Not all of Mitchell Elkind’s stroke patients are on social security. In recent years he has treated devastating attacks in people as young as 18. And he is not alone. A growing body of research indicates strokes among U.S. millennials—ages 18 to 34—have soared in recent years.

But an analysis by Scientific American has revealed significant differences in where these strokes are occurring, depending both on region and whether people live in rural or urban settings. The investigation, which used data from the U.S. Department of Health and Human Services’ Agency for Healthcare Research and Quality (AHRQ), was reviewed by five stroke experts and found that the West and Midwest have seen especially worrisome increases among younger adults. Moreover, large cities appear to have seen bigger increases than rural areas. The analysis employed hospital discharge data from 2003 to 2012 from the AHRQ’s Healthcare Cost and Utilization Project (HCUP) database.

The findings align with earlier studies that pointed to nationwide increases in strokes in this age group: In a study published earlier this year in JAMA Neurology, researchers at the U.S. Centers for Disease Control and Prevention concluded that in a nine-year span from 2003 to 2012 there was a 32 percent spike in strokes among 18- to 34-year-old women and a 15 percent increase for men in the same range. Scientific American’s analysis sought to dig deeper into the data by exploring whether the stroke trend differed by location.

The findings “are intriguing and interesting,” says Elkind, a stroke expert at Columbia University and New York–Presbyterian Hospital who reviewed the analysis.

“I would have expected it to be more uniform across the country,” Ralph Sacco, president of the American Academy of Neurology, notes that “data has been scant” about strokes among younger people.

“There has been mounting evidence from different studies suggesting that even though the incidence and mortality of stroke is on the decline, the rates may not be dropping quite as much—and even [may be] increasing—among younger populations,” Sacco says. “The reasons for these trends are not entirely clear but there are concerns about obesity, diabetes and physical inactivity having a greater impact in younger stroke victims.” Drug use may be another factor, he adds.

Infographic by Credit: Amanda Montañez; Data source: Agency for Healthcare Research and Quality

Infographic by Credit: Amanda Montañez; Data source: Agency for Healthcare Research and Quality

Although many of the details are murky, the potential impact is clear: In the short term severe strokes among younger adults are a big problem because disability in people in their peak earning years can severely impact their families and future lives, Elkind says. Longer-term, more strokes—even relatively mild ones—among younger adults are worrying because they portend an upcoming epidemic of worse attacks in another 30 years (since survivors’ second strokes are more likely to be stronger and potentially fatal).

“We are just seeing those little waves hitting the beach now but that tsunami will come in the future,” says Elkind, who notes that risk factors such as obesity and smoking are cumulative over time.

Scientific American’s number crunching found that not all the 18- to 34-year-olds’ stroke data mirrored trends seen in other age groups.

Unraveling the reasons behind the trend remains a complex matter. The earlier analysis from stroke expert Mary George and colleagues at the CDC, published this year in JAMA Neurology, found stroke risk factors such as obesity, smoking and hypertension are growing among younger adults. And Scientific American’s number crunching found that not all the 18- to 34-year-olds’ stroke data mirrored trends seen in other age groups. Younger adults, for example, saw statistically significant increases in stroke rates in the Midwest and West.

This is somewhat at odds with regional risks in the broader population, which are more concentrated in the southeastern U.S. In western cities with more than one million residents, for example, the analysis found strokes increased about 85 percent during the 2003 to 2012 time period. In the West as a region, strokes rose 70 percent at the same time. Across the Midwest they increased 34 percent. But in the South the relative increase was smaller and, unlike the spikes in other mentioned areas, this jump did not appear to be statistically significant.

What might explain the difference? The South, known for being part of a “stroke belt” with the highest stroke mortality numbers in the country, also had the highest younger adult rates to begin with—so the relative increase was lower. Other explanations for the regional trends remain to be explored. Pollution might be a contributing factor, which could help explain the higher rates in urban settings, because past studies associate strokes with long-term exposure to high levels of air pollution, Elkind says.

There could also be differences in detection rates. Perhaps an area like the Northeast—which saw less of an increase in strokes than some other areas—may have already been routinely using imaging technology to confirm strokes back in 2003, Elkind says. (The HCUP data did not have regional imaging information for an exploration of that question.)

Still, there are some preliminary indications this phenomenon is due to more than imaging differences: “If it was, then we would see a similar increase in strokes across all age groups,” notes Sacco, who is also chair of the neurology department at the University of Miami. (Nationally, overall stroke numbers have dropped in recent years.)

“I think this data is consistent with other data, and so whenever you have replication consistency across different data sets we begin to take it seriously,” Sacco adds. “I think the fact that we see this [trend] across all regions, and that we see the amount of relative increase for hospitalizations rising for stroke, is alarming.”

Moreover, a 2012 analysis involving younger adult stroke rates in Ohio and northern Kentucky from the mid-1990s through 2005 found rising numbers were not just due to better brain-scanning technology and related improvements in stroke detection.

“That’s the first thing everyone thinks of: increasing [magnetic resonance imaging] utilization. That is certainly a true phenomenon, but in our data we saw this [youthful stroke] increase independently of that,” says Brett Kissela, professor and chair of the Department of Neurology and Rehabilitation Medicine at the University of Cincinnati, who headed the 2012 work. It is likely drug use among the younger adult population also plays a role, he says. “But it really is terrible we don’t [have] an answer to this important question.”

…racial disparities in stroke risk could also muddle results.

Because the Scientific American analysis is based on hospital discharge data, differing levels of health care access could also play a role in these trends, says Virginia Howard, a stroke epidemiologist at the University of Alabama School of Public Health. With many rural hospitals closingin recent years, it is possible rural residents are seeking care farther from home in urban or suburban areas. This could make it appear that rates are going up in certain urbanized areas more than rural ones.

Finally, racial disparities in stroke risk could also muddle results. (For example, strokes generally occur more often among black Americans than among white Americans.) This type of disparity could also be at work in younger adults and could affect the results, Howard says. (The HCUP data were not adjusted for racial differences).

This kind of analysis “is exactly what we should be doing with the data—looking at it in different ways to tease it out,” Howard says. It helps clarify questions for researchers to explore and underscores the need for more prospective studies on stroke risk among younger adults, she adds: “Looking at this in different ways can help us figure out how to approach this issue, and promote community policies and discussions.”