by Monique Pasley, RHIT, CCS
Senior Manager, Clinical Analytics at AppRev
July 17, 2018
It’s no secret that coding is one of the most important functions of the HIM department. Code selection must be accurate to ensure proper reimbursement and depict a clear picture of a patient’s condition, medical history and they care they received .
It is important for HIM leadership to be aware of the strengths and weaknesses of their coding professionals, especially those leaders who haven’t had direct experience in coding. Compounding the complexity of this issue is a new challenge facing coders: computer assisted coding (CAC). While this software can increase productivity, coders should be carefully trained to understand it’s limitations.
CAC programs can recommend diagnosis or procedure codes from any of the code sets, but they only do so by using the words that are specifically documented in the record. Through natural language processing (NLP), these programs have the ability to fine tune and learn more specific recommendations. In order for the tool to learn, however, coders must be knowledgeable and diligent to make the consistent changes needed.
Diagnostic combination codes are a good example of the limitations of CAC. For instance, chronic kidney disease and hypertension are connected in ICD-10CM , but what if the provider states that the kidney disease is not due to hypertension? CAC can’t flag the coder to alert them of this scenario, yet combination codes like this one directly impact sequencing. On an inpatient account, this can impact the DRG.
When using CAC, coders need to be prepared for accepting procedure codes without completely analyzing the procedural report. On an operative report we often see acronyms for the device used, but those may conflict with the procedure title. For example: AICD is a cardio-defibrillator, but the title of the operative report says this is a pacemaker insertion. CAC will likely recommend the pacemaker procedure. A coder who is not indexing their own codes via the encoder or book may not even be aware there are different code choices for defibrillator vs pacemaker. Choosing one over the other will directly impact code selection and, ultimately, reimbursement.
It’s important for outpatient and inpatient coders to be well trained in the nuances of the coding rule exceptions, so they can be their own advocate and proactively research the codes recommended by CAC. In truth, CAC requires a coder to be more of an auditor and an expert at predicting possible errors than to simply be a coder.
In a study conducted by the American Health Information Management Association (AHIMA)1 to determine the impact on CAC accuracy, two teams were assigned to code inpatient accounts: one team coded with the help of CAC and the other coded without. The study found that while productivity can increase with the use of CAC and the capture of all related codes were the same between the two groups, the tool is most effective when utilized by credentialed and well-trained coders (those previously meeting or exceeding 95% accuracy rates).
While advancements in technology do improve efficiency, it’s important to remember that coding accuracy still requires human interaction and expertise. AppRev’s Charge Accuracy solution provides the perfect balance of software with expert consulting to ensure that quality remains the top priority of a hospital’s coding practices.
- Dougherty, Michelle; Seabold, Sandra; White, Susan E. “Study Reveals Hard Facts on CAC” Journal of AHIMA 84, no.7 (July 2013): 54-56.