How to Choose the Right Coding Audit MethodHealthcare has undergone several major upheavals in the past five years.
Many organizations adopted new electronic health record (EHR) technology while navigating the implementation of ICD-10-CM/PCS. In the years since the adoption of ICD-10, organizations have steadily navigated new systems and processes.
While ICD-10 coding productivity may not have fully rebounded to ICD-9 levels, it is steadily increasing. In February 2016, ICD-10 productivity was 22 percent below previous ICD-9 rates; as of July 2016, productivity had increased to 11 percent below ICD-9 rates.1 Now that productivity concerns have eased, organizations can resume ongoing coding audit programs to ensure optimal data and reimbursement.
Coding Audit Landscape Changing
In the years prior to ICD-10 implementation, many organizations focused on “hiring bodies” who had some knowledge of coding. ICD-10 was a new frontier and few, if any, coding professionals had experience with the system. As a result, health information management (HIM) departments were open to hiring sta with a lesser skill set and providing on-the-job training. Consequently, the primary focus of coding audits during that time was to ensure coder pro ciency with the new coding system.
Now that coding professionals are more skilled and familiar with ICD-10, HIM leaders are evaluating internal processes to ensure high levels of coding quality and productivity. Coding audits are now focused on what is meaningful to the organization rather than conducted in reaction to outside in uences.
More frequent coding audits are now more commonplace. Best practice involves ongoing, consistent audits focused on 3.5 to ve percent of total volume per month. Auditors adhere to a schedule to review sample cases every week. As part of the HIM work ow, this routine promotes e cient mitigation of any repetitive coding issues.
Coding Outcomes Based on Method
The two main coding audit methodologies utilized to measure ICD-10 accuracy—per code and per record—focus on assessing coding quality. Results can vary widely depending on the process.
Auditors look at every decision made by the coding professional. Some auditors call it the “code-for-code” method. is approach considers the entire coding picture, reviewing not only ICD-10 codes, but also discharge disposition, present on admission indicators, and other abstracting items. Codes that impact reimbursement are given twice the weight of those that do not.
This is also known as the “all right/all wrong” or “record-forrecord” method. Auditors review the record and look only at mistakes that impact reimbursement. If a mistake is deemed a reimbursement issue, the entire record is counted wrong. is method is very challenging for coding professionals.
Consider the following hypothetical audit for a department that requires coders to maintain a 95 percent accuracy rate. Coders falling below 95 percent accuracy are subject to disciplinary action.
For this audit, six records were reviewed for one coding professional. For each record, 10 codes were assigned by the coder, including two codes deemed to impact reimbursement. Each auditor found a reimbursement-impacting error in the same record. As illustrated in Figure 1 on page 19, the overall accuracy rate assigned to the coding professional for the audit varies greatly between the two approaches.
In this example, with the per code method, the coding professional’s correct code assignment was determined to be 70 out of 72 codes, resulting in a 97 percent accuracy rate—well above the average required. However, the results from the per record method indicates the coding professional has ve correctly coded records, resulting in an 83.3 percent accuracy rate. Using this method, the coder would be subject to disciplinary action.
Some prefer the per code method, believing it is more indicative of an individual’s coding acumen and reveals important coding quality data.
Types of Data Gleaned from Coding Audits
The most frequent data provided from an audit is an overall accuracy score for the coding team. Other elements identi ed include:
- Percentage of discharge disposition changes
- Missed query opportunities
- Changes to present on admission (POA) indicators
Identifying these items yields positive change throughout the organization. For example, one hospital experienced a high error rate in the collection of accurate discharge dispositions. Upon review, it was discovered there was no consistent location in the EHR where discharge disposition was collected. e clinical documentation improvement (CDI) team was deployed to ensure discharge disposition was consistently documented in the EHR, and the error rate plummeted.
Coding audit data also reveals speci c areas that coders struggle with. Depending on the issues identified, further action can be taken, including:
- Place the coding professional on 100 percent quality review
- Increase the coding professional’s audit volume from ve to 10 percent the following month
- Conduct a focused audit for the type of case causing difficulty, such as spinal surgeries
Auditors take twice the time to review cases as coding professsionals need to code them. erefore, HIM leaders are encouraged to maximize every auditor finding by requesting specific reports. HIM leaders should:
- Ask auditors to provide a full narrative that speci cally identi es errors
- Note exactly where auditors found documentation in the record to justify the nding. For example: “In progress note dated 1/1/17 Dr. Smith noted…”
- Ask coding professionals to review ndings and provide insight for their initial code assignment before the audit report is nalized
Sleep, Creep, and Leap!
Landscapers have a saying to describe newly planted gardens where the result of hard work is not immediately seen: “The first year they sleep, the second year they creep, and the third year they leap!” The adjustment to ICD-10 has been similar.
The industry dedicated signi cant e ort prior to October 1, 2015 and throughout the rst year adjusting to new ICD-10-CM/ PCS codes and ne-tuning systems. Seeds were being planted. In 2017, coder productivity slowly increased as the industry crept forward with ICD-10, bill holds decreased, and consistent payer payments were received.
2018 will be the year that ICD-10 coding leaps with even greater coder productivity, accuracy, and data outcomes. Organizations will mine speci c clinical data based on ICD-10’s granularity to track performance and identify national disease patterns.
Other predictions for how coding audits will progress in 2018 include:
- More internal auditing to ensure coding professionals are capturing all the speci city that ICD-10 allows
- An uptick in quality audits from outside agencies to ensure a deeper level of code accuracy
- Closer scrutiny of Medicare Recovery Audit Contractor (RAC) activity as the four contractors increase reviews and recoupments
- Focus on optimization of EHRs and legacy systems to implement new coding work ows and streamline the billing process
- Alakrawi, Zahraa et al. “New Study Illuminates the Ongoing Road to ICD-10 Productivity and Optimization.” Journal of AHIMA 88, no. 3 (March 2017): 40-45. http://bok.ahima.org/doc?oid=302058.