The dark psychosomatic history of cancer part III

Since ancient times, the onset of cancer has been linked to depression and melancholia. Hopelessness, despair, and sorrow were regularly mentioned as a trigger of cancer until modern research refuted such a connection.

In the second half of the 20^th century, the idea of a cancer-prone personality became widespread in both the scientific and popular press. People who repressed their anger and emotions were said to be at higher risk of developing cancer. Behavioral therapy was proposed as a treatment and prophylaxis. Our investigation shows that some of the researchers who introduced these ideas were financially supported by the tobacco industry to turn attention away from causal links between smoking and cancer.

Another psychosomatic myth states that having a positive attitude or “fighting spirit” leads to a better prognosis. The notion that patients could cure themselves of cancer if they had the right attitude or tried hard enough to overcome it, became popular in the 1970s and 1980s. These ideas were not supported by scientific evidence and led to stigma, guilt, and frustration.

As we will discover, the history of cancer has surprisingly many of these episodes. We will therefore divide it into three parts. Today, in part III, we dissect the Hans Eysenck affair and reveal how important psychosomatic research on cancer was funded by the tobacco industry. 

Dissenting voices

In part II of this series, we saw how the idea of ‘fighting spirit’ and the cancer-prone personality reached its way into popular culture. As psychosomatic theories of cancer became widespread, however, more people started to speak out against it.

The American Cancer Society, for example, issued a statement in 1981 titled ‘Unproven methods of cancer management’ in which they criticized the work of Carl Simonton. “There appear to be several conceptual flaws in Dr. Simonton’s thinking”, the statement read, “particularly the premise that a patient contributes to developing his cancer and has a direct personal role in curing cancer.” It cautioned that such views may induce unnecessary guilt and clarified that “there is no evidence that reducing stress enhances the body’s rejection or containment of cancer.”

Three years earlier, writer and breast cancer patient Susan Sontag published her classic book ‘Illness as Metaphor’. It condemned the mythology surrounding illnesses for causing blame, guilt, and stigma. Sontag criticized how “cancer is regarded as a disease to which the psychically defeated, the inexpressive, the repressed  – especially  those who have repressed anger or sexual feelings – are particularly prone.”

In the scientific literature, an influential critique of the psychosomatic approach to cancer was published in 1978 by Bernard Fox from the National Cancer Institute. The paper was 89 pages long and painfully pointed out how weak the evidence was. Psychosomatic studies suffered from a long list of deficiencies Fox showed. Unvalidated questionnaires, lack of a control group, inappropriate statistics, and confounding by other well-known risk factors such as smoking, were rampant.

Then in 1985, a large prospective study was published in the prestigious New England Journal of Medicine. The study failed to find a relationship between psychosocial factors and survival from cancer. The authors concluded that “the biology of the disease appears to predominate and to override the potential influence of life-style and psychosocial variables once the disease process is established.” The editorial accompanying the article stated firmly:

“It is time to acknowledge that our belief in disease as a direct reflection of mental state is largely folklore.”

At the time, it seemed that the popularity of psychosomatic theories of cancer would quickly decline. There were however two controversial events at the end of the 1980s that changed that directory.

The Eysenck and Grossarth-Maticek controversy

First, there were the incredible reports of the German sociologist Ronald Grossarth-Maticek most of which were published during the 1980s. Grossarth-Maticek claimed to have conducted three major prospective studies on personality and cancer survival, one in the former Yugoslavian town Crvenka, and two in the German city Heidelberg.

The findings were incredible. The personality aspects Grossarth-Maticek had measured at the start of the study, almost perfectly predicted who would succumb to cancer many years later. When all three cohorts are taken together, those with a cancer-prone personality type had more than 100 times the risk of dying from cancer than those in the control group. In comparison, the relative risk of lung cancer for smokers compared to non-smokers is approximately 13. The results byGrossarth-Maticek simply seemed too good to be true. There was also no official funding source or a list of scientific collaborators for each of the studies, which is strange considering how large an undertaking the data collection must have been. For most scientists such as Bernard Fox, the studies were unreliable and best to be ignored.

Eventually, however, the reports by Grossarth-Maticek found their way into the scientific literature. They became the center of attention because of the support from one of the most influential psychologists of the 20^th century: Hans Eysenck.

