Understanding Relative Disease Risk Using 23andme

 Click to Enlarge

There is a difference between "absolute disease risk" and "relative disease risk."  If you notice in the table above, my results showed that my risk of developing coronary heart disease is 39.6% vs an average risk of 24.4%. This puts my risk at 1.43x the average member of the population.  What's of interest to me, however, is my risk of developing a venous thromboembolism.  Despite my overall risk of 33.5% (lower than the risk of heart disease) I am 3.44x more likely to develop a venous thrombolism than the 10% risk for the overall population.  Thus, in comparison to the average, my relative risk is higher.

In an earlier post, I explained that I have the Factor V Leiden mutation, so this was not a surprise to me.  What I loved about my results is that their testing protocol picked it up!  I could not have been tested for this mutation for $99 alone much less received all of the additional information supplied to me.

Click to Enlarge

I love the simplicity of the graphics.  The Odds Calculator clearly shows what 33.5 members of a population of 100 would look like compared to an average of 9.7.  YOU CAN CLEARLY VISUALIZE YOUR RISK!  If  I didn't already understand what Factor V is, they provide me with a clear explanation.  In addition, they cite the articles that back up their conclusions.  You can see what years the research was published and determine for yourself how current it is.

Click to Enlarge

They provide a nice color graphic that shows the role that scientists believe is attributable to genetics vs. environmental and other factors.

You can see, that in this case, genetics plays a significant role.  As if all of this isn't enough information, you can click on tabs at the top of the page for even more information.

Imagine that all of this is provided on each of the diseases tested by 23andme.  I'm impressed.

Note:  All charts are part of my individual report from 23andme.

Looking at Disease Risk with 23andme

In recent posts, I have discussed the kinds of results you can get by submitting a sample to 23andme.  In this post, I'm going to discuss the one aspect of 23andme that seems to be more troubling than that used to identify ones genealogical heritage.  There is also a health component that can give you some idea of the role that your particular genes play in predicting disease.  Anyone who does this should be aware of a couple of a few facts:

1. This is a new and evolving science.  In many cases genes are clear markers for a given disease, while in others they are not.
2. Often it's not one gene that will predict your chances of eventually developing a disease but a cluster of genes.
3. Genes ARE NOT the total story when it comes to disease.  In many cases, environmental factors can play as large a role if not more of a role in whether or not you will eventually get a disease.
4. Often there are things YOU CAN DO to prevent disease.  
5. For the really sensitive diseases, like Alzheimer's, 23andme gives you the choice of looking at the results or not.

One of the things I like about their service is that you can look at the research on which their predictions were based.  You can see how recent it is, how large the sample size was and their level of confidence on what they are telling you.  Did I have any surprises?  Yes and no.  Here is a graphic for some of the diseases that my genes indicate may put me at an increased risk.

Click to enlarge

According to their results, I am most at risk for coronary heart disease.  This surprised me only because we think of ourselves as a "cancer" family.  My Dad and all four of his siblings developed cancer -- often at a relatively young age.  My mother and her sister did, too.  When I investigated the science behind their prediction, however, I really should have been paying more attention to this one.  I am treated for high blood pressure, common in my family, and I've been diagnosed with supraventricular tachycardia.  Interestingly, both of these conditions are easily treated -- so I can probably avoid what looks like a high risk.

I was NOT at all surprised by the venous thromboembolism prediction.   As I mentioned in an earlier post, I know I have the Factor V Leiden mutation (which they picked up in their testing) that puts me at a higher risk.  The cool thing about knowing, however, is that I doubt I'll ever have this problem.  Any time I've had surgery that puts me at a high risk (knee replacements), I've been treated with blood thinners.  Problem solved.  I also know to avoid certain behaviors (like sitting in one position too long, especially on a long flight).  Movement is essential.

I was really interested in the macular degeneration prediction as this, also, is prevalent in my family.  My opthamologist is aware of my concern and is on the alert for any signs.  So far so good.

I did open my Alzheimer's risk, and even though I have a higher chance of developing this than others in the population, it still is only a one in seven chance.  I'll go with those odds.  What surprised me was how low my genetic predisposition was to ulcerative colitis, since this is a disease I did develop at age 49 and has had a huge impact on my life -- proving once again that genes are only part of the story.

