Natural language processing is an important tool in value-based contracting, says an executive at a natural language processing company.
For payers and other risk-bearing organizations, accurately capturing patient disease risk is essential to ensure accurate reimbursement and control costs of care, especially for patients with underlying conditions.
However, capturing disease risk is challenging because it requires extracting critical information from patient records – and nearly 80% of medical data is unstructured. With messy, unstructured data, it is difficult for payers to create a holistic view of a patient in a semantically interoperable way that enables many downstream use cases.
In the past, payers relied on expensive and time-consuming chart reviews to discover and extract key structured data from patient records and claims. Recently, however, artificial-intelligence-based tools such as natural language processing (NLP) have played an increasingly important role in enabling payers to search records for key data needed for risk adjustment.
More value-based care, more risk adjustment
The importance of risk-adjustment for payers has increased as value-based care arrangements such as Medicare Advantage proliferate across the health industry.
Plans contract with CMS to provide benefits to Medicare Advantage enrollees and are reimbursed through a capitated system, in which plans receive a predetermined payment per member per month from CMS. These monthly capitated payments are risk-adjusted for each member to reflect their health status and estimate the appropriate monthly cost for Medicare-covered services.
According to the Health Care Payment Learning & Action Network, in 2020 overall, 60% of healthcare payments focused on quality and value, up from 53% in 2017 and 11% in 2012. Separately, 49% of practices responding to the American Academy of Family Physicians’ 2022 value-based care survey said they are participating in some form of value-based payment, and 18% are developing capabilities.
NLP is an essential tool for payers engaging in value-based arrangements, as it allows for processing patient records and additional risk-adjusted diagnoses and supporting documentation buried in typically unstructured data, promoting more accurate risk adjustment.
By giving computers the ability to read, understand, and interpret clinical language, NLP captures and manages data from patients’ episodic health records, enables payers to modernize chart review processes, and eliminates antiquated and bloated workflows associated with manually reading a patient record.
One of the critical benefits of NLP is that it enables organizations to process medical records at once, capture all key data points, and then filter and display relevant data on an as-needed basis, as opposed to multiple reviews by multiple physicians, researchers or auditors. For various information. In addition, NLP technology is a key element in the interaction between healthcare information systems, creating and validating health information from a wide variety of sources, including claims and patient charts.
3 Ways NLP Helps Payers
In addition to more accurate, efficient risk adjustment, NLP’s ability to rigorously extract key structured data from patient records has several other benefits for payers:
Controlling costs. Accurate risk capture is essential for reimbursement to payers for providing care to patients with multiple chronic conditions and comorbidities, for example, who by definition require more medical complications and more expensive care. Combined with predictive analytics, NLP can help payers identify patients, for example, with diabetes and stage-one kidney disease who are at risk of advancing to stage-two. Controlling costs is largely a question of foresight and prevention, and by enabling a more accurate picture of a patient’s health, NLP can help payers identify patients who need proactive interventions to prevent disease progression. Conversely, NLP can help payers identify patients who have been overtreated or received tests and procedures not justified by their diagnoses.
Developing a 360-degree view of patient health. A significant challenge for payers in developing a complete picture of patient health is that patients see multiple care providers and as a result their data is spread across multiple health records. For example, a diabetic patient may have frequent visits with a primary care physician, endocrinologist, dietitian, and other providers. By searching all patient records for structured information, payers can calculate the total cost of care for a diabetic patient.
Sharing insights with providers. As payers conduct analyzes of patient records, they can uncover many data points that can help providers close care gaps. By sharing these insights with physicians, payers can strengthen provider relationships around the shared goal of improving patient health.
Risk adjustment has become an increasing priority for payers as the market for Medicare Advantage and value-based care matures. In turn, NLP can serve as a critical tool for payers to gain hard-to-find insights from patients’ records, enabling them to reduce costs and promote more comprehensive care.
Ketan PatelMD is the Chief Medical Officer of SyTrue Inc., a natural language processing company.