Hyponatraemia is a common abnormality in hospitalized patients, with about 15% having levels below the lower limit of the laboratory reference range. Accepted wisdom is that hyponatraemia is a marker of poor prognosis. However, a critical analysis of the literature reveals significant problems. Researchers have used various cut-off levels for plasma sodium, often concentrating on more severely hyponatraemic groups. Many studies were small, and most did not include control groups. Nevertheless, the literature available does suggest an excess mortality associated with hyponatraemia. Whether this is a direct adverse effect of low serum sodium levels, or if hyponatraemia is simply a marker for 'sicker' patients, is not known. It is also uncertain whether mortality is increased with more severe hyponatraemia, or whether active correction of hyponatraemia will improve outcome. These issues should be addressed by adequately-powered, prospective, suitably controlled studies.