Dr. Mostashari's metaphor suggests using data that you already understand as the starting point of your "analytics journey". For most healthcare organizations that consists of information derived from administrative claims.
Now that may be somewhat of a surprise coming from a passionate advocate of EHR technology and someone responsible for encouraging its adoption. However, Dr. Mostashari's point was that effective analytics requires a solid foundation. And that foundation is only as strong as the integrity and reliability of the data used to construct it.
No doubt, the clinical data captured in certified EHRs holds a great deal of analytical promise. But it also represents a layer of information that requires a reliable foundation to be truly useful.
Here are a few reasons why Dr. Mostashari's suggestion makes sense:
1) Claims data uses well defined standards, such as ICD-9/10, CPT-4 and UB-04, to define data elements. It is also quite reliable, with high levels of data integrity, since it is used as the basis for financial transactions between payers and providers. These qualities makes it ideal for objective analysis.
2) Both government payers, such as CMS, and private payers use claims data for many of the quality and outcome measures used to evaluate provider performance. Understanding claims-based data from an analytical perspective is essential to proactively managing the performance of these measures.
3) Building, analyzing and optimizing a claims-based data set is an ideal way to develop a base set of analytical competencies within your organization.
Combined with information from various operational systems, this data set can be used to illuminate performance patterns in areas such as utilization, efficiency and productivity.
But the data elements that comprise standard claims data are just the starting point for laying the foundation. A great deal of value can be added by integrating information from readily available sources. Here are just a few of the possibilities:
The Healthcare Cost and Utilization Project sponsored by the Agency for Healthcare Research and Quality (AHRQ) provides a family of supplemental data sets that can be added to claims data, including:
- Clinical Classifications: Provides clinical meaningful categories using ICD-9/10 and CPT coded data.
- Comorbidities: Identifies clinically significant groups of conditions.
- Procedure Classes: Classifies procedures into major (OR-based), minor, diagnostic and therapeutic groupings.
- Utilization Flags: Reveals usage patterns for procedures and services such as ICU, diagnostic tests and therapies.
- Surgery Flags: Identifies surgical procedures and encounters.
Incorporating data from these sources into a basic claims data set, and refreshing it daily, will provide a powerful information asset that can be used to better understand utilization patterns, segment populations, and improve outcomes. Certainly a valuable first step to consider for your analytics journey.
