How the Framingham Heart Study Works

Director of the Framingham Heart Study program, Dr. William Castelli, stands in front of the Framingham, Massachusetts, house that served as research center and clinic for the program, on Aug. 22, 1984 . Nathan Benn/Corbis via Getty Images
Director of the Framingham Heart Study program, Dr. William Castelli, stands in front of the Framingham, Massachusetts, house that served as research center and clinic for the program, on Aug. 22, 1984 . Nathan Benn/Corbis via Getty Images

Twenty miles (32 kilometers) west of Boston lies the town of Framingham in Massachusetts, which in 1948 was tapped to become the site of the most influential longitudinal study ever done. Known as the Framingham Heart Study (FHS), the research project was set in motion by the National Heart Institute, known today as the National Heart, Lung, and Blood Institute (NHLBI), and is credited with uncovering much of what we know now about cardiovascular disease (CVD). CVD is a classification of heart and blood vessel diseases that includes coronary heart disease, pulmonary embolism, deep vein thrombosis, rheumatic heart disease and others. People with CVD often have heart attacks and strokes, and some 17.5 million worldwide die of CVD every year [source: Hajar].

At the time the study was created, little was known about the causes or risk factors associated with CVD, although by the 1940s, scientists did know that it was the leading cause of death in the U.S. Prior to this time, infectious diseases like tuberculosis and pneumonia killed more people, but advances in sanitation, antibiotic treatment and vaccinations put a stop to that [source: Framingham Heart Study].

In 1945, President Franklin D. Roosevelt died at age 63 from a cerebral hemorrhage (a type of stroke) due to untreated hypertension (high blood pressure). Following his death, it became glaringly obvious that, despite FDR's prior diagnosis of CVD, little was known about the causes, much less how cardiac issues should be treated. People thought dying from CVD was inevitable. Roosevelt's death inspired his successor, President Harry Truman, to recognize the prevalence of CVD as a threat to public health, and in 1948 he signed the National Heart Act into law, which established the FHS, a joint effort managed by Boston University and the NHLBI [source: Hajar].

The idea was to follow a large group of willing participants, most of whom initally lacked any obvious symptoms of CVD, over a significant period of time. By doing so, researchers hypothesized that certain commonalities or characteristics would eventually emerge, leading to conclusive evidence about why and how CVD develops. The initial funding commitment was for only 20 years, so the researchers were relieved when, in 1971 President Nixon established a federal contract to keep it going [source: Whitney]. More than 40 years later, the study is still chugging along, and now involves multiple generations and thousands of participants.

So, why was Framingham chosen for this noble goal, and how did the researchers collect the data, anyway?

Why Framingham?

Spectators cheer runners as they make their way past the 6-mile mark of the Boston Marathon on April 17, 2017, in Framingham, Massachusetts. Kayana Szymczak/Getty Images
Spectators cheer runners as they make their way past the 6-mile mark of the Boston Marathon on April 17, 2017, in Framingham, Massachusetts. Kayana Szymczak/Getty Images

To many, Framingham might seem like an odd choice for such an important study, instead of a better-known and more populated metropolis, like New York City or Boston. The city was actually selected as the site before researchers determined that this would be an epidemiological study — meaning one that looks at the history of the disease and its effect on people over time.

Ideally, the study would have been established in geographically separate areas at the same time, to represent a wide variety of racial, socioeconomic and regionally impacted groups. "The Framingham investigators have always been aware that the site may not be representative of the United States and have made repeatedly comparisons with other regions to test its generalizability," the scientists note on the study's website.

At the time the FHS started, the Framingham population of 28,000 was a mostly white and middle-class. However, the town did have some things going for it as a research site. The population was large enough to supply participants, yet the town was physically small enough that it wasn't too hard to observe or check in with them. The medical community and local hospitals were also very supportive of the study. The town had previously participated in the Framingham Tuberculosis Demonstration, so reseachers were pretty well assured cooperation from the locals [sources: Framingham Heart Study, Whitney].

Although not ideally diverse in nature, there were other ethnic groups represented in Framingham, thanks to an influx of post-World War II residents. This trend has only increased since the study's inception, with Framingham now home to about 70,000 residents [source: CityTownInfo].

Once the researchers had a location, they had to figure out how to enlist enough participants.

Choosing Participants for the Framingham Study

Dominic Verelli, an original participant in Framingham Heart Study, goes through his exam. Mark Peterson/Corbis via Getty Images
Dominic Verelli, an original participant in Framingham Heart Study, goes through his exam. Mark Peterson/Corbis via Getty Images

When the study began, the investigators settled on a target sample size of 6,000 people between the ages of 30 and 59. People in this age window more commonly develop CVD, yet a sample in this range would also include a significant number not already experiencing symptoms. Since the idea was to see why and how CVD develops over time, these two factors were key. Participants had to show up every two years to provide updated medical histories and have physical exams and lab tests done. Researchers hypothesized that there wasn't one cause of CVD, but many, so the medical histories and exams would be crucial to finding out what these causes were.

