Junkfood Science: The Cancer Personality Myth

August 22, 2008

The Cancer Personality Myth

Blame and guilt are the most tragic and hurtful results of today’s popular beliefs in mind-body wellness. No matter how many times research has shown that a “cancer personality” is a myth, the belief persists that a negative outlook, depression, anger, stress or fatigue can increase our risks for cancers... and that a positive mental outlook will keep us well.

The latest study in the news depicts one of the most common fallacies of logic and research biases, one often used to convince us of a correlation that doesn’t really exist. See if you can spot it in this news article, reviewed by a Harvard Medical School associate clinical professor:

A Sunny Disposition May Protect Against Breast Cancer

Women who are happy and optimistic appear to be less likely to get breast cancer than their gloomier counterparts, according to a case-control study... Ronit Peled, Ph.D., of Ben-Gurion University of the Negev here, and colleagues reported online in BMC Cancer...

To explore the issue, they undertook a case-control study in women ages 25 to 45 — 255 had been diagnosed with breast cancer at one of six centers in Israel and 367 were healthy and selected from one of two medical centers... All participants were interviewed with the Brief Symptom Inventory — evaluating depression, anxiety, and happiness and optimism — and a Life Event Questionnaire, in which they were asked to recall major stressful life events and their feelings. Breast cancer patients were asked for information from before their diagnosis (mean time from diagnosis to interview was one year).

Breast cancer patients reported significantly higher depression scores (P=0.04) and significantly lower happiness and optimism scores (P=0.00) compared with the controls. A greater percentage of breast cancer patients had been exposed to more than one negative life event compared with the controls (52% versus 43.9%), but the difference fell short of statistical significance (P=0.065)... "general feelings of happiness and optimism can play a protective role against the disease." The researchers were unable to identify the mechanism of action through which negative life events may influence breast cancer...

The biases which make this not a fair test of the hypothesis go beyond the obvious methodological problems, such as that the participants were not randomized, the associated relative risks (especially expressed as odds ratio) were untenable, and that the authors turned a correlation into a causation by concluding that happiness and optimism could help prevent cancer.

In this study, they administered questionnaires to women newly diagnosed (average within 7 months of diagnosis) with breast cancer at six major oncology units in Israel and to a control group of disease-free women visiting a small outpatient clinic for routine care. The breast cancer patients “were asked to report about their feelings prior to diagnosis.”

The authors reported that the women with cancer had slightly higher scores of depression and anxiety. Among specific stressful life events recalled by the women with cancer, there was no significant difference to the control group. But a computer regression model found a 62% odds ratio associated with having a cumulative number of life events and cancer, while a 25% lower odds ratio of general happiness feelings associated with the healthy women compared to those with cancer. This led the authors to carefully conclude that “experiencing more than one meaningful life event (severe and/or mild to moderate) is a risk factor for breast cancer among young women.”

In their introduction, the authors said that the relationship between “body and mind” is an old issue, since H.L. Snow reported back in 1893 that more breast cancer patients had experienced a traumatic life event. Since then, the belief has grown to include an association between cancers and chronic diseases with depression, anxiety and personality types, despite the fact empirical evidence continues to be lacking. Before we look at a sampling of the evidence which has repeatedly failed to support the 'cancer personality' myth, did you see the bias in this study?

Recall bias

If you’ve just been diagnosed with cancer or are facing any serious health problem, it’s understandable to feel more depressed and anxious than someone healthy. That does not mean that those feelings contributed to your cancer. More importantly, it’s common to ask “why me?” and search for an explanation. When you look back on your life, you’re more likely to remember being exposed to things or doing things you believe might have caused you to become sick.

Our subjective feelings and memories are vulnerable to “recall bias” and this phenomenon is problematic in epidemiological studies, especially “retrospective studies” that look back in time, trying to identify correlations (risk factors) among people already diagnosed with a disease.

Our memories are imperfect and unreliable, regardless of how certain we may feel them to be.

