Snowball in a Blizzard: The Tricky Problem of Uncertainty in Medicine, by Steven Hatch

Book of the week: In between what we know is good or bad for us is the rest of medical science, says Jennifer Rohn

June 23, 2016
Man smoking cigarettes with ears and nostrils
Source: Alamy
Buts and no buts: in medicine, there are things that we can be quite sure cause harm, such as cigarettes, but there are many other aspects of healthcare we know very little about

The world is plagued by ambiguities, and nowhere is this more true than in medicine. Advances in science and technology can seem almost miraculous, but beneath this glittering surface lies a vast sea of medical uncertainty unfamiliar to most people – including the very doctors who treat us. Steven Hatch, a physician and University of Massachusetts Medical School academic, aims to empower the general public, as consumers of medicine, with a toolkit for interrogating scientific and medical claims and for asking the right questions of healthcare professionals.

The underlying framework of Hatch’s thesis is the so-called spectrum of certainty. At one end are things about which we have strong evidence of health benefits – keeping fit and trim, for example. At the other are things that we can be quite sure cause harm, such as cigarettes. In between, we encounter things that we are reasonably sure are helpful or harmful: not slam dunks, but fairly good bets. In the dead centre are all the many things we know very little about. The vast majority of medicine, Hatch argues, lies somewhere between the two extremes, meaning that doctors very seldom operate on perfectly secure ground.

Misunderstanding this spectrum can land us in trouble, leading to harmful procedures when there is no need (such as mastectomies in groups at low risk for breast cancer), or avoiding something beneficial – such as vaccination – for fear of obscure harm. Our sense of risk is notoriously out of proportion at the best of times, as when we get hysterical about the latest rare tropical virus, while blithely ignoring the perils of that far more lethal double cheeseburger.

It turns out that we are wired as a species to make these miscalls. In a perilous world, it is far more beneficial to be on a hair trigger when it comes to risk. To use Hatch’s analogy, when an ancient ancestor saw a stick on the ground, he was much more likely to survive if he assumed it was a snake and ran than if he lingered to find out one way or the other.

We can all appreciate the stick/snake analogy. But in the far more complex arena of medicine, none of the risks can be understood without looking at the evidence and weighing the pros and cons on a population level – and then, importantly, attempting to map these generalities back to the individual. Statistics is not an easy subject, but Hatch transmits these tough concepts in conversational language, leavened with plenty of dry humour. (For the aficionados, he’s relegated the finer mathematical details to the appendix.)

The best way to appreciate abstract concepts is to see them at play in the real world. And indeed, the power of this book is its sheer number of examples, which will be close to home for any readers interested in their own health. There is something in here for everyone – prostate cancer screening, antidepressants, diets, infections, blood pressure, hormone replacement therapy and many others. Each example is presented as a study of what we know about the topic – and more to the point, what we don’t.

Take cancer, for example. The book’s title comes from an expression common among radiologists, who struggle to see the hallmarks of breast cancer amid abstract patterns of a typical mammogram. This signal-to-noise concept serves as a metaphor for medical uncertainty in general, where getting it wrong can lead to lawsuits or unnecessary deaths on the one hand, and brutal, wholly unnecessary interventions on the other.

You would think that all our advances in cancer diagnostic technology would lead to more successes. But as Hatch outlines in meticulous detail, this is simply not true in many cases. If you plot the number of people diagnosed over the past 30 years for various cancers, the line moves steadily upwards. Yet the overall death rate due to that cancer type remains stubbornly the same – a flat line that makes a mockery of all our know-how.

How can this be? The answer is as remarkable as it is strange. Although these ever-increasing cancers are “real” (that is, a skilled pathologist could see the evidence of a malignant tumour under the microscope), the unchanging population-wide mortality studies tell us that what we are diagnosing, in increasing numbers, is not the sort of cancer that actually kills you. This supposition is borne out in autopsies of older people who die of other causes; often, their tissues contain what any pathologist would class as cancer – but clearly not cancer as it’s commonly understood.

