The Framingham Heart Study watched a town for decades and invented the risk factor

Heart disease was already America's leading killer in the 1940s, yet medicine could not say who got it. The shock that concentrated minds was Franklin Roosevelt — told his hypertension was benign for years — dying in 1945 of hypertensive heart disease and stroke. In 1948 Congress funded the National Heart Institute to run the first large epidemiological study of cardiovascular disease, and the town of Framingham, Massachusetts won it: close enough to Boston hospitals, with enthusiastic local physicians. The plan was to recruit 6,000 of the town's roughly 10,000 adults.

The design was the innovation. Between 1948 and 1952, 5,209 residents aged 28 to 62 enrolled — women were 55 percent, more than half, unusual for the era. Everyone got a full workup and was re-examined every two years, indefinitely. Participants were given no health advice; diagnostic findings went to their personal physicians instead, so the study would not change the behavior it was watching. Cholesterol, blood pressure, weight, smoking and more were recorded on healthy people who mostly felt fine.

The bet paid on a delay. Early findings in 1957 tied hypertension and cholesterol to future heart disease, and in 1961 the study's six-year follow-up, published under director William B. Kannel, put two new words into medicine: risk factors. Elevated cholesterol, blood pressure and left ventricular hypertrophy each predicted coronary events, and — just as important — the risks compounded as their number grew. Preventive cardiology dates from that paper. The cohort never stopped: participants children formed a new cohort of 5,124 in 1971, and a third generation of 4,095 joined in 2002.

Measure risk before disease strikes, then wait decades

Case-control studies start after disease appears and depend on memory; a prospective cohort records exposures before outcomes, killing recall bias and reverse-causation doubts.

A whole town rather than a clinic supplied volunteers who would keep showing up for decades, in numbers large enough — and with enough women — that findings generalize.

Re-examination every two years made incidence itself measurable: rates, not just odds, so risk could be quantified and combined instead of merely associated.

Routing findings through participants' own physicians kept the study from altering the behavior it was designed to observe.

When causes hide in the past, build a panel that watches the future arrive. Fix the population, measure everything repeatedly, and let outcomes accumulate.

Framingham became the longest-running cardiovascular study in the world and the most productive single source of knowledge about heart disease; its vocabulary — risk factor — is now everyday language. Framingham-derived risk scores guided clinical guidelines for decades, and the Pooled Cohort Equations recommended in the United States today descend from the same idea. The study regenerated itself demographically, recruiting participants' children in 1971 and their children in 2002, and continues to examine cohorts more than 75 years after its first volunteer.