DNA Testing Is a Slippery Slope

by jeeg 14. April 2015 23:01


Money shark Mark Cuban set off a firestorm on Twitter after recommending blood tests for “everything available.” Genetic testing has value, but it’s not for commercial use.

Let’s say you have a whole lot of money.

Let’s say you have so much money that, even after you’ve purchased a basketball team and a chain of movie theaters, you still have enough cash left over to invest in a maverick approach to your own health care. Should you sink an unspecified sum into ordering every test you can find in the off chance that sooner or later you’re going to detect something valuable?

Mark Cuban says “yes.”

In a series of tweets posted at the beginning of this month, the media baron and basketball franchisee admonished his followers that, if they were sufficiently affluent, they should order blood tests for “everything available” on a quarterly basis in order to establish a baseline for their own health. In the event of some deviation from the any given person’s aggregate normal values, it might detect some abnormality earlier than it would otherwise be found when it started creating symptoms.

Looking at Cuban’s page now, it seems the tweets in question from April 1 have gone missing. But whether or not he has come to reconsider his advice, it’s a question that seems worth addressing.

Periodically, a parent will bring a patient in and request that I test his or her blood for anything wrong. Sometimes this is in the setting of a vague symptom, usually fatigue or malaise. Sometimes it’s more in keeping with Cuban’s recommendation to just check every once in a while to make sure everything looks okay. If the patient hasn’t had a screening test for high cholesterol or anemia in a while, I will often send those off. But usually I try to explain why taking a shotgun approach to lab tests isn’t likely to do any good.

First of all, there is a lot of “everything available” to be ordered. While some tests have merit as tools to screen for illness or assess risk of developing diseases later on (examples would include tests of liver function or fasting levels of sugar or blood lipids), others are only used as part of a more in-depth workup of a problem once it’s detected.

Can I say with absolute certainty that a small shift in a patient’s ferritin level, normally ordered when evaluating for iron-deficiency anemia, might not be an early sign of otherwise undiscovered disease? No. Inflammation can cause ferritin levels (along with many other substances in the body) to rise, so it’s plausible that some random patient’s illness might show up that way. But do I think checking levels every few months with confer any benefit for the vast majority of patients, to say nothing of the reams and reams of additional testing theoretically available? No, I do not.

When considering whether or not to start using any given intervention for patient health, a calculation medical providers need to keep in mind is called the number needed to treat. That’s the number of people who need to be given some kind of therapy to prevent a given bad outcome. An important part of that calculation is the baseline level of risk for that bad outcome.

It may make sense for patients with a history of heart disease to take an aspirin every day to prevent a fatal heart attack. For those patients, the number of heart attacks prevented is justification for taking the medication every day; even if many would never go on to have a heart attack anyway. However, the lower the risk factors for heart disease in the population you’re treating, the more patients have to take aspirin to prevent that one heart attack. For many healthy people, even in middle age, the risks may outweigh the benefits.

Similar kinds of analysis go into whether or not it makes sense to start using a test to screen people for an illness. How many times does the test need to be run before you detect the abnormality, and how bad is the outcome you’re trying to prevent? Annual CT scans of the chest in older patients with a long history of smoking might help detect lung cancer in earlier, more treatable stages. But you’d have to scan a lot more non-smoking young people to find that one rare cancer. For the overwhelming majority of them, the risk from the radiation far outweighs the benefit.

Considerations like these play a major part in deciding whether or not to recommend or perform screening tests. Angelina Jolie’s strong family history of breast cancer made testing her for the genetic mutation that increased her own risk of the same cancer worthwhile, and gave her the information she needed to take steps to lower that risk as much as she could. But it isn’t currently recommended that people without similar histories in their families get the same screening tests, and certainly not the same surgeries.

For people without elevated risk for any given condition, piling tests upon tests for “everything” is only going to produce more information without meaning for their health, amounting to nothing but noise. Worse, every test has the risk of generating a falsely positive result that then has to get investigated to no benefit. Even for well-established screening examinations, the risks of these false positives often lead medical providers to reevaluate their use and revise the recommendations surrounding them.

What Cuban recommended has the appearance of value, but for the overwhelming majority of people (even wealthy ones who can afford it) it won’t be worth the cost. As a panelist on Shark Tank he’s as ruthless as they come when dispensing with contestants whose product isn’t worth the money. But his business acumen exceeds his medical savvy, and I can’t recommend you follow his investment advice when it comes to your own health.

 

Russell Saunders, Daily Beast

 

Comments are closed
Log in