Two-thirds of the families in Framingham were approached to be part of the sample. Local civic committees and clubs personally contacted people to encourage them to participate. These efforts brought 4,469 people in for examinations, which wasn't enough. The study was expanded to include volunteer participants who hadn't been contacted initially. Another 740 of those volunteers were added to what's known as the "Framingham Cohort," numbering 5,209 men and women between the ages of 28 to 62.

Two issues resulted from this method: The sample group was not completely random, as the investigators had hoped, and the participant group was healthier than the general population. The researchers ended up adding people who had CVD to the Framingham Cohort [source: Framingham Heart Study].

In 1971, the study expanded to include offspring of the original cohort after President Nixon worked his research-extending magic. This "Offspring Cohort," comprised 5,124 people, and included spouses. This new group served a dual purpose. First, adding the children allowed researchers to look for evidence of familial clustering of CVD, while the extra layer of spouses allowed for examination of unrelated young adults [source: Mahmood].

In 1994, the First Omni Cohort (one of an eventual three) was established to address the increasingly diverse population of Framingham. The first of these cohorts was made up of 507 people with Native American, African-American, Hispanic, Indian, Asian and Pacific Islander roots who were local to Framingham or surrounding areas.

The study expanded yet again with the establishment of the Third Generation Cohort (Gen III), which began in 2002 and is ongoing and expected to close in 2019. It is currently following 4,095 people aged 19 to 79 at the study's onset, who have at least one parent enrolled already participating in the Offspring cohort. Yet another group added to the mix is the New Offspring Spouse Cohort, composed of spouses from the Offspring Cohort who never signed on before, for whatever reason, and who have two or more offspring taking part in Gen III.

Over time, the questions asked in the medical histories and the tests administered changed as the researchers learned more about the causes of CVD.

What We've Learned from the Framingham Study

Thanks to the Framingham study, we now know that certain behaviors increase a person's chances of developing CVD. Here are some of the findings that now seem commonplace but no one knew for sure before this study — as well as the date the finding was discovered [source: Framingham Heart Study]:

  • 1960: Cigarette smoking found to increase risk of heart disease.
  • 1961: Cholesterol level, blood pressure and electrocardiogram abnormalities found to increase the risk of heart disease.
  • 1967: Physical activity found to lessen the risk of heart disease while obesity increases the risk of heart disease.
  • 1970: High blood pressure found to increase the risk of stroke.
  • 1970: Atrial fibrillation (irregular heartbeat) found to increase stroke risk 5-fold.
  • 1976: Menopause found to increase the risk of heart disease.

"Much of our appreciation of the pathophysiology of heart disease came from the results of studies from the FHS," writes Dr. Rachel Hajar in the journal Heart Views. "It established the traditional risk factors, such as high blood pressure, diabetes and cigarette smoking for coronary heart disease. Framingham also spearheaded the study of chronic noninfectious diseases in the USA and introduced preventive medicine."

These findings led to a public health change from waiting for CVD symptoms to appear to trying to prevent them altogether. As a result, the Framingham Risk Score (FRS) — also known as the clinical risk score — was developed. Physicians continue to use them today to calculate a patient's risk of having or developing cardiovascular disease within 10 years. Each risk category has points assigned to it, which helps physicians calculate the risk specific to men and women. These categories include age, smoking status, blood pressure (treated vs. not treated), cholesterol level and whether or not the patient is diabetic [source: Davis].

The FHS has also identified other risk factors that, although not as impactful as smoking or diabetes, can nonetheless raise a person's risk for CVD. For example, menopause, as well psychosocial issues, like depression, stress and anxiety, are all linked to heart disease. Sleep apnea is also associated with a higher risk of stroke [sources: Hajar, Zoreh].

In addition to the personal risk of CVD, the study aims to give family members a better idea of their genetic odds. In 2010 researchers produced the first solid evidence that a person is three times more likely to have a stroke if one of their parents had a stroke by the age of 65. That same year, they determined that a patient's odds for having atrial fibrillation increase if they have a first-degree relative who has that condition [source: Hajar].

Criticisms of the Framingham Heart Study

Evelyn Langley, 87, an original member of the Framingham Heart Study, embraces a family member. Mark Peterson/Corbis via Getty Images
Evelyn Langley, 87, an original member of the Framingham Heart Study, embraces a family member. Mark Peterson/Corbis via Getty Images

Despite all the milestones, a study of this magnitude doesn't go forth without some missteps and criticism. As already noted, the lack of diversity in the study has long been a chink in the armor. This has called into question whether the results are truly generalizable, especially in today's melting pot society.

For example, one review found that the risk algorithm initially established was accurate for black and white Americans; however it estimated too high for Hispanics, Japanese-Americans, and Native American women [source: Davis]. Ideally, the FHS's continuing addition of more diverse groups will mitigate this issue and provide further insight into how genetics and lineage play into CVD. Some scientists have modified the FHS risk score to be more applicable to people of different ethnicities. For example, the ETHRISK calculator is tailored to British minority groups like those of Indian, Bangladeshi, Chinese, Irish and black Caribbean ancestry.

