Why You Need Entity Resolution for Health System M&A Analysis

Considering a healthcare merger or acquisition? Start with one of the most important metrics: the size of the shared patient population

Healthcare merger and acquisition activities continue to flourish in response to an increasingly competitive market. This introduces complexities around pre-integration planning as part of a broader due diligence process. A key metric that can drive planning activities is an understanding of the shared patient population between the merging organizations.

To capture that metric, consolidation and analysis of patient records must be done quickly and efficiently. When a merger or acquisition occurs with a health system serving the same region or specialty, there is a high likelihood that detailed and important clinical, registration and billing information for the same patient exist across the different systems. To measure patient overlap, it is essential to perform efficient and accurate analysis while maintaining patient confidentiality and privacy, and without disclosing PII to either party involved.

Let’s examine how entity resolution can help you conduct this analysis quickly and efficiently so that you can determine the potential impact of a merger or acquisition.

How to Analyze Patient Population Overlap Before a Merger or Acquisition

You may have intelligent data analysts who can quickly run comparisons to generate a decent picture of the overlapping vs. unique populations, but this approach results in only a point-in-time view and introduces other issues, including:

  • Ad-hoc analysis does not consider all historical data, which limits match accuracy.
  • It requires a lot of manual processing and individual decision making that lacks transparency.
  • The process is not easily repeatable but must be repeated to understand future delta files.
  • The final analysis does not always provide the proper contextual information required for business justification that verifies where the metrics came from. 
  • Manual comparisons increase the risk of exposure of critical personally identifiably to those doing the analysis.

For those organizations that have an existing EMPI or MDM solution, the ability to quickly load data and understand the shared patient population would seem easy. However, a lack of skilled and knowledgeable resources and scheduling and even licensing implications can quickly make this process daunting. Another option is to “pre-match” the records, which allows you to arrive at a good picture of how many patients are unique vs shared, before you load them into your systems. With this approach, you won’t unnecessarily create duplicates in your EHR that require future remediation or inadvertently overwrite current data with older data.

Reconciling Data During Merger & Acquisition Analysis

Once you understand your shared patient population, reconciling data can be an arduous undertaking that is dependent on the desired outcome. While completing a migration into a single EHR may be the desired end state, significant logistical factors can often impede or delay that process. Most often a phased approach, by region or market, is adopted to reduce internal resource contention. Iterative file matching by region ensures that decisions are timely and representative.

Quite often, maintaining independent platforms is the best you can hope for in the short term while planning continues to determine a desired end state. A common practice in healthcare is to link patient records across platforms by manually embedding a medical record number from one system into another or setting up a rules-based linking process to propagate the identifier. Both options can lead to erroneous linking of patient information and introduce clinical risk.

Health systems should consider new solutions that provide the speed, accuracy and privacy considerations needed for M&A analysis and allow them to focus on more important organization integration tasks instead of data analysis.

Senzing Entity Resolution On-Demand

Senzing® entity resolution allows your M&A team to accurately match patient records across disparate systems while protecting PII. The Senzing software is an easy to use, low-cost solution – for one-time, or ongoing on-demand use. IMT has partnered with Senzing to offer healthcare organizations of any size an entity resolution API that accurately matches data and detects relationships.

IMT’s deep expertise in healthcare data and integration patterns combined with entity resolution technology from Senzing ensures customers will achieve rapid analysis and results with minimal data preparation. Because Senzing is API-driven, you will need to apply some level of coding to utilize the capabilities, and IMT can help. IMT can offer on-demand services in the cloud to minimize the delay and cost of provisioning on-premise infrastructure, and allow you to only pay when you need to use it.

What’s unique about Senzing?  It’s Purpose-built AI for Entity Resolution.

Senzing delivers a simple yet sophisticated approach combining common sense with real-time active learning.  It does not require tuning, training, or experts. The software allows you to apply matching against real-time operations or to bulk loads of historical data, while processing daily transactions and queries.

Senzing is a great fit for merger & acquisition teams because the software can perform thousands of entity resolution transactions per second over hundreds of millions, or billions, of records, reducing analysis time from weeks to days. Senzing entity resolution is fast and scalable, new data sources can be added quickly with minimal data preparation and results are delivered immediately.

The highly accurate Senzing entity resolution algorithms understand inconsistent, ambiguous and messy name and address data effortlessly. The software also identifies commonly used generic data values and other issues in “dirty data” that impact matching accuracy. A Selective Anonymization option allows you to accurately compare records without disclosing personally identifiable data that compromises patient privacy. 

The results Senzing entity resolution delivers includes clear matches (or duplicates), possible matches or possible related entities. These details are critical to providing more conservative metrics as part of your analysis, preventing over-valuation of the patient population, and assessing guarantor or household relationships.

How Entity Resolution Can Boost Healthcare Processes

Fast, efficient, and inexpensive entity resolution in healthcare has valuable uses beyond M&A, including:

  • Population Health or ACO Rosters – for patients AND providers from payers and other networks
  • Marketing and outreach lists from care management organizations, both public and private
  • Ongoing relationship detection for households and guarantor groupings
  • Auditing matching performance and accuracy in embedded EHR patient matching algorithms

For more information on how Senzing entity resolution can help you with your on-demand, file matching needs, or ongoing health data entity resolution,  please contact Mark Holmes, VP Solutions at mark@imt.ca.