Insights (Essays)

Insights (Essays) January 2012
A review of “The Creative Destruction of Medicine: How the Digital Revolution will Create Better Health Care” (Basic Books, 2012) by Eric J. Topol

N of 1

Reviewed by Neil Seeman

In the future of medicine, the personal is less political than it is inevitable. In The Creative Destruction of Medicine, American cardiologist and geneticist Eric Topol invites us to consider the sunny side of the convergence of medicine and the digital revolution, drawing on a confluence of trends that should lead us to the era of personalized medicine.

Topol comes to the well-trodden issue of personalized medicine with some serious street cred. He was one of the first to notice that rofexobid (Vioxx) caused heart problems for some patients – it was eventually withdrawn from the market.  He was named Doctor of the Decade by the Institute for Scientific Information for being one of the top 10 most cited medical researchers.

He draws parallels between the large numbers of available digital information gigabytes and the large number of SNPs into which the newly sequenced human genome can be broken. He argues that DNA sequencing of individual genomes will soon become routine for clinical, research, forensic, and personal purposes. This will allow prevention and treatment of disease that is specific to the individual – the right drug at the right dose for the right person – as a consequence, there may one day be no more adverse drug reactions because the genome can be scanned beforehand to find out whether a particular drug will disagree with a specific individual.

Perhaps Topol’s greatest contribution to the debate is the recognition that this revolution will not be televised in medical schools. Because doctors are reluctant to move towards personalized medicine, consumers are charting the way. In the first section of the book, the author reviews the overall digital landscape – how the digital world has evolved and changed our lives outside of medicine; how our information in medicine is grossly deficient and population (not individual)-based; and how consumers, despite progress toward the convergence of masses of health information, are too often poorly informed.
In the second section, he discusses the four areas of digital medicine – wireless sensors, genomics, imaging, and health information – and how these can converge for human benefit. Then he previews the impact that digitizing humans will have on doctors and hospitals, on the life sciences industry and regulatory agencies, and, ultimately, on the individual.

As with any revolution, there are important downsides to consider. Here the concerns include the reduction of direct human contact and healing touch that may accompany increasing reliance on remote patient monitoring and the disappearance of hospitalizations or even in-person office visits. It will be increasingly tempting for physicians to treat the virtual human being – the scan, the DNA results, the biosensor data – instead of the real patient.

There is legitimate worry about adoption of new technologies before they have been adequately vetted and validated, or proven to be cost-effective and, ideally, cost saving. But there are also clear examples where looking at everyone as the same, buttressed by shoddy population-level data, permits the routine delivery of drugs to everyone when they are only really suitable for some. Statins (like Lipitor) are an example. Instead of identifying the one person or two people out of every 100 who would truly benefit, the whole population is treated. Topol points to Plavix (for clot prevention). It turns out that approximately 30 percent of people cannot normally metabolize Plavix to convert it to its active form. These people may need a higher dose for effectiveness. Some won’t respond even at the higher dose. And some, because of individual genetics, only need a half-dose (which isn’t available). The point is that this can be determined beforehand and, according to Topol, should be.

Or take mammography screening: Mass screening disregards individual variability and promotes unnecessary medical testing and procedures. For every 1,000 women who undergo mammography between ages forty and forty-nine, 98 (about 10 percent) will have a false-positive mammogram, 60–200 women (the range indicates variability in multiple studies) will undergo an unnecessary biopsy, and 84 per 1,000 women screened will have to undergo additional imaging, typically involving magnetic resonance or ultrasound. Only one per 2,000 women screened might avoid death from breast cancer with screening. In later decades of life, the numbers don’t change much. Even in women aged 70 to 79, false-positive mammograms occur in 69 out of every 1,000, and additional imaging is required in 64 per 1000. Proven over-diagnosis, representing unnecessary surgery, chemotherapy, or radiation, rises from 1–5 per 1,000 at ages 40 to 49, to 1–7 per 1,000 women aged fifty to fifty-nine.

Other policy dynamics in the current ecosystem of depersonalized medicine, he argues, make matters worse. The problems with Direct-to-consumer, or DTC, advertising of prescription drugs has been legal in the United States since 1997; DTC advertising is the epitome of population medicine. Millions of TV viewers are bombarded with infomercials, and some ask their doctors for a drug they don’t really need. The major prescription excess, as with many aspects of population medicine, adds up to a profound waste of resources.

The organic response to this one-size-fits-most mentality, argues Topol, will come from patient empowerment, which will drive personalized medicine forward at an unstoppable clip.  The World Wide Web is at the core of this movement. For example, targets more than 500 different conditions and has spearheaded some fascinating clinical research, such as response to medication by patients with migraine headaches. It was launched in 2008 to help people “anonymously track and compare health data, to better understand their bodies, make more informed treatment decisions, and contribute data to research.”  Diabetic Connect has close to 300,000 registered members, with monthly traffic of over one million unique visitors.  Many other health-networking sites are being created and are rapidly gaining membership, like, with different disease-specific targets.

Other technological examples abound throughout the book: sensors, for example, that monitor heart rhythm via the steering wheel. Topol is a geneticist, so he is at his best when he describes mining the genome “hypothesis free” – i.e., studying the genomes of a large number of people with a certain disease without any preconceived ideas as to where the troubled genes may lie –just going on a “fishing expedition” and turning up genes you never thought had anything to do with the disease. He notes that most of the genes that have been discovered in this way have never been previously theorized to have anything to do with their respective diseases, such as the CFH gene in macular degeneration, the FTO gene for obesity, and the TCF7L2 gene for diabetes. Like others who sing of the wonders of personalized medicine, he is in awe of the possibilities of pharmacogenomics – and I confess to subdued enthusiasm in this regard (a bias born of the fact that I’m the son of brain scientists who have impressed upon me the limited power of genes in predicting drug response).

Surprisingly, Topol hedges on whether the personal health record or the electronic health record will be the seed from which personalized medicine flourishes. I fall into the PHR camp, largely because the EHR camp (in Canada and the US) looks increasingly like the gridlock in Washington: self-important bureaucrats and lawyers bickering over procurement rules instead of what matters, such as investing smartly in primary care and home care (which may, in the Topol vision, become more virtual than face-to-face).

Where does this leave us? Topol emerges as an elegant champion of the “N of 1” school of medicine. N-of-1 or single subject clinical trials see an individual patient as the sole relevant unit of observation in a study investigating the efficacy of an intervention. Thanks to the science of individuality, we are learning about the uniqueness of each person and will be much more likely to unravel the root cause of the individual’s condition if we thoroughly study human beings one by one. What is both marvelous and paradoxical, he notes, is that our New Age capacity to understand an individual relies on network science and the trillions of data points on the Web — the more data that can be processed, the sharper the definition of a particular individual.

Topol makes the case that the masses of macro-data at our fingertips (literally), will unleash micro-level diagnostic and curative solutions never before imagined or hypothesized. It’s a remarkably bold vision that many experienced physicians will call naïve since it defies conventional wisdom – which is precisely why I think he’s on to something big.

About the Author(s)

Neil Seeman is CEO of the Health Strategy Innovation Cell, Senior Resident in health system innovation at Massey College in the University of Toronto, and Founder and CEO of The RIWI Corporation. He is co-author of four health policy books, most recently, XXL: Obesity and the Limits of Shame (Univ. of Toronto, 2011).


Publisher’s comment: for more on Eric Topol see:


Be the first to comment on this!

Note: Please enter a display name. Your email address will not be publically displayed