Health · 7 min read
What Sodium Reduction Trials Show About Blood Pressure and Cardiovascular Risk
Feeding and salt-substitution trials separate established blood-pressure effects from less certain population-wide outcomes.

Blood pressure is the clearest outcome
Public arguments about dietary salt often compress several questions into one. Does eating less sodium lower blood pressure? Does it prevent strokes and heart attacks? Does everyone respond by the same amount? Can a change shown in a controlled kitchen be sustained through ordinary food systems? The evidence is strongest for the first question and less complete for each question that follows.
Sodium is an essential nutrient, and the body regulates its concentration through the kidneys, hormones, thirst, and water balance. In many people, a sustained increase in sodium intake raises blood pressure; a reduction lowers it. The size of the response varies. Age, baseline blood pressure, kidney function, dietary pattern, medication, and genetic and environmental factors all contribute to what is often called salt sensitivity.
Randomized feeding trials establish the short-term blood-pressure effect more directly than observational studies can. Trials of clinical events require far more participants and longer follow-up, so much of the debate concerns how confidently a blood-pressure change can be translated into fewer cardiovascular events.
Controlled feeding isolates sodium
The DASH-Sodium trial, reported in The New England Journal of Medicine in 2001, assigned 412 adults to either a typical control diet or the Dietary Approaches to Stop Hypertension diet, rich in fruit, vegetables, and low-fat dairy foods. Within each diet, participants received high-, intermediate-, and low-sodium food for 30 days apiece in random order. Meals were supplied, and body weight was kept stable, allowing investigators to vary sodium while limiting other changes.
On the control diet, moving from high to intermediate sodium lowered systolic blood pressure by an average 2.1 millimeters of mercury; moving from intermediate to low produced a further 4.6-millimeter reduction. Effects were smaller but still present on the DASH diet. Compared with the high-sodium control diet, the low-sodium DASH combination produced a mean systolic pressure 7.1 millimeters lower among participants without hypertension and 11.5 millimeters lower among those with hypertension.
Those are group averages over short periods, not promises for individuals. The trial excluded people with several serious conditions and those taking medicines that affected blood pressure or nutrient metabolism. Participants did not have to shop, cook, or identify hidden sodium. The design therefore provides strong evidence of physiological efficacy under controlled conditions and limited evidence about long-term adherence.
Earlier, the 1997 DASH trial had shown that changing the whole dietary pattern lowered blood pressure even without making sodium reduction the sole intervention. Together, the trials caution against attributing every effect of a dietary pattern to one nutrient. Potassium, fiber, energy balance, alcohol, and food composition can matter alongside sodium.
Meta-analyses of randomized interventions generally find that sodium reduction lowers blood pressure, with larger average reductions among people with hypertension and with larger sodium changes. But trials differ in duration, urine collection methods, baseline intake, and co-interventions. A 24-hour urine collection is more informative than a spot sample for estimating an individual day, yet even 24-hour excretion varies from day to day. Measurement error can blur both dose and response.
Clinical-event evidence adds a second layer
The Salt Substitute and Stroke Study, published in The New England Journal of Medicine in 2021, moved beyond blood pressure. Investigators enrolled 20,995 people in 600 rural Chinese villages. Participants had a history of stroke or were at least 60 years old with high blood pressure. Villages were randomly assigned to use ordinary salt or a substitute containing 75 percent sodium chloride and 25 percent potassium chloride.
Over a mean 4.74 years, the salt-substitute group had lower rates of stroke, major cardiovascular events, and death. The trial is important because it measured outcomes that patients experience, not only a risk factor. Its cluster-randomized design and large sample make the result difficult to dismiss as a short laboratory artifact.
Yet the intervention was not pure sodium reduction: it both lowered sodium and raised potassium. Either change, or their combination, may have contributed. Participants were at elevated cardiovascular risk, lived in settings where discretionary salt was a major sodium source, and people known to use potassium-sparing diuretics, potassium supplements, or to have serious kidney disease were excluded. Results cannot be transferred without qualification to a younger low-risk population whose sodium arrives mainly in packaged and restaurant food.
The trial found no statistically significant increase in serious adverse events attributed to hyperkalemia, but blood potassium was not measured in every participant. For people with impaired potassium excretion, potassium-enriched salt can be dangerous. A population intervention therefore requires labeling, clinical guidance, and a way to protect those with kidney disease or relevant medications.
Why observational findings sometimes conflict
Some cohort studies report J-shaped or U-shaped associations, with higher risk at both low and high estimated sodium intake. Such results have fueled claims that reducing sodium below typical intake is harmful. The association is contested because sodium is difficult to measure and illness can change diet. A person with heart failure, for example, may reduce salt after becoming sick; the lower intake follows the disease rather than causing it. A single spot urine sample can also misclassify usual intake, and mathematical formulas used to infer 24-hour excretion may introduce systematic error.
Observational studies remain useful for long durations and broad populations that trials may not cover. But they cannot, by themselves, settle whether low intake caused an observed outcome. Randomization handles many forms of confounding, while controlled feeding measures a tightly specified intervention. The evidence should be read as a set of designs answering different questions, not as a vote in which every study has equal inferential weight.
There is also a difference between advising one patient and changing a food supply. Most sodium in many high-income countries is added before food reaches the table. Education can help consumers compare products, but reformulation, procurement standards, and gradual recipe changes may alter intake without requiring constant calculation. Those policies have their own evidence base and implementation costs; a feeding trial does not prove which policy will work.
Response varies, but variation does not erase the mean
Not everyone experiences the same blood-pressure change. People with higher initial pressure often show larger reductions. Individual response measured over a few days can be noisy because blood pressure, sodium excretion, sleep, exercise, medication adherence, and stress all fluctuate. Labeling someone permanently “salt sensitive” from one brief test may claim more precision than the measurement supports.
Variation is sometimes used to argue that population averages are irrelevant. The opposite mistake is to treat the average as an individual prescription. A modest shift in mean blood pressure can matter across millions of people even when some respond little and others respond substantially. For clinical decisions, a person’s conditions, medicines, overall diet, and repeated blood-pressure measurements matter more than the batch average alone.
Guidelines from the World Health Organization and national bodies generally recommend lowering high sodium intake. The exact target, the feasibility of reaching it, and the balance of sodium and potassium interventions remain debated. Severe, rapid restriction is not equivalent to a gradual reduction in a high-sodium diet, and medical states involving fluid or electrolyte balance require individualized care.
What the evidence permits
It is established that sustained sodium reduction lowers average blood pressure, especially among people with hypertension. It is also established that, in a large high-risk rural Chinese population, replacing ordinary salt with a lower-sodium, higher-potassium substitute reduced strokes and other cardiovascular outcomes.
It remains less certain how much benefit each person receives, which implementation works best in a given food system, and how event reductions generalize to low-risk populations or to places where packaged food supplies most sodium. Potassium-based substitutes require particular caution for people vulnerable to hyperkalemia.
The defensible summary is therefore specific. Sodium reduction is a validated way to lower blood pressure, not a universal explanation for cardiovascular disease. One major outcomes trial supports a practical substitution strategy in a defined population. The step from those findings to an individual target or national policy still requires information the trials were not designed to provide.
Sources and further reading
4 sources, 2 of them with a link you can follow.
- Frank Sacks and colleagues, “Effects on Blood Pressure of Reduced Dietary Sodium and the DASH Diet”, The New England Journal of Medicine (2001).
- Lawrence Appel and colleagues, “A Clinical Trial of the Effects of Dietary Patterns on Blood Pressure”, The New England Journal of Medicine (1997).
- Bruce Neal and colleagues, “Effect of Salt Substitution on Cardiovascular Events and Death,” The New England Journal of Medicine (2021).
- World Health Organization, Guideline: Sodium Intake for Adults and Children and subsequent sodium-reduction technical guidance.