Uncertainties in diagnostics, of cancer or other conditions, are especially problematic for those participating in screening programmes – the search for disease in seemingly healthy people. Mammograms are a key example: there is scant evidence that they do any good for women in their forties (when American women are screened), or even in their fifties (when countries such as the UK recommend testing).

So for breast cancer and some other ailments, Hatch makes a strong case that while screening low-risk people will undoubtedly lead to more positive diagnoses, it will not save lives to any appreciable level, and indeed may result in net harm. But people who are picked up by screening and are treated, and who then do not manifest disease, see strong personal evidence that the screening has “worked”. So do the doctors who ordered the treatments. Thereby is the myth of diagnostic value perpetuated, one patient at a time, while the impersonal population studies tell a completely different tale.

A similar story is seen with treatment. Hatch gives as one example the ongoing controversy of statin use for preventing heart disease. Statins can be hugely beneficial in people with high cholesterol levels, placing this treatment safely near the “good” end of the spectrum of certainty. But recent changes in US guidelines now recommend widening the pool of people who should be treated with statins by 30 per cent, which translates to some 13 million additional people. Here, as Hatch puts it, we “lurch towards the middle” of the spectrum, as few studies have tested the benefits of lowering cholesterol in people whose levels aren’t sky-high. So the benefits are unclear, while the possibility of side-effects is real.

Where does this leave us? In the face of Hatch’s statistical tough love, you might be tempted to stick your head in the sand. But his message is ultimately one of hope, as he rallies his readers to see uncertainty as a positive, a means towards achieving a good end. By embracing uncertainty, patients will feel comfortable engaging their doctors at the level of hard evidence, which will help them to navigate their own treatment. Likewise, they will be equipped to assess the conflicting and often incorrect health messages churned out via the media and the internet. Conversely, Hatch believes, if more doctors had the grace and humility to admit to themselves – and to their patients – that their advice is often only educated guesswork, patients would feel more reassured and less likely to turn to the pseudoscientific pedlars of potions, with their “ever-increasing yowls of the overly certain”.

Jennifer Rohn is principal research associate in nephrology, division of medicine, University College London.

Snowball in a Blizzard: The Tricky Problem of Uncertainty in Medicine
By Steven Hatch
Atlantic, 320pp, £14.99 and £9.99
ISBN 9781782399872 and 9889 (e-book)
Published 2 June 2016

The author

Author Steven Hatch, University of Massachusetts Medical SchoolSteven Hatch, assistant professor of medicine at the University of Massachusetts Medical School, lives “in the suburbs of Boston, Massachusetts. My wife is named Miriam; she‘s a full head of steam, so to speak, a ferociously competent and passionate woman who works as an architectural construction manager, does these huge projects where a hospital or university is adding on a big wing, $300 million projects that take a few years and involve interfacing with people from all kinds of specialties ranging from scientists to clinicians to public health people to lawyers to construction workers and so on.

“It’s highly demanding work that I’d never be able to succeed at, so I’m always amazed at how much she juggles,” he adds. “Plus she’s using those skills to develop sustainable healthcare projects in Haiti with the group she founded, Sustainable Healthcare for Haiti. We sleep once every few months...

“Our kids are twins, Erez and Ariella, age 14. My son passed me in height last year (I’m 1.85m) and he’s not done growing (yikes!). An ancient black cat, Nehemiah, also shares the house with us, and lets us know with loud yowls of his presence at, say, 3am, just in case we had forgotten.”

Hatch was born and raised in Mansfield, Ohio, “a town of about 50,000 people halfway between Cleveland and Columbus. Very blue-collar town with a lot of heavy industry that has, alas, largely been shut down since the late 1970s. Because I grew up Jewish in a very, very Christian city with a lot of fundamentalist churches, I got good at being able to look at problems from alternate angles, since I was typically the outsider. I think that's informed my worldview (and my appreciation for what it means to be an outsider, which is a big help when seeing how scared and alienated patients can become by the whole process of medicine).

