Subscribe to eTOC


KISSIMMEE, FL—When it comes to stroke trends, things are looking up. A new analysis of data from the Framingham Heart Study shows that the annual incidence of stroke has dropped significantly over the past 50 years. And at least for men, 30-day mortality rates have also declined significantly.

Raphael Carandang, MD, a resident in the Department of Neurology at Boston University, reported here at the American Stroke Association International Stroke Conference that the lifetime risk of stroke also dropped significantly. Investigators who discussed the data here gave credit to better control of blood pressure, cholesterol levels, and other stroke risk factors. Dr. Carandang collaborated on the research with Philip A. Wolf, MD, Professor of Neurology at Boston University.

The researchers compared stroke rates in three time periods: 1950 to 1978, 1978 to 1990, and 1990 to 2004. They focused on those epochs, based on when the use of brain CT and then MRI became widespread in the community.


Dr. Carandang said that the Framingham study, a population-based cohort that has been followed prospectively since 1948, provided the perfect opportunity to analyze temporal trends in incidence, lifetime risk, severity, and 30-day mortality rates of stroke for 50 years.

“While numerous studies have assessed temporal trends, the heterogeneity of study methods and differing populations yielded different trends and conclusions,” he said. Additionally, few prior studies had follow-up periods of more than 20 years.

“An obvious strength of the [Framingham] study is that over the last 50 years, it has used the same standardized and previously validated clinical diagnostic criteria,” he said. “It also benefits from excellent continuity by stroke neurologists.”

Overall, 8,636 participants from the original and offspring cohorts were included. All were stroke-free at age 55 and were followed until the first stroke or death. A subset of participants that was stroke-free at age 65 was used to estimate 10-year and lifetime risks of stroke at age 65.

“In addition to looking at age-adjusted annual incidence as many other studies did, we thought it would be more relevant to the individual to give estimates of the remaining lifetime risk of stroke. And 10-year cumulative incidence gives us an estimate of shorter-term risk, which we thought was a good intermediate between annual incidence and lifetime risk,” Dr. Carandang said.


Dr. Raphael Carandang: “While numerous studies have assessed temporal trends, the heterogeneity of study methods and differing populations yielded different trends and conclusions. …An obvious strength of the [Framingham] study is that over the last 50 years, it has used the same standardized and previously validated clinical diagnostic criteria.”


Among the findings (see “Trends in Stroke: The Statistics” for more details):

  • In men, the annual age-adjusted incidence of stroke dropped from 8.6 per 1,000 person-years in the pre-1978 era to 4.8 per 1,000 person-years in 1990 to 2004 (p = .001). In women there was also a significant decline, from 6.5 per 1,000 person-years to 5.0 per 1,000 person-years (p = .004).
  • The severity of strokes appeared stable over time with the percentage of strokes considered to be at least moderate on the National Institutes of Health Stroke Scale.
  • Thirty-day mortality rates decreased significantly in men. Women also showed a trend towards better short-term survival, but the differences were not statistically significant.
  • In men aged 65 years lifetime risk declined significantly.
  • In women, the 10-year cumulative risk decreased significantly from the first to second period studied, but it was stable in the period of 1990 to 2004. Lifetime risk did not change significantly over the three time periods.



Trends in risk factors may help explain the findings, Dr. Carandang said. For example, the percentage of smokers dropped significantly throughout the three periods, and the percentage of participants with hypertension, defined as a blood pressure exceeding 140/90 mmHg, decreased among men and women.

“While things are improving overall, men are faring better than women,” Dr. Carandang said. “Diverging trends in risk factor prevalence and better life expectancy in women may partly explain the different trends among genders, particularly in lifetime risk and mortality.”

For example, diabetes is still on the rise in women, but not in men. In women, rates rose from 4.8 percent in the first time period to 8.9 percent in the third (p < .05), while in men, there was a nonsignficant change, from 7.1 percent to 11.1 percent.

Also, women live longer than men, so they have more years in which they can develop strokes, he said.


Commenting on the study, Ralph L. Sacco, MD, Director of the Stroke and Critical Care Division at Columbia University in New York City, said that better control of blood pressure and cholesterol, “and perhaps diet,” are likely reasons for the improving trend.

But Daniel T. Lackland, DRPH, Professor of Epidemiology and Medicine at the Medical University of South Carolina in Charleston, and an American Stroke Association spokesperson, noted that there is still a long way to go.

“For example, if we look at the control of blood pressure in the time studied, we're doing better, but still not very good.

“We have more and better therapies for both blood pressure and lipids but we're not applying them on a population basis. We need to incorporate these therapies into both primary care and secondary prevention to effect a real change in temporal trends,” he said.

On the other hand, the improvement in 30-day mortality rates, at least in men, does suggest that neurologists are doing a better job at both getting patients to recognize symptoms of stroke as well as at treating stroke in the acute stage, Dr. Lackland said.

Dr. Carandang noted that limitations of this study are that it had a predominantly white study population and that researchers did not do a detailed analysis of risk factor prevalence and its effects on the trends. The group does plan to study further the gender differences in both lifetime risk and mortality.


  • ✓ An analysis of data from the Framingham Study from three time periods revealed that the annual and lifetime risk of stroke dropped significantly over the course of 50 years.