Sometimes the most effective approach to research is simply to take a closer look at existing data and analyze it in new ways. This is exactly the strategy Sara Paton, Ph.D., employed when called on to lead an investigation into alarmingly high breast cancer rates in Madison County, Ohio.
Paton is an assistant professor of community health with the medical school’s Center for Global Health Systems, Management, and Policy and an epidemiologist with Public Health —Dayton and Montgomery County. She was supported in the research by Marietta A. Orlowski, Ph.D., an associate professor of community health and of health, physical education, and recreation, and Sylvia Ellison, M.A., a research assistant with the center.
The need for the study became clear in 2007 when data from the U.S. Centers for Disease Control and Prevention (CDC) indicated that the incidence and severity of breast cancer in the county from 2001-2003 were the highest in Ohio, as well as being substantially higher than the national average. When state Senator Chris Widener, who was a state representative at the time, learned of the CDC data, he contacted officials at Wright State about investigating the problem. In response, University President David R. Hopkins, P.E.D., commissioned the study by Paton and her colleagues.
The first step in conducting the study, Paton explained, was to collect data from additional sources covering a longer span of time. In particular, she reviewed data from the Ohio Department of Health, the CDC, and the Ohio Cancer Incidence Surveillance System.
“We were able to look at all the breast cancer patients in Madison County from 1996 through 2006,” Paton said, “and do data analysis for those women, looking at descriptive factors and incidence and mortality.”
Paton then compared the numbers to those of five similar Ohio counties, the state as a whole, and the overall U.S. population. What she found was quite gratifying.
“For incidence, breast cancer in Madison County is actually very low,” Paton said. “We also looked at overall cancer and saw the same trend. Incidence of any type of cancer is low in this area.”
The fact that a more expansive, longer- term study yielded such different results didn’t surprise Paton.
“It’s a small county with a small number of cases,” she said. “This means that incidence varies a lot, year by year. There are maybe 20 cases a year, so the incidence could increase by 50 percent, and you’re really only adding five to 10 additional cases. With the data they were looking at originally, they picked one or two points, and they looked high.”
While any rise in the incidence or severity of cancer is lamentable, a handful of additional cases over the course of a few years can easily be due to natural variation and may not indicate a serious public health problem. Just to be sure, Paton and her team also mapped the data.
“Most cases are in the most populated areas,” she said, “which is what you would expect. There isn’t any indication this is an environmental issue.”
Unfortunately, the data also showed that while Madison County has a low incidence of cancer, the mortality rate is high. Mortality decreased significantly toward the end of the study period, based on data from 2005-2007, but it’s too early to consider the matter resolved.
“Is the low number a blip, or the start of a long-term trend?” Paton said. “We need a few more data points to say.”
With breast cancer, the timing of diagnosis can make a big difference. The study revealed that women in Madison County with breast cancer tend to be diagnosed at later stages. While the reasons for this are uncertain, it is clear that efforts to raise public awareness, promote prevention, and encourage women to do breast self-exams and get mammograms could have a positive impact. VS