As healthcare organizations continue to utilize data that is becoming increasingly complex, it is essential that the data is captured and documented properly. Clinical documentation improvement (CDI) helps ensure that the events of the patient encounter are captured accurately and the electronic health record properly reflects the services that were provided.
“Successful clinical documentation improvement (CDI) programs facilitate the accurate representation of a patient’s clinical status that translates into coded data,” the American Health Information Management Association (AHIMA) explains on its website. “Coded data is then translated into quality reporting, physician report cards, reimbursement, public health data, and disease tracking and trending.”
CDI also ensures that all members of the care team receive information on a patient, AHIMA added.
There can be great financial benefits from CDI as well. Nearly 90 percent of hospitals with more than 150 beds and outsourced clinical documentation functions saw gains of at least $1.5 million in appropriate healthcare revenue and claims reimbursement following CDI implementation, a 2016 report from Black Book Market Research found.
Of the nearly 900 healthcare leaders interviewed, 85 percent said their organization saw quality improvements and case mix index increases after implementing a CDI program.
Eighty-seven percent of hospital financial officers reported that case mix index improvement was the largest motivator for CDI adoption because of its potential to increase healthcare revenue and optimize high-value specialist utilization.
“The need for proper clinical documentation improvement driving quality outcome scores has never been more essential,” Black Book Managing Partner Doug Brown said.
Clinical documentation improvement can play a critical role in the move toward value-based care, but healthcare organizations must take advantage of tools and guidance put forth by stakeholders and the government.
Changing federal initiatives and evolving technologies involving artificial intelligence (AI) will also help support successful CDI programs that ensure patient care remains a top priority.
Using CDI to positively impact patient care
A 2016 study from the American Medical Association (AMA) found that for every hour clinicians spend with a patient, the clinician then spends two hours on EHR documentation. Providers spend 27 percent of their work time on direct patient interactions, and about 49 percent on EHR documentation.
Additionally, physicians spend an average of two hours working on EHR data entry outside of their office hours.
"This study reveals what many physicians are feeling – data entry and administrative tasks are cutting into the doctor-patient time that is central to medicine and a primary reason many of us became physicians," Steven J. Stack, MD, a past president of the AMA said in a statement.
“Clerical tasks and poorly-designed EHRs have physicians suffering from a growing sense that they are neglecting their patients as they try to keep up with an overload of type-and-click tasks," Stack added.
Workflow documentation tools, such as pre-structured data elements, can also help to improve the quality of documentation while reducing the time spent charting and free up more opportunities to connect with patients and devote attention to their needs.
Clerical tasks and poorly-designed EHRs have physicians suffering from a growing sense that they are neglecting their patients as they try to keep up with an overload of type-and-click tasks.
Phoenix Children’s Hospital recently opted for this approach at its outpatient clinics while looking for a more seamless way to track quality measures and monitor patient outcomes for improved chronic disease management.
Chief Medical Information Officer and VP Vinay Vaidya previously told EHRIntelligence.com that the hospital converted its full ambulatory outpatient clinic to an electronic system four years ago.
Around that same time, Phoenix Children’s came across an option to have more than 360,000 terms “structured and coded with every possible common coding schema, such as ICD-10.”
“What we got from the documentation solution was a way to almost have not the final form of the template, but an initial rough draft with the heavy lifting quickly, rapidly across our 30 divisions,” he said.
“At the end of last year when we finished our implementation, we then focused on getting this data back from their notes and creating real-time dashboards that impacted care.”
Vaidya added that overburdening physicians was a very real concern, but that Phoenix Children’s tries to regularly see the timeline of when the physicians complete documentation.
“Are they taking the work home on Saturday or Sunday? You can see small slivers, two people and three people, and are they completing the documentation by 5:00 PM on the date of service,” Vaidya said, discussing an example of the documentation program. “Almost 56 percent of them are completing that.”
The CMIO added that the hospital is seeing that nearly all documentation is completed within 24 hours, and that Phoenix Children’s is “not creating a lopsided data collection clerk out of a physician.”
What type of guidance on CDI exists for providers?
AHIMA HIM Practice Excellence Director Tammy Combs explained that patients must remain a top priority, even as physician workflow increases.
Documentation becomes more critical as providers continue to rely more and more on coded data, she said.
“You know the old saying: if it’s not documented, it didn’t occur,” Combs stated. “Now everything that's documented in the health record, is translated out into either an ICD-10 CM, ICD-10- PCS, or CPT® code.”
“These codes are how providers are recognized on the quality of care that they've provided,” she continued. “This also impacts their reimbursement. [The code] tells the payer that's looking at these denied claims or researchers out there what diagnoses occurred. It helps determine the outcomes. If the documentation is not in place, then the accurate code cannot be assigned.”
Provider education is key with CDI because there is truly a need for that higher level of specificity in the documentation process.
“It’s important to provide education out to those provider groups so they understand what CDI is and why it's needed,” Combs said. “Explain it's a resource for them to utilize, not a hindrance to them. It’s a resource to ensure that they get credit for the work that they're doing by validating that documentation is of high quality.”
