Dietary Sodium and Cardiovascular Disease Risk – Measurement Matters

2nd June 2016

CASH summary of publication released by The New England Journal of Medicine on Dietary Sodium and Cardiovascular Risk

A recent controversial study published in the Lancet has attempted to cast doubt on the role of salt reduction on health, despite many other papers disputing this and renowned health professionals highlighting major problems with the analysis.

This review in the New England Journal of Medicine written by Mary Cogswell and colleagues clearly and succinctly summarises the methodological problems found in these prospective cohort studies which alter the direction of the results. The authors evaluated the strength of the evidence using Hill’s classic criteria and made a number of conclusions:

  • Firstly, in studies where participants with pre-existing medical conditions such as cardiovascular disease were recruited, reverse causality may explain a non-casual association between low sodium intake and increased cardiovascular events.
  • In addition to this, studies employing spot urine tests to measure sodium intake are subject to overestimation of intake at low levels; multiple, non-consecutive 24-hour urine collections should be used as the gold standard. Samples from multiple days are also required to mitigate the attenuation of the effect-size estimate from day-to-day sodium excretion variability.
  • Studies with a short follow up and those relying on individuals to follow a low sodium diet are of particular concern, as a low sodium diet is uncommon and hard to follow in a Western country, where most of the salt we eat is already in the food we buy. Conversely, population-level strategies such as the setting of voluntary sodium targets in the United Kingdom and the monitoring of the outcome over some years (2003-2011), suggests a strong connection between reduced sodium in the food supply and decreased cardiovascular events.
  • Moreover, some authors have claimed that low sodium intake increases blood lipids during short studies. This fails to explain the physiological basis for increased cardiovascular risk and when limited to longer term interventions lasting more than 4 weeks, is found to be temporary and unlikely to have population effects. It might also be expected that use of blood pressure lowering agents, like diuretics, increase the risk however, analyses of 123 randomised trials have demonstrated reduced risk.  

The application of Hill’s criteria for the association between low sodium intake and increased cardiovascular risk suggests that the association is not causal. Measuring sodium intake presents a huge challenge on a short term individual basis and estimating population averages through public health interventions is far less prone to error.

As the author states, “Paradoxical findings based on inaccurate sodium measurements should not stall efforts to improve the food environment in ways that enable consumers to reduce excess sodium intake”. The argument now is not whether we should reduce salt, but how. A wealth of evidence exists to demonstrate that sodium reduction prevents cardiovascular disease  and has rightly led to salt reduction strategies taking place around the world.