A standard treatment for early breast cancer is endocrine therapy (ET), with drugs such a tamoxifen and aromatase inhibitors.
But the study found that ET was not being used in about half of the eligible patients.
For example, only 13,115 of 26,255 eligible patients (48.8%) initiated ET within 1 year of diagnosis, and only 13,944 (52.1%) continued with ET.
“This is remarkable, considering that ET confers an impressive one-third reduction in the risk of death from breast cancer in the first 15 years after diagnosis,” comment authors Michael J. Hassett, MD, of the Dana-Farber Cancer Institute, Boston, and colleagues.
The findings were published online on Jan. 27 in JAMA Oncology.
This study provides an “important and disturbing” glimpse of the hidden barriers patients face when seeking quality, guideline-concordant care, says Kathy Miller, MD, the Ballve Lantero professor of oncology at Indiana University School of Medicine and associate director of clinical research at the IU Simon Comprehensive Cancer Center, Indianapolis, who was approached for comment.
Geographical variations
In their study, Dr. Hasset and colleagues set out determine the extent to which geospatial variations in early breast cancer care are attributable to health service area versus patient factors. They analyzed Surveillance, Epidemiology, and End Results (SEER) Medicare data for 31,571 patients with newly diagnosed with stage I-II nonmetastatic breast cancer between 2007 and 2013 who were followed for at least 3 years.
The patients had a median age of 71 years, and 61.4% had stage I disease at diagnosis.
Geospatial density maps (heat maps) in the paper highlight regional performance patterns. For initiation of ET within 1 year of diagnosis, the regions that appeared the worst (with less than 50% of patients getting this treatment) were parts of California, Utah, New Mexico, Louisiana, Georgia, Kentucky, Washington, and an isolated patch in Michigan.
In addition to the striking finding that nearly half of all women who are eligible for ET did not receive that therapy, the investigators found that 81.6% of 21,190 eligible patients received radiation therapy and 72.8% of 9,903 eligible patients received chemotherapy.
This also varied across the graphical regions, with the heat maps showing that the areas that were delivering radiation and chemotherapy to 70% to 80% of women were similar to the areas that were not initiating ET in about half of these women.
The authors found that the geographical region and health service area (HSA) explained more observed variation (24% to 48%) than patient factors (1% to 4%).
“While patient characteristics, such as race and ethnicity, were significantly associated with variation in breast cancer care, they explained a relatively small proportion of the total observed geospatial variance,” the authors comment.
“In fact, most of the total observed variance was owing to randomness or unexplained factors,” they add. The largest share of variation – 35% to 45% – was unexplained.
“The ET metrics demonstrated the largest total observed variance, the lowest absolute performance (only 49% of patients had an ET prescription within 1 year of diagnosis), and the strongest association with region/HSA,” they conclude.
Though limited by factors inherent in a retrospective review of SEER-Medicare data, the “unexplained nature of most geospatial variation in initial breast cancer care is not likely to change,” they comment.
Future quality improvement efforts should focus on reducing this unwarranted geospatial variation, particularly through the use of ET in eligible patients and with strategies that work across health care delivery systems, they suggest.
Approached for comment on the new findings, Dr. Miller posits that “many factors may be at play.”
“Unfortunately, the SEER database doesn’t allow us to sort out the impact of poverty/cost of care, distance to medical care, availability of specialty and subspecialty care, and payer/provider networks that may limit choices and options for second opinions,” Dr. Miller told this news organization.
She said that patients should be encouraged to consult reliable patient-focused information, such as that provided by the American Society of Clinical Oncology through its disease-specific sites, and to seek a second opinion from a university center. In many cases, major centers have become more accessible through virtual visits made available in the wake of the COVID-19 pandemic, she noted.
This study was supported by Dana-Farber Cancer Institute and the American Cancer Society. The authors and Dr. Miller have disclosed no relevant financial relationships. Dr. Miller is a regular contributor to Medscape with her Miller on Oncology column.
A version of this article first appeared on Medscape.com.