The global market size of NLP in health care is projected to reach USD 7.2 billion by 2027 from USD 2.2 billion in 2022.1
Value-based care model offered by the Centers for Medicare and Medical Services (CMS) is an emerging alternative health care program to the traditional fee-for-service model and focused on offering quality care rather than quantity.
The first CMS Hierarchical Condition Category (HCC) model was implemented in 2004. This model produces different risk scores for beneficiaries who reside in either the community or an institutional setting, or who are new enrollees.
In 2006, the Prescription Drug Hierarchical Condition Category (RxHCC) risk adjustment model was implemented. Similar to the CMS-HCC risk adjustment model, the RxHCC model uses demographic characteristics and disease variables to predict costs.2
Hierarchical condition category (HCC) coding is a risk-adjustment model originally designed to estimate future health care costs for patients. The Centers for Medicare & Medicaid Services (CMS) HCC model was initiated in 2004 and is becoming increasingly prevalent as the environment shifts to value-based payment models.3
Therefore, any error in documenting and reporting the International Classification of Diseases, Tenth Revision, Clinical Modification (ICD-10-CM) codes will disturb the RAF score, which in turn will disturb the overall value-based reimbursements.
The current challenge in the Value-based Risk Adjustment
The risk adjustment factor (RAF) score also known as risk score is assigned based on the patient diagnosis that maps to an HCC and demographic factors, its documentation will require the highest specificity for maintaining accurate ICD-10-CM codes and accurate RAF scores.
Each medical record correctly submitted with a matching sampled enrollee CMS cover sheet is evaluated independent of all other submissions and is reviewed for both validity and diagnosis coding. The entire medical record is reviewed before making a final decision on validity and coding. Only RADV coding results from valid medical record submissions are used to substantiate payment. Invalid or a lack of a medical record submitted will potentially impact the payment error calculation. It is critical to understand the ‘Medical Record Reviewer Guidance’ provided by CMS pertaining to documentation issues that will be considered on a case-by-case basis.4
How NLP technology is helping Health Care payers to enhance Risk Adjustment Coding accuracy
With NLP-enabled coding technology, MA organizations can seamlessly and accurately risk-adjust their members, prioritizing the ones that are at high-risk, by identifying the most missed diagnosis coding conditions and the ones that are over-coded, or lacking documentation, allowing them to focus on the high-risk members.
In addition, an NLP-powered HCC coding technology offers a set of diagnosis code books that automatically will reduce the amount of time taken to perform the first-pass review while reducing the amount of data they must initially enter.
Ensuring first-pass review accuracy reduces the need for organizations to deploy multiple levels of quality analysis (QA), or chart audit, which is usually performed to ensure codes are accurate and none have been missed.
NLP-powered risk adjustment HCC coding solution, particularly highly accurate NLP that is tried and tested on real charts, will ensure your HCC coders with utmost accuracy in the first-pass review.
This way, MA organizations, and HCC medical coders save an ample amount of time that can be used in audits to safeguard from penalties.
RAAPID is powered by tried and tested NLP technology in healthcare.
Our NLP-enabled coding solutions can be customized and personalized based on an MA organization’s requirement, all in an effort to reduce clicks and speed up chart reviews. This way, a higher number of chart reviews can be processed in a short time, while improving risk adjustment coding accuracy.
We understand it is challenging for MA organizations to understand and manage the ICD-10-CM guidance while performing risk adjustment, therefore, we are here to offer HCC coding solutions that leverage explainable AI that presents the context and the code evidence along with the output, unlike a black box, designed to match your risk adjustment workflow for improved efficiency.
Today, RAAPID clients are benefiting from our HCC CAPTURE (Chart Review) and HCC Compass (Chart Audit) solutions that are helping them to perform risk adjustment coding processes at reduced costs, whilst positively impacting their revenue cycle management.
To know more about how RAAPID’s NLP-enabled HCC coding solution can help you improve your risk adjustment results, feel free to Contact Us.