As with many studies, there was good news and bad news.
The good news: Public health researchers from the University of Washington looked at cancer’s mortality rate county by county and found that overall, deaths from the disease dropped 20 percent during the last 35 years, falling from 240 deaths per 100,000 people in 1980 to 192 deaths per 100,000 people in 2014.
The bad news: There are definite “winners” and “losers” when it comes to surviving cancer. And as usual, the losers are our most vulnerable citizens: the poor, the uneducated, and the marginalized.
“As a country, cancer’s mortality rate is coming down and that’s something we should be proud of,” said Dr. Ali Mokdad, lead author of the study released Tuesday by the University of Washington Institute for Health Metrics and Evaluation. “But when we start examining it at the county level, where public health policies begin, we see many counties are not sharing the same benefits. They are falling behind when it comes to the progress the country is making.”
Mokdad said he and his colleagues had expected some health disparities to be reflected in the data, but they were “shocked by the levels between the highest and the lowest.”
“Vulnerable populations are the underserved and they have long had higher mortality and, in some cases, incidence rates compared to middle-class white populations,” she said.
The UW researchers used de-identified death records from the National Center for Health Statistics and population counts from the U.S. Census Bureau and other sources to model county-by-county death rates for 29 cancers between 1980 and 2014. They also corrected for age so they could get a clear picture of why people with certain cancers — particularly those thought to be survivable — were dying while others were cured.
All told, 19.5 million people died of cancer in the U.S. during the 35-year period, including 5.7 million people who died of lung (and related) cancers, 2.5 million people who died of colorectal cancer, 1.6 million women (and men) who died of breast cancer, 1.2 million people who died of pancreatic cancer and 1.1 million men who died of prostate cancer.
But cancer mortality rates varied widely, depending on the region. (An online visualization tool, created by the researchers, can be found here.)
In 1980, overall mortality ranged from 130.6 deaths per 100,000 people in Summit County, Colorado, to 387 deaths per 100,000 people in North Slope Borough, Alaska. Summit County’s overall cancer rate was the lowest in 2014, as well, with 70.7 deaths per 100,000 people compared to 503.1 deaths per 100,000 in Union County, Florida.
The study showed distinct cancer clusters. Breast cancer deaths were more prevalent in the South and along the Mississippi River. Liver cancer was higher along the Texas-Mexico border. Kidney cancer clustered in North and South Dakota and in counties in West Virginia, Ohio, Indiana, Louisiana, Oklahoma, Texas, Alaska, and Illinois.
Mokdad said these cancer clusters were not linked to environmental carcinogens (think contaminated water at North Carolina’s Camp Lejeune) but were more about socioeconomic factors, access to care, inconsistencies in care, lifestyle choices and other addressable issues.
“If you overlay our map of lung cancer mortality, you see it is clustered around West Virginia,” Mokdad said. “Overlay that map with [one on] smoking prevalence in the country and you see people smoke much more in that area.”
The same thing applied for obesity or alcohol use, he said. Mapping out deaths county by county also illustrated how better access to care, prevention, and research can drive cancer deaths down — at least for those who are able to tap into these resources.
“You can spot the universities and medical centers easily on the maps,” he said, referencing regions where there cancer deaths were lower.
An accompanying editorial said the most striking finding of the study was that cancer deaths were highly variable across counties and across the decades. Survival for breast and lung cancer was consistently worse in counties in the South and Appalachia while survival for breast, colorectal, and lung cancer consistently improved in the Northeast and the West.
“What the study shows you is really the health disparities and how many people are falling behind and where cancer is increasing in communities,” Mokdad said. “It also raises the question, ‘Why?’”
Public health research has shown there are four big reasons why people are more likely to die from cancer, Mokdad said, and none of them have to do with the devious nature of the disease or lack of available treatment.
The first two, not surprisingly, have to do with socioeconomic status and access to care.
