Improving Patient Matching: Part 2 - State of Patient Data and Strategies for Improvement

Dan Carotenuto's picture
 By | January 29, 2014
January 29, 2014

Overcoming patient matching challenges starts with getting it right within a healthcare organization first and that requires with understanding how accessible, accurate, timely and meaningful patient data exists in the enterprise.

This blog is part of a four part series that explores overcoming patient matching challenges by improving patient identity and data management within healthcare organizations first and effective and enduring approaches to patient matching that providers can take to ensure their short and long term clinical success, financial stability, and, ultimately, their survival.  In part 1 of this blog series, "Improving Patient Matching:  Using Healthcare IT and Data Strategy to Create a Single Patient View," I discussed the risks and history of the patient matching challenge.
 
In this blog I will discuss the current state of patient data management in healthcare organizations and why it contributes to patient matching complexity.  It will also introduce approaches to overcome patient identity and data management challenges within healthcare organizations to facilitate better HIE and population health.
 
State of Patient Data
 
Effective patient matching is critical to HIE and starts with aggregating patient data from within the organization.  One might suggest all that is necessary, therefore, is simply a data aggregation solution.  That approach would be shortsighted, short-term and ultimately inadequate.  It is prudent for healthcare organizations to first provide optimum patient data consistency and accuracy within and across their internal business domains and systems.  This approach ultimately yields better HIE as it facilitates timely and more accurate information exchange such as with the creation of Continuity of Care Documents (CCDs).  This is easier said than done.  A look into the state of patient data will show us why.
 
Within healthcare organizations, patient matching complexity and inconsistencies are born out of years of inconsistent and inadequate patient data management and has resulted in the following:
 
  • Disparity in the storage and maintenance of patient recordsclinical to financialwithin and, subsequently, across providers
  • Inconsistent adoption of healthcare IT
  • The proliferation of system and application silos deployed as quick fixes to resolve immediate departmental and individual practitioner needs
  • Interoperability inadequacies between systems and applications—exacerbated by mergers and acquisitions
  • A lack of enterprise-wide data management standards.  
 
The result is duplicate patient records, multiple interdependent systems like doctor portals, clinical systems, insurance applications, bed utilization applications and healthcare analytics with data that is out of sync, incomplete or incorrect.  The consequences include lower patient satisfaction scores, non-compliance, fines, reduced funding, privacy violations and, most importantly, inadequate quality of care.
 
What we are talking about is poor quality of patient data.  It significantly impacts the algorithmic-based patient matching approaches—deterministic and probabilistic matching—discussed in part 1 of this blog series, which are only as good as the data on which they are based.  Although it may sound a bit harsh, as the axiom goes:  garbage in, garbage out, or in our case, "bad patient data" in, "bad patient data" out.  
 
Quality, Accessibility, Timeliness, Meaningful
 
There are three key challenges in effective patient data and identity management within the healthcare organization:  ensuring the quality, accessibility and timeliness of patient data throughout the patient journey across the continuum of care in a single healthcare organization.  If these challenges are not addressed, the result is a fragmented and inconsistent view of a patient as opposed to a single view (figure 1).
 
Addressing these data management challenges will minimize patient matching complexity and promote optimal HIE and population health.  It will also foster opportunities.  Patient data that is synchronized, complete, correct and unique across all systems and applications lets us go from a fragmented view of patients to a single view of patients that is substantially more meaningful.  The consequences go beyond HIE and regulation compliance to facilitate better quality of care and more opportunities for insight from healthcare analytics and clinical research.
 
How do we fix it?
 
Making patient data more accurate, timely, accessible and more meaningful demands a twofold approach:  a combination of technology and the right business strategy—a data strategy—that both need deployment across the healthcare organization, however big or small.  
 
Enterprise Master Data Management 
 
The technology is in the form of enterprise Master Data Management (MDM)—not a revelation—where patient data is mastered for related business domains and subject areas.  The mastering process will create a single view, a single version of truth if you will, of patients.  Vital parts of the mastering process are data quality management to de-duplicate, complete, validate and correct patient data and data integration technologies to deliver or pull patient data to or from the right systems, in the right forms and at the right time.  
 
While the need for enterprise Master Data Management may not be a surprise, the following are essential and should not be overlooked:
 
  • Integrate the data quality and mastering services with healthcare operations.  This is also known as operational MDM.  The idea is to process and master patient data as it enters the enterprise at the point of service/care.  This is a goal that can start with a longer synchronization period, say 24 hours, progressing to every 12 hours to 1 hour to ultimately real-time or near real-time.  This will require a data integration hub and the result would be a real-time master data hub that will make available the most current, accurate single version of truth about a patient.
  • Take advantage of data stewardship components and workflow tools for manual remediation and integrate manual remediation into the mastering process.
  • Perform the following functions in an integrated fashion to promote patient data integrity:  data integration, data profiling, data quality, remediation for exceptions, document access, and collaboration/Interoperability.  It is imperative that an organization’s data management technologies not only support all of these functions, but that all of these functions interoperate with each other and facilitate collaboration among subject matter experts (SME) and data contributors—the producers and consumers of patient data.  
  • Ensure integration of the mastered repository with reporting repositories for accurate and more meaningful healthcare and performance analytics.  This can influence improvements to the business of healthcare operations that can translate to improved patient satisfaction, timely billing and financial and regulatory filings and quality of care.
 
Data Strategy
 
A prudent business strategy is in the form of a data strategy that establishes an enterprise-wide culture of data governance and trust.  This means ensuring through corporate to departmental policies and procedures that all applicable stakeholders, people and processes participate in maintaining as high a level as possible the quality, accessibility, timeliness and meaningfulness of all data throughout the enterprise.  Admittedly, this is no small task, but it is necessary for both short and long term success.  
 
It is important to know where the healthcare organization stands with regards to a culture of data governance.  Does it have any governance at all?  Is it data driven where data drives success?  Is it policy driven?  Or are they close to being fully governed where policies are auditable, data lineage across systems is available and custodians (data), stewards (business process) and committees are all in place?  
Finally, it is important to note that enterprise MDM and a culture of data governance and trust go hand-in-hand and are interdependent. 
 
This brings us to the next installment in this blog series where I will discuss in greater detail enterprise master data management within healthcare organizations for managing patient identity and data management how it will help them improve healthcare delivery and operations, manage costs and performance and deliver better HIE and population health.
 
References
 
Improving Patient Matching:  Using Healthcare IT and Data Strategy to Create a Single Patient View
 
Copyrighted and published by Project HOPE/Health Affairs as Arthur L. Kellermann, and Spencer S. Jones, “What It Will Take To Achieve The As-Yet-Unfulfilled Promises Of Health Information Technology,” Health Aff (Millwood). 2013, volume 32, no. 1, 63-68. The published article is archived and available online at

Improve the Patient Journey With Omni-Patient: Creating a Single Version of Truth Across the Continuum of Care, Dan Carotenuto, John Backhouse, Information Builders, 2013,