“In terms of my parents and their influence in my upbringing, I think the most important element was that they didn’t really push me one way or another to pursue certain studies or interests. They didn't say to me, ‘We want you to be X’; they just wanted me to do what pleased me and so I went to college without any preconceived notions of my future calling and followed my nose.

“I discovered I had a knack for writing, and it was only much later, when I had decided to pursue medicine (another accidental interest) that I learned my father really wanted me to be a writer. And, while he wasn’t exactly crestfallen that I was pursuing medicine, he was a touch disappointed. (That may give you some context for my dedication in Snowball, as I apologise to Dad for finally having a published book to my name, but it got published six years after he passed away.)”

Of his school days, Hatch recalls not being especially studious “in the narrow sense of accomplishing assigned tasks for a class or studying for tests. I was not a particularly disciplined student. I never cared about grades and so my average from primary school all the way through medical school was shockingly uniform at the mark of B – definitely not outstanding. I think teachers mostly liked having me as a student, but I was very inconsistent, since my level of engagement depended very much on the subject matter.

“Other students who were more disciplined and focused had an easier time with advancement in the academic world, and when I finally decided to get serious about medicine, I had to learn some hard lessons about studying, because in order to get into med school, you really do have to care about the grades and the numbers and jumping through the various administrative hoops. But I have always maintained a sense of being ‘undisciplined’ in that I can get interested in lots of subjects instead of being highly focused on just my specialty. In the UK there’s a long and proud tradition of that in academia, but in the US that more multidisciplinary approach remains the exception.”

A professor of literature, Karen Klein, “was the first person who really made me see that thinking was something more than what you did as part of a job (English and American lit was my area of study as an undergraduate). Growing up in Mansfield, academic pursuits were pretty much strictly thought of in terms of what kind of a college you got into, and what that would mean for your job prospects. The idea of swimming in ideas as an end in and of itself was not something I really grasped until I studied with Karen.

“Of course, I think my parents would be offended by this! They were, in fact, people who cared deeply about ideas; it’s just that Karen got me to be conscious of it, as I had just absorbed my parents’ way of thinking about things without realising what I was doing (if that makes any sense). From there, I did learn from some very talented professors at Brandeis University (and a little at Case Western Reserve, where I received a master’s in English before deciding to change tracks and pursue medicine instead), but mainly my influences were writers who got me to look at the world in different and new ways.”

Who were those writers? “I love fiction, but I became an avid nonfiction reader from early on and that’s still true. The writers that shaped me initially were Bertrand Russell, Paul Fussell, Stephen Jay Gould, Richard Dawkins, Hunter Thompson, Susan Sontag, Simon Winchester, Joyce Carol Oates, and Jared Diamond among others. I tried to dabble in academic stuff like Derrida, Lacan, Foucault and Terry Eagleton, and went through this jag where I was reading a lot of literary theory.

“While in graduate school, I went to a conference on early modern period English literature and I was listening to all these people talk about these writers, and I had this revelation that although I was a reasonably bright guy, I had absolutely no idea what anyone was talking about. And that’s when I realised that I couldn’t do lit theory or the humanities in the academy – that I had to do something much more practical, and that whatever dent I might make in the world of ideas was not going to be in theory.”

He adds: “I don’t mean to say that as a put-down of that style of academic discourse, only that it’s not really for me. I’m just too concerned with the effects of ideas in the real world and can’t think about ideas without simultaneously exploring their implications. I’m just too concrete a thinker. So that was when I made the decision to leave the study of literature and do something different, and I accidentally stumbled into medicine. It was something that never interested me previously, but I got a job working in a hospital, and I realised it combined all sorts of things I liked: science, teaching, pastoral care. Looking back, I think that job was the luckiest thing that ever happened to me, because I wouldn’t trade being a doc for anything in the world.”

Twenty-eight by the time he entered medical school – “old enough that I didn't require professional influences in the same way that you do when you're 10 years younger” – Hatch nevertheless recalls there being people “who served as role models for particular aspects of my work; for instance, in the acknowledgements, I mention a man named Alan Rothman, who is the last person who can truly be called my mentor, and while he really was important for my development, I already knew how I wanted to approach patients and practice medicine.”

