Stewart & Lewis ^{Reference Stewart and Lewis1} have shown that adolescent psychiatric in-patients typically have vitamin D deficiency and hypothesise that this may be due to reduced exposure to sunlight. This may well be true, but it is much more likely that the reduced vitamin D levels are due to changes in the immune system linked with psychiatric disorders. For example, it is now well known that depression is associated with inflammation ^{Reference Berk, Williams, Jacka, O'Neill, Pasco and Moylan2} and that immune markers are typically raised. What is less well known is that vitamin D is a negative acute-phase reactant, ^{Reference Waldron, Ashby, Cornes, Bechervaise, Razavi and Thomas3} which means that its levels drop in response to inflammation. Therefore, a low level may not indicate a deficiency, but rather the presence of inflammation. It would be interesting to reanalyse the data and see whether there are any links with particular diagnoses.