The examples are prolific in the medical literature, such as people with diabetes will remember eating more sweets as a kid than someone who believes themselves to be healthy, or those with cancer will be more likely to remember being exposed to electromagnetic fields from power lines, even when neither are really true. In a classic 1991 epidemiological study, women in the Nurses Health Study had been asked in 1982 to remember how often they colored their hair and their sun tanning. When the same women were re-examined years later, those diagnosed with melanoma were more likely to report remembering tanning (which they believed a major risk factor for melanoma), where there was no increase in their memory of hair coloring.

This bias in epidemiological research was examined by Eman Hassan at the Department of Health Care and Epidemiology at the University of British Columbia in Canada, in an article in a 2006 issue of the Internet Journal of Epidemiology. Recall of information is unreliable and inherently inaccurate, he reminded researchers and medical professionals. As he noted, research has shown that 20% of critical details of an event are irretrievable after just one year, and 50% are irretrievable after 5 years.

Not only do we forget things, but “memory tends to distort perception in systematic ways,” he said. Also, “repeated retrieval of already stored events may add new information as facts and thus events are re-stored in the brain in an altered fashion.” Past events have a way of being rewritten in our minds.

“People with a disease tend to think harder about their prior exposures than disease free people,” professor Hassan wrote. “The presence of disease is presumed to act as a stimulus that affects both the patient's perception of the causes and his search for possible exposure to a hypothesized risk factor.” The more significant or scary the condition is perceived to be, the more likely recall bias can be expected, he said.

In contrast, people among the healthy control group, are more likely to under report their exposures or behaviors, because they’re less likely to remember them or see them as being relevant. The end result is that the relative risk of the exposure is inflated in these studies comparing the two groups and looking for correlations.

New mothers of babies with serious congenital problems are naturally filled with guilt and seeking answers, and numerous studies have shown that as they desperately look back trying to recall something that might have caused their babies to be born with problems, they are more likely to over report exposures. The body of evidence has shown that studies using actual medical records and clinical data are considerably more accurate than studies using self-reported information, such as that recalled by parents of children with childhood diseases.

Recall bias is most pervasive in case-control studies, he said, but can also effect randomized controlled clinical trials when the participants are not blinded to whether they’re in the intervention or control group. While researchers can make efforts to design studies to minimize recall bias, the fact remains: “Research including reported data about past experiences will always be threatened by the limitations of the individual's memory and the influence of disease/exposure status on the recalling process in humans.”

The “cancer personality” myth

We’ve already examined that the soundest studies continue to show that a positive mental outlook and surrounding oneself with upbeat people doesn’t make cancer patients live longer, nor does feeling sad or angry shorten survival. The largest study of its kind, a randomized clinical trial of patients with head and neck cancers published last December in the journal Cancer, found “no statistically significant association between well-being and survival.” Their results held even after factoring for every possible confounding factor, such as stage of cancer, smoking, marital status and support, and income. Despite exhaustively attempting to show a link between mental outlook and prognosis, they could find “no support for the hypothesis that negative emotional well-being predicts poorer survival.”

The New Age belief in a mind-body connection has become so popular, that the "cancer personality" myth persists despite the body of evidence, and people believe that chronic diseases are under their control and be prevented by doing everything right and visualizing wellness and positive energy. Perhaps, learning how diligently researchers have studied and debunked this belief, more people can be empowered with the knowledge that can help free them from guilt and blame.

Researchers in The Netherlands, for example, looked at 9,705 women enrolled in a breast cancer surveillance program in 1989-90. The women were given personality questionnaires and then followed for 13 years to see if there was any link between personality traits and those who later got breast cancer compared to those who didn’t. The researchers published their results in the January 2008 issue of the Journal of the National Cancer Institute. They could found no correlation between any personality factor and a higher risk for breast cancer, with or without adjusting for medical risk factors. Note that this is a stronger study because it wasn’t looking backwards (retrospective) among a group of people already with cancer and trying to find a correlation, but was prospective.