In the next post I'll show you some of the information that is available for each of the diseases.  I'd love to hear your comments on whether or not you would feel comfortable getting results such as these.

Health Results from 23andme

When I first discussed with my daughter, Liz, that I wanted to submit a sample to 23andme, we discussed her concerns.  First of all, she is a Genetic Counselor.  Secondly, there is a bit of a debate within the Genetic Counselor community about the wisdom of "direct to consumer" DNA analysis.  Will the typical consumer be able to understand the results (much less be able to deal with potentially negative results)?  After all, that is what Genetic Counselors do.  There are genes that have been identified that are associated with breast cancer.  Worse yet, what if an analysis shows that you have a gene that will definitely lead to an early death -- like Hungtington's.  Do you really want to know that your genes predict that you are likely to develop Alzheimer's?  What about Parkinson's?  All of these are possibilities and something you should consider.

I am 62 years old.  I already have experienced some significant health issues.  Would the results support what I already know?  In my family, there is a significant history of cancer.  We all "assume" that that is what will eventually kill us.  Quite a few relatives have developed macular degeneration.  And the bigger question -- how strong is the science behind 23andme's health predictions?

I told my daughter that if she didn't want to know the results, I wouldn't share them.  She said she didn't think I could keep my mouth shut.  Even if I could, she wasn't sure that she could stand "not knowing" if I knew.  I decided I was going to do it anyway.

Well the results came.  I was pleasantly surprised.  For one thing, they let you know whether or not you are a "carrier" for genes that are responsible for things like cystic fibrosis, sickle cell, etc.  I found out that I was not a carrier for any one of them!  What a nice thing to know.  You can click on each disease for a description of what it is.  The yellow stars indicate the level of confidence in the research that backs up the prediction.  Not only that, but you can be referred to the specific studies that identified the gene.  Here are some of my results.

Needless to say I was consoled by these results.  Of course if my results showed that I was a carrier, I may have felt differently.  In my next post, I will discuss a few of their predictions for disease and how closely the results correlated with what I know about my health status.

Anticipating DNA Results

So you send your sample in for testing and then you wonder what kind of results you might get.  Everyone knows that we inherit half of our DNA from our mother and half from our father.  Knowing this, it is logical that 25% of our DNA would come from each of our four grandparents.  The waters get a little muddier with siblings, aunts, uncles and cousins.  I found this great graphic on Wikipedia that helps make it clear. (All graphics can be enlarged by clicking on them).



Graphic Credit: Dimario, Wikipedia Commons

The Human Genome Project determined that humans share 99.9% of their DNA in common, although this exact percentage has recently been challenged.  This means that all of the diversity among humans comprises a relatively small part of our entire genome.  Since I'm not aware that any of my close relatives have been tested (yet) and I know most of my close relations, I really didn't expect to find any "cousins" that would fall within the second cousin category or closer -- and I didn't.  Here is what I did discover.

Unlike FamilyTreeDNA, 23andme scans all of your genome to look for patterns that may be used to predict your ancestry.  Mine turned out to be 100% European.  No surprise there.  Here are my results.

I really didn't understand this graphic until I compared it to a graphic for someone that had more of a mixed inheritance pattern.  Here is a sample for an African-American man.

I find it interesting that, as is the case with so many African-Americans, a lot of this man's ancestry is European in origin.  You can go to 23andme, create a free account and explore other sample groups under "Ancestry Painting."

When 23andme compared me to others in their database, the closest relationship they could identify for me was a potential 4th cousin. You can see from the table posted above that I would not anticipate sharing much of my DNA with this person.  23andme computed that we shared 0.48% of our DNA -- big whoop!!! (Note:  I've since found out that the 36 cM we share is a big deal.  Who knew).  Here is what we share in common.

You are reading it correctly.  We share one little section on chromosome 20.  I'll bet I share that much with my husband.  I guess I'm going to have to have him submit a sample.  So come on, cousins, submit a sample!  I want to have some fun.

In my next post, I'll discuss what I found out from a medical point of view.