Saturday, December 29, 2012

Vitamin D for Prevention

This blog posting is discussing the recent news about using Vitamin D as a preventative for type-1 diabetes (not a cure).  You can read more about the study here (and many other places on the net):

More personal and emotional news article:

Quick Summary:  The researchers measured Vitamin D levels in the blood of people who were later diagnosed with type-1 diabetes, and compared that to levels in the blood of people who were not diagnosed.  The levels in the diagnosed group were significantly lower.  There was an obvious correlation.  Vitamin D was not an absolute preventative.  There was not a specific amount of Vitamin D where if you took more than that you would avoid type-1.  Rather there is a change in probabilities.  Higher levels of Vitamin D led, on average, to lower levels of type-1 diabetes.    Some people with high levels of Vitamin D still got type-1 diabetes, it was just less likely.  The reverse was also true: some people with low levels of Vitamin D avoided type-1 diabetes, but it was less likely.

A Little Background: It has been well known for decades that people who live near the equator have lower rates of type-1 diabetes than people who live nearer the poles.  However, it is not known why this is true.  Some people believe that a lack of sunlight or Vitamin D increases the rate of type-1 diabetes.  Other people think it might be wealth, genetics, diet, or any one of a huge number of differences.  (In the Americas, for example, Mexico is closer to the equator and has a lower rate, while USA is farther from the equator and has a higher rate.  But there are also large differences in wealth, genetics, diet, and so on.)

A Little More Background:  Research studies can broadly be put into two groups: population based studies and intervention studies.  Population based studies are studies that take two groups of people and compare them in some way.  Hopefully, the two groups should be as similar as possible, except for the one thing being studied.  Intervention studies take one group of people, and gives some of them a treatment (the intervention) and not the others, and then compares them.

It is important to remember that there is a clear difference in quality between the two types of studies: intervention studies are generally much higher quality and are much easier to interpret.  Population based studies are often apples to oranges comparisons where the differences seen have nothing to do with the change being studied.  The Mexico to USA comparison is an example.  Is Mexico's lower type-1 rate due to genetics?  sunlight?  wealth?  cleanliness?  With a population based study, it is usually impossible to know.

This study was a population study, and so it is not as strong evidence as an intervention study.

Understanding This Research

For the recent Vitamin D study, the researchers followed people in the military.  When they entered the military some of their blood was frozen.  So later (on average 1 year later), if they were diagnosed with type-1 diabetes, the blood was tested for Vitamin D levels.  Soldiers who were similar, but who did not come down with type-1 diabetes served as a control group.  It's a very resourceful experimental design, because normally it would be very hard to test Vitamin D levels months or years before diagnoses, so finding the store of available frozen blood was brilliant insight.

This research was unique in several ways.  For one thing, it is the first study I have seen that looked at relatively old people.  Previous studies that I have seen have dealt with infants.

This study has avoided many of the common pitfalls of population based research.  In particular, population based studies often compare people from different countries or different regions, who often have many differences.  In this study both the control group and the diabetes group were taken from the same pool of people (American service members), which is a huge advantage over many population based studies.

Should People Without Type-1 Diabetes Take Vitamin D?

Remember this study says nothing about the effects of Vitamin D on people who already have type-1 diabetes, so it provides no support for the idea that people who already have type-1 should take Vitamin D.  However, the open question is this: should people who don't yet have type-1 diabetes, especially brothers and sisters of people with type-1, take extra Vitamin D?

That depends entirely on your personal beliefs about how much evidence is required, before you will pay money for a treatment.  Right now there are two population based studies that suggest that Vitamin D has a protective effect (one is here:  Only you can decide if that is enough support for you to change your behavior and spend your money.  There have also been studies on Vitamin D that have shown no difference, and I'm sorry I don't have time for a full review of all the studies.  But the link below will take you to the 13 studies listed in clinical trials for "type-1 diabetes" which study vitamin d, if you want to review them all:
link to for type-1 and Vitamin D

For comparison, the FDA generally requires 4 intervention studies (and there are other quality requirements on these studies, as well) to approve a new drug.  Population studies don't count.  Of course, Vitamin D isn't a new drug.  Nevertheless, if it were, the FDA would say there is not yet enough data to approve it's use.

If you are considering extra Vitamin D, I strongly recommend you discuss it with your doctor first.  There are blood tests for Vitamin D that your doctor can order.  These are the exact words of Dr. Garland, who worked on this study:
“While there are a few conditions that influence vitamin D metabolism, for most people, 4,000 IU per day of vitamin D3 will be needed to achieve the effective levels,” Garland suggested. He advised interested patients to ask their healthcare provider to measure their serum calcidiol before increasing vitamin D3 intake. “This beneficial effect is present at these intakes only for vitamin D3,” Garland said. “Reliance should not be placed on different forms of vitamin D and mega doses should be avoided ..."
This research was funded by the US government via a grant to the Diabetes Research Institute (DRI).  It was published in Diabetologia, a first rate European diabetes medical journal.

I have blogged twice before about Vitamin D, here:

Other researchers are studying Vitamin D, and in particular Dr. Taback is organizing a large, intervention study.  If you are the patient type, you might want to wait until it is complete.  That is the first intervention study that I know of, and intervention studies are a much stronger form of evidence than population studies.  Unfortunately, Dr. Taback's work will take years to complete.

Personal note: I rarely blog on population based studies, like this one.  In general, I'm very nervous about their level of quality.  (I've seen some particularly bad studies in the area of nutrition, Vitamins, and related fields.)  I think one of the main problems with science reporting is that it is far too optimistic in reporting the results of population based research, much of which turns out to be wrong.  I'm also sensitive to the fact that they do not help get a drug approved.  A treatment supported by 10 or even more population based studies will not get approved by the FDA, unless intervention trials are done.  However, I am blogging about this study, because I think these researchers did a particularly good job of designing their study.  But it is still just one population based study.  Even for intervention studies, I don't consider one study alone to be definitive, and even less so for a population based study.  This is a step down a path, not the end of a journey.

Excess Vitamin D can accumulate in the body, and you can overdose (especially smaller children, if given adult doses).   Do not think "It's a Vitamin, so it's always safe" or "It's a Vitamin so everyone can take it" or "if taking X amount is good, then taking 10 times that much must be better!".  None of these things are true, and all of them can be dangerous.

For examples of study where giving vitamin D to people who already had type-1 did nothing:

No protective effect of calcitriol on beta-cell function in recent-onset type 1 diabetes: the IMDIAB XIII trial.
No effect of the 1alpha,25-dihydroxyvitamin D3 on beta-cell residual function and insulin requirement in adults with new-onset type 1 diabetes.

Joshua Levy
All the views expressed here are those of Joshua Levy, and nothing here is official JDRF or JDCA news, views, policies or opinions. My blog contains a more complete non-conflict of interest statement. 
Clinical Trials Blog:
Cured in Mice Blog: