Our office struggled for years with coding patient encounters to correctly capture all the reimbursement to which it was entitled. One solution that worked well for us was sending someone from our business office to “coding school.” Her five-week stint has in turn helped us code more accurately, increase revenue and become more Medicare compliant. It has also proven cost-effective.

Worth the investment

We are a family medicine residency program in Asheville, N.C. Every year we lose about a quarter of our physician staff to graduation and start fresh with new graduates who have no training in how to code. We have 24 residents and 10 faculty physicians – the equivalent of seven full-time physicians conducting approximately 35,000 patient visits per year. We have an electronic health record, and our documentation is usually more thorough than handwritten notes. But with the complexity of the coding rules and the inexperience of our residents, our physicians tend to undercode office visits.

In 2000, we elected to enroll one of our business office employees in a $900 course to become a certified coding specialist. (See “Courses available.”) She already had her certification as a medical insurance specialist, with expertise in medical terminology and anatomical terms. To maintain coding certification, she’ll need 12 hours of continuing education yearly.

Once she had completed the course, we began a prebilling audit on selected Medicare notes done by the faculty physicians. During the next six months, she audited 998 charts, which represented about 13 percent of the Medicare patients seen. Of those notes, 45 (or 4.5 percent) needed to be downcoded for a subtraction of $1,350, and 338 (or 33.9 percent) needed to be upcoded for an addition of $10,809. The net benefit was $9,459.

In theory, the benefit could have been much more. The 998 charts constituted about 3.25 percent of all our patient charts. If she had audited all of them, the benefit extrapolates to $291,046 for that six-month period.

She initially took about nine minutes to review each chart. Now with thousands of charts behind her, she has shaved that time to six minutes.

Impressed with this success, we expanded her audits to include residents. She continued to do selected prebilling audits so that we could adjust the code to the appropriate level. During the next year, we continued to see an actual benefit of about $500 per month from the audited charts. She reviewed about 150 charts per month. Our physicians received a monthly report of their coding accuracy. Physicians who had a significant discrepancy from the coder’s level were encouraged to meet with her to review their notes and see where they could improve. We also instituted coding cases consisting of actual notes that she found to be inappropriately coded, which were reviewed at monthly educational sessions with her explanation of the proper coding.

Our coding specialist has found that her expertise – auditing charts and educating our physicians on coding – takes about 20 percent of her workweek. In essence, we have added a coding specialist to our staff without hiring an additional staffer.

Coding with confidence

The change within our practice has been dramatic. (See “Cracking the code.”) In 2000, our coding profile showed that we coded 27 percent of our established patient visits as 99212s, 59 percent as 99213s, 9 percent as 99214s and 3 percent as 99215s. This past year, 2 percent of our 26,448 evaluation and management codes for established patients were 99212s, 56 percent were 99213s, 38 percent were 99214s, and 3 percent were 99215s. The difference between a level-2 and level-3 charge and between a level-3 and level-4 charge is $20 and $40, respectively. Compared to our coding level in 2000, this would result in an annual increase in charges of $457,590 for those 26,448 established patient visits, or $65,370 per physician equivalent.

The faculty physicians are now much more confident in their coding ability and pass this on to the residents while precepting. Our coding accuracy over the past six months has been 75 percent, with fairly equal distribution of upcodes and downcodes.

In conclusion, we found that training a staff member to be a coding specialist has been very cost effective, has increased our revenue and has improved our Medicare compliance. She has been an invaluable resource for us in improving our own ability to code correctly and in helping our residents to develop this skill as well.