Online DNA Test: A Consumer's Experience

When our Professor Greg Fowler, PhD brought up the idea of having a student take a DNA test so that our class could critique the process, I sat back thinking this wasn't such a great idea. Not only would the student have to deal with privacy in the sense of negotiating privacy statements online, but they would also be under some pressure to disclose test results in class and at a public forum.

But, the test would be free thanks to a connection Professor Fowler had with DNAdirect.com. He suggested a hemochromatosis test and until the moment he brought it up, I had never heard of hemochromatosis. As we looked up more information about this potentially devastating disease, I realized that on the surface of it, I was a very likely candidate to have the genetic markers for hereditary hemochromatosis. Consider the following fact set presented at the DNAdirect.com website:

• One in ten people of northern European descent have the markers for hereditary hemochromatosis
• One in four Celtic descendents have the genetic markers for hereditary hemochromatosis
• Hemochromatosis theoretically is a causal agent of about 10% of all type II Diabetes cases
• Hemochromatosis is a disease that is under diagnosed due to a wide array of symptoms.
• Hemochromatosis is the most common genetic defect in America and 1 to 1.5 million people currently suffer from it.

My mother is of mostly Irish heritage and my father is mostly Norwegian. My grandmother, father, sister and brother as well as I are all type II diabetics and none of us were morbidly obese when we were diagnosed suggesting a hereditary element to the disease. I realized that each one of the conditions above was relevant to me and it led me to think that if I was indeed a carrier of the genetic defect, I would be doing my family a great service in discovering a causal agent for both our diabetes, and possibly, for the untimely deaths of my grandfather and great-grandfather on my father's side.

Also, as a type II diabetic I take a medication that suppresses liver function and I imagined that hemochromatosis would be an especially devastating disease considering that it is associated with catastrophic organ failure if left untreated. I was swayed by these arguments and I decided to take the test.

DNAdirect.com was reasonably informative at their website and promised a full report on my results afterwards. After clicking around on the net on various sites, it was apparent that not all companies took privacy very seriously. DNAdirect.com to their credit, did provide a reasonably complete privacy statement, but what is lacking there and at virtually every site I looked at was what exactly happens to the information after it is provided to a laboratory. Some online companies say they destroy the sample, others do not even say what they do with it at all after the test. At DNAdirect.com, they promised to remove my information from the sample theoretically making it anonymous - at least in their system. The actual test was sent to Myriad Labs in New Mexico and is subject to their privacy laws.

While the idea of my DNA being used for future research doesn't particularly worry me, the idea that digital files that can identify me probably exist at a laboratory that also has my sample of DNA. DNA, no matter the level of security, is still an absolutely personal source of identification - does CSI ring a bell? I bring this up because it was one of the most anxiety causing lines of thinking regarding the taking of this test. But, it is also true that if I had a test done through my doctor, my HMO would probably have a similar arrangement with the lab doing the test. The difference then, by going direct was that the HIPAA privacy and sample usage protections only kick in if I use health insurance to get the test.

Still, I moved forward through the process until I was confronted with the final screen - the agreement with DNAdirect.com. Anybody who has installed software on their computer has seen the equivalent of this screen. It was a gray two by three inch window that contained an eight page document full of legalese, much of which closely resembled the privacy statement. However there was some new information that made me pause. When applying for health or life insurance or employment, some agencies may request information about genetic testing. Failing to provide relevant information upon request from employers, health or life insurance providers may be considered fraudulent or other violation of the law. (DNA Direct test agreement, 2006)

This was an aspect of privacy I hadn't thought of at all. If a healthcare provider, insurance company or employer request this information and I don't disclose it, I am committing fraud. One criticism that I have regarding the marketing of such tests is that if the information is legally available from me upon request, then the company should have to avoid wording in their advertising that suggests that the test is absolutely private. Still, even after reading this, I hit the "accept" button.

The next steps were to talk to a doctor and a genetic counselor before taking the test. My doctor was gracious enough to talk to me about this test. I had been sure about the test being indicated for me. She said that she never would have ordered the test since there was no history of liver disease in my family and I was not expressing any of the symptoms of the disease. Instead, if she were in any way suspicious of hemochromatosis, she would order a series of simple and inexpensive blood tests and look for elevated iron content.

When the test arrived, I opened the packages carefully and laid it out. I took out my testing checklist and cleared off the kitchen table. The envelope containing the sensitive swabs and the foam holders for them were opened last and laid out in a solemn orderly way. I remember hearing the kittens playing in the background and not thinking anything of it. Did I mention that my wife and I have two six month old kittens? The male kitten named Richard Parker jumped up on the kitchen table slid into my DNA test sending everything to the floor. Since the test did not identify me as half Burmese and half Tabby, I assume it was my DNA that was tested after all. But, I think this anecdote brings up an interesting point about home tests - the testing environment is far less controlled than a doctor's office and the lab must assume the sample was mine. Still, the test was easy to take. Just as the advertising for online DNA testing promised; anyone could do it, even Richard Parker and me.

The most difficult aspect of the test was figuring out what the possible outcomes really meant. This test screened for three genetic markers, but these markers are not created equal. In fact, there are an estimated 40 markers associated with the disease. According to NIH, a double expressing of the HFE 282y marker was associated with roughly 85% of all hereditary hemochromatosis cases. In order to have a double expression of the gene, one would have to inherit one copy from each parent for it to be on both copies of the gene. The odds of that happening for me were that my father of northern European descent had a one in ten chance of having the defect and my mother had perhaps a 25% (she is probably not 100% Irish) chance of having it producing a 2.5% chance for me to have a double expression of the defect. But wait, even if both my parents had the marker to give, any offspring would only have a 25% chance of getting a double gene expression making it a little more than a .8% chance of having what is by far the most common associated marker to the actual expression of symptoms of hemochromatosis. That is an 8 in one thousand chance of having the double gene expression associated with 85% of all hereditary hemochromatosis.

The actual expression of the disease for someone with the most likely gene expression occurs in 50% to 93% of those who have the double gene expression (Toland 2000). This means that .4% to .75% of the population who are considered as at risk for hemochromatosis as I was, ever develop symptoms.

Flying in the face of privacy concerns, I will now brazenly tell you one last bit of my medical history. I do not have any of the three markers, indicating that my chances of ever developing symptoms from hemochromatosis is roughly .3% to .5%, or the same as everyone else in the general population. The point I am trying to make here is that if tests are marketed directly to consumers, then we should find ways to make clear a reasonable prognosis of what the test is really going to mean. As a genetic counselor pointed out to me, these tests are only part of a much larger picture, and doctor's offices are more likely to have resources available to them that can interpret results in a meaningful way. If I were to make a single recommendation for the marketing of these medical screens, I would say that something akin to an FDA label should be available, giving in concise relatable terms several data fields. The following should be available to anybody considering the test:

• The genes being tested
• The likelihood a person from the general population has the marker
• The likelihood a person has the marker based on specific indicated factors
• The likelihood that a person with said marker will express actual symptoms if left untreated.
• The likelihood of a person without these markers would still get the disease.

I am sure that if I were confronted with the knowledge that I had a .2%-.5% chance of developing hemochromatosis if I were from the general population, but nearly a one percent chance given my ancestry, I might not have been so impressed with the idea that taking the test was worthwhile. Perhaps I should not be so smug, after all, we may have only proven that the handsome kitten named Richard Parker, is at not a high risk for developing the symptoms of Hemochromatosis.

Vern