At Diameter Health, we are passionate about the potential for upcycled clinical data to drive better health and a more efficient healthcare system. When data is truly interoperable and actionable, it can provide timely, precise insights critical to understanding member health, care gaps, and risk factors.
Risk adjustment is a great example of a use case for which Upcycled Data presents a tremendous opportunity for improvement both in efficiency and accuracy. Risk adjustment is the process used to quantify an individual’s health or disease status. The resulting risk score, or compilation of weighted member risk factors, can have many applications, including establishing reimbursement under payment models such as Medicare and Medicaid and identifying patients for disease management programs and care coordination.
Historically, and still, in many cases, risk adjustment processes have relied heavily on claims data to identify and analyze member risk factors. However, claims data presents some challenges and limitations in risk adjustment:
- Time lag: lacking information up through the most recent months of a member’s history
- Incompleteness: lacking precise clinical information, such as lab results, vital signs, and notes
- Rigidity: lacking historical information for new enrollees and aspects of member journey only captured in the EMR
By contrast, clinical data sourced from the EMR is:
- Timely: updated with recent and historical member information
- Comprehensive: contains rich clinical elements, such as lab test results, vital signs, and clinician notes
- Dynamic: contains historical information for both new and retained members as well as supports more informed prospective suspecting
Not only do these elements increase the completeness of member risk scores, but they also allow plans to think about the whole member. Chronic and acute conditions, lab results, clinical readings, social determinants, and more are all critical elements captured in the EMR that inform how and when plans should deliver care. For example, suppose a member is recently diagnosed with diabetes, and PHQ9 scores (a tool used to measure depression) from a follow-up visit indicate they may be at risk for depression. In that case, the two risk factors will only be present in the clinical data. Thus, leveraging clinical data for risk adjustment coding allows for a more accurate representation of this member’s disease burden. It also enables proactive outreach to discuss their mental health and the resources to manage a new diagnosis. Clinical data is a strategic asset that allows health plans to deliver the right care to the right members at the right time.
Why is clinical data “out of the box” inherently hard to use?
Clinical data offers the potential for greater insight into member risk and health, as outlined above. But the shift towards strategic use of clinical data for risk adjustment requires an investment in clinical data acquisition and technology to transform raw clinical data into an asset that is standard, organized, and actionable. The inherent variation in source documentation, plethora of custom and standard terminologies, and fragmented storage of data across the healthcare system all contribute to clinical data’s complexity. To solve this usability challenge, Diameter Health’s Fusion technology normalizes inbound risk factors to industry-standard terminologies, including SNOMED and ICD10, and produces member level and population level outputs prepared for risk adjustment coding workflows.
How can Diameter Health help maximize the ROI of clinical data?
Investing in data quality is a recipe for maximizing both clinical and financial ROI. Compared to raw source data, we’ve seen up to a 30% increase in standard, interoperable diagnosis codes with Diameter Health’s Upcycled Data. More standard diagnosis codes translate into more complete disease reporting and more accurate risk scores. For example, in a real-world study with a national health plan customer, Diameter Health technology increased the capture of the diabetic population by up to 52% by normalizing diagnoses to SNOMED and ICD10. Looking across all diseases and in multiple health plan studies, Upcycled Data has the potential to increase RAF scores by 0.234-0.337 on average across a plan’s membership. When applied across millions of members within a health plan, this increase has significant revenue implications and the potential to transform how and when members receive care.
To learn more about how quality data leads to accurate risk scores and how to cost-justify an investment in clinical data, download our whitepaper, “The Strategic Role of Clinical Data in Risk Adjustment.”