for complimentary registration now.
