On Cancer and Risk

By / December 12, 2011

Cancer is much in the news of late.  The Canadian Task Force on Preventative Health Care has recently released new guidelines recommending fewer mammography screenings and an elimination of breast self-exams, much to the consternation of patient advocacy groups.  Meanwhile, debates rage about PSA tests for prostate screening, with the U.S. Preventive Services Task Force recommending that PSA tests be removed from routine physicals.

The link between the two sets of guidelines is the risk of overtreatment.  Early screening leads to false positives and aggressive treatment for cancers that may never have grown or spread.  This has real cost, and the benefits are limited.  As Andre Picard outlined in the Globe and Mail, there are four types of tumours:

  • Slow-growing tumours that would likely be found and treated without screening
  • Aggressive cancers that are deadly, whether they are detected by screening or not;
  • Innocuous lesions or growths that are not deadly and frequently result in overdiagnosis and overtreatment;
  • Potentially deadly tumours that are detected at just the right time by screening to allow for lifesaving treatment.

Screening is really only helpful in the fourth case, which is also the least likely scenario.

Cancer, in essence, is a catch-all word, encompassing a myriad of different types of tumours that can have radically different effects.  Cancer treatment is similarly complicated.  There are many possible treatment options for the many types of tumours (including no treatment at all), with different likelihoods of survival or recurrence or side effects.  Moreover, cancer treatment is very uncertain.  While the pathology can seem exact — Gleason Scores, Bloom-Richardson, Trus Prostate Volume — it is not predictive.  All those numbers merely represent what has happened, on average, in previous cases.

At the best of times, people are very bad at understanding such a complex web of statistics and probabilities; standing on the precipice of your own mortality does not enhance the rational, mathematical mindset.  As Gina Kolata wrote in the New York Times last month, “The word “cancer” is so powerful it overwhelms any conversation.”

We do not think, “given my age, the tumour size and degree of lymph node invasion, I seem likely to survive.  No need to panic here.”


We think, “Am I going to live?”  “What will happen to my children?”  “Just cut it out.”

It is no wonder, then, that we react so violently against the idea of reduced screenings, fewer tests.  The numbers may add up, but the reaction comes from a deeper, more visceral place, where probabilities have no standing.

If an emotional reaction is understandable, that is not to say it is helpful.  ‘Just cutting out’ a cancer can be counterproductive — introducing significant pain and suffering without prolonging life.

Drawing by Julie Man

Consider the case of radical mastectomies.  For decades, breast cancer patients underwent radical mastectomies, removing the pectoralis major, even ribs, chest cavities, and collarbones in an effort to prevent cancer’s recurrence.  It felt right — just cut out it out — but it rather than extend life, the procedure left women disfigured, disabled, and in pain.  As early as the 1930s, evidence began to emerge that radical mastectomies might be inflicting severe pain for no benefit, but it wasn’t until the 1970s that breast cancer patients stopped undergoing the treatment on a routine basis.  Such is our need to feel like we are doing something, that we are fighting the disease, that we can control the outcome.

I suspect this phenomenon is an important reason that cancer screening is something less than an unalloyed good.  If your doctor tells you that your cholesterol levels are high, you are unperturbed if she recommends monitoring your levels for further change.  If your doctor tells you that you have cancer, even a stage 0 cancer, you demand intervention.  Treatment feels like taking control of the situation; you are fighting the good fight against the big C.  Any side effects seem abstract and unimportant by comparison.

But for the best outcomes for patients, it’s important that treatment doesn’t just feel like it’s helping — that it is actually extending life.  And it’s important that a treatment’s side effects are understood and weighed against potential benefit the treatment can provide.  For this to happen, patients need to understand risk — their personal risk profile, the risks of treatment options, the risk of recurrence.

All of which brings me to one of Bridgeable’s latest projects, TheraChoice.  We are collaborating with the department of Biomedical Communications at the University of Toronto to build a risk communication prototype for cancer patients.  TheraChoice’s goal is to empower patients to make therapeutic choices by giving them a better understanding of the benefits and drawbacks of treatment options.  With funding from the Applied Research and Commercialization Initiative of the Federal Economic Development Agency for Southern Ontario, the University of Toronto has hired the lovely and talented Julie Man to work on risk visualizations.  Nick Woolridge, Jodie Jenkinson, and Shelley Wall are investing their time and risk communication expertise, while we at Bridgeable are investing in ethnographic research and user evaluations.  Already we have completed an academic literature review and an ethnographic study.  We have drafted design principles to guide risk communication in oncology, and are currently compiling all of this work into a design guide.  And, work is underway on a web-based prototype that we can evaluate with patients early in the new year.

We’re very excited about the project.  It has been a treat working with the team from U of T, and we are uncovering some very interesting findings that we’re anxious to put into practice in our prototypes.  Our current focus is breast and prostate cancers, but we think there is a lot of scope to expand the approach to other cancers or disease states.