Why did 43 percent of South Dakota women diagnosed with early-stage breast cancer surveyed choose mastectomy rather than lumpectomy?
"According to research evidence, survival rates are considered equal," explained dean Nancy Fahrenwald of the South Dakota State University College of Nursing.
Working through the South Dakota Women's Cancer Network, the South Dakota Comprehensive Cancer Control Program and seven cancer treatment centers in the state, Fahrenwald got responses from 1,093 breast cancer survivors who had been diagnosed in the last five years.
To determine which of the nine independent variables tipped the scales toward mastectomy, she turned to associate professor Chris Saunders of the mathematics and statistics department.
"The statistics were not meant to be innovative but to provide an answer that is robust and rigorous," said Saunders, who used logistic regression to analyze the survey results. His background in epidemiological statistics includes predicting whether certain cancer types will respond to a given treatment based on their RNA expression profiles.
Considering distance, age
Fahrenwald had divided the sample into three groups based on where the women lived—urban, large rural and small rural. The small rural women didn't have access to care in the nearest town, while those classified as large rural, such those living near Aberdeen, might have a cancer treatment center nearby, she explained.
Before beginning the regression analysis, Saunders and graduate student Danica Ommen combined distances to surgery and to treatment into one variable because the covariants were very highly correlated.
The statistical analysis reinforced Fahrenwald's conclusion. "Distance did not seem to account for a significant amount of the variation in their choices," Saunders said.
"This finding dispels prior evidence that women who must travel further choose a mastectomy," Fahrenwald said, noting that "this will impact people's thinking about rurality."
When looking at age as a determining factor, the nursing researchers found that 20- to 30-year-olds and those older than 75 tended to choose the more surgically invasive treatments, Fahrenwald explained. Young women with early-stage cancer in one breast chose bilateral mastectomy, often with reconstruction, then chemotherapy and even gene therapy.
However, Ommen found that statistically, age was not a major factor in treatment choice.
Considering both options
After narrowing the analysis down to those women who discussed both lumpectomy and mastectomy with their surgeons, Ommen concluded that the most significant factor to predict treatment choice is the stage of the woman's breast cancer. She also verified the logistic analysis using reverse stepwise procedure for regression.
Among the 415 women who considered both options, 109 were at Stage 0, 186 at Stage 1, and 120 at Stage 2. The majority of those at Stage 0 and Stage 1 chose lumpectomy, 57.7 percent and 65.6 percent respectively, which shows that a statistically significant relationship was observed, Ommen explained.
However, the women with Stage 2 cancer had a tendency to choose the more surgically extensive treatments with nearly 58 percent opting for mastectomy, according to Ommen.
"Getting the disciplines together means you get a better product," said Saunders, who views statistics as a supportive discipline.
This collaborative work will provide "statistically rigorous justification for the relationships that are leading to perceptions that rural community members are choosing mastectomy," Fahrenwald said. That will then help formulate public health policies to encourage patient-centered treatment decisions where options are provided and women make an informed choice and are highly satisfied with that choice.
Study links air pollution to increased colorectal cancer risk through DNA changes