Another interesting prospective study examined 29,595 twins enrolled in the Swedish Twin Registry. They all completed the Eysenck Personality Inventory test, which measured their degree of neurotic tendencies (levels of anxiety, self confidence, depressive outlook, obsessive-compulsiveness, phobias, etc.) and extroversion, when they were enrolled in 1973. They were then followed through 1999 for six different types of cancers: hormone-related organ cancers, virus-related and immune-related cancers, digestive organ cancers (excluding liver), respiratory organ cancers, cancers in other sites, and all cancer sites. The researchers could find no correlation between the personality traits and risks for any cancer group.

It might seem intuitively correct that feeling depressed, grief and exhaustion could lower our immune system defenses and make us more vulnerable to cancer, but clinical research has not been able to support this fear. In another prospective study of 8,527 people in the Copenhagen City Heart Study, they had been examined in 1991-1994 and evaluated for their degree of depressive feelings and exhaustion. After about 8.6 years, the researchers compared the 976 people who got cancers to those who didn’t. They found those with the highest levels of exhaustion had a 20% lower risk for developing cancer at all sites, a 36% lower risk for developing smoking-related cancers, and about half the risk of developing virus and immune-related cancers. And no significant difference was found for cancers related to alcohol consumption or hormones. They simply were unable to find any support for the hypothesis that fatigue and depression, as measured in the vital exhaustion index, increased the risk for cancers.

Personality, grief or depression do not make you more vulnerable to cancer, said Dr. Jimmie Holland, M.D., a psychiatric oncologist at Memorial Sloan-Kettering Cancer Center in New York, and author of The Human Side of Cancer. In the chapter, “The Tyranny of Positive Thinking,” she wrote that society was placing an undue and inappropriate burden on patients that seemed to come out of popular beliefs about a mind-body connection.

I would find patients coming in with stories of being told by well meaning friends, "I've read all about this - if you got cancer, you must have wanted it." Others said, "I've been told that my personality must have caused my cancer and I guess I just didn't handle stress right in my life." Even more distressing was the person who said, "I know I have to be positive all the time and that is the only way to cope with cancer-but it's so hard to do. I know that if I get sad, or scared or upset, I am making my tumor grow faster and I will have shortened my life."

This pop psychology fills books, tabloids and talk shows, she wrote, and comes from an interest in mind-body connection. “It is important that people know that research simply does not back up these ideas,” she said. Obviously, having a rational personality and seeking appropriate and timely medical care when you have significant suspicious symptoms is important, but the idea that people can give themselves cancer or cause their own cancer from negative thoughts, internalized anger, or attitudes is not based on facts. As she wrote:

Studies do not support the myths about psychological causes of cancer and the role of emotions in tumor growth. The mind-body connection is fascinating because people hold such strong beliefs about it. I have come to feel that it is very much like religion. There are people who are "believers" and all the data in the world couldn't shake their faith. "Nonbelievers" simply are those who would not likely believe in the mind-body connection, even if the data were produced.

This belief is the source of the common response to blame the victims, though, and used to conclude when something bad happens to someone that they must have brought it on themselves. It’s a psychological method, she said, that allows us to say to ourselves:

"Well, it couldn't happen to me--it was his own fault." It produces a false sense of security that we can prevent events that are actually beyond our control.

So rule number one in coping is: "Don't believe you brought cancer on yourself." The research does not show that either personality or how you handle stress in your life raises risk of developing cancer. This is one of the myths that makes coping more difficult these days.

Rule number two is: "Don't believe that you have to have a positive attitude all the time and that sadness or worry will shorten your survival." This tyranny of positive thinking is also related to the "mind over matter" ideas of our society.

Things like cancer, diseases of aging and death often just happen, and dutifully following some ideal diet, lifestyle or mind-body wellness modality cannot change that. Rather than letting that be a scary thought, it’s empowering knowledge that can free us from the tyranny of living obsessed by fear of dying or from being taken advantage of by those promising optimal wellness.

Science can help you enjoy living life to the fullest and have a lot more fun doing it, just being you.

© 2008 Sandy Szwarc.

Bookmark and Share