The original DASH trial: what controlled feeding established
With sodium and weight held steady, the 1997 trial tested whole dietary patterns over eight weeks.

Illustrative food image, not a photograph of the study intervention.
A controlled test of dietary patterns
The original DASH trial enrolled 459 adults with systolic blood pressure below 160 mm Hg and diastolic pressure between 80 and 95 mm Hg. Everyone first received a control diet for three weeks. Participants were then randomly assigned for eight weeks to continue that diet, receive a fruit-and-vegetable-rich diet, or receive a combination diet. The combination emphasized fruits, vegetables and low-fat dairy while reducing saturated and total fat. Sodium intake and body weight were kept constant. That design tested dietary patterns without making weight loss or sodium reduction the assigned explanation for the blood-pressure differences.
Appel LJ, Moore TJ, Obarzanek E, Vollmer WM et al.. The New England journal of medicine, 1997. PMID 9099655 · DOI
The overall comparison with control
At baseline, average blood pressure was 131.3/84.7 mm Hg. Compared with the control diet, the combination diet lowered systolic pressure by an additional 5.5 mm Hg and diastolic pressure by an additional 3.0 mm Hg. Both comparisons had P values below 0.001. The fruit-and-vegetable diet produced a smaller systolic difference of 2.8 mm Hg, also with P below 0.001. Its diastolic difference was 1.1 mm Hg, with P equal to 0.07. That last comparison did not meet the conventional 0.05 threshold, so the two fruit-and-vegetable findings should not be described as equally conclusive.
Appel LJ, Moore TJ, Obarzanek E, Vollmer WM et al.. The New England journal of medicine, 1997. PMID 9099655 · DOI
Baseline hypertension changed the size of the difference
Among the 133 participants classified as having hypertension, the combination diet lowered systolic and diastolic pressure by 11.4 and 5.5 mm Hg more than control, respectively. Both comparisons had P values below 0.001. Among the 326 participants without hypertension, the corresponding differences were 3.5 and 2.1 mm Hg. These subgroup averages show why a single DASH number cannot promise the same response for every person. They also remain comparisons within this trial population and feeding period, rather than direct estimates of an individual reader’s future blood pressure or long-term response to a self-selected diet.
Appel LJ, Moore TJ, Obarzanek E, Vollmer WM et al.. The New England journal of medicine, 1997. PMID 9099655 · DOI
What this trial did not isolate
DASH demonstrated a blood-pressure effect from a combination of dietary changes over eight weeks while sodium and body weight were maintained. It did not isolate a single responsible ingredient, and the combination diet included low-fat dairy. A vegan adaptation therefore should not be described as the exact intervention tested in 1997. Nor did this abstract report a reduction in strokes, heart attacks or deaths: the measured outcome was blood pressure. The useful conclusion is specific and substantial enough without embellishment: a controlled dietary-pattern change lowered blood pressure, with larger differences in the subgroup that began with hypertension.
Appel LJ, Moore TJ, Obarzanek E, Vollmer WM et al.. The New England journal of medicine, 1997. PMID 9099655 · DOI