Regulatory requirements related to clinical documentation have been among the top challenges cited by the clinical community, ONC Chief Medical Officer Dr. Thomas Mason told EHRIntelligence.com.
“Documentation requirements that focus more on billing have taken away some of the ability for clinicians to document what's most appropriate for care,” he explained. “We've also heard that the health IT itself has been challenging to use in terms of efficiency and effectiveness and usability.”
“Clinicians also believe that reporting for a variety of quality programs has been an issue in terms of fulfilling the requirements for a variety of quality programs.”
Documentation variation also exists between different types of clinicians, and focusing on different aspects of the documentation is essential for providers to formulate their diagnosis and their assessment and plan, Mason acknowledged.
“Depending on the clinician, there is going to be a variation in what's most essential for them to determine the most appropriate diagnostic and treatment plan for their patients,” Mason said. “A one-size-fits-all for the documentation templates within electronic health records was really problematic in the earlier adoption phase of EHRs.”
ONC is working towards a solution though, he added. The agency is working closely with CMS on a report to Congress required under the 21st Century Cures Act. That report calls for HHS to establish a goal, strategy, and recommendations with respect to reducing regulatory or administrative burdens relating to the use of EHRs.
The patient is becoming more of an active participant in the documentation process. The sharing of information also improves the accuracy of the documentation process.
“We've been working side by side with CMS and looking at what are the challenges, what are potential recommendations and solutions,” Mason said.
The first priority is addressing documentation requirements, which was a top clinician concern. The second focus area is health IT usability and user-centered design, Mason said.
This includes looking at the actual functionalities, capabilities, and the usability of the software. It also involves seeing what the challenges are and how ONC approves the health IT that clinicians are using.
EHRs and quality reporting is another major area of concern raised by providers.
Finally, other governmental requirements, such as burdens related to the use of prescription drug monitoring programs (PDMPs) and burdens with electronic prescribing (e-prescribing) of controlled substances, are also generating comment.
“How do we make those technologies more user-friendly for end users, and how do we integrate those better into the clinical workflow?” Mason posited.
OpenNotes is one initiative that is gaining traction and allowing patients to be able to view their clinical notes, which in turn is also impacting – and improving – the clinical documentation process for providers.
“There is an increasingly open environment in the exam room, where now clinicians are more often sharing information and have, in some cases, even modified their exam room to be able to have the patient more engaged,” Mason said. “The patient is becoming more of an active participant in the documentation process. The sharing of information also improves the accuracy of the documentation process.”
How will natural language processing impact CDI?
Natural language processing (NLP) holds a lot of promise in terms of being able to extract salient data points from unstructured data, such as the narrative of progress notes, Mason stated.
“This could drastically improve the amount of data available for analytics and a host of other initiatives, and could significantly reduce the need for manual chart review and abstraction, which could reduce cost, and also reduce time,” he explained.
“One of the challenges is really the accuracy of the NLP algorithms,” Mason continued. “There's really very little room for inaccuracy. When leveraging AI and NLP, it's important to assure that any emerging technologies are adequately tested before they enter the production domain to ensure patient safety.”
There's a significant amount of promise in NLP, Mason maintained. There is also potential in improving upon existing voice recognition technologies. This would better allow or enable clinicians to interact with their EHR using voice commands and voice recognition that is better integrated with the current EHR documentation functionalities.
Customization has been a problem with this approach, he added.
“The model has been to allow more customization in terms of documentation templates and less standardization around what are the best practices for a particular clinical workflow, and what are the technologies that best support those workflows using natural language processing or voice recognition,” Mason explained.
“That's something that [ONC has] heard and has addressed in terms of the report that we're writing for HHS, and our section on usability and health IT. It’s thinking through what are the best ways to better leverage AI and documentation, capabilities and functionalities to improve the clinical documentation experience.”
When leveraging AI and NLP, it's important to assure that any emerging technologies are adequately tested before they enter the production domain to ensure patient safety.
Dictation and NLP can help reduce physician workloads, including clinical documentation time demands, according to a 2016 study published in The Journal of Medical Internet Research.
Researchers tested four different clinical documentation approaches, finding that clinical documentation approaches incorporating NLP were more effective than pure standard approaches.
“We found that a pure protocol of NLP Entry as well as hybrid protocols (involving both NLP Entry and Standard Entry) showed promise for EHR documentation, relative to Standard Entry alone (Standard-Standard Entry),” the researchers said.
CDI approaches took an average of 16.9 minutes for cardiologists, while the NLP model took 5.2 minutes.
Even so, there is still room for improvement with ensuring accuracy with NLP, the team suggested.
“These findings suggest that, pending further study, EHR documentation methods using a combination of dictation and NLP show potential for reducing documentation time and increasing usability while maintaining documentation quality, relative to EHR documentation via standard keyboard-and-mouse entry,” researchers explained.