June, 2011 marked the 10th anniversary of the great ‘Mission Accomplished’ moment of the Human Genome Project, when President Clinton, with no regard whatsoever for his own personal safety, stood directly in between Francis Collins and Craig Venter to announce the completion of “the most important, most wondrous map ever produced by humankind.” In November, in honor of the occasion, the American Museum of Natural History, in partnership with the Council for Responsible Genetics, hosted a panel discussion entitled “The Human Genome and Human Health – Will the Promise Be Fulfilled?” This was an opportunity for four very smart people to recap the discussion of everything that hasn’t happened as predicted in the last 10 years (Oh, the missing heredity! Oh, the shortcomings of personal genomics!) and why, in retrospect, this was all entirely predictable, as things generally are in retrospect. And then the two scientists on the panel predicted that we are on the cusp of great things and the two social scientists on the panel warned that great things come at a steep price and we all agreed, and why not? After all, we almost always are, and they almost always do.
The thing about making predictions is that it is hard to get it wrong if you go with generalities (it’s always something!) and even harder to get it right, if you are going after specifics. Thousands of people will have a heart attack this year. The guy sitting in front of you with the ten pounds of jelly donut hanging over his belt buckle? Hard to say. So we in the prognostication business cling to certain reliable, gospel truths. Technology will get faster and cheaper every year! Understanding pathophysiology will lead to cures! You will meet a tall, dark, handsome stranger! No – sorry, that one isn’t us. New studies will illuminate the relationship between genotype and phenotype! Hallelujah.
Here’s another one: personalized medicine will save us money! Can I get an amen? It’s something we hear all the time, in medical journals and newspapers and political speeches. “The savings from personalized medicine,” said a man in the audience at the panel discussion, nodding his head with conviction, “how soon we will see that?”
“Well,” said Dr. Robert Green, renowned neurogeneticist from Beantown (Hah-vard, of course) “I am not convinced that it will save us money. I think it might cost us money.” You could almost hear the band stop playing.
Is he right? The Personalized Medicine Coalition cites savings as one of the intrinsic advantages: “The cost of health care in the United States is on an unsustainable upward climb. Incorporating personalized medicine into the fabric of the health care system can help resolve many embedded inefficiencies, such as trial-and-error dosing, hospitalizations due to adverse drug reactions, late diagnoses, and reactive treatments.” (The Case For Personalized Medcicine, 3rd Edition.)
But think about it. Someone comes into your office carrying their personal genomic printout from 23andMe or Navigenics or whoever comes next. They have an increased risk of Condition X. What do you suggest? Step 1: increased screening and testing. Well maybe the testing modalities are not that good. Too bad. You suggest them anyway, because it is downright cruel to tell a person they have an increased risk of the dreaded Condition X and that THERE IS NOTHING TO DO ABOUT IT. Why do we send people who are BRCA 1 or 2 positive for bi-annual screening of their ovaries? Because it is a great screening test? Noooo. Because it is all we have to offer? Bingo.
And remember, that printout is going to contain multiple increased risks. So, step 2: return to step 1, and repeat.
Now, conversely, someone comes into your office with a paper saying that they have a decreased risk of Condition Y. Do you tell them to stop doing screening? Skip their annual physical? Start smoking cigarettes? Noooo. Because you know perfectly well that SOMEONE with exactly this genotype is going to get Condition Y, and you don’t want to be responsible if it turns out to be THIS GUY (see Paragraph 2 on the challenges of prognostication).
We are forgetting the medical equivalent of Moore’s Law: that visits to the doctor result inevitably in EXPONENTIALLY MORE VISITS TO THE DOCTOR. Call it Dr. Moore’s Law: Medical Care Generates Additional Medical Care at a Rate that is Exponential.
Now, please, don’t get me wrong. I realize that, at times, personalized medicine is going to save us money. Pharmacogenetics improving the use of medication will save time, money, and lives. Preventing certain forms of chronic disease like diabetes, if we find a way to intervene for those most at risk, will save a fortune. But right now, the savings are much more speculative than the costs. A reflexive adherence to the dogma that personalized medicine saves money creates a hype that can only lead to disappointment. Making medicine better is a huge goal: making medicine solvent is too much to ask of any fledgling field. Feeding the hype is tempting, because it generates the sort of excitement that brings in attention and funding. But ultimately, propagating a dogma that generates unrealistic expectations will snatch defeat from the jaws of victory, as our real-life success stories are weighed against the myths of our own making.
*Gratuitous Kant reference. Philosophy students: please enjoy.
6 responses to “On The Old Saw: That Personalized Medicine Will Cost Money In Theory But Will Be A Money-Saver In Practice*”
Brilliant! Everytime I think about the amount of people getting their SNPs done and deciding that they have a predisposition to increased amounts of earwax or whatever and need to call Genetics- well, I need to reach for the wine. Loved this concise and clever article about limitations to personalised medicine.
I’m not sure that cost saving was ever really a major promise of personalised medicine – its easy to take a single sentence out of a large document. It could save money under certain circumstances but the main plank was “making medicine better”. I don’t suppose MRI’s save money, perhaps they make medicine better. On the shortcomings of personailsed medicine there have been several publications over the last couple of years which indicate that genetic variants (present at birth of course) are just as predictive as family history and trad risk factors (and when the latter appear – the patient is already unwell) – see http://bit.ly/sYzFvt . This is not a bad situation only 4 years after GWAS studies began to be published – even the PHG foundation said “it shows that genetic data can have genuine predictive value” http://bit.ly/tVUQyS
Technology will, and does, get faster and cheaper and better every year, but not the use and the application of it
Well said and very interesting! I’m a graduate student, and we just took a final on this topic. I wasn’t looking at personalized medicine from this angle, so thanks for keeping my head straight!
Well said, as always, Laura. However, as Genetic Counselors we always deal with the uncertainty, or “gray area” that follows any new genetic test and/or technology. We will see patients and we will use our counseling skills to help them understand the risks, benefits, and limitations of the information–because that is what we do best. We are not just test interpreters but need to address the issues behind the questions. We will follow Personalized Medicine’s evolution because someday it will pay off. Perhaps not in dollars but in better health for us all.
Why do we not look at personalized medicine a different way. For those folks who already have a disease and the one size fits all approach is doing nothing to help them. Isn’t that the promise of personalized medicine? We would be able to identify the genes that are corrupted and devise a treatment plan based on that information. I am a prostate cancer survivor ( so far) and would love to be told that base on your profile, we have a treatment plan that will cure you of prostate cancer 100%. But that is not how it works in todays world of medicine.
Pingback: how can a 9 year old make money