Medical coding and clinical documentation conditions have become more complicated in recent years. This added complexity can produce a notable administrative burden for doctors and contribute to operational incompetence.
Clinics and health practices often leverage technology solutions, such as natural speech processing, to help alleviate the burden of the doctor and improve clinical documentation efficiency.
Becker’s Hospital Review lately spoke with Jason Mark, director of Data Science and Natural Language Understanding teams for 3M Health Information Systems, about how natural language understanding (NLU) — the most advanced development in natural language processing (NLP) technology — is streamlining workflows for clinicians, clinical documentation development professionals, and coders.
He received how language recognition, more complete patient information, and better pattern exposure can improve performance and empower clinicians with enhanced patient insights. The latest COVID-19 pandemic offers a unique backdrop to show the power of these technologies.
In-Depth Coding
NLP is a comparatively mature technology. With NLP, computers can read the language, see parts of speech like nouns and verbs, handle entity recognition such as the names of doctors or medications, and get message intent. Natural language understanding (NLU) enhances NLP, augmenting it with deep content sources, such as clinical data and in-depth coding experience. This generates a robust therapeutic information design.
“As we implement NLU to a patient’s electronic health report, we can receive a complete picture of the person’s health,” Mr. Mark stated. “We aren’t just picking out a few hints and stating, ‘We see this plus that, so we think the sufferer has diabetes.”
NLU offers data on how many health states interact with one another in addition to delivering more precious insights into specific health requirements.
NLU also can discriminate between clinical expression and the administrative and financial language used for revenue cycle and coding.
Another benefit of NLU is the ability to explain the full patient encounter, not just a particular record. This is important, especially when working with hierarchical condition categories that require a longitudinal registration.
The Clinical Picture
“We have a more general model which helps us appreciate the patient’s clinical picture across all of the data available to us,” Mr. Mark stated. “For example, we know that three visits ago, you were checked for a state that needs follow-up, and that hasn’t arrived yet.”
NLU-based solutions eliminate additional responsibilities for healthcare specialists. By adding layers of deep thought to electronic health records (EHRs), these policies recognize the possible data and limit queries to providers as they manage information.
A natural limiting factor in adoption by providers is “alert fatigue.” Whenever a clinician speaks the word ‘diabetes,’ for example, some systems help the provider to ask the sufferer five questions. In contrast, NLU methods know whether these issues have already been answered elsewhere in the case record.
The Lab Results
“We check provider overload by only asking about what is relevant and what can’t be answered with existing data,” Mr. Mark responded. “When we get documentation right from the beginning, it degrades query rework loops, which saves time and effort for everyone downstream.”
When problems arise inpatient records, NLU-based practices address them based on various user roles. For instance, if a patient record has a documentation gap or if inequality exists between lab results and medication, the order might nudge the provider and remind them to document the point upfront.
If the user is a clinical documentation development specialist, the system will help them craft a query. Taking a forward approach to user tasks, lowers game levels for everyone.
Social Determinations
The advantages of NLU extend beyond the revenue cycle. Thanks to more powerful computing potential and access to larger volumes of data, NLU-based solutions conduct first pattern recognition compared to care relationships and strong demographics. “We can now be more sensible about social determinants of health,” Mr. Mark stated.
“When a doctor directs a medication, systems will usually be able to define whether the sufferer will be able to fill it based on their insurance, benefits, access to transport, and more.”
