Freis led the team that designed and carried out the study, which was the first randomized, placebo-controlled, double-blind and multi-institutional drug efficacy trial done in cardiovascular medicine. A total of 523 patients and 17 VA hospitals participated in the study, which began in January 1964. Previous VA trials had indicated that a combination of thiazide diuretic, reserpine, and hydralazine was effective in controlling the blood pressure of the majority of hypertensive patients, so this was the regimen chosen for the study. Patients were randomly assigned double blind to either active drugs or placebos.
The patients were chosen from those treated at VA hospitals, and had a greater prevalence of cardiovascular disease than might be found in the general population of hypertensive patients. The median age was 49 years, with about 20 percent between the ages of 60 and 75. Patients were excluded if they had curable forms of hypertension or had a history of a severe hypertensive complication, such as a cerebral hemorrhage, malignant hypertension, severe congestive heart failure requiring continuous use of diuretics, or acute hypertensive encephalopathy. (However, patients with a history of cerebral thrombosis, congestive heart failure controlled with digitalis alone, or healed myocardial infarction were permitted in the trial.) Freis also chose to exclude noncompliant patients, i.e., those who were unreliable in taking their medications as directed. He identified such patients in the initial screening by prescribing an innovative placebo to which riboflavin (vitamin B2) was added; when excreted, riboflavin causes the urine to fluoresce when viewed in ultraviolet light. To qualify for the study, patients' urine tests had to show, on two follow-up visits, that they were indeed taking the pills as directed.
A total of 523 male patients were randomly assigned to treatment and control groups. One hundred forty-three of those had average diastolic blood pressures over 115 mm Hg. The study was terminated in this subgroup after only 18 months of treatment because of the great difference in morbidity and mortality in the control versus the treated patients. Four of the 70 patients in the control group died as a result of cardiovascular complications as compared to none in the 73 treated patients. An additional 17 patients in the control group developed non-fatal complications, such as the malignant phase of hypertension, severe congestive heart failure, cerebral hemorrhage, or disabling cerebral thrombosis requiring their removal from the trial. An additional 6 control patients developed non-terminating events, including mild congestive heart failure, myocardial infarction, and non-disabling strokes. One of the 73 treated patients was removed from the trial due to drug toxicity. Thus over the period of follow-up the incidence of cardiovascular complications was 27 in the placebo group versus only one in the treated group. It became ethically untenable to allow the control group to continue on placebos when the consequences of not offering drug treatment were so clear.
The other 380 patients, with diastolic blood pressures ranging from 90 to 114 mm Hg were followed for an average period of 3.3 years, the longest follow-up being 5.5 years. At the end of this time 19 deaths due to cardiovascular causes had occurred in the control group versus 8 in the treated group. The most frequent cause of death was either myocardial infarction or sudden cardiac arrest, of which 11 occurred in the control group and 6 in the treated group. Stroke was the next most common fatal event, with eight in the placebo group versus only one in the treated group. Overall, 56 (28.9 percent) of the control group developed cardiovascular complications during the trial, compared to 22 (11.8 percent) of the treated group.
The trial demonstrated conclusively that antihypertensive drug treatment was effective in patients who had any evidence of cardiovascular damage and in those with diastolic blood pressures averaging 105 mm Hg or higher. Treatment dramatically reduced the incidence of strokes, congestive heart failure, and progressive kidney damage, although it did not prevent myocardial infarction and sudden cardiac arrest.
When these results were published in 1970, there was very little reaction from the media or from physicians at first. More interest was generated when Freis received the prestigious Lasker Award for Clinical Research in 1971. Meanwhile, philanthropist and health policy advocate Mary Lasker, armed with copies of the VA study, prevailed upon Elliot Richardson, Secretary of Health, Education, and Welfare, to establish a national program for high blood pressure education, to alert physicians and the general public to "the silent killer." In 1972 the National High Blood Pressure Education Program was started, under the direction of the National Heart Institute at NIH. The campaign it launched made hypertension and its treatment a household word.
The VA Cooperative Study was also a turning point in medical thinking about high blood pressure treatment. The Framingham Heart Study, a large-scale, longitudinal study begun in 1949, had by the early 1960s shown a strong correlation between high blood pressure and heart attacks, congestive heart failure, strokes, and kidney damage. With the VA study results in hand, Freis and a growing number of converts could argue persuasively that treating even moderate hypertension would save lives, even if the ultimate cause of the condition remained unknown. Subsequent studies conducted during the following decade supported this argument. Meanwhile, hypertension screening and treatment became much more routine. And though cause and effect cannot be proved, the national death rates from stroke decreased at a rate of 4.2 percent per year between 1973 and 1977.