USPSTF's Draft Recommendations re Vitamin D: Huh?

OK, first, let me own up to my goof.  This past Christmas holiday, I was out of stable & reliable cellular service, much less Internet access, for about 2 weeks.  During this time, I missed the US Preventive Services Task Force's meta-analysis & evidence report published in the Annals of Internal Medicine in which the authors concluded that "combined vitamin D and calcium supplementation can reduce fracture risk . . . Evidence is not sufficiently robust to draw conclusions regarding the benefits or harms of vitamin D supplementation for the prevention of cancer."

Second, let's recall that less than 2 weeks ago, the USPSTF concluded that "vitamin D supplementation is effective in preventing falls in community-dwelling adults aged 65 years or older who are at increased risk for falls (B recommendation)."  So what changed over the last 14 days?  And why the apparent dramatic 180o turnaround announced yesterday to the lay press?  Let's take a look at the draft summary of recommendations one by one.

USPSTF concludes that the current evidence is insufficient to assess the balance of the benefits and harms of vitamin D supplementation, with or without calcium, for the primary prevention of cancer in adults.  In fact, this is not any different from the conclusion drawn & published 6 months ago.  But what about all the studies that I've reported on?  Remember that observational studies are useful only for developing hypotheses but useless for proving cause & effect, the latter which is required for making public policy recommendations.

USPSTF concludes that the current evidence is insufficient to assess the balance of the benefits and harms of combined vitamin D and calcium supplementation for the primary prevention of fractures in premenopausal women or in men.  When generalizing results from specific data sets, make sure the conclusions apply to you.  For instance, positive pregnancy tests can confirm pregnancy in appropriate age women but are most likely erroneous in men, right?  Likewise, most, if not all, the data reviewed was on postmenopausal women, thus the conclusions can't be generalized to men or premenopausal women.

Always remember that correlation does not imply causation.  And in the meantime, stay tuned for an analysis of the rest of the draft recommendations in tomorrow's post.

Follow @alvinblin
Tweet