Geospatial variation in breast cancer care across the United States is often due to unexplained or random factors rather than patient characteristics, according to research published in JAMA Oncology.

Researchers noted that breast cancer mortality rates can vary by more than 4-fold across counties, and this suggests a need to implement interventions to target health care delivery system factors.

Because the potential influence of geospatial variations on breast cancer management has not been well studied, the researchers conducted the current study to “determine whether quality improvement efforts should target specific patient subgroups vs regions and to determine whether they should vary based on the type of breast cancer treatment.”

Continue Reading

The retrospective study included 31,571 patients diagnosed with stage I to III breast cancer between 2007 and 2013. Patient data were derived from the Surveillance, Epidemiology, and End Results-Medicare database.

The researchers looked at variations between regions for 5 metrics:

  • Stage I disease at diagnosis (a proxy for screening mammography and timely surgery)
  • Receipt of chemotherapy
  • Receipt of radiation therapy
  • Initiation of endocrine therapy within 1 year of diagnosis
  • Continued endocrine therapy 3-5 years after diagnosis.

For each of these metrics, the researchers determined whether geospatial variance was a result of region or health service area (HSA), something random, unexplained reasons, or patient factors (such as age, race, ethnicity, and socioeconomic status).


A majority of the cohort (61.4%) was diagnosed with stage I disease. The most common treatment was radiation (81.6%), followed by chemotherapy (72.8%), continued endocrine therapy at 3-5 years (52.1%), and initiation of endocrine therapy within 1 year (48.8%).

Regional patterns for these 5 metrics varied. For all metrics, the greatest share of variation was due to unexplained reasons (35%-54%), followed by region/HSA (24%-48%). Patient factors were rarely the cause of variation (1%-4%).

However, reasons for variation were not consistent across the different metrics. Variation in stage I disease at diagnosis was largely due to random factors or unexplained reasons, roughly in equal measure.

For treatment patterns with chemotherapy or radiation, variations were largely due to unexplained reasons, followed by random factors. Region/HSA and patient factors played a larger role for radiation than for chemotherapy.

Endocrine therapy (both initiation and continuation) had the largest proportion of variance due to region/HSA (28% and 39%, respectively).

“This is remarkable, considering that endocrine therapy confers an impressive one-third reduction in the risk of death from breast cancer in the first 15 years after diagnosis,” the researchers wrote.

They concluded that this study suggests “substantial unexplained geospatial variation in initial breast cancer care” and that variance due to region/HSA “was multifold larger than that explained by patient factors.”


Hassett MJ, Tramontano AC, Uno H, Ritzwoller DP, Punglia RS. Geospatial disparities in the treatment of curable breast cancer across the US. JAMA Oncol. Published online January 27, 2022. doi:10.1001/jamaoncol.2021.7337

This article originally appeared on Cancer Therapy Advisor