He was, he says, “definitely gregarious as an undergrad. I lived with a group of guys, all of whom were considerably smarter than I am, and I learned a tremendous amount simply by living with them and watching how their minds worked. They were all voracious readers and they thought out loud. Having idle conversations in our common room turned out to be sessions in which my mind was sharpened by their depth of thought and their intellectual standards. If you had an opinion, you had better be prepared to back it up and think carefully about what you said. I don’t just mean opinions in the sense of political or theological ideas – I mean even about topics that might seem mundane or dull. 

“For instance, one roommate, a guy named Ted Frank, had thought about sports not like all the breezy-but-silly prose of sportswriters, but from the perspective of statistical analysis, which is now something that is considerably better understood (in the US, we call this the Moneyball approach after a famous book written about a baseball team that built its club using advanced statistical metrics), but in the late 80s, that way of understanding baseball or any other sport was still very much on the cutting edge.

“So I learned a lot from Ted, and my other roommate Glenn Branch, who studied philosophy and was always way above my head; and two other guys, one of whom was an early computer geek and used to take apart Macintoshes and put them back together just for fun, and the other was a chess player and read widely. I probably learned two-thirds of everything I learned in college just by living with them and having them drive me to think carefully, and read really good writers (most of those writers I mention above were not assigned by a teacher, but were being read by my roommates, just because they seemed interesting to them).

“That may not sound gregarious, but they were only part of my life. I also worked at the cafeteria [on campus] a lot and enjoyed doing the behind-the-scenes physical labour, and made a lot of friends both among the other students working the jobs as well as working with the kitchen staff.

“My other major activity was writing for a magazine called The Watch that tried to bridge the gap between two different sets of people who wanted to write: one was an explicitly political group who weren’t quite revolutionaries but were very committed to a left-wing agenda, which at the time included the end of apartheid, the decay of the inner city, Reagan’s and Bush’s policies both at home and abroad; the other group, to which I was more attracted, also leaned left but were not quite as focused on causes per se but thinking about cultural issues. For instance, I wrote a good review of Spike Lee’s film Do the Right Thing, a piece I’m still proud of more than 25 years after I wrote it.

“So between the roommates, the cafeteria work, and the people I met at our magazine, I interacted with a variety of groups. Fun years. You sent these questions as I enter into my 25th reunion weekend, so you have me in a rather nostalgic mood!”

What gives him hope?

“Professionally, my students. I’m incredibly lucky to be in the presence of passionate, committed, energetic, brilliant people who have chosen caring for others as their calling in life. When patients have a good experience as an inpatient, they routinely talk about nurses first, and then students. I try to remember that every time I talk to a patient or a family member – that the reason why they look to nurses or students is because they sit there and talk with them and take them on their own terms. If we can figure out a way of not pounding that precious quality out of medical students during residency, we'll do better as a profession.”

Karen Shook

You've reached your article limit.

Register to continue

Registration is free and only takes a moment. Once registered you can read a total of 3 articles each month, plus:

  • Sign up for the editor's highlights
  • Receive World University Rankings news first
  • Get job alerts, shortlist jobs and save job searches
  • Participate in reader discussions and post comments


Print headline: A healthy dose of ambiguity

Have your say

Log in or register to post comments

Featured Jobs

Assistant Recruitment - Human Resources Office

University Of Nottingham Ningbo China

Outreach Officer

Gsm London

Professorship in Geomatics

Norwegian University Of Science & Technology -ntnu

Professor of European History

Newcastle University

Head of Department

University Of Chichester
See all jobs

Most Commented

men in office with feet on desk. Vintage

Three-quarters of respondents are dissatisfied with the people running their institutions

students use laptops

Researchers say students who use computers score half a grade lower than those who write notes

Canal houses, Amsterdam, Netherlands

All three of England’s for-profit universities owned in Netherlands

sitting by statue

Institutions told they have a ‘culture of excluding postgraduates’ in wake of damning study

A face made of numbers looks over a university campus

From personalising tuition to performance management, the use of data is increasingly driving how institutions operate