We already met Eysenck in part II of the psychosomatic history of cancer as a collaborator of the Scottish physician David Kissen. Eysenck believed that personality features determined who would get cancer and that other risk factors, including smoking, were of lesser importance. He wanted to test his theory, but despite being one of the most cited psychologists of all time, he could not find collaborators. When David Kissen suddenly died of heart disease in 1968, Eysenck discovered that nobody was willing to continue his work on cancer. As he explained in one of his publications:

“Kissen’s untimely death caused me to look around for oncologists with whom I could work to continue and improve these early studies. The reaction of the medical establishment was entirely negative; nobody had any interest in working with me, or even in allowing me to approach and test their patients. I also wanted to extend these studies from patients already complaining of difficulties (although not yet diagnosed) to healthy people showing the cancer-prone personality. This too proved impossible through lack of interest on the part of the establishment, and I had to give up for the time being.”

When Grossarth-Maticek came forward with results that supported his maverick theory, it must have been like a godsent gift to Eysenck. This was the data he wanted, and he was going to use it.

Many of the findings of Grossarth-Maticek were published in journals that Eysenck had founded such as ‘Behaviour Research and Therapy’ or ‘Personality and Individual Differences’. Eysenck frequently appeared as a co-author and sometimes presented the findings on his own, even though he was not involved in the actual study design or data collection.  

With Eysenck as co-author, the claims and statements were no less extreme. Grossarth-Maticek and Eysenck claimed to have developed a form of behavioral therapy – ‘creative novation therapy’ – that could prevent people from dying of cancer. They suggested it should be used as a prophylactic in patients with a cancer-prone personality. They even went so far as to state that a leaflet and brief explanation could also prevent cancer. This simple intervention – which they called ‘bibliotherapy’ – reduced cancer mortality by 80%.

At this point, even proponents of a psychosomatic approach started to raise their eyebrows. In 1991, an entire issue of the journal ‘Psychological Inquiry’ (Volume 2, Issue 3, 1991) was devoted to dissecting the data of Grossarth-Maticek. One piece after the other argued why the data was unreliable. Most damaging was the report by the Dutch psychologist Henk Van Der Ploeg which had the fitting title: ‘What a Wonderful World It Would Be.’ Van Der Ploeg was part of an independent investigation of Grossarth-Maticek’s findings. When he reanalyzed the data, he could not replicate the original results. In his report he writes:

“Given the result that in this prospective longitudinal study the psychological and/or somatic data lost most of their predictive power after the moment at which the data were deposited and the moment at which the collection of mortality data was started, it was difficult not to suggest that ‘some data manipulation could have taken place.’”

Van Der Ploeg also found strange repetitions in the data, suggesting either manipulation or grave error. Hermann Vetter, the statistician who had collaborated with Grossarth-Maticek, also found suspicious patterns in the data. “I cannot avoid the conclusion”, he wrote, “that these data also have been produced artificially.” Lastly, an attempt to replicate the findings by Grossarth-Maticek, led by Manfred Amelang at Heidelberg University, failed to reproduce the findings. The questionnaires, for example, did not function as claimed.

That should have been it. In the early 1990s, it was clear to everyone involved that the data presented by Grossarth-Maticek was unreliable and misleading. But nothing happened. There were no retractations and no further investigations into research misconduct.

For the next chapter in the story, we must fast-forward to 2016, almost 20 years after Eysenck’s death. In that year, psychiatrist Anthony Pelosi received an invitation to write a contribution to the journal ‘Personality and Individual Differences’, to celebrate the centenary of Eysenck’s birth and his contributions to the study of personality. Pelosi, however, was one of the researchers who had drawn attention to the flawed data of Grossarth-Maticek and Eysenck back in the 1990s. In his contribution, he recounted this shameful history without holding back, calling it “one of the worst scientific scandals of all time.” Needless to say, this text was not accepted by the journal Eysenck founded. It was eventually published in the Journal of Health Psychology, whose editor called for an independent investigation into the matter.

