Response to 'Letter to the Editor' regarding the article 'The seasonal importance of serum 25-hydroxyvitamin D for bone mineral density in older women' Dear Sir, We thank Dr Sugiyama for his thoughts regarding our publication [1]. These comments mainly deal with differences in physical activity as an underlying explanation for our findings and a null effect on BMD with vitamin D supplementation in metaanalyses of RCTs. We agree that a more accurate definition of true vitamin D deficiency is needed. At what concentrations of serum 25-hydroxyvitamin D (S-25OHD) should physicians recommend supplementation?The response from Dr Sugiyama indicates a conservative approach but very low concentrations of S-25OHD inevitably do lead to rickets and osteomalacia. The dramatic effects of vitamin D treatment on BMD in these (vitamin D deficiency) diseases cannot be explained by physical activity. On the other hand, without stringent indications, that is supplementing those without true deficiency, there is a legitimate fear that vitamin D supplementation might actually cause net harm [2,3]. The report from the Institute of Medicine [4] also emphasized that there may be risks from both low and high levels of vitamin D and that there may be a Ushaped curve of risk, 'which has been seen with many other nutrients as well ' [4]. We showed a clear seasonal variation for the importance of S-25OHD on bone mineral density (BMD) in women [5]. Our results are in agreement with at least 40 nmol/L measured by LC-MS/MS for optimal bone health if the blood sample is drawn during summer whilst those with a low S-25OHD during winter were not, on average, at higher risk of osteoporosis. Our observational design is indeed a major limitation, but differences in leisure time physical activity are nonetheless unlikely to have distorted our results. We did, in fact, adjust our results for differences in reported physical activity levels without major influences on our estimates. Have they existed, as an explanation, such differences in physical activity would also be observed in women during winter time, but during that season, we did not find any major difference in BMD dependent on S-25OHD status. Nota bene, differences in risk factor patterns between those with low and those with higher S-25OHD were similar during summer and during winter (including no major dissimilarities in physical activity). Moreover, physical exercise in older women provides only modest positive effects on bone accrual [6-8] -effects on BMD much smaller than those observed by us.A thorough meta-analysis from Ian Reid's research group displayed on average no major effect of vitamin D supplementation on bone mineral density [9]. No effect modification by higher dose, duration of treatment or low baseline 25-hydroxyvitamin D levels (<50 nmol L À1 as suggested by Institute of Medicine [4]) was noted. The problem is, however, differences in assay performances of the studies included in the meta-analysis, seasonal variation, fat mass, nutritional status (e.g. calcium intake) and ...