Objective: The aims of the study were to determine the frequency at which each emergency medicine evaluation and management (E/M) code is used, to identify factors associated with their use by academic pediatric emergency departments (PEDs), and to compare PED E/M code utilization rates with rates reported by Centers for Medicare and Medicaid Services for general emergency departments (EDs).
Method: A 24-question survey was sent to 42 academic PED medical directors. Questions pertained to PED demographics, physician staff, records/documentation, billing education, and E/M coding data for 1 year. The general ED E/M code utilization rates were obtained from the published Centers for Medicare and Medicaid Services database. Descriptive statistics and odds ratios were used to report and compare the data.
Results: Twenty (48%) of the surveys were returned, and 9 (21%) completed the E/M coding data questions. From these 9 departments, the mean PED annual census was 46,065 (range, 23,531-92,910). The methods of PED medical record documentation were template (6), handwritten (2), and dictation/transcription (1). Charge documents were completed by the PED physician (3), professional service coders (4), and hospital coders (2). Coding/documentation in-services were provided to the physicians of 7 PEDs, and billing audits were performed in 5 PEDs.
The total number of charges for the 9 PEDs was 325,129, 78.4% of the census. Multiple reasons were given for the discrepancy between census and charges. The percentage of each of the 5 levels of service billed was calculated for each of the 9 PEDs. The 2 lowest levels of service were used 38.3% of the time, whereas the 2 highest were used 19.2% of the time. The range for the highest level of service varied widely from 5.3% to 53.3%. Approximately 65% of E/M codes used by general EDs were for the 2 highest levels of service. The PED with the highest percentage of upper level charges (53.4%) was the only PED that used dictation/transcription for documentation.
Conclusion: Although the response rate was low, and thus the validity of the results was limited, the findings may serve as a benchmark for E/M code utilization in PEDs. The large variation in use of the E/M codes among the PED in our study and the lower rate of using the highest E/M codes by the PEDs compared with the general EDs suggest potential opportunities for academic PEDs to improve billing practices.