Kings College London, the academic institution where Eysenck worked, responded by conducting a – rather limited – investigation in 2019. In short, it concluded that the critics were right. The report called the papers by Eysenck and Grossarth-Maticek “unsafe” and urged the responsible scientific journals to retract them. Several journals have done so but not all. There are still misleading reports by Eysenck and Grossarth-Maticek in the scientific literature and these are occasionally cited or included in meta-analyses.

Funding by the tobacco industry

There is one more chapter to the Hans Eysenck affair. From internal corporate documents released in US litigation, we now know that Eysenck received secret funding from the tobacco industry.

Since the 1960s Eysenck had questioned the link between smoking and cancer. He argued that heritable personality features were more important. “The importance of personality factors in the causation of disease”, he wrote in 1985, “has been disguised in recent years by rather simple-minded assertions regarding the importance of factors such as smoking and diet.” Eysenck even went as far as claiming that “lung cancer, in its early stages, may cause smoking.” In another publication, which has now been retracted, he investigated prejudice against smoking, arguing that “prejudice is much more dangerous to health than smoking!” He also complained about the media, claiming that “stress increased by constant repetition of evil consequences of smoking might constitute a self-fulfilling prophecy.” Eysenck was a perfect ally for the tobacco industry. At the time of his death, he was the third most cited psychologist of all time, and during his career, he happily used his influence to cast doubt on the harmful effects of smoking.

Eysenck, however, is not the only familiar name that shows up in internal documents of the tobacco industry. David Kissen, the Scottish physician who claimed that lung cancer patients had a poor outlet for emotional discharge, was funded by the Tobacco Manufacturers’ Standing Committee since at least 1958. Dr. Claus Bahnson who described cancer as “a regressed – effort on the part of the psychological organism to replace a recent loss” received funding from the Council for Tobacco Research in the early 1970s. Dr. Maria Blohmke, the first female president of the Deutsche Sozialmedizinische Gesellschaft, obtained research funding from R.J. Reynolds Tobacco Company for her study on “psychosomatic aspects of lung cancer.” H.J.F. Baltrusch, former president of the European Working Group for Psychosomatic Cancer Research, also received funding British American Tobacco.

The inescapable conclusion is that many of the key researchers who developed and promoted the idea of a cancer-prone personality received secret research funding from the tobacco industry.

Grossarth-Maticek was also on the radar of the tobacco industry, but they initially rejected him because his research was seen as untrustworthy (one internal document calls him a charlatan). Eventually – likely because of the support from Eysenck, – tobacco companies were willing to take the risk. There is even an internal document where one statistician by the name of Peter. N. Lee says to have ghostwritten a paper for Grossarth-Maticek. “…My name does not appear at all on the paper”, Lee explained, “it being felt that my tobacco industry association may affect its acceptance.”

The name of Carl Simonton also pops up. He had sent R.J. Reynolds a research proposal but it is unclear if it eventually got funded or not. One internal document states that Simonton      believed that “there was unduly exclusive emphasis on chemical or physical carcinogens in studying the origins of cancer.” It also states that he “was not interested in getting U.S. Government grants because he disliked the restrictions regarding the conduct of research which were imposed on such grants.”

The main goal of the tobacco industry seems to have been to draw attention away from the link between smoking and cancer. Psychosomatic research was useful for them in that regard because it generated a lot of attention in the media even if its scientific foundations were shaky. A letter from executives of Philip Morris to Gotthard Booth gives insight into their line of reasoning. It states:

“Our hypothesis that smokers come from a cancer prone population would either be confirmed or denied. The former would help us and the latter would leave the situation just where almost everybody thinks it is anyway.”

The Spiegel study in the Lancet

There was a second event that revitalized interest in a psychosomatic approach to cancer. In 1989, David Spiegel and his colleagues at Stanford University published the results of an unplanned survival analysis. The results were a big surprise and have since been cited more than 3000 times.

In the 1970s, Spiegel and colleagues had conducted a randomized trial of supportive group therapy for patients with metastatic breast cancer. The goal of the therapy was to support patients to live life as fully as possible in the face of death. The study was published as a success. After treatment, patients in the intervention group reported less anxiety and depressive symptoms than those in the control group.

