Datashop Blog

How Marrying Clinical and Claims Information Creates Magical Value-based Results

“Combining clinical and claims data lets providers manage a patient’s health and understand what they are doing,” said John Cuddeback, MD, Ph.D., Chief Medical Informatics Officer for the American Medical Group Association (AMGA). “It allows the provider to move from being reactive to being proactive.”

There may be a whole lot of data systems out there in the healthcare space, but the power of data remains grossly underutilized. Traditionally, healthcare organizations rely only on the most basic data to analyze patient population. Amidst the multiple disparate data systems, healthcare organizations need to find an accurate, data-driven understanding of population health. For better clinical and financial outcomes, they might require something more to create a holistic picture.

How clinical and claims data can complement each other?

Claims data have the financial information related to health insurance claims, basically an overview of the list of procedures done, providing deep insights into how the population health resources are being utilized. Being highly structured, claims data can be easily assembled and evaluated in only a few days.

But to go beyond that, clinical data is needed to understand the health status of a patient.  Approximately, 60% to 80% of the clinical data is stored in the form of lab & results data, which is not included in claims. Claims data can be useful for retrospection purposes as its time frame is usually 3-6 months, while clinical data can play a crucial role in near real-time actions with their timeline of 1-30 days. Moreover, managing population health solely based on the claims data is ineffective, as there is at least a 30-day lag before claims data is available again.

Why bringing them out together is a necessity?

Both the claims and clinical data have their own advantages and shortcomings, but the silver lining in this situation is that they both complete each other like puzzle pieces. A survey revealed that at most 22% of claims filed by providers get rejected, out of which 15.5% trace back to improper or missing diagnostic codes (ICD-10) and procedure codes (CPT code). Adding to this, reimbursement claims submitted by providers to Medicare comprise of about 13% of all denials.

Clinical data, on the other hand, gives a holistic picture of services provided by the physicians to the patients. Even if claims data can be collected and assessed timely, it would not provide the complete picture of a patient’s health, down to every little detail.

The integration of claims data with clinical data provides a very specific value- from vital information captured in population level data to the level of detail captured in the EHR. It could determine how a patient at risk of congestive heart failure is falling through his medications and needs proper attention. It could also help providers analyze if there are any underlying risk factors which may have a significant impact in the future. Plus, as the clinical data is directly validated by the clinicians, it would require less time and effort to attest to specific medical conditions and accelerate proactive interventions to enhance care.

Overcoming the challenges of clinical and claims data integration

The process ingesting volumes of clinical and claims data and putting them together to deliver meaningful, actionable insights is a complex one. There are some challenges that limit the integration of clinical and claims:

  • Both clinical and claims data could be inaccurate, incomplete, or erroneous.
  • Variations in data schema and architecture of data systems.
  • Matching patient identifiers and linking patient records across various tables to create a comprehensive record.
  • Picking the most accurate and recent version of claims data to reflect updated information.

The first step to overcoming these challenges is to create a scalable and agile data repository that has the power to ingest all the varying sources of data together. At Innovaccer, we created a unified healthcare data platform that deploys a Hadoop-based Integrated Data Lake. The data lake can ingest multiple data sources, not just from CCDs, X12 837/835 files, HL7 feeds, but also right from actual EHR tables, and integrates these varying data sets together.

Once all the necessary data is gathered and normalized into longitudinal records, they are run through an Enterprise Master Patient Index module. Every record is assigned common patient attributes such as name, date of birth, gender, address, phone and offers quick accessibility to records with configurable search-and-match algorithms. The EMPI module leverages a combination of fuzzy logic and exact logic to match the records based on different data fields and to rectify gaps and redundancies.

The platform endorses its own very structured form of documentation that leverages HBase, Elastic, and MongoDB, to store records separately and longitudinally. The platform nests data for every activity, every encounter, and every patient, and reconciles updated information with both clinical and claims data to reflect real-time outcomes.

Delivering value-based outcomes across the network

Working with various healthcare organizations, we realized that data serves two most important objectives of any healthcare organization- one, to drive efficient operations and two, to enhance their reporting capabilities. Not to mention how integrated clinical and claims data can be leveraged to drive effective, real-time outcomes improvement.

Once clinical and claims data are linked together, it can be thoroughly analyzed to determine if there are any looming gaps in the network and what could be the best way to proceed with quality and cost improvement. The providers can observe their adjudicated as well as pre-adjudicated claims to observe their performance and enhance it. And if they find a growth opportunity they can tap into, they can always drill down to understand their progress on a deeper level and take the best measures. Once the anomalies are eliminated, the providers can accelerate towards a network driven by information transparency.

The best part about integrating clinical and claims data together is that it not only provides a holistic view of the population, it also allows monitoring of the impact of clinical decisions providers undertake. An organization can create a central utility to compare cost, quality and efficiency metrics to track trends and identify care gaps for every patient, at the right time. Plus, with two-way interoperability, this information can also be viewed right on their screens- without having to leave the EMRs.

The Road Ahead

In short, quality of the healthcare systems can be greatly improved by the intelligent use of clinical and claims data coupled in a coherent manner. The right kind of infrastructure that can ingest claims and clinical data together and is scalable enough to grow as the amount of data increases could be a start. When data is bound together, connecting the dots, providers can have access to the most recent data that can identify unique patients as well as offer insights into population health. Together, this knowledge will lead us into the dream of better care delivery in a world driven by quality, aimed to make us all healthier and happier.

To learn how integrating clinical and claims data together can drive meaningful improvements in population health, get a demo.

For more updates, subscribe.

Join Team Innovaccer at booth #649 this HIMSS ’18, from March 5 to 9, at Venetian Palazzo, Sands Expo Center in Las Vegas, and learn how we can assist you in delivering an efficient, data-driven healthcare.

Abhinav Shashank

Abhinav Shashank

Chief Executive Officer & Co-Founder - Innovaccer.

Add comment