this 2011 randomized, controlled, prospective study from the British Journal of Psychiatry shows that vitamin d supplementation did not improve well-being in a group of over 1000 elderly women compared to a similar-sized control group.
This is a good study, with negative results. I don't think it means that vitamin d is of no use, but rather that it cannot be assumed to have obvious positive effects for everyone. Some of the effects measured in other vitamin d studies may be the result of non-causative associations (e.g. those with various healthier habits and health paramaters may be more likely to have higher vitamin d levels, but the vitamin d is not the cause of this healthiness, it results from it)
However, the data on this issue continues to evolve. There is some good positive data on vitamin d as well, though not enough in terms of randomized, prospective studies. It will be important, for example, to look at whether vitamin d could obviously be an effective adjunct to other therapies for treating depression. Or whether vitamin d alone has little effect, unless combined with other positive factors.
Meanwhile, I still believe that the standard recommended daily dose of 400 IU for vitamin D is too low, and that 1000-2000 IU per day is better.
See my previous post on vitamin d, http://garthkroeker.blogspot.com/2009/02/vitamin-d-other-vitamins.html