Thu, 11/01/2018

In the Marine Corps, one of the first things I learned about how to prioritize and follow various orders I received was the phrase “Stars over Bars”. “Stars” refers to the rank of generals and “bars” indicates the silver or gold bars that makeup the insignia for captains and lieutenants. In the corporate world, we might refer to this as “C’s over VPs”. Clear chain of command is vitally important in the life and death operations of both the military as well as healthcare organizations, but it may not be the most efficient way to order the priorities of an enterprise analytics program. 

In this blog, I will discuss an immediate, short term approach to managing analytic requests that most organizations can easily implement and harnesses your data governance program (from the last blog post). I will also provide a long term approach that optimizes the efficiency of your executive leadership, analytics staff, and performance improvement groups across the enterprise.

In one organization I belonged to, we called the “C’s over VPs” situation the “wild west” method.  Sometimes requests came through a committee in a formal manner, but often executive leadership made requests directly to the analytics leadership. Sometimes, as executives were learning the names of various analysts and developers, they made requests directly to the staff level. The obvious problem with this is that needs are coming in asynchronously, with no knowledge as to other requests made from various parts of the organization, how those impact each other, or how they may impact current projects in motion, or vital maintenance, or operational analytic work. At a minimum, organizations who find themselves in this situation can quickly implement three simple processes of demand management to help control the chaos: 

  • First, setup a regular time for analytics leadership to meet and discuss what requests have come in, the channels through which they arrived, and coordination of priorities and assignments.
  • Secondly, implement a simple request form (document, web form, etc.) that can help formalize the process and categorize key elements of the goals, objectives and ROI (clinical, operational and financial) of each project. It is amazing how helpful a form can be in diverting marginal requests, but also in assisting the requestor in being more deliberate and descriptive in their ask. For example, asking for data related to diabetes isn’t much more helpful than just asking for data – a certain level of specificity is important. 
  • Finally, use your data governance committees as conduits for all requests.  They will benefit by seeing the types of work completed with the high fidelity data they govern, and they can also add value to the process by illuminating the key data availability, and appropriateness for certain parameters that may be vague or missing. 

Another approach to setting analytic priorities is more introspective and should live alongside the demand management process described above. In our modern healthcare environment, there is hysteria related to the need to improve – low cost, happy patients and staff, better outcomes. It is important to keep in mind “workflow eats technology for lunch”. We want to avoid having a bunch of performance improvement projects going on that disrupt business process and clinical workflows, unless they are well vetted, thought through, and consider what has been done before. I once had a nurse tell me that if one more “suit” tries to tell her a new way to “scrub the hub”, she’s going to stick the needle in their eye!

The key to looking inward for answers is the use of comparative data and benchmarks. Create a matrix that lists rows of companies that provide public ranking data for your organization such as NCQA/HEDIS, CMS/Hospital Compare, Healthgrades, US News & World Report, as well as any privately contracted companies you use for metrics and comparative data such as Optum, Truven, Vizient, Premier, etc. Now, create columns for all of the common metrics between these sources (e.g. – Prevention and Screening, Meds Management, CAUTI, CLABSI, SSIs, PSIs, Readmissions, Mortality, Patient Satisfaction, Access, etc.) Analyze which metrics are hurting your organization the most as graded by the highest number of public ranking sources and list them according to severity of underperformance, number of ranking bodies impacted, and penalties associated with each. Factor in how large the gap is between your organizational actuals when comparted to state and national averages or benchmark rankings.  Now, focus your performance improvement resources on enhancing the top 2-5 issues until you are satisfied. A very effective colleague of mine developed this approach in her role of directing clinical quality performance, and it got results. These priorities can also serve as an operative “high water mark” that can be used to compare against new requests that come in, making it easier to push back, if needed.

With good communication to leadership and following proven change management principles in implementing these request/approval/prioritization processes, you can begin to tame your organizational “wild west” and confidently pursue scalable and sustainable value based care for your organization.