“We know that people who are more educated are more likely to seek medical advice when they see a danger sign and to adhere to advice because of their education,” he said.
But money — and good insurance coverage — play a key part, too.
People who are uninsured or underinsured (their policies don’t cover prevention, for instance) don’t have the same access to care. Some may go without regular physicals or preventive screenings because they can’t afford the doctor’s visit or the tests. Some may have to choose between buying food for their kids and paying for a mammogram.
“If you catch it early, your survival will increase,” Mokdad said. “Not a lot of men die of testicular cancer because if it’s detectable, it’s curable. So when you see people dying of it, it’s about access to quality medical care.”
Case in point: Mokdad and colleagues found that testicular cancer mortality decreased nationally by nearly 40 percent during the 35-year period. But not everyone benefited. In Nantucket County, Massachusetts, where the per capita income is $53,410, death rates for this very treatable cancer dropped 72 percent. In Union County, Florida, where there are two prisons and the per capita income is a little over $12,000, the death rate rose about 40 percent.
Quality of medical care is another big factor that influences survival, said Mokdad, pointing to prompt treatment, proper follow-up and regular monitoring of patients as “very important.”
“It makes a big difference when you have good doctors paying attention, devoting time and doing the research,” he said.
Finally, there are all those behaviors that drive cancer — from smoking and obesity to lack of exercise and excessive use of alcohol, among other culprits. These preventable risk factors are often curbed by prevention messages (as long as everyone can read and understand them) and/or effective public policy (think “sin” taxes and smoking bans). But prevention strategies and public health policies vary from county to county.
Thompson and her health disparities colleagues at Fred Hutch have created a number of strategies designed to better reach these underserved populations.
“In the Yakima Valley, we tailor messages to reach the Latino population by asking participants what messages resonate with them and then tailoring our messages to meet their needs,” she said. “Through doing this we have been able to increase screenings for breast cancer, cervical cancer and colorectal cancer among Latinos. We have found that people respond very well to tailored messages.”
But while researchers have strategies, she said, “policies to implement such practices on a national level are lacking.”
“Over 30 years ago, Margaret Heckler, then secretary of Health and Human Services, published The Heckler Report in which she deplored the disparities of death and illness in the U.S. minority populations. Unfortunately, the disparity remains and we in public health should make every effort to erase [them],” Thompson said.
This gap between the health haves and the have nots is even more disconcerting, she said, “in an era when the Affordable Care Act will be repealed, leaving the vulnerable even more uncertain about their access to screening and treatment.”
Mokdad, who is the vice-chair of UW’s new Population Health Initiative, said breaking down cancer death rates county by county will help public health officials in a number of ways.
“Public health is local,” he said. “It occurs at the county level. But many times counties don’t have the means to gather these types of metrics, which are necessary to justify funding. By providing counties with this data we’re empowering people to come to leadership and say, ‘Hey, we have a problem. We need to deal with it.’”
The metrics also provide a “road map to address the health issues in these communities,” he said. Counties can compare their performance with that of others, figure out what’s worked well elsewhere then implement policies that will bump up survival rates and quality of life for the underserved — and everyone else.
“We’re hoping that people will look at this and say, ‘What did this county do? What did this state do?’" he said. “Lessons can be shared.”
Overall, the study is much more than just metrics, he said. It’s a call to action.
“It’s more than just counting the bodies, how many have died,” he said. “It’s the first step in dealing with this problem. It’s very important for us, as we debate health care, not to forget people who are disadvantaged — in rural or remote areas, in big cities, in poor neighborhoods, in places where there is more smoking or more obesity. We can’t work in silos anymore. This is how we deal with health problems so everybody can improve.”
Diane Mapes is a staff writer at Fred Hutchinson Cancer Research Center. She has written extensively about health issues for NBC News, TODAY, CNN, MSN, Seattle Magazine and other publications. A breast cancer survivor, she blogs at doublewhammied.com and tweets @double_whammied. Email her at firstname.lastname@example.org.