April 17, 2012
Eric Topol, MD, has written a book about the convergence of the digital revolution and medicine. It is full of fascinating information and prognostication, but I wish he had given it a better title. He called it The Creative Destruction of Medicine: How the Digital Revolution Will Create Better Health Care. Medicine will not and cannot be “destroyed.” It will be improved and transformed, perhaps, but not destroyed. And any new developments will have to be evaluated for safety and effectiveness by the good old time-tested methods of science.
The future world of medicine is really exciting: science fiction is becoming real. As I read Topol’s book I serendipitously found it paraphrased by a character in another book I was reading, Chop Shop, by Tim Downs.
I see a world where no one ever dies from an adverse drug reaction; where physicians have an entire range of medicines to choose from to treat a deadly disease; where medications target tumors like smart bombs and leave surrounding tissues unharmed; where genetic susceptibility to disease can be determined in childhood, and possibly even prevented.
(If you haven’t yet discovered Downs’ hilarious “Bug Man” detective series about a crazy forensic entomologist, you have a treat in store.)
But back to non-fiction. Our world is changing almost too rapidly to comprehend: the Internet reaches everywhere, and there are far more mobile phones in the world today than toilets. We have hardly begun to tap the current potential of new technologies, and unimagined further developments await us. Topol is a qualified guide to this new world: he is a respected cardiologist and geneticist who ha s been on the forefront of wireless medicine and who was a major whistleblower in the Vioxx fiasco. He knows whereof he speaks, and he writes lucidly and accessibly.
Monitoring. Sensors now available or in development can do amazing things. They can monitor chronic diseases and provide remote electronic early warning of medical crises. Biosensors in cars could prevent accidents by detecting impending seizures, heart attacks, diabetic reactions, etc. Monitoring is a good thing, but you can have too much of a good thing, and some of what Topol advocates smacks of overkill. One of his patients sends him e-mails with 3 or 4 daily measurements of everything from blood pressure to oxygen saturation. Topol monitors his own sleep witha Zeo device. This is “nice to know” information but its impact on health outcomes is not so clear.
Lab on a Chip. A biosensor can be incorporated into a cell phone’s SIM card to do everything from detecting malaria to analyzing electrolytes. A phone camera can take a picture of a skin lesion and use a sophisticated algorithm to determine if it is melanoma. Digitizing breath might detect lung cancer. A high-tech “tattoo” worn on the skin can be read by your cell phone to measure your blood sugar.
The Office Visit of the Future. Virtual office visits may replace face-to-face encounters. Tools like video chats, telemedicine, and e-mail are already available.
Electronic Health Records as a Research Platform. Electronic databases have enormous potential: they can be useful in drug development, post-marketing surveillance, gathering statistics about disease, and monitoring adequacy of treatment. Just one example: if every patient on a drug were entered into a single database along with comparable patients not on the drug, even a rare adverse effect could be detected, and less rare effects could be spotted earlier. If we could include DNA sequencing in that database, we might learn which genotypes were susceptible to a certain side effect and could avoid prescribing the drug to such patients. We have the capability to do this today, although implementing it would be far from easy.
The Privacy Issue. Hacking into databases is a danger to patient privacy today, and the danger will grow in proportion to the amount and value of the data. No one has yet built a truly hack-proof system, and it’s unclear that if it will ever be truly possible. As with medicine itself, the benefits must outweigh the risks. But consider what it would be worth to the large insurance companies and employers of the future, if they could discriminate based on genetic profiles and other private medical information.
Genomics. Our ability to sequence patients’ genomes opens up whole new worlds. Topol characterizes genome analysis as “hypothesis-free” research, but I don’t think that’s quite accurate. We can screen lots of data looking for the unexpected, but we are still working with hypotheses about how that screening can produce results. Topol is enthusiastic about currently available direct-to-consumer genetic testing; I’m not so sanguine. He says if his genome showed a high risk of blood clots he would be more inclined to get up and walk around on long flights. Maybe. Does genomic testing really change behavior? At least one study showed it didn’t, but Topol was impressed that these patients expressed an increase in the intent to undergo screening tests like colonoscopy. I think supervision and interpretation of these tests by doctors is a reasonable precaution; Topol thinks it would constitute unfair interference with health freedom.
Genetic analyses can assist in drug development by teasing out who benefits and who gets rare side effects. With knowledge of gene specific effects, some rejected drugs might have been approved for a subset of patients.
Topol recognizes that it will not be a simple matter of finding a gene for every disease. Genetics is far more complicated: many conditions are multifactorial, genes interact with each other, and environmental factors affect gene expression. David Gorski’s recent article explains the complexity of genetic factors in cancer. In one study he cites, multiple biopsy samples revealed different genetic profiles in different parts of the primary tumor and metastases. Personalized treatment based on genetic analysis of a single biopsy would fail.
The Nicholas Volker Case. Topol describes this as the first instance of the life-saving power of genomic medicine. A child with a unique bowel disorder was found to have a mutation and was treated with an umbilical cord blood stem cell transfusion. His recovery was attributed to DNA sequencing, but doctors had already been contemplating this stem cell treatment before the mutation was detected. The outcome might have been the same without DNA testing.
Treating the Individual. CAM providers claim to tailor their treatments to the individual, but they are mainly making things up or relying on pseudoscience. In the medicine of the future, we will have truly individualized treatment based on scientific reality.
Doctor Bashing Topol criticizes current medical practice for relying too heavily on randomized trials and using population-based rather than individualized treatments. He speaks of “Resistance from the priesthood of medicine.” He says “Of all the professions represented on the planet, perhaps none is more resistant to change than physicians.”
I think that’s demonstrably false. Medical practice is constantly changing and evolving in response to new information and new technologies. It is true that it takes an average of 17 years from medical discovery to daily clinical practice and that this can be accelerated. We can do better. But so far there are very few instances where we have the knowledge to tailor prescriptions to a patient’s genome, and treatments based on studies of large groups are surely better than guesswork.
He quotes a review in The Economist that concluded “Virtually all alternative medicine is bunk” and characterized the CAM industry as a placebo-delivery service. I don’t disagree with that. But Topol is careless about specific CAM facts. Some of his errors:
- “To date, the only randomized, rigorously performed trial of a supplement was of glucosamine’s effect on knee arthritis.”
- Glucosamine is “quite effective.”
- His description of Simon Singh’s book and legal troubles is inaccurate on several counts.
- He claims that acupuncture, hypnosis and guided imagery lead to better outcomes, and that this is supported by good studies.
I noticed this amusing typo: “Chronic disease…[affects] more than 140 individuals in the United States.” Right, a lot more!
The future of medicine holds great promise. I don’t mean to be a wet blanket, but the challenge will be to temper our enthusiasm with good judgment. We can’t assume unalloyed benefit from every technological advance. Just because we can do something like constantly telemonitoring everything from our serum potassium level to our blood pressure doesn’t mean we should do it, or that it would be a good use of limited health care funds. We don’t want to create a world of cyberchondriacs. Data overload is a problem. Privacy is a major concern. Forgoing face-to-face human interaction may have significant downsides. What’s called for is what scientific medicine has always called for: cautious enthusiasm with scientific testing. Not “the destruction of medicine,” but the natural continuation of it.
This article was originally published in the Science-Based Medicine Blog.