Professional Fee Ratios for US Hospital Discharge Data

Background:US hospital discharge datasets typically report facility charges (ie, room and board), excluding professional fees (ie, attending physicians' charges).Objectives:We aimed to estimate professional fee ratios (PFR) by year and clinical diagnosis for use in cost analyses based on hospital discharge data.Subjects:The subjects consisted of a retrospective cohort of Truven Health MarketScan 2004-2012 inpatient admissions (n=23,594,605) and treat-and-release emergency department (ED) visits (n=70,771,576).Measures:PFR per visit was assessed as total payments divided by facility-only payments.Research Design:Using ordinary least squares regression models controlling for selected characteristics (ie, patient age, comorbidities, etc.), we calculated adjusted mean PFR for admissions by health insurance type (commercial or Medicaid) per year overall and by Major Diagnostic Category (MDC), Diagnostic Related Group, Healthcare Cost and Utilization Project Clinical Classification Software, and primary International Classification of Diseases, 9th Edition, Clinical Modification (ICD-9-CM) diagnosis, and for ED visits per year overall and by MDC and primary ICD-9-CM diagnosis.Results:Adjusted mean PFR for 2012 admissions, including preceding ED visits, was 1.264 (95% CI, 1.264, 1.265) for commercially insured admissions (n=2,614,326) and 1.177 (1.176, 1.177) for Medicaid admissions (n=816,503), indicating professional payments increased total per-admission payments by an average 26.4% and 17.7%, respectively, above facility-only payments. Adjusted mean PFR for 2012 ED visits was 1.286 (1.286, 1.286) for commercially insured visits (n=8,808,734) and 1.440 (1.439, 1.440) for Medicaid visits (n=2,994,696). Supplemental tables report 2004-2012 annual PFR estimates by clinical classifications.Conclusions:Adjustments for professional fees are recommended when hospital facility-only financial data from US hospital discharge datasets are used to estimate health care costs.