People with higher vitamin D levels in their blood are less likely to die of colorectal cancer, U.S. researchers said Tuesday.

But the researchers also made the surprising discovery that the vitamin did not appear to affect the chances of dying from any other type of cancer.

The study's results question the conclusions of several other recent studies that have found that vitamin D can lower the risk for many types of cancer. The Canadian Cancer Society was so impressed by the growing body of evidence about the health benefits of vitamin D that it recommended earlier this year that the entire adult population of Canada begin taking vitamin D supplements.

The Society noted that Canadians are at particular risk of vitamin D deficiency since we typically don't produce enough vitamin D from sunlight.

For this study, Dr. Michal Freedman, of the National Cancer Institute in Bethesda, Md., and colleagues analyzed data from the third national Health and Nutrition Examination Survey. The researchers tracked 16,818 people who were part of the health survey between 1988 and 1994.

The participants provided blood samples that the researchers used to determine the level of vitamin D in their blood. The researchers then followed the subjects through 2000. The study group was not asked about any vitamin D supplementation they might be taking.

Among the group, 536 participants died of cancer during the study period.

People with higher levels of vitamin D when they entered the study had an impressive 72 per cent reduced risk of dying from colorectal cancer, compared to those with the lowest levels of vitamin D, the researchers reported.

But the researchers saw no link between vitamin D levels and the overall risk of dying from cancer, including from lung, prostate or breast cancer, non-Hodgkin lymphoma and leukemia.

The results are published in the Journal of the National Cancer Institute.

NIH experts Cindy Davis and Johanna Dwyer, in an editorial accompanying the study, noted that the six to 12-year period of the study may not have been long enough to demonstrate an effect of the vitamin on cancer deaths. This would be particularly true if vitamin D plays a role in preventing cancer in its earliest stages, as some suspect.

As well, the commentary notes that the vitamin D blood levels were measured at only one time point and "may not have been representative of long-term chronic levels."

They add: "These findings must be put into the context of total diet and lifestyle. There are many risk factors other than diet for colorectal cancer, and there are many possible dietary risk factors other than vitamin D that have been linked to cancer risk."

They also caution that "the public should not, in a rush to judgment, assume that vitamin D is a magic bullet and consume high amounts of vitamin D. More definitive data on both benefits and potential adverse effects of high doses are urgently needed."