Exclusive Lumere Research: What does the clinical evidence say about brexanolone? Review the Research
Get Started Sign In


Bonnie Lai, PhD, Vice President, Lumere

Over the past decade, health systems have invested heavily in EHR infrastructure, and as a result, clinicians have needed to meticulously collect, input and document data like never before. But how can they put insights from that data to meaningful use, to examine care utilization patterns and identify opportunities to improve clinical practice and outcomes? Too often we hear variations on the same theme: “We have so much data and so few insights.”

Clearly, some barriers need to be overcome. At times, common approaches to analyzing drug data can unintentionally create boundaries between pharmacists and other clinicians (e.g., prescribing physicians or P&T committee members).

Below are three recommendations to help hospital pharmacists from making missteps when conducting analyses:

  • Don’t focus exclusively on spend
    Many hospital pharmacy leaders regularly look at 80/20 reports (showing the products that comprise the top 80% of drug spend) from their wholesaler or GPO to monitor their drug budgets. However, using this data to justify formulary deletion of an expensive drug often alienates physicians—they suspect that administrators are simply trying to save money without simultaneously considering patient outcomes.
  • Reduce use of traditional drug classification
    The commonly used AHFS drug classification is based on chemical structure or mechanism of action. These drug classes aren’t always helpful for choosing the best drug to treat a specific condition and often don’t allow for meaningful tradeoffs or therapeutic interchanges— prescribing physicians select medications based on patient indication, not the drug’s chemical structure.
  • Avoid relying on benchmarks to identify opportunities
    Benchmarking drug cost per DRG is a common analytical approach to identifying clinical initiatives. Getting this analysis right can be challenging; it involves mapping charges to NDCs, identifying the appropriate peer set, and removing confounding cases with comorbidities or complications. Even if the analysis is accurate, benchmarks don’t reflect evidence-based best practice, nor do they provide actionable direction on what clinicians should be doing instead.

Achieving “actionable insight” requires that those who are responsible for prescribing have the evidence that they need to make an informed decision. Forward-thinking health systems have been able to analyze data in ways that support formulary management based on an understanding of ”total cost of care”. Drug costs are weighed against clinical outcomes (e.g., length of stay), operational outcomes (e.g., turnaround times), and financial outcomes (e.g., reimbursement) along clinical pathways across the continuum of care. The result is evidence-based formulary management.

For example, Vanderbilt University Medical Center’s specialty pharmacy team has analyzed data from their clinics to show that multiple sclerosis patients had improved outcomes in movement (e.g., quicker time to walking). “These analyses have helped Vanderbilt deliver better patient care and—in some cases—helped us to secure supply of drugs that have limited distribution from manufacturers,” said Jim Hayman, chief pharmacy officer at Vanderbilt.

Best-in-class pharmacy leaders are also directly engaging physician champions in these analyses. For example, a pharmacy administrator at an IDN in the Northeast worked with an orthopedic surgeon to define outcomes “measures of success” for determining whether Exparel should be added to formulary. The surgeon decided to discontinue the trial because results weren’t trending well. According to the administrator, the key to success was agreeing on the methodology upfront, including which measures to focus on and how they would be calculated.

Taking these approaches to generate actionable insight is imperative for pharmacists who seek to advance their practice as essential members of healthcare teams.