Do billing codes accurately reflect pediatric emergency physician workload? A cross-sectional study.

BACKGROUND

Measuring physician workload in the pediatric emergency department (ED) could help optimize staffing, improve department efficiency, and provide a metric to assess interventions aimed at improving pediatric ED flow. However, no accepted measure of physician workload exists. Billing codes, which reflect the perceived complexity of treating a patient, may serve as a surrogate for physician workload. Our objective was to evaluate whether billing codes are a valid surrogate for pediatric ED physician workload.

METHODS

We conducted a health records review to determine if billing codes were associated with measures of pediatric ED physician work. Visit information was extracted for 150 pediatric ED visits. We used multivariable ordinal logistic regression models to assess the association between pediatric ED physician-assigned billing codes, with measures of visit complexity, and measures of pediatric ED physician work. We also completed a sensitivity analysis considering a billing auditors-assigned billing codes.

RESULTS

Three measures of pediatric ED physician work were independently associated with increased physician-assigned billing codes: receiving labs (OR 5.6, 95% CI 2.2-15.4), ordering medications (OR 2.3, 95% CI 1.1-5.1), and having specialist consultation (OR 4.4, 95% CI 1.6-12.5). We did not find any statistically significant associations between physician-assigned billing codes and measures of physician work after adjusting for visit complexity, age, and sex. Visit acuity (PaedsCTAS 1-3) was associated with increased billing codes (aOR 5.1 95% CI 1.9-15.7). These results were largely consistent with our sensitivity analysis considering billing auditor-assigned codes.

CONCLUSIONS

Overall, we found limited evidence supporting the content validity of billing code as a surrogate of pediatric ED physician workload. These results, coupled with the potential value of tracking physician workload, highlight the necessity to develop a valid and reliable measure specifically considering pediatric ED physician workload.