Originally published in July 2019, updated in May 2022
In our latest data quality blog series, we introduced the importance of interoperability and how healthcare organizations can achieve a framework of data normalization by mapping disparate data to industry standards. The first blog in the series explored the value of terminology mapping for lab data.
In this second installment, we review the complexities of normalizing medication data that comes from disparate sources and why mapping these variations to a common standard language is critical for driving accurate clinical decisions support and quality patient care. Mapping to a common standard, such as RxNorm, can also help improve data accuracy for care and disease management initiatives, optimize quality measures reporting, and achieve reliable analytics.
The importance of data normalization in healthcare
In the age of electronic patient records, pharmacy IT systems, and other supporting technologies, medication terminology standards are critical for advancing interoperability to optimize patient care and support high-level analytics initiatives. Accurate and complete data aggregation provides value for research, population health, and medication best practices for care and disease management.
The multitude of medication standards used within the industry—both standard and proprietary—creates challenges in reconciling all available data in a meaningful way. Most hospitals use at least ten disparate IT systems, most of which contain some level of drug information and rely on different terminologies. Multiple access points make the consolidation and normalization of data a significant obstacle.
Additionally, some medication standards are updated daily, requiring time-intensive ongoing maintenance. Accurate and complete data aggregation doesn’t happen by accident. For example, your EHR probably uses Medi-Span for prescribing medications and RxNorm for recording a medication list or allergies. Your dispensing system is probably based on National Drug Codes (NDCs). All of this information is required to accurately tell your patient's story and needs to be harmonized to avoid duplicative information that feeds into your downstream high-value initiatives, like a chronic care management program.
What does it mean to normalize medication data?
Without a data infrastructure in place to support data normalization, healthcare organizations risk experiencing negative downstream impacts resulting in revenue loss, misidentified gaps in patient care, inaccurate analytics, or incorrect clinical decision support.
The business case for leveraging an infrastructure that automates medication mapping to RxNorm is an easy one to make due to the sheer volume of data that exists across a health system. I am lucky to be using the Health Language data quality solutions to automate the mapping that I do on a daily basis.
Without this added efficiency, I would never be able to complete mapping hundreds of thousands of medication concepts. It combines the efficiency of machine learning with the deep clinical knowledge of the Health Language experts to help organizations address the burdensome, error-prone processes traditionally managed across numerous spreadsheets and departments. Most organizations simply do not have the staff to tackle this problem manually, and even if they did, clinical staff should be working in their areas of expertise and analyzing the data rather than normalizing it.
Specifically, Health Language solutions allow healthcare organizations to collaboratively map, search, and distribute data throughout the enterprise. Clinical auto-mapping powered by domain-specific algorithms ensures the highest map rates and accuracy, minimizing the need for manual review. Flexible workflows allow cross-departmental collaboration, with management dashboards to alert teams of project progress. In addition, audit trails provide the traceability needed to follow the lifecycle of a code for HEDIS® measures or just for effective governance.
How data mapping makes a difference beyond EHR systems
Health Language clients use the applications to normalize data and provide the level of governance needed to support an organization’s semantic interoperability requirements for data.
One biopharmaceutical company is using Health Language to power a learning health system (LHS) to improve disease outcomes. The LHS leverages the solution to harmonize vast amounts of medication data from over 10,000 patient records to create a single source of truth. By mapping proprietary codes to industry standards for improved analytics, the LHS is curating real-world data to improve research.
In another example, a vendor organization uses the Health Language data quality solutions to power a real-time clinical surveillance solution. Through this partnership, the surveillance solution can break down data silos that exist across hospital systems by normalizing medication EHR feeds from over 500 different hospitals. This allows them to close patient information gaps and overcome roadblocks that hinder real-time analysis of a patient’s condition.
To learn more about how Health Language data quality solutions can help your organization, speak to a Health Language expert today.
Read on to the third installment of this data normalization blog series, in which we explore the importance of allergy mapping in clinical data.
HEDIS® is a registered trademark of the National Committee for Quality Assurance (NCQA)
