by Mark Bowden
“I think he felt he wasn’t really being taken seriously,” says Sandborn. “So he came over and we looked, and we ended up finding some degree of inflammation that was pointing in the direction of Crohn’s disease, but he wasn’t really having many symptoms. So the questions then became: Is this some kind of early subclinical Crohn’s disease? Should we even go as far as treating it, or just wait?”
Larry’s impressive quest to fine-tune his body had led him to this: an early diagnosis of Crohn’s disease, an incurable condition. It isn’t fatal, but it has a long list of uncomfortable and sometimes painful symptoms that tend to flare up from time to time; they center on the GI tract, but may include eye inflammation, swollen joints, fever, fatigue, and others. Apart from that one episode of abdominal pain, Larry was still feeling fine. But the graphs showed, and his new doctor more or less confirmed, that he was sick.
And that part about its being incurable? Let’s just say that in Larry, Crohn’s disease has encountered a very dedicated adversary.
If past thinkers leaned heavily on the steam engine as an all-purpose analogy—e.g., contents under pressure will explode (think of Marx’s ideas on revolution or Freud’s about repressed desire)—today we prefer our metaphors to be electronic. We talk about neural “circuitry,” about “processing” information, or about how genes “encode” our physical essence. In this worldview, our bodies are computers, and DNA functions as our basic program, our “operating system.”
This is certainly how Larry, the computer scientist, talks about the human body. In this context, all of human history can be seen as a progression from a world that was data-poor to one that is data-rich. Starting with those early summers working in secrecy at Livermore, Larry has witnessed firsthand the exponential progress of computing power posited by Moore’s law, which states that processing power should double roughly every eighteen months. So when Larry talks about the potential for computers to help us understand our bodies, he isn’t talking about their showing us more isolated details about an unfathomably complex system; he’s talking about knowing everything.
“We are going to know—once you know each of your cells’ six billion genome bases, with all the imaging down to the micron level, and when you know every damn gene and every bacterium—at a certain point, there are no more data to know,” he says. “So certainly by 2030, there is not going to be that much more to learn…. I mean, you are going to get the wiring diagram, basically.” Once they are armed with the wiring diagram, Larry sees no reason why individuals cannot maintain their health the way modern car owners maintain their automobiles.
Larry actually concedes the point made by Dartmouth’s Welch—that presented with enough data, pretty much everyone is going to find something wrong with himself or herself. He just disputes that this would be a bad thing. “All of us do have something beginning to go wrong, but then, so do our automobiles,” Larry says. “In today’s world of automobile preventive maintenance, we don’t wait for our cars to break down and then go to the ‘car doctor.’ Every ten thousand or twenty thousand miles, we go in and get an exhaustive look at all the key variables since the last check. If they find something wrong with my car—which will be different from what they find about yours—then they take appropriate action and I go back to driving a ‘healthy’ car. Occasionally, something is discovered that indicates a bunch of cars need to be called in and get a certain item replaced. I can imagine that occasionally, as a new DNA segment is related to some disease, people with that DNA signature will be called in for ‘preventive maintenance.’”
If Larry is right, then our descendants may view early-twenty-first-century medical practices, which we consider a triumph of reason over superstition, in the same way we now view eighteenth- and nineteenth-century folk remedies. A particularly likely candidate for scorn in an age of “quantified” health care is our one-pill-fits-all approach to prescription drugs. In his book The Creative Destruction of Medicine, the physician-author Eric Topol cites such dosing as an example of medicine that is “population-based,” rather than “patient-centered.” He notes the widespread use of statins to lower LDL cholesterol, a factor in heart disease. Topol doesn’t deny the cholesterol-lowering effect of these drugs, but he argues that double-blind testing also shows that this effect benefits only a tiny fraction of those treated. One of the most effective statins, Crestor, has been found to reduce the incidence of stroke, heart attack, or death from 4 percent of patients in the placebo group to 2 percent of the group taking the statin. And yet these drugs are widely administered to patients considered at risk. Topol writes:
Instead of identifying the 1 person or 2 people out of every 100 who would benefit, the whole population with the criteria that were tested is deemed treatable…. What constitutes evidence-based medicine today is what is good for a large population, not for any particular individual.
Pharmaceutical companies don’t mind. And as long as the harmful side effects are within acceptable limits, the Food and Drug Administration doesn’t mind, either. Some patients will be helped. All of them will be buying the pills, and all will be subjected to follow-up tests, some of them uncomfortable and most of them unnecessary. What if there were a way, Topol asks, of knowing, before prescribing the drug, which 2 percent would be most likely to benefit from it? In an observation that Larry wholeheartedly endorses, Topol writes:
Fortunately, our ability to get just that information is rapidly emerging, [and we are] beginning an era characterized by the right drug, the right dose, and the right screen for the right patients, with the right doctor, at the right cost.
