Improving Diabetes Outcomes Takes Teamwork
Educate providers on how to paint a complete clinical picture in the medical record to ensure patients get the help they need.
Diabetes comes at a tremendous cost, from both a health and human standpoint and a financial standpoint, explained Colleen Gianatasio, MHS, CPC, CPC-P, CPMA, CRC, AAPC Approved Instructor, CCS, CCDS-O, at RISKCON 2021. She emphasized that because of the enormous impact of diabetes, “it’s something that we really have to get our arms around, from not only our industry’s standpoint but from a healthcare standpoint, as well.”
Thus, nowhere is specificity more crucial than in coding for type 2 diabetes, which Gianatasio described as “a chronic condition that runs through all of our different value-based reimbursement structures.” When we assign diagnosis codes, they affect not only risk adjustment, but also health outcomes and quality metrics. This is why “it’s important for us as coders and professionals in the business of healthcare to get this right and to tell these patient stories,” she told attendees, “so that we are part of the solution and not part of the problem.”
We must identify when providers fail to paint a complete clinical picture in the patient’s medical record and intervene. The key to fixing this problem, according to Gianatasio, is provider education. To help healthcare business professionals better understand this devastating disease and the documentation necessary to fully tell a patient’s story, Gianatasio broke down and demystified diabetes documentation and coding in her presentation, “Advanced Coding: Diabetes.” Here’s a recap of that session.
Leverage CDI to Improve Clinical Outcomes
Before delving into what to look for in clinical documentation and official guidance, Gianatasio discussed the evolution underway in risk adjustment. “Historically, risk adjustment was done retrospectively,” she explained. “We’re seeing that move closer and closer to the point of care, so we can really change that provider’s behavior … And we’re seeing a whole new world evolve around outpatient CDI [clinical documentation improvement].”
CDI is the bridge between the doctor and the coder — the CDI specialist looks for clues within the record and queries the provider for more robust information if they think “documentation might be missing and the doctor may not be giving herself credit for the work she’s already doing,” said Gianatasio.
With the shift toward the point of care, rather than reporting to providers about how they performed the year before, we’re able to interact with them before they see the patient and give them quality data that is “clean and actionable.” Gianatasio went on to say that CDI “is all about provider engagement, we have to get in there, we have to partner with our providers and give them the tools that they need to do the best job that they can to take care of that patient.”
Diagnosing Prediabetes and Diabetes
Results indicative of prediabetes:
- Hemoglobin A1C, which shows average blood glucose over the past two to three months, of 5.7-6.4%
- Fasting plasma glucose (FPG) of 100-125 mg/dL
- Oral glucose tolerance test (OGTT) two-hour blood glucose of 140-199 mg/dL
How is diabetes diagnosed?
- A1C: 6.5% or higher
- FPG: 126 mg/dL or higher
- OGTT two-hour blood glucose: 200 mg/dL or higher
- Random plasma glucose test: 200 mg/dL or higher
“Generally, your provider will do this twice to make sure that they have a true diagnosis of diabetes. Some providers, if it’s through the roof, might choose not to test twice. They might just get the definitive diagnosis with one test,” Gianatasio noted.
Recognize Query Opportunities
CDI specialists must be familiar with the symptoms and lab values indicative of diabetes to know when a query is needed. They are tasked with looking through progress notes for mention of conditions commonly seen in patients with diabetes such as:
- Polyuria – increased urinary frequency, excessive urination
- Polydipsia – excessive thirst despite drinking plenty of fluids
- Post-prandial hunger
- Blurry vision
- Weight loss, even when eating more (type 1)
- Paresthesia – tingling, pain, or numbness in the hands or feet (type 2)
After reviewing common symptoms, Gianatasio walked attendees through the various laboratory tests and corresponding results used to diagnose prediabetes and diabetes. She stressed that when you’re doing retrospective risk adjustment, the purpose of knowing this information is not for diagnosing, but rather identifying CDI query opportunities.
