Clinical Natural Language Processing for Medical Necessity Review
In our first installment of this four-part blog series, we explored how Clinical Natural Language Processing (cNLP) can impact risk adjustment. In the second installment, we discussed how cNLP improves the outlook on quality measures reporting—and specifically how it can impact the bottom line. This third blog segment will uncover the challenges of medical necessity reviews and why the ability to extract data from free text is critical to improve the process.
Medical Necessity Reviews: A Necessity in Healthcare
Medical necessity reviews have become an important part of documentation improvement and compliance programs in today’s healthcare organizations. Often conducted concurrently or retrospectively, these assessments ensure that coding accurately matches severity of illness on the provider side and that payers are reimbursing for the appropriate level of care.
Documentation is the foundation that supports medical necessity, and specificity is essential. Because more than 80% of valuable clinical information is captured in unstructured data types, such as free text fields, healthcare organizations need an efficient method of extracting insights to support medical necessity from the various, unstructured data sources.
Medical Necessity: A Deeper Look
Medical necessity denials continue to be an industry problem, especially with the introduction of ICD-10, and the need for greater specificity in documentation. Without the right blend of people, processes, and technology, healthcare organizations often find themselves facing revenue loss and compliance issues—whether it is coming from the payer or provider side of the house.
When it comes to Medicare reimbursement, the Social Security Act says that services are not covered if they “are not reasonable and necessary for the diagnosis or treatment of illness or injury or to improve the functioning of a malformed body member.” Notably, the Centers for Medicare & Medicaid Services (CMS) has named medical necessity the primary criterion for reimbursement, in addition to the requirements of a CPT® code.
The reality is that most private payers have policies that identify what procedures and diagnosis codes support medical necessity, as well. That means if the criteria are not found in clinical documentation and codified on a claim, payment will not be rendered. Clinical NLP can add value immediately by reducing review time, increasing staff efficiency, reducing administration costs, and improving the quality of your data for accurate analytics by unlocking data that might otherwise be missed.
The Challenge of Accurate Medical Necessity Reviews
Without a method for extracting key data from free text documentation, healthcare organizations run the risk of incomplete information when using analytics to support medical necessity reviews. This data can be found in a variety of places including history and physicals, medication lists, admission and discharge documents, and transfer orders.
For example, consider the medical necessity requirements for inpatient stays. Coding for MS-DRG requires that the diagnostic, procedural and discharge status matches the attending physician’s description and other information contained in the medical record. To validate MS-DRG code assignments, review teams must locate the supporting documentation in the unstructured text in various areas of the patient’s medical record. The record must meet the inpatient criteria as well as the specifics of the codes to ensure compliance.
Because concurrent medical necessity review ensures that coding is accurate before a claim is submitted, many healthcare organizations now rely on these proactive processes. The problem is that the manual methods many reviewers currently use to find data in free text is simply not conducive to real-time reviews, and it opens the door to oversights that may put a provider out of compliance or impact the bottom line. Today, this review process is commonly done by teams of highly skilled, highly paid physicians utilizing keywords to comb through patient charts and documents. Manual chart reviews to support medical necessity in a single patient record can take anywhere from 30 minutes to three hours depending on the information collected. Furthermore, time-intensive, repetitive manual processes by nature are error prone.
How Health Language Solutions Help
The Health Language Clinical Natural Language Processing (cNLP) Solution delivers the industry’s most comprehensive approach to identifying and leveraging unstructured data and is easily integrated into health IT environments through application programming interfaces (APIs). Through functionality that enables extraction, conversion, and mapping of free text to industry standards, the application ensures interoperability and meaningful exchange of data assets that have the power to provide deeper analytical insights.
The solution features industry-leading functionality such as:
- Proprietary cNLP Lexicon library contains 1+ million provider-friendly terms including clinical synonyms, acronyms, and common misspellings
- Embedded contextual awareness to support proper identification of the context surrounding clinical information
- Best-of-breed tool sets configured to desired client use cases, all coded to a single API
- Knowledge of multiple terminology domains out of the box
In addition, clients have access to the Health Language team of clinical informaticists who can rapidly assess needs and recommend optimal configurations for immediate results.
Be on the lookout for our fourth and final blog where we will explore the use of cNLP to empower predictive analytics.
CPT® is a registered trademark of the American Medical Association (AMA).