As psychosomatic ideas received more attention because of the books by Simonton, and Siegel, the Stanford group was tempted to see if their intervention also increased survival rates. This was never the goal of the treatment but it was thought that negative results would be useful, for example, to debunk false ideas. To their surprise, however, there was an effect and not a small one either. Patients in the intervention group lived significantly longer. The results were published in The Lancet and became the hot topic of the day. Eysenck, for example, frequently referred to this publication to argue that the data of Grossarth-Maticek was not that extreme after all…

There were, however, some problems with the data. Most notable was the survival curve of the control group. People who did not receive the intervention seemed to have died quickly and around the same time. As explained by Bernard Fox: “the curve dropped rapidly to zero survival, with no patients having relatively longer survival than the rest. This was of interest because most survival curves are skewed, having a tail that indicates a few people with unusually long survival compared with the rest of the group.”

When Fox compared the results with survival rates from a regional database of patients with metastatic breast, it became clear that the control and not the intervention group was unusual. Fox didn’t think these results would be replicated. He was right.

Several large studies were initiated, trying to match the design of Spiegel and colleagues as closely as possible. These studies convincingly showed that supportive group therapy did not prolong survival. Even Spiegel’s own replication study came to that conclusion. The hype was based on nothing.

Because the trials were measuring survival, it took almost a decade before the results were in. In the meantime, patients were frequently told that joining a support group to encourage the expression of emotions, would help them live longer. It caused unnecessary pressure and guilt in fragile patients who were already struggling with a horrific disease.

Dr. Pamela Goodwin, one of the authors of a failed replication attempt, explained that rather than being disappointed by the negative results, she saw them as liberating. Her team started the study because they saw how patients were pressured by friends or family to join support groups. Another doctor explained in the New York Times how a woman, whose husband died of lung cancer, was in agony because she felt more could have been done to prevent his death if he only joined a support group: ”She said, ‘I feel so guilty. I never got him to go to one of those groups.’”

Impact on patients

Although there are no scientific studies on the impact of psychosomatic myths on cancer patients, anecdotal evidence suggests that it caused a lot of damage. When discussing the psychoanalytic literature of the 1950s and 1960s, one review noted:

“In retrospect, one cannot help but feel that insult was added to injury by suggesting to cancer patients that immature personality development or a real or symbolic loss precipitated their malignant illness.”

Many of the psychosomatic myths still linger on to this day. A 2001 survey of breast cancer survivors found that “Despite lack of evidence substantiating stress as a cause of breast cancer, many breast cancer survivors believed stress caused their cancer.” When asked what prevented cancer recurrence 60% attributed it to a positive attitude, and only 3.9% to tamoxifen.

In his 2011 book “When the body says no, the cost of hidden stress”, Hungarian physician Gabor Maté enthusiastically refers to the dubious research mentioned in this blog series. Maté uses the work of Kissen, Temoshok, and Gossarth-Maticek to claim that “repression of anger increases the risk for cancer for the very practical reason that it magnifies exposure to physiological stress.” Despite the lack of scientific evidence, this idea remains attractive. Maté’s book was an international bestseller and was translated into 15 languages.

Myths about positive thinking also impact patients to this day. “It’s a real problem when cancer patients feel they must stay positive”, Dr. Keith Block explained in an interview, “an example of a metaphor that causes more self-blame. When patients can’t stay positive, they think this will worsen their recovery, so their despair gets even worse.”

Back in 1985, when the belief in ‘fighting spirit’ was popular, Neil A Fiore, a former cancer patient, and psychotherapist wrote an angry letter in the New England Journal of Medicine, saying:

“It is criminal to suggest to patients that they could cure themselves of cancer if only they had the right attitude or tried harder.”

Not much seems to have changed since then. An online testimony from a cancer patient in 2020 explains how sometimes “the good vibes, encouragement and positivity can go too far” it reads:

“What happens when dark feelings are met with toxic positivity is tremendous guilt. At least in my case, when I share the challenges I’m feeling and I’m met with toxic positivity, I feel so much guilt that I just completely shut down. I start internalizing my feelings and begin to worry that all these negative emotions will bring back cancer;”

In one of our future blog posts, we hope to delve into the problematic history of positive psychology. If you would like to receive a notification each time a new blog post appears, you can enter your email address in the ‘Subscribe’ section below.

Read part I and part II of the psychosomatic history of cancer on this blog.

An overview of all articles in this series can be found at the bottom of the introductory article.