Independent Project Part II

I’ve made a good bit of progress this past week on my independent project part of my fellowship. I have been examining the relationship between race and rates of utilization for various cardiac care procedures. Many papers have discussed the finding that heart attack patients who are black undergo a number of heart procedures at a lower rate than those who are white. Much of the literature on this topic is 5, 10, or 15 years old. As such, I thought it would be interesting to see if anything has changed in recent years. I have access to some data which allows me to look at the years 1997-2008, so I can see if trends have changed over time.

I started off simple and calculated the rates of utilization for the procedures I am inspecting (I briefly mentioned these in the last blog; CABG, PTCA, and cardiac catheterization) by race. My findings are fairly consistent with past literature. White heart attack patients underwent each of these three procedures at significantly higher rates than blacks. The magnitude of these differences varied, however. For cardiac catheterization, blacks were no more than 10% less likely to receive the procedure than whites. This gap decreased somewhat over the period of time I investigated, and actually became statistically insignificant in 2008.  However, for the CABG procedure,  utilization rates for blacks were around a third less than for whites. Further research will be necessary to figure out why differences in utilization vary so much by procedure.

In order to calculate these rates, I first had to narrow down my dataset to individuals that were diagnosed with a heart attack. Diagnoses are recorded by ICD-9 codes. The main ICD-9 code for a heart attack is 410, but there are many subcategories. Hence, I had to set up a short program which extracted all the relevant codes for heart attacks. From there, I calculated the utilization rates for each of the three heart procedures, conditional on the patients being diagnosed with a heart attack.  The most time intensive part of this process has not been the final calculation of rates, but rather first extracting the necessary data.

Later, I will describe my next steps in attempting to model just how big an impact race has on the utilization of cardiac procedures.