During COVID-19 and beyond, a foundation for health value starts with data integrity

Featured in Fierce Healthcare

The ability to manage population health and public health starts with data integrity. 

Access to reliable, trustworthy data enables healthcare organizations to achieve complete patient views—bringing together all data, including social determinants of health, to more effectively treat patients and help them manage complex conditions. And, as the industry is quickly learning, access to accurate data during a public health crisis is crucial to contacting individuals who test positive for disease and containing the outbreak.

But recent surveys show providers, health plans and government agencies struggle to access the data they need to help vulnerable populations even in non-crisis scenarios:

  • 30% of healthcare organizations struggle to exchange patient health records with other providers.
  • 52% of hospitals do not use patient data from outside their electronic health record (EHR) because their systems’ workflows don’t support external data.
  • 1 in 5 patient records are duplicate records, limiting the ability to establish accurate case histories.

And as COVID-19 ups the ante for data integrity, a Duke-Margolis Center for Health Policy report shows up to 50% (PDF) of COVID-19 lab reports are missing key contact information needed to alert individuals to their test results. This obstructs local and state health agencies’ efforts to investigate newly diagnosed disease, identify infection clusters, localize disease hot spots and match patient identification for clinical queries, according to the report (PDF), written by faculty and researchers for Duke as well as national leaders in healthcare and health policy.

With the long-anticipated data interoperability rule now delayed due to the COVID-19 outbreak, how can healthcare organizations and public health agencies overcome data integrity challenges that present hurdles in health management—both publicly and at an individual level? Here are three vital approaches.

1. Strengthen data demographic verification using existing systems. The Duke-Margolis report points to three data elements that are commonly missing from medical records of individuals who undergo testing for COVID-19 through clinical laboratories, which perform 83% (PDF) of positive test results reported by the Centers for Disease Control and Prevention:

  • Race/ethnicity
  • Telephone number
  • Address

At Verato, the experience of one of our colleague’s parents, who was exposed to COVID-19 while visiting an assisted living community, illustrates how easy it is for these data to fall through the cracks. The parent, whom we’ll call “Marie,” was directed to a drive-by testing site. While Marie completed a form with her date of birth, address and telephone number to give to the testing facility, the lab technician was hesitant to accept it, stating that she was told only the individual’s name and date of birth were needed. But these two identifiers alone don’t guarantee the right record will be matched to the right patient. Especially with common names like “Jennifer Smith,” providers can often have multiple instances of patients with the same name and a similar birth date in their system. 

When minutes count, fail-proof patient identification matching is essential. One strategy is to pair the EHR (or any clinical repository) with referential matching, a plug-and-play solution that leverages a continually updated, highly curated reference database of identities that span the entire U.S. population. With a referential matching data service, the right association is made immediately, strengthening the validity of patient data in downstream applications like the EHR or electronic lab repository (ELR). Because referential matching enriches existing systems rather than replacing them, there is no danger of disrupting operations through use of this tool—especially critical during a public health response.

2. Invest in data interoperability—even without a government mandate. Amid COVID-19, the value of data interoperability is abundantly clear. However, the pandemic reveals the extent to which the healthcare industry and government agencies have underinvested in this key capability. Even within systems, match rates vary widely—especially during an era of consolidation. One study found match rates between healthcare organizations can be as low as 50% even when organizations share the same EHR vendor due to variability in technology and processes.

Closing the healthcare data gap requires that organizations invest in data management infrastructure that not only eliminates misfires in patient matching but also strengthens data integrity at every point in the system.

For example, a recent Black Book Research survey shows 67% of all payer data and 90% of all provider data go unused for data analytics. Issues with data quality commonly prevent some organizations from performing meaningful analyses. At Intermountain Health, efforts to address data quality issues through data validation checks, data normalization and data-cleansing steps pushed match rates with the University of Utah above 60%. Later, focus on additional technical and operational updates, like data algorithms, supported match rates of 95% while strengthening the organization’s data analytics capabilities. 

During a public health crisis, a foundation of data integrity and interoperability helps avoid situations where public health agencies must “cobble together information from many disparate systems,” which has prompted the need for “feasible, short-term steps to improve interoperability and exchange of key data for COVID-19 containment,” according to the Duke-Margolis report.

3. Enrich commercial lab reporting data. Beyond the pandemic, integrity of medical claims for all healthcare organizations depends on accurate capture of key data elements such as the patient’s date of birth, health plan group number, member number and address.

A Black Book Research survey shows 33% of denied claims stem from inaccurate patient identification or information, costing the nation’s healthcare system $6 billion annually, or $1.5 million, on average, per hospital each year. Sometimes, these data are not inputted at the point of service; other times, it may be inputted incorrectly. Increased focus on integrity of patient data not only enhances quality of care coordination and patient safety but also strengthens an organization’s financial health—critical given the hit many organizations’ balance sheets have taken during COVID-19. 

Supporting a timely approach to improved data integrity

Data integrity has never mattered more in healthcare—and now, we have the technology to achieve it. By using existing systems to bolster data integrity at the point of service or the point of research, healthcare’s key decision-makers can eliminate gaps in patient data that not only threaten the public health response in times of crisis but also weaken the ability to manage population health and ensure highly reliable care.