Investing in an audit can reveal costly coding errors
By Brian Whitman
You probably select the level of an evaluation and management (E/M) visit or service more than 20 times a day. But are you choosing the right code for the service you provide? If not, join the many internists struggling to get E/M coding right. Here are some strategies to help you find out how you're doing—and tips for what to do if you find problems.
Q: What should I know before I pick a code?
A: First, be familiar with all the components that go into selecting a code level for an E/M service: history, examination and medical-decision making.
Next, know the Medicare E/M documentation guidelines so you can be sure to provide the right documentation to back up the code you select. E/M codes must correspond to medical services performed and documented.
To help guide your decisions, you can use the online “Evaluation and Management Service Codes: Selecting and Documenting Appropriate Levels of Service” tool from ACP's Practice Management Center.
Q: How do I know if I'm picking the right codes?
A: The only way to truly know if you‘re making the right choices is to do a coding audit. An audit would review your documentation to make sure it matches the selected level of service.
To do a meaningful audit, you need to complete a chart auditing tool for a sample of your charts. (Two such tools are online. Click on “Coding/Documentation Auditor Worksheets.”) Choose a small random sample and have a staff member determine the appropriate coding level for the services documented on those charts. Compare those codes to the ones selected by physicians.
Q: How many coding errors are too many?
A: If more than 20% of your codes are incorrect, you need to take further action and thoroughly review coding and documentation guidelines. Medicare contractors frequently offer free seminars to physicians and staff on how to select E/M codes. Check your local contractor's Web site for a seminar schedule.
Q: The audit sounds useful, but time consuming. Do I have other options?
A: One less time-consuming approach is to compare your coding levels to members of your specialty using national and publicly available Medicare data. You can access that Medicare data online.
Such a comparison won't give you insights into problems with individual charts, but comparing your billing nationally to your peers can be a useful proxy. This is also one of the tools used by Medicare audit contractors in determining which physician or practice to audit.
Q: How do I use the Medicare data to make a comparison?
A: In the past, practices have compared their own data against Medicare's by creating a spreadsheet.
But the College has now created a tool to make such a comparison easier: the "E/M Coding Assessment Tool" is available to ACP members online. (The ACP site currently uses figures for general internists, so subspecialists may wish to gather the national Medicare data directly.
To use the tool, you have to enter how frequently you provide new patient office visits (99201-99205), established patient office visits (99211-99215) and outpatient consultations (99241-99245). Office staff should be able to provide you this information by running a report from the practice management system.
Q: What if I have significantly more high-level codes than the Medicare average?
A: It could mean several things. You may be coding incorrectly, or your patients may be sicker than average. Or it may mean your documentation patterns are different from your colleagues.
Do not assume on the basis of frequency that you are doing anything wrong. However, if you find a significant variance, you should perform a coding audit on a selected sample of charts.
Q: What if I have significantly more low-level codes than the Medicare average?
A: Again, there are a number of reasons why this could be happening: Your patients could be healthier than most or you may be trying to “play it safe” by billing lower codes. But that's a bad idea for two reasons. Medicare contractors consider systematic undercoding as problematic as systematic overcoding. And two, you're losing a lot of revenue. Again, a coding audit is the best way to pinpoint the problem.
Q: If I code around the average level, should I bother with an audit?
A: Any practice may benefit from completing an occasional audit. For instance, if your frequency data showed that you were in line with Medicare averages but you had sicker patients and better documentation than the average physician, you are probably undercoding.
Q: Can Medicare use comparison coding information to trigger an audit?
A: Yes, especially if such a comparison shows a pattern of outlier billing. However, Medicare rules require contractor audit actions to be commensurate with the extent of the perceived problem.
A description of these audit rules, known collectively as the progressive corrective action program, was published in the July-August 2004 ACP Observer. Private payers may also use these data for auditing purposes.
Brian Whitman is Senior Analyst for Regulatory and Insurer Affairs in ACP's Washington office.
Internist Archives Quick Links
New Leadership Webinars
The ACP Leadership Academy is offering FREE webinars covering the core tenets of leadership, leadership in hospital medicine, finance, and more.
Join ACP Today!
ACP membership connects you with like-minded colleagues and provides access to a variety of clinical resources, practice tools, and ways to earn MOC and CME.