Ultimately, the goal is clinical outcome improvement. And a key part of that is working with providers to ensure the medical record conveys the patient’s complete clinical picture. Our industry is data-driven, Gianatasio said, but we have to remind ourselves “there are people behind this data, and if we do it right, it’s going to tell the story of these people, and it’s going to get them the help they need.” For example, if we can get our providers to accurately document and code prediabetes, we’re able to identify these at-risk patients and provide care management interventions such as educational programs, nutritionist appointments, and free gym memberships.
Learn more: Watch part of the RISKCON 2021 session where Colleen Gianatasio reviews the major diabetic code categories and explains the impact on patients in the digital version of Healthcare Business Monthly, available online when signed in to your My AAPC account.
Be Careful Not to Overcode
Screening for diabetes and accurately coding those encounters is important “because that’s our preventive care; that shows we’re taking really great care of our patients,” Gianatasio explained. Watch your providers to make sure they are classifying the patient’s condition correctly, she advised.
For example, if a patient is diagnosed with prediabetes, make sure to report it with R73.03 Prediabetes. If a patient has hyperglycemia in the emergency room and the doctor does a rule out to make sure they don’t have diabetes before initiating treatment, it’s vital to use the correct code to avoid erroneously miscategorizing a patient. When it’s just an abnormal glucose in a patient who does not have diabetes, it’s imperative to capture that by coding R73.09 Other abnormal glucose.
Keep in mind the big picture: What happens downstream if we erroneously code E11.9 Type 2 diabetes mellitus without complications, placing that patient into a diabetic category? “That not only puts the patient into a risk adjustment category that’s incorrect, which we’re going to try to recapture year after year, but from a quality standpoint, our HEDIS® and our STARS, people are going to say from a payer standpoint, we have to make sure all our diabetic patients have an A1C check every year, they have their eyes checked every year, and their kidneys checked every year,” Gianatasio pointed out. And as payers and providers start collaborating and reaching out more to better educate and manage, “if that patient doesn’t have diabetes, now we’ve not only created a problem for the provider who’s asked to recapture it year over year, the payer who’s responsible now for these quality metrics, but now we’re creating an abrasion with our members and our patients who are at the heart of all we do. So, follow these codes through; again, there’s a face behind the data.”
Strive for Precision
Diabetes is no longer classified as uncontrolled, a change that occurred with ICD-10. Pay close attention to make sure your doctors document the specific blood glucose abnormality and avoid using the term uncontrolled, especially when they’re trying to indicate hyperglycemia.
“If you go to the Index in your ICD-10 book and go to Diabetes, uncontrolled, it’s going to bring you to another decision point: hyper- or hypo-,” Gianatasio explained. “So, if that’s all your doctor documented, all you can code is E11.9, which is risk-adjusted at a different level than E11.65 [Type 2 diabetes mellitus with hyperglycemia].” Documentation that supports E11.65 includes “inadequately controlled,” “out of control,” and “poorly controlled.”
Pro tip: Make sure your organization’s decisions and practices are based on official guidance; sources of official guidance include the Centers for Medicare & Medicaid Services (CMS), ICD-10-CM guidelines, and AHA Coding Clinic. That way, “if you’ve been chosen for a risk adjustment data validation (RADV) audit, you could fall back on that guidance, and that would support your argument,” Gianatasio explained, “You can’t pull random books or random curriculum and use that in a RADV audit.”
Coding Education Best Practices
The key to documentation improvement is provider education.
1. Start with HCC awareness. Summarize why all chronic conditions caused by diabetes are lumped together and the impact from risk adjustment and care management standpoints. Emphasize that it’s important to document and code all systems affected by a diabetic manifestation as they illustrate the need for different care management and a higher level of severity.
2. Give ICD-10 code starters. Narrow down the list of over 300 diabetes codes by breaking down the major E11 Type 2 diabetes mellitus code groups and acquainting providers with the use of fourth digits to allow them to pull applicable codes quickly and easily.