It's also been suggested that the FHS-established risk factors that apply so well to men, may not be indicative of CVD risk in women because as many as 20 percent of women who experience coronary events have none of the associated risk factors [source: Davis]. In a similar move to the ETHRISK calculator, a Reynolds Risk Score has been developed, which includes some other ingredients, like family history and the high-sensitivity C-reactive protein (hs-CRP) test to better determine a woman's risk of heart disease.

Other critics insist that high saturated fat levels — pinpointed by the FHS long ago as related to getting heart disease — are less to blame than the excessive intake of trans fats, high carbohydrates and sugars [source: Healy]. Indeed, the FHS originally found that there was no relationship between a participant's intake of calories from fat and his or her cholesterol level. It also seemed as if a drop in cholesterol levels correlated with an increase of CVD death for those over age 50. These findings puzzled the researchers and were not included in their official report [sources: Eades, Malhotra].

"A lower cholesterol is not in itself the mark of success, it only works in parallel with other important markers, like a shrinking waist size and diminishing blood markers for diabetes," writes cardiologist Aseem Malhotra.

Other Findings of the Framingham Heart Study

Over time, the FHS has expanded to look at medical issues other than those involved with CVD. Neurological disorder research has been a focal point for several decades, particularly after a program began in 1997 where FHS participants could donate brain tissue after death.

This has allowed researchers to look at the impact of the aging process on the brain, and to learn more about neurological illnesses, like Parkinson's or Alzheimer's diseases. This research is also helping to develop a better understanding of the genetic risks of these diseases.

A street sign in Framingham touts its place as the home of this famous heart study.
A street sign in Framingham touts its place as the home of this famous heart study.

In addition, a recent examination of Offspring Cohort participants found a possible link between consumption of artificially sweetened beverages and dementia and stroke. Specifically, those who consumed one or more of that type of beverage per day were 2.9 times as likely to eventually be diagnosed with Alzheimer's disease, and three times as likely to experience ischemic stroke (stroke as a result of blood clots) as those who didn't [source: Bachert].

On a more lighthearted note, another study used FHS-gleaned genetic data to determine that many married couples look alike because of conscious or unconscious preference for certain characteristics, like height, weight, social class and religion. The FHS data showed that, while this practice was pretty much a given post-World War II, it has declined consistently in the decades since [source: Sebro].

At least 1,200 articles have been published in prominent medical journals using data from the FHS [source: Hajar]. And as this remarkable study continues, no doubt there will be many more to come.

Author's Note: How the Framingham Heart Study Works

Just about all of us have been impacted by some form of heart disease, whether firsthand or by way of a loved one. As someone who lost one of her favorite people without warning to catastrophic myocardial infarction, I hope that this valuable research continues for years to come, making sure to tweak it along the way to be more generalizable to all types of people.

Related Articles

More Great Links


  • Bachert, Alexandria. "Study finds diet soda associated with stroke, dementia." MedPage Today. April 20, 2017 (June 18, 2017)
  • "Framingham's TB Study 1916-1923 — Paving the way for the Framingham Heart Study (1948)." 2017 (June 18, 2017)
  • CityTownInfo. "Framingham, Massachusetts." 2017 (June 18, 2017)
  • Davis, Margot MD and Margot Davis, MD, Jason Andrade, MD, FRCPC, Carolyn Taylor, MD, MPH, FRCPC, Andrew Ignaszewski, MD, FRCPC. "Cardiovascular risk factors and models of risk prediction: Recognizing the leadership of Dr Roy Dawber." British Columbia Medical Journal. Sept. 2010 (June 18, 2017)
  • Framingham Heart Study. "Epidemiological Background and Design:
  • Framingham Heart Study. "Epidemiological Background and Design:The Framingham Study." 2017 (June 18, 2017)
  • Framingham Heart Study. "New Offspring Spouse Cohort." 2017 (June 18, 2017)
  • Framingham Heart Study. "Omni Cohorts." 2017 (June 18, 2017)
  • Framingham Heart Study. "Postmortem brain tissue donation program." 2017 (June 18, 2017)
  • Framingham Heart Study. "Third Generation Cohort." 2017 (June 18, 2017)
  • Hajar, Rachel. "Framingham Contribution to Cardiovascular Disease." Heart Views. April-Jun 2016 (June 18, 2017)
  • Mahmoon, Syed S. and Daniel Levy, Ramachandran S. Vasan and Thomas J. Wang. "The Framingham Heart Study and the Epidemiology of Cardiovascular Diseases: A Historical Perspective." The Lancet. March 15, 2014 (June 18, 2017)
  • Whitney, Kathy. "Framingham Heart Study's landmark impact examined." Vanderbilt University News. Oct. 3, 2013 (June 18, 2017)
  • Zohreh Khayyam-Nekouei and Hamidtaher Neshatdoost, Alireza Yousefy, Masoumeh Sadeghi, and Gholamreza Manshaee. "Psychological factors and coronary heart disease." ARYA Atheroscler. Jan. 2013 (June 18, 2017)