By Vivian Ho, Meei-Hsiang Ku-Goto, Hui Zhao, Karen E. Hoffman, Benjamin D. Smith and Sharon H. Giordano
Background: Little is known about regional variation in cancer treatment and its determinants. We compare rates of adherence to treatment guidelines for elderly patients across Texas and whether local specialist supply is an important determinant of treatment variation.
Methods: Previous literature reviewed indicated 7 recommended courses of treatment for colorectal, pancreatic, and prostate cancer. We analyzed Texas Cancer Registry data linked with Medicare claims for the years 2004 to 2007 to study patients with these cancers. We tested for unadjusted and adjusted differences in treatment rates across 22 hospital referral regions (HRR). We tested whether variation in the local supply of specialists treating each cancer was an important determinant of treatment.
Results: We found significant differences in adjusted treatment rates across regions. For removal and examination of 12+ lymph nodes with colon cancer resection, 13 of 22 HRRs had rates significantly different from the median region. For adjuvant chemotherapy for regional colon cancer, five HRRs significantly differed from the median. For prostate cancer treatment with a favorable diagnosis, nine HRRs differed from the median HRR. Of the 7 treatments, only the local availability of surgeons was an important determinant for excision of lymph nodes for colon cancer patients.
Conclusions: There are significant variations across Texas for seven recommended cancer treatments. No one region has consistently higher or lower treatments than other regions, and local specialist supply is not an important predictor of treatment. Different factors may be determining regional variation in treatment rates across cancer types and treatment options.
Read the full article at BMC Health Services Research.