Getting there will mean essentially dismantling the health care industry as we know it. (Thus the creative destruction of Topol’s title.) Or, as Larry puts it: “A lot of enormously wealthy, established, powerful institutions in our society are going to be destroyed.” And why not? Over the past twenty years, computers have been toppling and rebuilding industries one by one, from retail sales (Walmart and Amazon) to banking (ATMs and online services) to finance (high-speed online investing) to entertainment (Web streaming, downloads, YouTube, etc.) to publishing (e-books and news aggregators). We’re just babes in this new digital era, and it will eventually upend almost every field of human endeavor.
Larry sees medicine as a stubborn holdout. Current efforts to reform the system—for instance, the Obama administration’s initiative to digitize all health records by 2014—are just toes in the water. Medicine has barely begun to take advantage of the millionfold increase in the amount of data available for the diagnosis and treatment of disease. Take the standard annual physical, with its weigh-in, blood-pressure check, and handful of numbers gleaned from select tests performed on a blood sample. To Larry, these data points give your doctor little more than a “cartoon” image of your body. Now imagine peering at the same image drawn from a galaxy of billions of data points. The cartoon becomes a high-definition, 3-D picture, with every system and organ in the body measured and mapped in real time.
Indeed, a very early prototype of this kind of high-definition image already exists at Calit2. It is, of course, of Larry.
Inside a “cave” fashioned from large HD screens (each with dual rear projectors) and linked to eighteen gaming PCs to create a graphics supercomputer, Larry and I step into a stunning image assembled from an MRI scan of his torso. The room, the size of a walk-in closet, is lined with giant screens, front, sides, and back. More screens angle from these walls toward a floor that is illuminated from above. Two curved, waist-high metal railings offer support, because viewers at the center of this visual world can easily lose their balance. A sensor strapped to your forehead tells the computer where you are looking, so as you turn your head it smoothly blends the images on the screens to create a seamless 360-degree alternative world. (This is clearly the future of video games and cinema.) I had to lean on the metal bars to remind myself I was not someplace else. Once we were in position, Jürgen P. Schulze, a Calit2 research scientist, punched up a di
splay of Larry’s own coiled, sixty-three-year-old entrails. I felt as if I could reach out and touch the wrinkled contours of his intestines and arteries.
Larry’s inner ten-year-old rejoices. “Look!” he says, lifting and opening his hands. “This is me!”
He points to the source of his health concerns, the precise six-inch stretch of his sigmoid colon that is visibly distorted and inflamed. This is Larry’s discovery, and his enemy.
I note that the display breaks new ground in the annals of self-disclosure: Larry is literally turning himself inside out for a journalist. He does worry a little about making public such intimate details, but this openness is part of how he believes medicine ought to be—and ultimately will be—practiced. The current consensus that medical records should be strictly private, subject to the scrutiny of only doctor and patient, will be yet another casualty if Larry’s health care vision comes to pass. “A different way to organize society is to say it is human-focused, human-centered, and patient-centered, and that there are no legal or financial repercussions from sharing data,” he says. “There is a huge societal benefit from sharing the data, getting data out from the firewalls, letting software look across millions of these things.”
The way the system works now, when a technician examines the MRI of a patient’s abdomen, in two dimensions, on a single screen, she compares and contrasts it with perhaps thousands or even tens of thousands of other images she has seen. She then writes a report to the physician explaining, on the basis of her memory and experience, what is normal or abnormal in what she sees.
But “software can go in, volumetrically, over, say, a million different abdomens,” says Larry, gesturing at the image of his own innards, “and come up with exquisite distribution functions of how things are arranged, what is abnormal or normal, on every little thing in there. In my case, what I have found is inflammation. Unaddressed, it may lead to structural damage and maybe eventually surgery, cutting that part out. So I am going to have another MRI in three months, and that will tell me whether the things I am doing have made it better, or if it is the same, or has gotten worse.”
It’s that sense of control that appeals to Larry as much as anything.
“The way we do things now,” he says, “the technician will examine it and write up a report, which goes to my doctor, and then he explains it all to me. So I am disembodied. Patients are completely severed from having any relationship with their body. You are helpless.”
Shedding that sense of disembodiment and helplessness is, in theory, one of the most attractive features of Larry Smarr’s quantified self. Individuals will understand their own bodies and take care of themselves; doctors will merely assist with the maintenance and fine-tuning. With that sense of personal ownership established, Larry believes, the average American won’t continue to drink five hundred cans of soda a year, or ingest some sixty pounds of high-fructose corn syrup. After all, educational campaigns about cigarettes have helped lower the proportion of smokers in America to below 20 percent. If we made such inroads into the obesity epidemic, Larry says, “we would have a national celebration.”
For his part, Larry is no longer disembodied. He has had key snippets of his DNA sequenced, and will have the whole thing completely sequenced by the end of this year. In just what he has seen so far, he has discovered telltale markers linked with late-onset Crohn’s disease. He has developed his own theory of the disease, based on his reading of the most recent medical literature and his growing perception of himself as a super organism. In a nutshell, he suspects that some of the essential bacteria that should line the walls of his intestine at the point where it is inflamed have been killed off, probably by some antibiotic regimen he underwent years ago. So he has begun charting, through stool samples, the bewilderingly complex microbial ecology of his intestines.
