Intended for healthcare professionals

Observations Medicine and the Media

Are schoolchildren unhealthily underhydrated?

BMJ 2012; 344 doi: (Published 19 June 2012) Cite this as: BMJ 2012;344:e4196
  1. Margaret McCartney, general practitioner, Glasgow
  1. margaret{at}

Headlines claiming that children aren’t drinking enough water were based on a study that was unpublished and not peer reviewed. It was also funded by Nestlé, a big player in the bottled water market, writes Margaret McCartney

“Nearly two thirds of children are not drinking enough at breakfast time to be properly hydrated, according to a study,” began a story in the Independent.1 So too read a headline on the Daily Mail’s website.2 The Scotsman published a shorter piece,3 and news networks in New Zealand, Ireland, and India picked up the story.4 5 6 The Independent’s article contained a large chunk of a press release that Nestlé had released under embargo to coincide with the presentation of research it had funded at a conference in San Diego.

Nestlé Waters, which describes itself as “the healthy hydration company,” manufactures several bottled waters, including Perrier, San Pellegrino, Vittel, and Nestlé Pure Life, saying on its website that its brand is “the right choice for healthy family hydration.”7 The UK bottled water market is worth £1.5bn (€1.9bn; $2.3bn).8 Is it really true that children are lacking in water, whether from tap or bottle?

The press release was headlined, “British children start the day not sufficiently hydrated.” It went on, “Nearly two thirds (60%) of children aged 9-11 arrive at school not sufficiently hydrated. The figure is higher for boys at 68.4%, with girls faring slightly better—although 53.5% are still arriving at school with an insufficient level of hydration.” But what is “insufficient hydration?” The press release states that “the term ‘not sufficiently hydrated’ references the stage before clinical dehydration; at this stage you could benefit from drinking more fluids so that you maintain the balance of water in your body.”

This unpublished research was a study of 452 children in Sheffield aged between 9 and 11 years who had been asked to complete questionnaires on how much food and fluid they consumed before leaving for school in the morning. Each child had to obtain a urine sample 30-60 minutes after breakfast, which was handed into school for measurement of its solute concentration.

Accompanying the press release was a document “Research Details,” which contained a chart that gave “the categories for urine osmolality—the main evaluation criteria.” This stated in red that urine osmolality of 801-1000 mOsmol/kg showed that cellular hydration status was “high = dehydrated cells” and that an osmolality of >1000 mOsmol/kg correlated with “highly concentrated = very dehydrated cells.”

Do we know the normal urinary osmolality in children? Nestlé gave me three references to support its assertion that 800 mOsmol/kg is “the limit between a euhydrated [with normal body water content] and a slightly dehydrated status,” and I will use the units specific to each paper.

Osmolality (mOsmol/kg) and osmolarity (mOsmol/L) are different measures of solute concentration that are comparable to within a few per cent in biological systems because the density of dilute solutions such as urine in such conditions is about 1 kg/L.9 For non-dissociating solutes these units are also numerically equivalent to mmol/L.

The first of the papers, from the European Food Safety Authority, was titled “Scientific opinion on dietary reference values for water.”10 This states, “Normal hydration status is the presumed condition of healthy individuals who maintain water balance,” and the paper is clear that “normal hydration status can be achieved with a wide range of total water intakes because of homeostatic control mechanisms.” It gives the reference range for urine osmolarity as 50-1400 mOsmol/L.

In circular fashion, this paper also cited another of the papers separately referenced by Nestlé, from the Journal of Paediatrics.11 Here, urine osmolality, 24 hour urine volume, and water intake were measured in 479 healthy children taking part in the Donald (Dortmund nutritional and anthropometric longitudinally designed) study. The data were used to calculate “a range of euhydration and ranges of risk of hypohydration.” These children were by definition healthy; this classification was capable of expressing statistical variation in urine osmolarity but was not designed to elucidate the meaning of these results.

The third paper referenced by Nestlé used some of the data from the Donald study to produce a definition of euhydration12; however, the authors also state that as well as urinary osmolality “additional clinical or biochemical signs of hypo(hyper)-hydration are necessary to diagnose hypo(hyper) hydration in a subject.”

It seems, therefore, that we have reference ranges for urine concentration in children, but we do not necessarily know what is unhealthy. Nestlé, however, said in a statement that a morning urinary osmolality of >800 mOsmol/kg showed that the children were “not sufficiently hydrated.”

But does a morning urinary osmolality of >800 mOsmol/kg really represent the “stage before clinical dehydration,” as Nestlé asserts? Neil Turner, chairman of Kidney Research UK, says not. “This is nonsense,” he said. “If you found serum osmolality to be outside normal limits you might be able to come to that conclusion, [but] you can’t say that from a single urine specimen.”

So why did Nestlé say that children with a urinary osmolality of >800 mOsmol/kg are “not sufficiently hydrated”? I suggested to Nestlé that this should be regarded as physiological. The company responded, “We agree that children can easily concentrate their urine if they are not drinking sufficiently,” and, “Our research was not designed to evaluate the consequences of high urinary osmolality on health.”

I asked Gerard Friedlander of Paris Descartes University Medical School, who presented the research in San Diego, whether “not sufficiently hydrated” was meant to imply something harmful. He told me by email that it did not: “We have said that two thirds of these children have nearly reached their maximal renal concentration capacity. They are of course still in the physiological range.” However, the data released by Nestlé showed only that 18.6% of children had a urine osmolality >1000 mOsmol/kg; how near they were to the maximum was not disclosed. A BMJ paper from 1996 found that healthy children aged 3-18 had, after a night’s fast, morning urinary osmolalities ranging from 329 to 1218 mmol/kg, indicating that children are able to concentrate their urine well above 1000 mmol/kg.13

So why did Nestlé publish a press release saying that children aren’t drinking enough water, and why did journalists report this as fact? The lack of critical peer review did not help. Nestlé said, “We publicised this research for the same reason that the abstract was submitted . . . to the Congress. We considered that it is of concern that the study found so many children to be close to the upper limit of urine concentration before they go to school.” Yet this is contentious and a potentially biased view of results that haven’t been published. Nestlé says that this study is “part of a research project commissioned by Nestlé Waters, which also commissioned studies in France, Italy, and the USA.” Children, of course, need free access to water, and tap water is better for the environment than bottled water. But none of us need feeding with hype from the bottled water industry.14


Cite this as: BMJ 2012;344:e4196


  • Competing interests: None declared.

  • Provenance and peer review: Commissioned; not externally peer reviewed.

  • Feature: Waterlogged? (BMJ 2011;343:d4280, doi:10.1136/bmj.d4280)