Saving Time and Reducing Cost with cNLP

Aug 18, 2023

The review of medical records is critical in determining whether claims are billed and reimbursed accurately, and DRGs are the primary methodology for both hospital billing and health plan reimbursement. Although the overall methodology is well constructed and documented, several challenges are associated with a DRG review due to the reliance on the unstructured medical record, which can often be hundreds to thousands of pages.

While health plans have made tremendous progress in developing workflow and processes to create a more efficient business process, the medical record review process is still entirely manual. According to health plan executives, this factor inhibits the desired strategy of moving these reviews to a pre-pay environment or “shifting them to the left.”

Due to this challenge, health plans are looking to AI to assist in automating the costly administrative requirements associated with the DRG review. Technical leaders have confirmed these shifts to AI—including data integration, natural language processing (NLP), business intelligence (BI), and data annotations—were their top priorities for implementation.

What is cNLP and How Does it Help the Review Process?

Natural language processing (NLP) is an AI process used to increase the speed of traditional medical record reviews. NLP teaches computers to understand text datasets, including their contextual nuances. A subset of NLP, Clinical Natural Language Processing (cNLP), is tailored to understanding clinical text. For example, a myocardial infarction is another name for a heart attack; cNLP knows this contextual information and can find records with both terms. cNLP can also divide data based on context; it can take a myocardial infarction and determine if the record describes a family history of heart attack or if the patient was admitted to the hospital with a heart attack.

Benefits of cNLP in DRG Review

Searching and Reading to Validation

Reviewing medical records is a high-cost skill set, and finding individuals with these skills is expensive since most DRG audits require both coding and clinical experience. In addition to expertise, manual reviews require considerable time and additional resources to complete.

cNLP can be used to streamline laborious medical record reviews. Instead of requiring skilled staff members to perform low-value tasks such as type, search, read, and repeat, cNLP automates that process, allowing reviewers to focus on high-value record validation and final determination. AI technology reduces your team’s time searching for information because cNLP assists organizations in streamlining their manual review process, leaving them more time for higher-value, more critical tasks.

Risk Management

cNLP finds appropriate data throughout the medical record and leverages it to examine factors such as ventilator time. Additionally, it can quickly identify baselines and trend lines by recognizing the dates and values of key clinical data. This way, the auditor can focus on the validated diagnosis instead of duplicated facts or muddled information—facilitating decision-making and reducing the risk of human error.

Streamlining the Reviews Process

Consistency is key, so streamlining internal processes is essential for teams of all sizes. Using cNLP to streamline your review process improves consistency among auditors twofold. First, by integrating evidence-based criteria for validating diagnoses for novice users, and second, by bridging the experience gaps between members within your internal team. Differing views are common and understandable. Still, auditors should be able to pick up a caseload where someone else left off with ease and clarity.

The utilization of cNLP represents a significant leap forward in revolutionizing the medical record review process. The challenges posed by unstructured medical records and the manual nature of the review process have impeded the desired progression toward increased accuracy and efficiency. By harnessing AI’s abilities to automate tasks, enhance risk management, and streamline processes, cNLP ushers in a new era of productivity. As health plans increasingly turn to AI-driven solutions to overcome the challenges of manual review processes, cNLP stands as a beacon of transformative potential.

How TREND Can Help

CAVO® is a revolutionary technology platform that supports various complex clinical claims review processes performed by health plans and services companies. CAVO empowers highly skilled clinical and coding resources with AI-driven functionality that shifts the focus from “low value”  tasks and requirements such as document access, search, analysis, and determination support within medical records and other unstructured data to “high value” tasks and requirements such as validation and final determination. The technology enables clinical and coding reviewers to easily access and structure medical records, itemized bills, and additional clinical data efficiently, consistently, cost-effectively, and profitably.

With over one million completed case reviews, CAVO delivers technology proven to scale up your medical record reviews and increase your team’s productivity by 300%-500%.