In response to Dr Mozaffarian, it is correct that our prospective observational study1 does not demonstrate causality. We do not suggest at any point that we have demonstrated causality. Indeed, our main conclusion is that further prospective studies and randomized controlled trials are now needed. Similarly, we identified that vitamin D may have therapeutic potential in this context. Again, supplementation trials are now needed to establish whether this is the case. We also acknowledge the possibility of reverse causality—in short, people with preexisting dementia may have low levels of vitamin D at baseline and also be susceptible to further cognitive decline. This hypothesis is testable, and we established that excluding people with dementia at baseline and controlling for impaired mobility did not change the pattern of associations. Furthermore, there was no statistical interaction between baseline cognition and vitamin D levels, suggesting that the association was not driven by people with poor cognitive function at baseline. There were a number of differences between participants grouped by serum 25(OH)D levels at baseline as we would expect, including total energy intake. These differences were controlled for in fully adjusted models and did not change the pattern of results. Given the wealth of evidence from animal and in vitro experiments to suggest that vitamin D has a wide range of biological functions throughout the human body, including neuroprotection, we do not accept that an association with cognitive decline or nonskeletal diseases is “intuitively unlikely.”
Llewellyn DJ, Langa KM, Lang IA, Melzer D. Investigating Factors of Decline in Cognitive Function or Dementia—Reply. Arch Intern Med. 2011;171(3):267. doi:10.1001/archinternmed.2010.530
Customize your JAMA Network experience by selecting one or more topics from the list below.
Create a personal account or sign in to: