by Masha Gessen
Before our awkward conversation ended, Dr. Garber did mention the possibility of a preventive mastectomy. She said it quickly, as though in passing, but I tried to grab this conversational straw by making an immediate objection. “I’m still using them to feed my daughter,” I said, smiling. The counselors seemed to take this comment as a categorical refusal to discuss the subject, and moved on, leaving me confused. After all, that same study had credited prophylactic mastectomies with increasing life expectancy by as much as 5.3 years.
Svenya and I walked silently to the parking garage, where I flashed my distinctive small blue Dana-Farber card: I was a cancer-center patient now, and this qualified me for free parking. The air smelled of February cold and spilled coffee. I felt nauseated and lonely, as one does on a gray day after a party.
***
What I did over the following couple of months was, I imagine, normal under the circumstances. Sometimes I was obsessed with the subject of cancer and my decision. Sometimes I forgot all about it. Sometimes I lay down with my daughter on a futon on the floor in her room and, as she nursed and drifted off to sleep, I thought that sacrificing physical parts of myself and even my youthfulness was a small price to pay for continuing this happiness. “If you are anything like me,” my physician had said, “you are looking at your kids and thinking you just want to be around for their college graduation.” I had had much bigger plans, but I was learning to think small.
I had the MRIs. I got the phone call from my doctor telling me there was a lump. I waited for my biopsy date. When it came, I spent a day walking around Cambridge with a chemical icepack under my arm; it ended up leaking and damaging a good silk shirt. I waited for my results. I got used to the idea that I had cancer. I imagined surgery, chemotherapy, and radiation. I considered having an affair to give my body a proper send-off. Then none of it happened. I had to return to trying to make my impossible choice: live and wait for the cancer to come or start carving up a body that felt utterly healthy.
We were in Cambridge that year because I had a fellowship at Harvard University. The Nieman program is designed to give so-called midcareer journalists a chance to spend a year at the university in the pursuit of intellectual enrichment. I had spent nearly twenty years writing articles, and I had recently had a baby, so I viewed the year as a chance to jump-start a brain that had effectively ground to a halt. It worked: Everyone around me was actively, energetically, and even expressively thinking, and very soon I felt the wheels start to churn inside my head. Now that I felt myself paralyzed by an impossible choice, I decided to try to harness some of the best brainpower in the world to make my decision for me. It seemed clear enough that the genetic counselors had catastrophic tunnel vision. It seemed I was not up to the task. What would the economists say?
Economics, as a field, had been one of my discoveries as a remedial college student. Like most journalists, I had at times been compelled to write about financial matters, and, like most journalists, I labored under both the fear that my ignorance would be discovered and the misapprehension that economics was the boring and confusing science of money mass and aggregate values. As it turns out—handily—the dismal science is in fact a study of the way people make decisions. Also handily for me, economics was at that time undergoing a revolution. The 2002 Nobel laureate in economics, the psychologist Daniel Kahneman, and his coauthor Amos Tversky had shown that for nearly three centuries economics had operated under the blatantly incorrect assumption that people made decisions rationally and in their own best interest. Rather, they argued, people relied on intuition and immediate perceptions, distorted by built-in bias—and only sometimes, if they were intelligent and aware of their mistakes, could they correct their decisions rationally. Kahneman’s Nobel lecture was devoted to the concept of “bounded rationality.”
Naturally, just over a year later the Harvard course called Psychology and Economics was oversubscribed. I was auditing it, struggling with the calculus but reveling in reading about beautifully designed experiments. The two instructors talked about things like lotteries, the Milgram obedience experiment, and an experiment run on eBay, the online auction site, showing that people were more likely to bid—and to bid faster—on a lower-priced CD with a higher shipping and handling fee than a higher-priced CD with a lower fee, even though the total amount of money spent would be the same. Most of the studies they examined exposed some of the mechanisms through which people regularly make decisions that are demonstrably wrong. It was the best kind of course. Twice a week, it made you feel smarter than other people, and, because it concerned itself with things like lotteries, purchases, and fateful decisions, it carried the promise of helping you become a winner. So now I went to see one of the instructors. It seemed like it would be a very good idea to sit down with him and do some math.
David Laibson was sitting, slightly hunched, in front of his computer monitor, scrolling down a spreadsheet and talking on the phone. He was saying things like, “This one is a point-four chance” and “She is no more than a point-five chance.” It turned out he was scrolling down a list of prospective graduate students, each of whom had a rating quantifying the chances of his or her accepting an offer from Harvard. I found this extraordinarily appealing. I felt that if David and I could find a way to frame my own predicament in similar terms—a 0.4 chance of cancer weighed against, say, a 0.5 chance of salvation—my own rationality would know no bounds.
I quickly explained that I was working on an article about the way people go about making decisions based on genetic information—using the breast cancer genes as an example. David immediately confounded my expectations by doing what anyone would have done: try to get out of answering the question. He told me about two New York University economists who have demonstrated that people do not really want to know what will happen to them, especially when it is not clear what to do with the knowledge. I countered with two other studies, specifically concerning the BRCA mutations, which showed that women do want to know. “They just say they want to know,” David responded. So I told him, instantly eliciting the cocked-head look. It was an awkward, even ridiculous moment. I had met David as a fellow auditing his class. I had come to his office as a journalist interviewing an expert. Now these seemed like false pretenses: I was a woman feeling very lost, asking a man I barely knew what I should do with my life and my health.
Later I read the studies David had mentioned. Two New York University economists, Andrew Caplin and John Leahy, looked at the usefulness of information given to patients by doctors: in essence, whether the patients benefited from knowing more or less. The ultimate answers were intuitively predictable: It depends on the patient, and on the news. Some people prefer to know more, while others like to know less: A much earlier study classified them as “monitors” and “blunters,” concluding that the former benefited from information about an upcoming stressful medical procedure while the latter suffered from it. Along the way, the economists concluded that doctors are not very good at handling bad news, or perhaps even information in general, and made the titillating suggestion that things like test results may better be conveyed by a machine.
A third economist, Botond Köszegi at the University of California at Berkeley, reclassified “monitors” as “information lovers” but made an important new observation: Regardless of whether patients are inclined to seek information or avoid it, they prefer the same sorts of tests—ones that, in the economist’s lingo, have “an upside potential surprise more than a downside potential surprise.” It may sound obvious that good news is preferable to bad news, but it was actually an important insight into the motivation of people who go for medical tests. Köszegi pointed out that one is highly unlikely simply to go “check for cancer” but will seek a second opinion if diagnosed with cancer. In the first instance, there would be either no news or bad news; in the second, there was always the hope, however slim, for a pleasant surprise. So it made sense that women like me—those who had watched their mothers die of cancer—
would opt for genetic testing. Not so deep down, we all believe we will develop it, so we go in hoping for a negative test result, which, in our minds, would be akin to winning the lottery. I remember being almost elated when, before my test, the genetic counselor pointed out that my chances of having the mutation were fifty-fifty—compared to near certainty, these had seemed like terrific odds.
I also reviewed the studies with which I had countered David’s argument. I had unintentionally overstated my case. One of the studies, conducted in Israel, looked at people’s attitudes toward being tested for Huntington’s disease or for a nameless disease very much like it: rare, incurable, debilitating, diagnosed by genetic testing with absolute certainty (everyone who has the mutation will go on to develop the symptoms), and generally developing in midlife. About half the subjects said they would want to know they had the mutation. In the second stage of the study, people were asked whether they would want to know if they carried a mutation correlated with a disease very much like breast cancer: common, treatable, and sometimes curable, less than certainly diagnosed by genetic testing (survey subjects were given theoretical odds of 60 percent), and also generally diagnosed in midlife. The respondents did not appear to care that this disease was far more common or even that in this case the presence of the mutation was an imperfect predictor; all that mattered to them was that this hypothetical disease could be cured. Now the contingent of those who would want to know grew to 80–93 percent.
The other study, conducted in Boston, was a telephone survey of two hundred Jewish women, who were asked whether they would want to be tested for one of the BRCA mutations. The results were less dramatic: Only 40 percent said they would want to know—as many as said they would not. This seemed to make sense, though: The Israeli study was a little more recent, and Israel has applied genetic testing more widely, so it had penetrated the culture. Still, even 40 percent is a very high rate.
Did that mean the economists were wrong? Not necessarily. In fact, probably not. The Israeli paper also reviewed older studies of willingness to be tested for the Huntington’s gene. Before the gene was actually discovered, about half of those who had Huntington’s in their families said they would want to be tested. Once the test became available, only about 15 percent of potential subjects chose to take it. It is much harder to calculate how many of the potential carriers of a BRCA mutation are choosing to be tested: There are too many variables to be able to estimate the pool of subjects. But when I went to see one of the authors of the Boston study a couple of days later, she answered my question: “We don’t have women stomping down the door to get tested.”
Finally, I even found a study that justified my own choice. A group of researchers at Georgetown Medical Center asked relatives of carriers of the BRCA mutations whether they wanted to know their own status. A majority agreed—but among the minority who did not, depression frequently took hold. The findings were so striking that the researchers called their paper “What You Don’t Know Can Hurt You.” Remarkably, the New York University economists had used this exact study to bolster their point that some people really do not want to know. I got the point that even if they did not, they perhaps should.
But then, I am an information lover. Now, though, sitting across from the young economics professor and waiting for him to tell me what to do, I wondered whether I’d been too passionate, even rash, in my pursuit of knowledge.
He sat back down at his computer and opened a new Excel file. In the left column he entered ages, year by year, starting at thirty-seven (which happened to be his age as well as mine). Across the top, he placed the options: “Oophorectomy,” “Mastectomy,” “Oophorectomy and Mastectomy,” and “Do Nothing.” Now we had to devise the formulas for figuring out the value of life under each possible decision for each possible year. “Let’s normalize not being alive as having a utility of 0,” suggested David, reasonably enough. “Further, let’s normalize a year of healthy life to have a utility of 100. Should we assume that life without breasts is equally good?” That was unexpected, but it sounded right. Sure, I said, provided the surgery went well.
Now what would be the value of life with cancer? I could recognize that question as a classic problem: The value would get higher as probability got lower. The longer a woman lives after the initial diagnosis, the more likely it is that her life will once again approach normalcy. David asked if we should simply equate a diagnosis of cancer with death, which had a value of 0. Now this was a clear cop-out: If we did that, the correct choice would instantly become obvious—do everything in order not to die—but we would intentionally have ignored all complexity. So then David decided to calculate the values more precisely—that is, to perform the act of alchemy for which I hoped: to express life in numbers.
He switched from his computer screen to the huge dry-erase board opposite his desk. After a couple of false starts, he wrote:
V(G) = 100 + ß [(.98)V(G) + (.02)V(illness)]
Translated, that said, the value of a year of life in what we consider a “good state”—apparent health with none of the possible disadvantages of surgery—is equal to 100 (this is the utility we had assigned to a year of normal healthy life) corrected by “the discount factor” B, which consists of the normal mortality risk for young people, which is about a quarter of 1 percent, and the 2 percent annual risk of getting breast cancer. In brackets on the one side is the 98 percent likelihood of a good life and on the other the 2 percent likelihood of a diagnosis of cancer.
Now we had to decide what the value of life with cancer would be. David asked me some of the sorts of lottery questions I had concluded he most enjoyed: Would you take this gamble if, say, your chances of developing cancer were 80 percent? 60 percent? 40 percent? Any percent? He wrote my answers on the board and, stepping ever so lightly over some logical steps, we came up with an answer that was perhaps random but also reasonable: The utility of life with cancer, upon first diagnosis, would be 70. This seemed generous—certainly more generous than our original attempt to pose a diagnosis of cancer as equal in value to death—but also somber enough: You do not die right away, but life is never quite the same either.
V(illness) = 70 + ß [(.8)V(illness) + (.2)X0] = (skipping the algebra here) = 347
What that means is that while life—actual life—with cancer might have a utility flow of 70 utils per year, which is not all that far from normal healthy life, the likelihood of death makes the value of one’s entire life with cancer just 347. Plugging this value into the first equation, we find out that life in the “good state” has a value of 4,752, even allowing for the 2 percent annual risk of breast cancer.
I already knew that if I got a mastectomy, the utility of one year of life would be essentially the same. The value of my entire life, however, would actually increase because the risk of breast cancer would be cut by 90 percent. Still, after the age of forty the risk of ovarian cancer would kick in, also to the tune of about 2 percent a year, bringing the value of my now breastless life right back down to 1,500. What if I got an oophorectomy? Sure, that would cut my ovarian cancer risk by 95 percent. But it might make me instantly old, susceptible to heart disease, high blood pressure, osteoporosis, and so forth. I suggested David look at my collection of all the age-specific risk statistics of these diseases. David’s solution was far more elegant: He just changed our “discount factor” to reflect the mortality risk of an older person—someone over sixty-five. The assumption here was that the mortality risk is a good enough expression of all the disadvantages of life after menopause. That mortality risk is about 1.5 percent a year, rather than a quarter of a percent, as it is for young people. The utility of life in a state we termed “Good/old” turned out to be 95 (as compared to 100 for good/young). And the value of my entire life?
V(G/old) = 95 + ß'[V(G/old)] = 9,500
Huh? Did that mean it was better to be old and cancer-risk-free than young and at risk?
“Hey,” said David, “being old isn’t so bad. I can’t do many
of the things I could do when I was twenty, and I don’t miss them.”
I thought about this. Physically I did not feel that different from when I was twenty. True, to achieve this state of well-being I had to drink a lot less, smoke not at all, and work out a lot, but these did not count as costs. (On the other hand, when I was twenty, I could actually follow this kind of math with pleasure.)
“You know what?” continued David. “Having a 1 percent chance of dying from being old just doesn’t compare to having a 2 percent chance of getting cancer.” Actually, it seemed to me that it did: It was only half as bad.
“It seems the doctors are right,” David concluded. Surgery was the solution. We stared at each other for a minute; this was an outcome neither of us expected. I had been almost sure we would prove that the absurd-sounding recommendation of a preventive oophorectomy was wrong. The intimacy of the interview grew uncomfortable again.