Otolaryngology Coding Alert

Mythbuster:

Ditch These Coding Habits to Diminish Denial Distress

Increasing first-pass payment success might be easier than you think.

Submitting clean claims is the goal of any practice, but if you’re making even the simplest errors, such as coding from memory, you are putting yourself at risk for denials. One way to steer clear of claim issues is to adopt the phrase “quality over quantity” as your mantra. While a little cliché, in the world of medical coding, quality records can mean the difference between claim acceptance and denial.

Keep reading to learn about three workflow myths that might be hindering your claim acceptance rate, along with examples to help you kick your bad coding habits.

Myth 1: Coding From Memory Boosts Efficiency

Chances are, you’ll make more errors when coding from memory. In otolaryngology, you might see a handful of codes frequently, and you may have even made a cheat sheet with commonly used codes. However, skimming a record and entering a code from memory or a cheat sheet is likely to cause problems in the long run.

Kick the habit of skipping steps and plugging in codes from your memory bank. “Experienced coders can definitely make mistakes due to complacency,” notes Brooke Hullett, CPC, CRC, CFPC, an independent clinical coding specialist with The Veranda Medical Group in Nashville, Tennessee. “It is important that we not only stay up to date on code and policy changes, but also continue to challenge ourselves by creating new goals,” Hullett adds.

Myth 2: When in Doubt, Downcode

This myth must be busted because it is not only incorrect, but it can also lead to investigations and allegations of fraud. “Coding lower to avoid problems is an old school mentality to avoid problems, but CMS [Centers for Medicare & Medicaid Services] has been clear that any inaccurate coding, high or low, is inappropriate,” explains Chelle Johnson, CPMA, CPC, CPCO, CPPM, CEMC, AAPC Fellow, billing/credentialing/ auditing/coding coordinator at County of Stanislaus Health Services Agency in Modesto, California.

“The importance of the medical records is more than just accurate coding,” explains Johnson. “The medical records are a history of what was addressed and treated during the visit. If the documentation is too vague to code, then most likely there are significant gaps in the documentation that need to be addressed to have an accurate record of what occurred,” Johnson continues.

Example: A patient with diabetes and recurrent middle ear infections presents with suspected otitis and concerns over hearing loss. The otolaryngologist documents “chronic otitis media” and “diabetes without complications.

Section I.C.4.a.2 of the ICD-10-CM Official Guidelines tells you that if the provider doesn’t include the type of diabetes, you should default to E11.- (Type 2 diabetes mellitus). Since the provider noted the diabetes was without complications, that guideline leads you to E11.9 (Type 2 diabetes mellitus without complications). But you’ll still want to check with the provider to verify whether it’s type 1 or type 2, as treatments and costs vary dramatically, potentially having a significant impact on the patient’s health and insurance coverage. You should also ask them for additional information on the chronic otitis media because details about type and laterality will affect code selection, enabling a higher level of specificity than H66.90 (Otitis media, unspecified, unspecified ear).

Education and communication are the keys needed to ensure an accurate claim. “By having open conversations with our providers in regard to compliance guidelines, appropriate leveling, and documentation, these issues can be eradicated or at the very least drastically reduced,” says Johnson.

Additionally, the term “without complications” implies not only that there are no complications (which is rare), but also that the condition is actively being controlled. Without Z79.4 (Long term (current) use of insulin) or another code to indicate a specific treatment modality, the insurance company could deny the claim.

Coding tip: For E10.- (Type 1 diabetes mellitus), reporting Z79.4 is not required. Type 1 patients are insulin-dependent, so insulin use is automatically implied by the code. You’ll notice E10.- is the only diabetes mellitus category without a “Use additional” note for insulin.

Myth 3: Always Take Your Provider’s First Word on E/M Levels

Most providers are not coding experts, which is why it’s so important for coders and auditors to continually educate providers. “Just because the boss said to do it, does not make it correct,” says Johnson. “I work with my coders, billers, and auditors to ensure that not only do they know the appropriate action to take, but why they are taking that action … without the why, then the staff is just acting mechanically,” Johnson adds.

Example: Suppose your provider spent 42 minutes with a patient and tells you the time spent justifies reporting the office visit as 99215 (Office or other outpatient visit for the evaluation and management of an established patient … high level of medical decision making … When using time for code selection, 40-54 minutes of total time is spent on the date of the encounter.). However, you disagree.

The medical record states that the patient presented with a runny nose, postnasal drip, and itchy eyes, and was ultimately diagnosed with allergic rhinitis. If the provider spent 40-54 minutes on that encounter, the insurance company is going to wonder why. On paper, the encounter looks simple and low-risk, and the diagnosis raises questions about the medical necessity of reporting 99215. The coding may be appropriate from a CPT® perspective and still be inappropriate and result in denial based on the medical necessity perspective as defined by the payer.

From the expert: Taking a few extra seconds to double-check a code or proofread your work is usually enough time to catch a careless error or two. “It is much quicker to double-check your work and send the claim correctly the first time. This keeps everyone paid and happy,” says Hullett.