He showed me a detailed analysis of one such sample on his computer, drawing my attention to the word firmicute. “So, what the hell is a firmicute?” he asks rhetorically. “And in particular, it is in these two groups, Clostridium leptum and Clostridium coccoides. So I go back, and I go, ‘Clostri-Clostri-Clostri, that rings a bell. I had it in my last stool measurement.’” He pulls up an older chart on his screen. “Here is my stool measurement from January 1, 2012. And here are my bacteria. Lactobacillus and Bifidobacteria: that is what you get in, like, a yogurt and stuff like that, right? Clostridia: you can have them from zero to four-plus. Four-plus is what they should be. And you can see I am deficient here on a number of them,” he says, pointing to low numbers on the chart. “So then I went back over time and got them plotted, and they never were above two, and now they are collapsed down to one. So it looks like I am losing. So what do Clostridia do? Because I am missing them—I am missing that service.”
You may note the Alice in Wonderland quality of all this. Every question Larry seeks to answer raises new questions, every door he opens leads to a deeper level of bewildering complexity. One could easily conclude that these levels never bottom out, that the intricacy of the human body, composed of its trillions of cells—each dancing to the tune of a genetic program but also subject to random intersections with outside forces such as radiation, chemicals, and physical accidents—is for all practical purposes infinite, and hence permanently beyond our full comprehension. But Larry, with his astrophysics background, is utterly undaunted by complexity. This is the gift of the computer age: things once considered too numerous to count can now be counted. And Larry believes that questions about how the human body functions are ultimately finite.
In his own case, Larry has zeroed in on what he believes is the specific missing bacterial component behind the immune-system malfunction causing his bowel inflammation. He’s begun a regimen of supplements to replace that component. If it doesn’t work, he’ll devise a new plan. He isn’t aiming for immortality—not yet, although, as far as he is concerned, it’s not out of the question. As we develop our ability to replace broken-down body parts with bioengineered organs, and as we work toward a complete understanding of human systems and biochemistry … why not?
Reflecting on Larry’s vision of a patient-centric, computer-assisted world of medical care, Dr. Welch allows: “I can conceive of this happening. But is this the model we want for good health? What does it mean to be healthy? Is it something we learn from a machine? Is it the absence of abnormality? Health is a state of mind. I don’t think constantly monitoring yourself is the right path to that state of mind. Data alone are not the answer. We went through all of this with the Human Genome Project. You heard it then: if we could just get all of the data, all of our problems would be solved. It turned out that the predictive power of mapping the genome wasn’t all that great, because there are other factors at play: the environment, behavior, and chance. Randomness has a lot to do with it.”
And these are not the only reasons to be skeptical of Larry’s vision. Researchers will certainly continue to map the human body in ever-greater detail, enabling doctors to spot emerging illness earlier and to design drug treatments with far more precision. But in the end, how many people will want to track their bodily functions the way Larry does, even if software greatly simplifies the task? Larry says the amount of time he has spent monitoring and studying himself has grown a lot, but that it still adds up to less time each day than most Americans spend watching television. But even if that time is radically reduced by software, how many of us, understanding that our decrypted genome may reveal terrible news about our future—Alzheimer’s disease, crippling neuromuscular diseases, schizophrenia, and so on—will even want to know?
When I ask Larry this question, he frowns and says, “I can’t understand that.” The very idea stumps him. To him, not wanting to know something—even bad news—just doesn’t compute. His whole life is about finding out. He’s a scientist to his core.
“I hear it a lot, but I don’t understand it. Because whatever it is, if you suspect that you are going to have, say, Alzheimer’s disease within five years or t
en years, then that should focus your mind on what it is you want to accomplish in the days that you have left.” Then, after a moment for more thought, he adds, “And if you don’t know, those days are going to just slide by, in which you could have done something that you always meant to do.”
He knows that the way he lives and works might seem eccentric or even a little crazy to others. “Most of my life, people have thought I was crazy at any given point,” he says. “Maybe being crazy simply means you are clear-sighted and you are looking at the fact that you are in a period of rapid change. I see the world as it will be, and of course, that is a different world from the one we live in now.”
Larry is in a hurry to get there. He sees himself ten years down the road as someone healthy and active and strong, instead of someone struggling to manage the increasingly uncomfortable and debilitating effects of Crohn’s disease. As he makes his way down the supplements aisle at his Whole Foods Market, looking for a very specific assortment of probiotics with which to mix his remedial cocktail, he’s not just trying to save himself. He’s trying to save you.
SPORTS
The Silent Treatment
Sports Illustrated, September 2013
You might not expect a long-retired, much-dinged, memory-impaired NFL quarterback to recall the details of a single play from twenty-seven years ago, but when I began describing the one that interested me to former Browns play-caller Bernie Kosar, he interrupted me. “Unfortunately, I can finish this story, but go ahead,” he said. “I’ll be nice and pretend I don’t know. I could claim, ‘Hey, I’ve had concussions; I don’t know what you’re talking about.’ But … go ahead.”