Clinical documentation integrity

How AI-driven CDI programs give people more of what we need most: time

The transition to value-based care has made doing more with less increasingly important, especially through healthcare business management. Patient documentation must accurately reflect not only the patient population’s health and history, but also the level of care required to appropriately care for patients. Today’s technologies and artificial intelligence (AI) solutions are designed to support CDI teams in their quests for efficiency, and a well-coordinated approach gives people more of what they need most: time.

This article was originally posted in ICD-10 Monitor, on September 9, 2019.

The concept of “doing more with less” is far from uncommon. You hear it every day in nearly every aspect of life: at home, we strive to live simply; in the community, we build resiliency; and at work, we are asked to boost productivity while improving effectiveness.

The transition to value-based care has made doing more with less increasingly important. That is, clinical documentation is a core function across healthcare business management, because documentation must accurately reflect not only the patient population’s health and history but also the level of care required to appropriately care for patients. Documentation that is either inaccurate or lacking in appropriate specificity (or both) can have a negative impact on reimbursements, quality ratings, denials, and—most importantly—patient health and wellbeing. On the other hand, we know from extensive available research that healthcare organizations with robust CDI programs consistently rank high in terms of quality, especially around observed-to-expected mortality ratings, an area that is most directly affected by the quality of documentation.

For example, if a patient’s documentation reflects a urinary tract infection, but that patient died, the documentation was clearly wrong. There was a missed diagnosis within the record, which not only affects reimbursements, but it also is publicly reported that healthy patients may die while in the care of that organization.

In short, we need to do more to work smarter and more efficiently. Through the focus on clinical documentation improvement (CDI), a great responsibility falls to coders and CDI specialists, as they’re charged with applying their skills and experience in new ways. Not only must these professionals possess the appropriate knowledge to help ensure excellent clinical documentation, but they must also have the finesse to collaborate with physicians and executives along the way. That is, their responsibilities have expanded, but they have no more time in the day.

Today’s technologies and artificial intelligence (AI) solutions are designed to support CDI teams in their quests for efficiency. Consider a clinical documentation specialist (CDS) which has 100 patient charts in their workflow. Research suggests that only 30 of those charts require follow-up or clarifications—but which 30 should you spend your time on? In the past, each record would need to be manually reviewed to uncover only those charts that represent quality improvement opportunities. Today, the right technology can help CDSs home in on those more complex cases automatically, allowing them to apply and use their skills more effectively and spend their time in ways that can make the most impact.

Similarly, AI can be applied to the front end of the documentation process, supporting providers and physicians as they add specificity where it matters most. A healthcare virtual assistant, for example, is an AI-powered platform that augments providers’ knowledge, recognizing natural language to make chart searches, EHR navigation, and CPOE more intelligent, simpler, and voice-powered. By helping providers create more complete and compliant documentation upfront, these technologies are creating efficiencies on the backend, giving even more time back to CDI specialists and coders. Now, they’re able to dive deeper into patient records, free from the weight of retrospective queries and rework.

Engaging the entire organization in denials prevention

Denials remain a costly burden for healthcare organizations, both in terms of financial impact as well as the amount of time coding and CDI specialists must dedicate to addressing denials as they arise. The rules around denials are fluid, and it can be difficult to keep up with them; this is complicated by the fact that once a claim has been denied, there is typically a short timeframe in which to address it. The process chews up time and other resources and requires extensive knowledge of the appeals process.

It is in this area that today’s clinical documentation technologies are well-positioned to support organizations’ denial prevention strategies. CDI and coding specialists can rely on tools that capture greater specificity in clinical documentation, mitigating the risk of denials upfront. On the backend, the right technologies will also include appeals templates to help CDI teams effectively and efficiently address denials as needed.

While the technologies themselves are designed to support any organization’s denial prevention strategy, adding a guidance and educational program will optimize the impact of these investments. These programs can drive toward organizational goals by keeping all clinicians and specialists on top of changing payor requirements and processes, helping continuously improve the quality of documentation, and engaging the entire organization in quality improvements.

In other words, the combination of educational programs and technology solutions help improve the quality of documentation upfront, which alleviates the burden of rework while also helping to prevent denied claims going forward.

The future of CDI: education, process changes, and AI-powered improvements

In December 2018, McKinsey published its research on the “promising” effects of natural language processing (NLP) in healthcare. Indeed, from the ability to identify appropriate and accurate ICD-10 diagnosis codes to the creation of time-savings opportunities, the influence of NLP—itself a subset of AI—can have a significant and positive impact on CDI teams’ ability to expand case coverage while maintaining resources.

This progress, of course, is made more powerful by the advanced analytics only available through innovative technologies that can aggregate and process large amounts of data. We hear consistently that one of the most challenging aspects of healthcare leadership is benchmarking: how are we doing? How are we doing relative to our own performance, or that of other organizations like ours? Where are our problem areas? Has our patient population changed and affected the case mix?

Getting answers to these questions used to be incredibly difficult. Today, however, with real-time intelligence that’s available within advanced CDI technologies, it has never been easier to uncover areas where education, process changes, and other interventions can have a positive impact and give all of us more of what we need most: time.

Optimize your CDI program

AI‑driven documentation integrity solutions that meet your needs

Learn more