- Diabetic nephropathy (E11.2-) for complications of chronic kidney disease (CKD) and end-stage renal disease
- Diabetic retinopathy (E11.3-) for vision complications
- Diabetic neuropathy (E11.4-) for nerve complications that commonly accompany type 2
- Diabetic circulatory problems (E11.5-) for vascular problems such as peripheral vascular disease (PVD)
- Other diabetic complications (E11.6-) for arthropathy, skin ulcers, and oral complications, which will require an additional code to document the complication
Stressing to the provider that these codes justify the need for specific — and very different — treatment modalities is tremendously important for risk adjustment purposes, Gianatasio argued. This means the unspecified type 2 diabetes code E11.8 should be avoided at all costs.
3. Teach what needs to be documented to justify each code. For diabetes coding, that means documenting elements such as fundoscopic, vascular, and neurological examinations; monitoring of HbA1C, glucose levels, urinalysis results; as well as treatments. While acronyms such as MEAT and TAMPER are helpful, Gianatasio suggests using a more provider-friendly version: DSP (diagnosis, status, plan).
“It’s clinically relevant to the patient that you capture all the ‘diabetes with’ complications because a patient with diabetic neuropathy is going to go into a different care management direction than a patient with diabetic retinopathy,” said Gianatasio. She emphasized that it’s vital to tell the whole story; if a patient is dealing with multiple complications, ideally, all of those conditions should be coded.
In a perfect world, providers would give us every single bit of specificity, Gianatasio said, but we’re not in an ideal world. What she tells providers is to code what they know for certain at the time of the encounter and also code for the body system the diabetes is affecting. So, if it is affecting three different body systems, you’re going to need three different diabetes combination codes. She pointed out that it is appropriate to code more than one ‘diabetes with’ complication code for a patient and that certain diabetes codes require additional codes to further identify the manifestation.
Gianatasio walked attendees through a slide that breaks down the coding for diabetes and brings things back to the physician’s perspective. She stressed that if you can get them to document information detailed enough for categorization to the fourth digit level of specificity, you now know what that patient requires, the direction their management needs to be headed in, and which specialists they might need to see.
What Does the 4th Digit Tell Us?
Diabetic nephropathy E11.2-
Patient Impact: Diabetes is the top cause of CKD and end-stage renal disease worldwide.
Coding Specificity: Use additional code for stage of CKD if known (N18.1-N18.6).
Diabetic retinopathy E11.3-
Patient Impact: One of the most important causes of visual loss worldwide.
Coding Specificity: Type, location, severity, and proliferative and nonproliferative.
Diabetic neuropathy E11.4-
Patient Impact: Peripheral polyneuropathy and autonomic neuropathies are some of the most common complications of diabetes.
Diabetic circulatory disease E11.5-
Patient Impact: Includes peripheral vascular disease (PVD), which is severely underdiagnosed and coded.
Coding Specificity: Carotid, aortic, renal, mesenteric, and coronary arteries are central arteries, not peripheral.
Other specified complications of diabetes E11.6-
Patient Impact: Arthropathy, diabetic skin ulcers, and oral complications of diabetes.
Coding Specificity: There is an instructional note to include additional code to identify the complication.
Work Together to Improve Outcomes
After discussing AHA Coding Clinic guidance and going through a few chart review examples, Gianatasio circled back to her take-home messages. Healthcare business professionals need to focus on educating providers on why detailed documentation that fully captures the clinical picture is so important and how it impacts patient management. CDI is all about provider engagement and education, which can be more easily achieved by speaking the doctor’s language. Brush up on pathophysiology and pharmacology, she advised attendees, because that will facilitate you being able to work with providers to get them to transfer the work they are doing mentally to the medical record. Remind them that comprehensive clinical documentation enables proper code selection, accurate data collection, and quality patient care, and will ultimately result in better clinical outcomes.
- CMS Releases 2022 Physician Fee Schedule Final Rule - November 11, 2021
- Relative Value Units: The Basis of Medicare Payments - November 1, 2021
- Improving Diabetes Outcomes Takes Teamwork - November 1, 2021