The average practice submits half of its codes wrong, while some practices rarely exceed more than one code right out of every five codes. Inexact and inconsistent coding increases the risks of undercharging, overcharging, and post-payment audit. This article outlines evolution of coding from individualistic art towards disciplined and systematic process.

It is convenient to review the role of coding in the context of the entire claim processing cycle, which consists of patient appointment scheduling, preauthorization, patient encounter note creation, charge generation, claim scrubbing, claim submission to payer, and followup, which in turn includes denial or underpayment identification, payment reconciliation, and appeal management. The importance of thorough knowledge and correct application of coding rules at the charge generation stage of claim processing cycle are well known and have been frequently discussed. Less obvious but no less important is the ability to make correct interpretations of the same rules at the claim followup stage during denial or underpayment analysis and upon receiving payment and explanation of benefits.

Coding is difficult because of a four-dimensional complexity. First, the sheer volume and intricacy of coding rules make it difficult to select the right procedure code, correct modifier, and necessary diagnosis code for the given medical note. For instance, a claim will get denied if you charged for two CPT codes but provided an ICD-9 code that shows medical necessity for one CPT code only. Next, the payer-specific modifications exacerbate the complexity of coding, creating the need to code or process differently the same procedures depending on the payer. For example, some payers require medical notes attached to some CPT codes in addition to standard ICD-9 codes. Third, the codes and regulations change over time, necessitating continuous coding education and re-education. Finally, charge generation and claim followup are disconnected in space and time and often performed by different people, adding to confusion and costs of the claim processing cycle.

Only experienced coders can handle such complexity but experience too often turns into handicap as, in the absence of a reliable self-correcting process, the coder or the followup person may repeat the same mistake over and over. Hence ad hoc coding is error-prone and expensive. Paper superbill-driven coding improves upon traditional coding because it allows fewer errors and eliminates some of the costs. Computer aided coding with integrated superbill completes the transformation of coding from individualistic art towards disciplined and systematic process and is the most reliable and least expensive solution.

Traditional Coding

Since the practice owner is ultimately responsible for coding quality, it behooves the physician to manage personally the coding process. But traditionally, in the absence of systematic practice management, the physician looked for a coding approach to avoid the burden of coding. Such an approach to coding is error-prone and expensive. According to the Healthcare Financial Management Association's "Tip Sheet: Medical Claims Denial Management," the average error rate for CPT coding is 45%-55%. Some specialties (e.g., interventional radiologists) have trouble exceeding even 18% of correct coding, according to the March 2003 issue of "Healthcare Biller: The Communication Network for America's Health Care Billers," a monthly newsletter from Aspen Publishing.

Traditional coding involves the doctor, data entry personnel, and certified coder. The doctor dictates, types, or handwrites descriptions of diagnosis and procedures, without listing actual codes. The data entry personnel enter codes based on reading doctor's descriptions, and the certified coder supervises and audits the quality of coding by the data entry personnel.

Traditional coding process is error-prone because the certified coder does not audit 100% of entered codes and because such process does not have a vehicle for context maintenance between the charge creation and claim followup stages. The errors may become especially expensive upon post payment audit of the charges by the insurance company. This process is also expensive because multiple people are involved in the coding process and because the errors, if discovered at all, will be discovered only downstream, rising the costs of error correction.

Paper Superbill-driven Coding

Pre-compiled superbill-driven coding process places the doctor in control of coding, ties together claim creation and followup stages, and avoids many shortcomings of traditional coding. Such a process delivers two-fold advantage of lower cost and improved communication. First, the doctor codes at the end of patient encounter without involving data entry personnel in the middle. Second, the paper superbill serves the role of a formal vehicle for coding information communication between charge creation and claim followup stages. Additionally, a pre-compiled superbill improves coding consistency across the doctors within the same practice.

Superbill creation process has four stages:
1. List the codes used most often first. Use CPT frequency report.
2. List the diagnosis codes
3. Leave room for ancillary services
4. Include patient's information

Along with the advantages over the traditional coding process, the paper-based superbill still has four shortcomings. First, the data must be re-entered into the system from the paper superbill, introducing potential for errors. Next, the superbill must be reviewed periodically to adjust for changes in practice operations. Worse, it is difficult to keep up with changes in coding regulations, necessary modifiers, and bundling decisions that differ across various payers. Finally, the paper superbill contributes nothing to upfront coding error identification and correction, delaying potential error identification and resolution to post-submission, or worse, post-payment phases. Obviously, the later in the process the error is identified, the more expensive is its correction.

Computer Aided Coding with Integrated Superbill

Computerization and integration overcome most of the problems of paper superbills, eliminating duplicate data entry, automating code review and adjustment for frequency, practice operations, and payer idiosyncrasies, and shifting much of the error identification and correction from post-payment stage to claim pre-submission stage.

Computer aided coding with integrated superbill offers multiple advantages:
1. Dynamic - Adjusts for changes in practice operations and payer specifics. For instance, adds automated alert to satisfy unique payer demands, such as requests for paid drug invoices in addition to injection CPT code and J code for supplies.
2. Precise - Matches codes to EMR and alerts in real time about potential coding errors, such as confusing modifiers 59, 76, 77, and 91 for repeat procedure or test, or not coding the ICD-9 code to the highest level of possible digits in spite of specific diagnostic available in EMR.
3. Defensive - Allows for real-time profiling of coding patterns to alert about potential audit flag.
4. Reliable - Facilitates end-of-day juxtaposition of visits with charges, avoiding unpaid visits.
5. Inexpensive - The doctor can use it directly, eliminating extra data entry step and associated costs.

In summary, coding is a mission-critical responsibility of practice owner. Computer aided coding with integrated superbill places the doctor in control and enables dynamic, precise, reliable, consistent, defensive, and inexpensive coding process. Superbill digitization and integration overcome the four-dimensional coding complexity, tie it to EMR, patient scheduling, and billing (i.e., to the entire spectrum of practice management functions), and require powerful Vericle-like computing platforms.