Context: Choosing the correct E&M code for most providers is a difficult and misunderstood process. As a result improper coding occurs and can result in fraud charges. Many providers are unaware mistakes are being made and as a result improper coding occurs on a daily basis. More than 10% of total Medicare payments in the fiscal year 2004 were improper, according to a CMS report. 1 This resulted in either too much or too little money going to doctors and other program participants.1
Objective: This project was designed to evaluate the coding accuracy and the documentation performed by four family medicine providers in a community setting in northern Idaho to determine the frequency at which improper coding occurs in a typical family practice setting. Determining the frequency of improper coding will demonstrate if there is a need for providers to become more familiar with coding guidelines. In addition this project attempted to determine the potential financial impact incurred by a practice as result of improper coding. The ultimate goal would be to identify methods where by providers could be advised where coding and documenting mistakes can be avoided in the future.
Design: The clinic identified in the project, has a policy where each individual provider assigns his or her own E&M code after each patient visit. Ten chart notes from the first office day in May, 2005 were randomly selected from each provider. I chose to evaluate only one day's chart notes, as this sample would represent a typical day. Ten chart notes from four different providers were used, totaling forty notes. The chart notes were evaluated first according to documentation. Specifically, I evaluated the inclusion of the following: chief complaint, history of present illness, pertinent past medical history, physical exam and medical decision-making. All the chart notes were reviewed in the same fashion using a checklist computerized palm program called Stat Coder. This program assigns the correct E&M code according to current documentation as identified by the Health Care Financing Administration.2,3,7 Stat Coder makes this process more straightforward with automated checklists that count the documentation elements of a patient visit for you. 2,3,7 A tally was kept of how many charts were coded improperly, either up-coded or down-coded and how many were coded correctly.
Setting: This project was conducted at a busy family practice in northern Idaho.
Subjects: Charts were selected at random. Subjects eliminated from the study were new patients, patients who had a physical, all pregnant patients and patients who were under the age of 18.
Results: Ten chart notes for four providers were reviewed, totaling forty charts. Eighteen (45%) chart notes were coded correctly. Fifteen (37.5%) chart notes were upcoded and seven (17.5%) charts were down-coded. Two hundred and twenty seven dollars was lost due to seven chart notes being down-coded. Whereas, the up-coding of fifteen chart notes led to the over billing of three hundred and thirty two dollars.
Conclusions: It was shown that improper coding occurs greater than 55% of the time in a typical day. These errors led to financial losses of $227.00 and $332.00 was billed out when it shouldn't have been. Imagine the losses incurred over the entire year. If you assume there are 260 working days in a year and if this happens daily, than $59,020 was lost and $86,320 was over-billed. It wasn't expected that down-coding would occur at a greater rate that up-coding, but this study showed providers tend to up-code rather than down-code. It is important to note that the results could vary according to practice and provider and this design lacked a large sampling pool, which could change the results dramatically. More research needs to be done, to determine where the majority of errors are made and what can easily be corrected to prevent these errors in the future.
Files are restricted to Pacific University. Sign in to view.