Artificial intelligence (AI) is transforming prospective risk adjustment in value-based care by helping payers and providers overcome the historic challenges of understanding patient health status in real time.
Everyone benefits from accurate patient RAF scores – payers, providers, regulators and certainly patients. No one questions the necessity of retrospective risk adjustment, but data volumes, deadline pressures and the urgency to capture and recapture all create friction.
The promise of prospective risk adjustment programs is that diagnosing and coding conditions annually at the point of care make retrospective risk audits more accurate and run more smoothly. Unfortunately, prospective risk programs fall short when health plans rely on fragmented or limited data sets and require clinicians to upload data or review insights outside their workflows.
For prospective risk programs to succeed, value-based care organizations must pull together the full picture of patient care journeys, surface new, relevant insights and deliver those insights when and where they are most impactful.
This can feel like an impossible mission without technology, but AI allows providers to navigate the three biggest challenges to successful prospective risk programs.
1. Capturing the full picture of patient health
How can an industry drowning in data need more data? Despite generating massive amounts of data, payers and providers can lack visibility into complete patient care journeys in a health system as fragmented as the US.
AI excels in aligning vast amounts of disparate data, both from electronic health records (EHRs) and external sources such as pharmacies and labs. Technology such as Optical Character Recognition (OCR) combined with NLP takes a picture of every page of clinical documentation and aligns it under a specific patient.
Once AI organizes patient data, the technology can align each record and encounter into a longitudinal view of patient health, which is crucial for accurate and complete risk adjustment.
2. Finding insights in a mountain of medical data
Without technology, the volume and complexity of medical data can overwhelm those who need it, if reviewed manually.
In contrast, AI and natural language processing (NLP) can address this challenge by automating data processing and analysis. NLP models extract relevant information from unstructured clinical notes and uncover evidence of chronic conditions through medication information, patient history data and interview notes, or past diagnoses that could predict the presence of undocumented conditions.
NLP “reads” patient data to identify suspected undiagnosed conditions, extract clinical evidence, and highlight where in the source document to find it. This allows clinicians to focus on patient insights faster than by manually scrolling through hundreds of pages of medical charts.
3. Reducing provider abrasion through EHR integration
Value-based models rely on clinicians to document chronic conditions on an annual basis, but plans can struggle to even recapture known conditions without engaging providers. This puts pressure on retrospective audits and creates urgency to review documentation, manage additions or close care gaps after the patient encounter. Throughout the year, plans can inundate providers with requests for documentation or to address open diagnoses.
The volume of requests and the steps providers must take to complete them can cause provider abrasion. Even online gap closure systems can struggle with low participation when the assessments take providers outside of their workflow and require updating both a third-party external system and the provider’s own EHR.
With AI, clinical insights can be organized and displayed in ways that make sense to providers and become actionable for risk adjustment. In this scenario, clinicians access patient insights at the point of care within their EHR, allowing them to quickly review suspected diagnoses, the associated clinical documentation and add to a recapture workflow with one-click.
A new evolution in prospective risk adjustment
Not all prospective risk solutions are created equal. The right ones overcome historic challenges to successful prospective risk programs by building a complete view of patient care, surfacing new clinical insights, and delivering those insights where they have the most impact – the point of care.
Our Clinical Intelligence solution curates insights that identify gaps in care in a clear and understandable way for providers. By synthesizing clinical data and predicting potential diagnoses or care gaps, AI empowers clinicians to accurately document patient conditions during patient encounters without disrupting existing current workflows.
Real-time support from AI enhances the quality of care in a way that could never be done manually. It also ensures that clinical coding for reimbursements reflects the true complexity of patient care and makes patients healthier. As a result, healthcare payers can achieve more accurate reimbursements while simultaneously improving the lives of patients.
To get more insight on how AI is fueling success in value-based care, read our Guide to AI-Powered Risk Adjustment.
OIG is auditing more health plans for Medicare and Medicaid Overpayments at a time when payers are already concerned about lower reimbursements.
Learn More →AI solutions are transforming provider workflows through clinical system integration and by driving insights to the point of care.
Learn More →Value-based care companies are using AI to more completely understand patient health status and find undiagnosed conditions through prospective risk adjustment programs.
Learn More →