Over at Mind Matters, Chadrick Lane reviews a fascinating experiment that revealed the rewarding properties of information, regardless of whether or not the information actually led to more rewards:
In the experimental design, monkeys were placed in front of a computer screen and were trained to perform a saccade task, in which they learned to direct their gaze at specific areas. The monkeys were first given the option of choosing between one of two colored targets. One of these targets would give the monkey advance information about its future reward. The advance information came in the form of visual cues, one representing a large reward and the other a small reward. Choosing the other initial colored target revealed cues that were randomly associated with reward size, thus possessing no informative value. After only a few days of training, the monkeys showed a clear preference for choosing the informative colored target.
The researchers then tested to see when the monkeys wanted the information. In this scenario, the monkeys were again initially presented with two colored targets. One of these targets had informative value while the other did not. The difference was that the monkeys always received informative cues just before their rewards. The choice each monkey had to make was whether to see an earlier informative cue. Despite always having a delayed informative cue, regardless of which initial target they selected, the monkeys preferred to have advance information as soon as possible. Like high-school seniors waiting on their SAT results, the monkeys wanted to know, and they wanted to know right now.
Forgive the far-fetched connection, but this experiment makes me think about health care costs. In recent years, it has become clear that an important driver of health care costs concerns unnecessary medical tests and imaging. From CT and PET scans to MRIs, Americans are given these tests at a far higher rate than citizens of other developed nations. The end result is tens of billions of dollars squandered on useless technology.
Obviously, many factors play a role in the overuse of high-tech imaging. There’s the problem of defensive medicine, in which doctors prescribe tests that they don’t think are relevant just to avoid a potential lawsuit. And then there’s “technology creep,” in which shiny new tools always seem better, even if they actually work worse. (This helps explain why spending on new health technology – from surgical robots to proton beam therapy – makes up more than 65 percent of the more than 6 percent annual increase in healthcare costs.)
But I think our need for more information, even if the information doesn’t increase our understanding, also plays a role. Just look at chronic back pain, a medical conundrum that I explore in my book. When MRI was first introduced in the late 1980s, the medical community believed that the imaging tool would transform the diagnosis of back problems. For the first time, doctors would have access to stunningly accurate images of the interior body. Since doctors could finally image the spine and surrounding soft tissue in lucid detail, they should be able to offer precise diagnoses, locating the aggravated nerves and structural problems that caused the pain in the first place. This, in turn, would lead to better medical care.
Unfortunately, MRI’s haven’t solved the problem of back pain. The problem is that the machine sees too much. Doctors are overwhelmed with information, and struggle to distinguish the significant from the irrelevant. Take, for example, spinal disc abnormalities. While X-rays can only reveal tumors or problems with the vertebrae, MRI’s can image spinal discs⎯the supple buffers between the vertebrae⎯in meticulous detail. After the imaging machines were first introduced, the diagnosis of various disc abnormalities began to skyrocket. The MRI pictures certainly looked bleak: people with pain seemed to have seriously degenerated discs, which everyone assumed caused inflammation of the local nerves. Doctors began administering epidurals to quiet the pain, and, if the pain still persisted, would surgically remove the necessary disc tissue.
The vivid images, however, were misleading. Those disc abnormalities are seldom the cause of chronic back pain. A 1994 study published in The New England Journal of Medicine imaged the spinal regions of ninety-eight people with no back pain or any back related problems. The pictures were then sent to doctors who didn’t know that the patients weren’t in pain. The end result was shocking: two-thirds of normal patients exhibited “serious problems” like bulging, protruding or herniated discs. In 38 percent of these patients, the MRI revealed multiple damaged discs. Nearly 90 percent of these patients exhibited some form of “disc degeneration”. These structural abnormalities are often used to justify surgery and yet nobody would advocate surgery for people without pain. The study concluded that, in most cases, “The discovery by MRI of bulges or protrusions in people with low back pain may be coincidental.”
Medical experts are now encouraging doctors not to order MRIs when diagnosing back pain. A recent report in The New England Journal of Medicine concluded that MRIs should only be used to image the back under specific clinical circumstances, when doctors are examining “patients for whom there is a strong clinical suggestion of underlying infection, cancer, or persistent neurologic deficit.” In the latest clinical guidelines issued by the American College of Physicians and the American Pain Society, doctors were “strongly recommended…not to obtain imaging or other diagnostic tests in patients with nonspecific low back pain.” In too many cases, the expensive tests proved worse than useless. All of the extra detail just got in the way. The doctors performed better with less information.
And yet, we all understand the motivations behind such medical practices. If it’s my back that’s in pain, I want to know everything possible about my abnormal spinal discs – the extra information feels essential. And if I’m a doctor, I want to see what’s happening on the inside. In other words, people have strong information-seeking instincts, which tend to overwhelm the statistical evidence. It doesn’t matter what the NEJM says – we know what feels right.
The larger point is that it’s never easy cutting health care costs. In the abstract, the waste looks easy to fix – we should dramatically reduce the number of MRIs given to patients with back pain. The data couldn’t be clearer. But when you look at the problem in detail, and try to imagine the decision-making process from the perspective of the individual patient or doctor, you often find powerful instincts behind the reckless spending. These high-tech tests, after all, have given us a new source of information. And because such information is inherently rewarding – we always want to know – we prescribe the tests, even if they’re not actually informative.