The amount of health-related data generated on an annual basis continues to rise. In fact, a single patient produces nearly 80 megabytes of data each year. These data points, from clinical indicators captured in the electronic health record (EHR) to claims encounters to activity readings tracked by wearables, contribute to the complex fabric of healthcare information.
Weaving a patient’s vital health information into a meaningful dataset can transform care delivery to be more personalized, preventative, and proactive. In addition, this data fabric can drive meaningful change through:
- Personalized care delivery to treat the whole patient while improving their experience.
- Prevention of latent disease and avoidable cost and utilization to reduce the total cost of care.
- Proactive outreach and implementation of interventions at the right time to improve outcomes.
However, as the data informing this fabric continues to explode in volume, it also increases in variety and complexity. Thus, as many organizations begin to invest in obtaining access to various forms of health information, they realize that the journey to harnessing its power has just begun.
Barriers to Value: What Contributes to the Complexity of Clinical Data and how can this be Addressed?
Clinical data, sourced primarily from EHRs, health information exchanges (HIEs), and labs, contains rich information about an individual’s health journey. From diagnoses to lab results to vital signs, clinical data sources capture timely, precise indicators that are not always present in other sources, such as claims data. Because of these unique characteristics, many healthcare organizations aim to leverage clinical data to stratify risk, supplement quality measures, close coding and care gaps, and effectively manage population health.
However, the reality is that up to 50% of clinical data in its raw native form may not be usable, standardized, and interoperable, preventing organizations from successfully integrating and applying this information to capitalize on its potential. It is notoriously messy and inconsistent, and systemic data quality issues must be addressed before the data can be leveraged.
Variations in source terminologies, including non-standard and standard codes, and unique documentation practices, varying from practice to practice and EHR to EHR, are contributing factors. This is compounded by unpredictable inaccuracies contributing to clinical data complexity—information documented in the wrong place within the medical record, duplicate entries of the same prescription or diagnosis, the plethora of custom terminologies used, and fragmented data storage across healthcare systems.
As one example, there are at least 100 distinct ways clinicians may document hemoglobin A1C, the laboratory test for blood sugar. The scale of this problem increases significantly when looking at not just one lab test but all lab tests, then all medications, diagnoses, procedures, immunizations, and across other domains. The challenges of clinical data content complexity are incredible.
While regulatory agencies, such as the Centers for Medicare & Medicaid Services (CMS) and the Agency for Healthcare Research and Quality (AHRQ), maintain industry-standard terminologies for healthcare coding, the use and consistency at which they are applied in practice are limited. Healthcare must begin organizing around these standards to ensure a common language for data exchange and insights generation. Healthcare information is incomplete, inaccurate, and unpredictable without this alignment to terminology standards.
Additional data consumption challenges that contribute to clinical data cacophony include:
- Competing standards, both terminology and format (local versus national standard coding; FHIR versus C-CDA versus flat file format).
- Challenges impacting effective data collection at the point of care, resulting in data gaps.
- Data fragmentation across siloed systems and various EHRs, (consider that the average Medicare patient sees 11 providers across five care sites, and no one is privy to the whole story).
Healthcare Needs Upcycled Data: How can the Totality of Clinical Data Potential be Unlocked?
Clinical data has the potential to transform business processes, accelerate time to insights, and improve health outcomes with critical information and timeliness that other datasets lack. However, the complexities introduced by various coding and format standards prevent stakeholders from successfully tapping into this value. To solve these challenges, organizations may look to hire, train, and deploy in-house resources and teams of data analysts. However, manual terminology mapping and similar approaches yield modest data usability returns and are costly, time-consuming, and inefficient, preventing organizations from being prepared to scale, grow, and innovate effectively.
Upcycling is a powerful process to transform raw clinical data from disparate sources into a standards-based, interoperable asset, accelerating the deployment of clinical data and allowing organizations to redeploy resources strategically and operate efficiently. Upcycling data means semantically normalizing source codes to industry standard terminologies (e.g., LOINC, CPT, ICD10), enriching with metadata such as drug classes and CCS codes, and aggregating across multiple sources per patient to generate a longitudinal, deduplicated record.
The resulting data asset is positioned to digitally transform population health, public health, value-based care, and analytics initiatives across the industry. To further encourage the exchange of quality, standard clinical data, Upcycling and real-time conversion of legacy formats, including CCD, HL7v2, and flat files, to FHIR resources ensures the use of high-quality data in FHIR for Patient Access, Payer to Payer, and analytics use cases.
The Benefits of Upcycled Data: Why and Why Now?
Patients, providers, and plans are eager for the shift to value-based care (VBC) and a more optimal patient experience. Understanding the medical, social, and behavioral needs of each person within a population is critical to improving the quality of care and health outcomes. Organizations must leverage data as a strategic business asset with expertly generated insights to achieve a more personalized, preventative, and proactive care delivery model. Upcycled clinical data helps accelerate the shift to VBC and aims to increase healthcare system efficiency and improve clinical outcomes and patient experience while reducing the total cost of care.
Upcycled data as an asset drives impact across the following:
- Better care quality. Healthcare providers and plans benefit from complete information to inform decision-making, enabling targeted care delivery to address care gaps, coding gaps, disease management, and social and behavioral needs.
- More accurate analytics. Accurate clinical data directly contributes to more precise analytics, predictive modeling, and AI/ML model development. Whether used as input into a tool for predicting mortality risks or population health analytics, upcycled clinical data delivers an average of 30-50% more usable data across clinical domains to increase the accuracy of generated insights.
- Lower costs. Through enterprise efficiencies as well as workflow-level efficiencies, organizations benefit from automation, speed, and scale to reduce admin burden and associated costs. Additionally, with clinical data assets applied across critical use cases, including risk adjustment and quality measurement, duplication and waste are reduced.
Clinical data is a powerful, strategic asset when it is actionable, consistent, and useful.
When successfully upcycled, it will fuel innovations in value-based care and whole-person care programs, provide seamless data exchange between payers, providers, and individuals. Data is a force to drive meaningful change in healthcare, and better data means better decision-making across the care continuum.
Author:
Mary Lantin is President and Chief Operating Officer, Diameter Health and General Manager, Clinical Solutions, Availity .
Mary has an established track record of successfully operationalizing organizations’ strategic visions. At Diameter Health, a national leader in clinical data quality and interoperability, she led the business and was responsible for driving the corporate strategy, designing product portfolio capabilities, and delivering products and services to customers across multiple markets. She oversaw the commercial, technology, product, finance, and operations teams. Diameter was acquired by Availity earlier this year. Prior to joining Diameter Health, Mary served as President and General Manager at Optum Analytics for Payer and Provider Solutions. Mary also held senior leadership roles at various health information technology companies, where she developed and executed strategies for the rapid deployment of enterprise-wide analytics solutions.
Mary earned a bachelor’s degree from Princeton University, and a master’s in public health from Harvard University.
Diameter Health is now part of Availity
