[ad_1]
Artificial intelligence (AI)-enabled tools such as natural language processing (NLP) have been integrated into a wide range of applications, including risk adjustment coding tools, for greater efficiency and accuracy in the healthcare industry. For Medicare Advantage (MA) plans, these tools can significantly improve the accuracy of diagnostic and hierarchical condition category (HCC) data needed to support risk adjustment programs and help ensure appropriate reimbursement.
Preparing for new RADV changes
With NLP-driven tools, MA plans can uncover errors during retrospective chart reviews before a risk adjustment data validation (RADV) audit. Once required only for roughly 10% of MA plans each year, RADV audits will now affect all MA plans as the Centers for Medicare & Medicaid Services (CMS) step ups efforts to reduce overpayments.
As part of its aggressive strategy, CMS will also audit a larger number of records — as many as 200 records per plan. The policy change underscores the need for both accuracy and efficiency for MA plans.
The RADV audit expansion follows other significant policy changes, which now allow CMS to extrapolate its audit findings from the sample of medical records reviewed to the entire plan contract — potentially putting a single contract at risk for millions if the agency decides the records do not adequately support enrollees’ diagnoses. The elimination of the fee-for-service (FFS) adjuster also increases the burden on plans to ensure accurate and complete HCC reporting or risk extrapolated penalties.
How AI can help MA plans
For MA plans that have not previously undergone a RADV audit, these changes offer a timely opportunity to integrate AI into their coding practices and establish appropriate policies and procedures with the technology.
By incorporating AI-enabled tools into their workflows, MA plans can prioritize critical documentation and ensure their coding teams concentrate on the most relevant areas of lengthy, complex medical records. For example, these tools can easily identify common mistakes such as HCCs reported in the wrong setting (inpatient versus outpatient) or by the wrong specialty. NLP-enabled tools can also help coders quickly find instances in which retrieved medical records for two different members were accidentally merged, which creates inaccuracies for retrospective chart reviews or RADV chart submission processes.
Strategies for rolling out AI-enabled tools
Following are best practices for plans to consider as they implement AI-enabled tools to improve the accuracy of their coding and risk adjustment programs.
Launch an AI governance committee for human oversight. Plans should establish a framework for vetting and overseeing new uses of AI or NLP in their organizations. By creating a governance committee of clinical, technical and coding experts, plans can review different use cases for AI and have a forum to raise concerns about potential inappropriate uses. To guide organizations in healthcare and other industries, the Responsible AI Institute offers best practices for AI governance structures, as well as principles for reviewing AI projects. Adhering to guidance from industry advocacy groups can help leaders ensure ethical implementation of AI in coding and other areas.
Create a “sandbox” environment for coders to test the tool. Providing coders with test documents so they can experiment with the tool can help them practice workflows they will experience in real life. Plans can also provide a user checklist to help coders simulate various scenarios and record any issues related to performance or usability.
Publish a scorecard with metrics to measure performance holistically. Leaders should maintain an ongoing commitment to evaluating the performance of AI-enabled tools. Plans should view their performance holistically and track overall and individual productivity and accuracy metrics. Potential red flags are coders who, when compared to peers, are exceptionally slow or fast when using AI tools. Plans should also look for signs suggesting an overreliance on AI, such as a coder who accepts AI-generated suggestions nearly 100% of the time. The specific benchmarks set by plans should depend on factors such as their line of business, the type of software used and whether the data is pulled from electronic medical records (EMRs) or scanned PDF records. Plans should review their metrics at least monthly to identify opportunities for improvement and share the results with key stakeholders.
Leverage end-user feedback for continuous improvement. Soliciting feedback from coders is essential to ensure a positive user experience. Sometimes, coding tools that generate excess recommendations for coders can slow them down, hamper productivity and create frustration. Having coding “superusers” submit suggestions to managers and leadership can help continuously refine the technology and procedures.
Align on performance expectations with vendors. If plans leverage AI-enabled software through a coding partner, they should have performance guarantees related to system performance, uptime/downtime metrics, and NLP accuracy with deadlines and potential penalties for delays. This can help safeguard plans against system outages and other issues that could potentially derail their project deliverables and reporting deadlines.
Preparing for new CMS auditing efforts for MA plans
As CMS ramps up its RADV initiatives in the months ahead, plans should ensure their risk adjustment programs meet the highest standards of accuracy and compliance. Prospective and retrospective analytics enhanced by AI can help plans work with providers to optimize documentation at the point of care and identify coding errors during audit preparation. Plans may also want to consider conducting a second-level review of coding results, which allows them to correct unsupported HCCs that could easily be overlooked during the first-level review. By combining AI-enabled tools with expert oversight, plans can improve the success of these efforts as they encounter greater regulatory oversight in the future
Photo: Thanakorn Lappattaranan, Getty Images
Katie Sender, MSN, RN, PHN, CRC, is vice president of clinical and coding services for Cotiviti. With more than 25 years of healthcare experience, Katie is responsible for leadership and management oversight of teams spanning the globe to ensure optimal client outcomes and service delivery through management of key performance indicators related to Clinical and Coding Solutions.
This post appears through the MedCity Influencers program. Anyone can publish their perspective on business and innovation in healthcare on MedCity News through MedCity Influencers. Click here to find out how.
[ad_2]
Source link
