Rural–Urban Gaps in Potentially Excess Deaths From the Five Leading Causes, 2010–2017

In 2017 the five leading causes of death in the United States — heart disease, cancer, unintentional injury, chronic lower respiratory disease (CLRD) and stroke — together accounted for 1,723,085 deaths, about 61% of the total. This CDC surveillance report asked how many of the deaths before age 80 were potentially excess: more than would be expected if every state had the death rates of the three states with the lowest rates for that cause (the benchmark). Every state is held to the same benchmark, whatever its demographics.

Not every potentially excess death could be prevented by public health alone — distance to specialty or emergency care, environmental hazards and other local conditions matter — but many might be, through better health care access, healthier communities and action on education, jobs and housing.

Six kinds of county

About a fifth of Americans, some 60 million people, live in nonmetropolitan (rural) areas covering 97% of the country's land. Rural areas have more smoking, high blood pressure, obesity and physical inactivity, less access to care, and more concentrated poverty: 31% of nonmetropolitan counties are places where at least a fifth of residents are poor, compared with 15%–19% of metropolitan counties.

An earlier CDC report compared simply "metropolitan" and "nonmetropolitan." This one used all six levels of the NCHS urban-rural classification, from most urban to most rural: large central metropolitan, large fringe metropolitan, medium metropolitan, small metropolitan, micropolitan and noncore.

The NCHS 2013 urban-rural classification of U.S. counties. CDC, MMWR.

National trends, 2010–2017

Annual percent changes in the number of potentially excess deaths:

In 2017, 54% of unintentional injury deaths, 30.1% of heart disease deaths and 9.2% of cancer deaths before age 80 were potentially excess.

The rural gap

In every year from 2010 to 2017, nonmetropolitan counties had higher percentages of potentially excess deaths from the five causes than metropolitan counties, and the most rural (noncore) counties had the highest of all six groups. In 2017 in noncore counties:

The most urban counties had the lowest shares — for heart disease, 18.5% in large fringe metropolitan counties against 44.9% in noncore counties. The widest gap was for CLRD: 57.1% of deaths in noncore counties against 13.0% in large central metropolitan counties, and the gap for CLRD grew the most over the period.

Even as the population under 80 in nonmetropolitan counties fell slightly, from 44.3 million to 43.9 million, their potentially excess deaths from heart disease, unintentional injury and CLRD rose, while cancer fell. Medium and small metropolitan counties followed similar trends. The most urban counties saw cancer and CLRD fall. Heart disease is an example of what the two-way split hid: excess deaths fell in large fringe metropolitan counties (−1.1% a year) but rose in medium and small metropolitan counties and most in micropolitan counties (+2.5%).

Annual percent change in potentially excess deaths before age 80 from the five leading causes, by urban-rural county type, 2010–2017. CDC, MMWR.

Injury: the urban exception

Unintentional injury ran the other way. Its potentially excess deaths rose across most county types, fastest in large central metropolitan (18.3% a year), large fringe metropolitan (17.1%) and medium metropolitan (11.1%) counties, and slowest in noncore counties (2.2%) — narrowing the gap with rural counties, where around 55% of injury deaths were already excess. Across all county types, age-adjusted death rates rose 89% for poisonings, 19% for falls and 7% for motor vehicle crashes from 2010 to 2017. Drug poisoning death rates rose 119.3% in large fringe metropolitan counties, 91.0% in large central metropolitan, 86.3% in medium metropolitan and 30.8% in noncore counties.

Share of deaths before age 80 that were potentially excess, by cause and urban-rural county type, 2017. CDC, MMWR.

Regions and states

The shares varied widely among the 10 HHS regions. For heart disease in noncore counties in 2017, they ranged from 55.8% in Region 4 (Alabama, Florida, Georgia, Kentucky, Mississippi, North Carolina, South Carolina and Tennessee) to 13% in Region 1 (Connecticut, Maine, Massachusetts, New Hampshire, Rhode Island and Vermont); for cancer, from 31.2% in Region 4 to 0% in Region 9 (Arizona, California, Hawaii and Nevada). States varied too: in metropolitan counties, excess cancer deaths fell in every state, while excess injury deaths rose in most states, with Wyoming the only exception.

Limitations

Using the lowest state rates as benchmarks could inflate excess-death counts, though applying the same benchmark everywhere keeps comparisons fair. The analysis deliberately did not adjust for risk factors. Conservative statistics may have missed some trends. Benchmarks from 2008–2010 may understate what could be achieved later. The five causes should not be added together, since preventing one death may only shift it to another cause. And not every death is equally preventable.

Tracking potentially excess deaths by the six county types can help health departments spot and monitor problems and aim interventions where they are needed.

Sources

• Garcia MC, Rossen LM, Bastian B, et al. "Potentially Excess Deaths from the Five Leading Causes of Death in Metropolitan and Nonmetropolitan Counties — United States, 2010–2017." MMWR Surveillance Summaries 68(10). CDC. The report calls the national CLRD trend "stable" in its abstract but "increased" in its results, so only the annual change is given; it also says the rural–urban gap widened for three causes in one place and four in another, so no count is given; and its abstract describes increases "from the five leading causes" in figures that its results give for unintentional injury, which is how they appear here.