Comment on US Geographic Distribution of Recombinant Tissue Plasminogen Activator Use by Hospitals for Acute Ischemic Stroke
To the Editor:
We read with great interest the recent article by Kleindorfer et al1 on the geographic distribution of recombinant tissue plasminogen activator utilization, and we applaud the authors for addressing this important question and agree with their conclusions. We amplify 1 point made by the authors. The authors used the MEDPAR (Medicare Provider and Analysis Review) administrative database to determine the number of patients potentially eligible for treatment with intravenous recombinant tissue plasminogen activator using ICD-9 codes 433, 434, or 436 within diagnosis-related group (DRG) 14, 15, and 524. Using ICD-9 procedure code 99.1, they estimate that 2.4% (11 884/495 186) of all United States ischemic stroke admissions received recombinant tissue plasminogen activator during the study period (July 1, 2005–June 30, 2007).
This method is likely to grossly underestimate treatment rates in comprehensive stroke centers, which are often the largest providers of stroke therapy but do not routinely document the 99.1 code. Despite treatment rates of >20% at each of our 2 centers,2 few faculty can recall coding a 99.1. Further, it is our understanding that for our coders to bill 99.1, the attending physicians have to document that they were present for the treatment (Sheryl Martin-Schild, MD, PhD, personal communication, 2009). This also has the potential to result in a gross underestimate of treatment rates at academic centers where intravenous recombinant tissue plasminogen activator is more likely to be administered by a resident or fellow.
Because policy-makers are often forced to use administrative data to make important public health decisions, it may be wise for stroke doctors to think twice when it comes time to coding 99.1. In this instance, good documentation has the potential to result in higher-quality statistical data and more accurate needs analysis.