A new study breaks down death rates from 29 cancers in every county in the U.S.
Credit: Getty Images

First, some good news: The overall rate of cancer deaths in the United States dropped 20% from 1980 to 2014, according to a new study published in JAMA. Now, the not-so-great news: Some places in the U.S. are much riskier than others when it comes to your chances of dying of cancer. Huge disparities exist across states and regions, with deaths from cancer still increasing in many counties across the country. In other words, the study authors say, where people live could have a big influence on how they die.

The new research examined more than 19.5 million deaths in the U.S., from 29 different types of cancer, over three and a half decades. And despite the nationwide trend of declining death rates, researchers found that 160 of the 3,000-plus counties they studied actually showed increases in cancer deaths during the same period.

For example, if you live in certain counties in Florida or Kentucky, your chance of dying of lung cancer is higher than anywhere else.

“This is unacceptable,” says lead author Ali Mokdad, PhD, professor of global health at the Institute for Health Metrics and Evaluation at the University of Washington. “We know other countries have done a better job eliminating these disparities, and we need to do something about it. We spend more money than anybody else and we debate health care more than anybody else, but at the same time we don’t get similar results.”

The results by county, year, and cause of death can be viewed on an interactive map on the Institute for Health Metrics and Evaluation’s website. They show that, even in the most recent data from 2014, large inequalities persisted. For all cancers combined, death rates per 100,000 residents ranged from 71 in Colorado's Summit County to 501 in Florida's Union County.

Those same two counties also ranked highest and lowest for 2014 deaths specifically from lung cancer—the cancer that kills more Americans than any other. This rate was 20 times higher in Union County (231 deaths per 100,000) than in Summit County (11 deaths per 100,000). Union County was followed by four counties in Kentucky.

It’s not surprising that parts of the South and rural areas have higher rates of lung cancer, say the authors, since previous research shows that smoking rates area also higher in these regions.

Most counties across the country saw a decline in deaths from breast, prostate, cervical, colon, testicular, and Hodgkin lymphoma cancers. The lowest breast cancer rates were reported in parts of the West, Midwest, and Northeast, while high mortality rates persist in counties along the Mississippi River.

Deaths from kidney cancer were also highest along the Mississippi River, as well as in Oklahoma, Texas, and parts of Alaska and North and South Dakota with large Native American populations. Rates of prostate cancer deaths were highest in Alabama, Mississippi, Arkansas, Georgia, South Carolina, and Virginia.

Notably, deaths from liver cancer and mesothelioma (a relatively rare type of lung cancer that can be caused by asbestos exposure) increased in nearly every county studied. For liver cancer, clusters of counties in California, Oregon, Washington, New Mexico, and Texas all saw large increases. (Rates of alcoholism, a well-known risk factor for liver cancer, aren’t particularly high in these counties, suggesting that other causes are likely to blame.)

The study did not compare prevalence of risk factors, diagnosis and treatment rates, or access to care, so it cannot draw definite conclusions as to why these big differences exist—so it's not clear if you are more likely to get cancer in those counties, or just die of it (say, due to lack of effective treatment). But based on previous research, the authors say that inadequate prevention programs—and higher rates of smoking, obesity, and other unhealthy habits—in certain regions can explain at least part of the results.

Socioeconomic factors, such as income and education level, also affect the quality of a person’s cancer care, Mokdad told Health. “An educated person is more likely to understand the danger, and the need for proper care,” he says. Whether a person has health insurance, and how good that insurance is, also matters.

Differences in death rates should also be a red flag that certain communities aren’t getting the benefits of screening, early detection, and specialized treatments that others are, he says—especially for cancers with normally high survival rates, like breast and prostate cancers.

These findings are particularly useful, the authors write, because they break results down by county and not just by state (like similar studies have done), which can mask these hyper-local patterns. They hope their findings will provide insight for the creation of new health policies, which are often implemented at the county level.

“You cannot change something unless you are able to measure it, so we wanted people in these communities, poor or rich, to have their numbers,” says Mokdad. “They can take this message to their community leaders and ask for policy support to address their specific problems—or look to similar counties where rates have gone down, and study what worked for them.”