Does group therapy mean longer life?

Researchers examine the impact group therapy may have on cancer patients.

Psychologists don't need to be convinced of psychotherapy's power for helping people deal with the stress, trauma and anxiety of deadly diseases such as cancer. But what about its ability to prolong life?

That's the critical question being asked by varied research teams around the United States, Canada and Europe. They're trying to replicate the 1989 report by Stanford University's David Spiegel, MD, and his colleagues that women with metastatic breast cancer who participated in their study lived an average of 18 months longer if they participated in supportive group therapy than if they didn't (Lancet, Vol. 2, No. 8668, p. 888_891).

Two other studies in the early 1990s reported similar findings. Fawzy Fawzy, MD, and his colleagues found that patients with malignant melanoma lived longer if they participated in a brief group psychoeducational course. And psychologist Jean Richardson, PhD, reported that an educational, supportive intervention improved survival rates in people with lymphoma.

But none of these early studies set out to test directly whether therapy could prolong the lives of cancer patients--the survival analyses were conducted many years after the original studies ended and all three have methodological limitations making generalization of their findings unwarranted, say researchers.

Many of the new studies, including one by Spiegel and his colleagues, are larger, better designed and some, along with examining survival rates, are searching for mechanisms by which psychosocial interventions may promote survival.

To date, only one research team has reported any findings--late last year psychologist Alastair J. Cunningham, PhD, and his colleagues at the Ontario Cancer Institute in Canada found no difference in survival between breast-cancer patients who participated in group therapy and those who did not (Psycho-Oncology, Vol. 7, p. 508_517).

But his study was smaller than Spiegel's and, along with not enhancing survival, his intervention provided no psychological boost to participants. Therefore, most researchers agree that the jury will remain out until several large-scale, well-controlled studies are complete.

"I think we have enough data to ask the question 'Do psychosocial interventions prolong life?'" says Ohio State University psychologist Barbara Andersen, PhD, who is conducting a study of her own. "The studies that are out there now weren't designed to look at this question. It's very different to find a survival effect by happenstance versus designing a study to see if you get a survival effect."

Although Cunningham's study of 66 women with metastatic breast cancer didn't find a survival effect, he hasn't ruled out the possibility that psychosocial interventions can help prolong life. But, he believes, it is only a subset of women who benefit.

In fact, in a study that's now in press, Cunningham finds that women who are motivated and participate actively in his psychosocial intervention--a small subset of all women with cancer--live twice as long as predicted by oncologists based on their medical records. And among those motivated women, those who were most motivated live more than three times longer than predicted. "When you separate out the super stars from the people who drag their feet, you see a very strong effect" of the intervention, he says.

He believes that researchers must begin to examine the intensity of people's therapy participation if they ever want to find significant increases in survival, he says.

As for Spiegel's study, psychologist Bernard Fox, PhD, has suggested that the control group was unintentionally biased--the women died at a far faster rate than women in a national sample of women with breast cancer (see Psycho-Oncology, Vol. 7, p. 361_370). Spiegel disagrees (and wrote an accompanying response in the same journal issue), noting that it was the sample as a whole that may not have matched national norms. Thus Fox's critique speaks to the generalizability of the finding, not its validity. However, he adds that his one study was never intended to be definitive.

Indeed, before ruling out the idea that all people can benefit from psychosocial interventions, researchers such as Ohio State's Andersen and the University of Rochester's Gary Morrow, PhD, want to see the results of many more studies. Both have their own research under way--Morrow with a study of more than 400 women with early-stage breast cancer and Andersen with, by study's end, 230 women with stage II or stage III breast cancer.

If people in their studies benefit from the interventions, Andersen, for one, hopes she will have something to say about why. In her study, she and her colleagues are looking for mechanisms that might account for any improvements they may see in survival. In particular, they're measuring changes in the women's immune and endocrine function, and preliminary results indicate that immune function in women who attend group therapy is more active than in women who don't participate in group therapy. But this line of research is not without controversy.

In fact, oncologists are still debating whether or not the immune system is even involved in fighting cancer. And if it is, it may be because the immune changes that occur with psychosocial interventions may not be of sufficient magnitude to affect the disease process, says Andersen.

"To think a biologic response measured 4, 8 or 12 months after psychosocial treatment would be correlated with survival outcome two, three, four years hence is stretching it a bit," she admits.

In the study by Fawzy and his colleagues, for example, the researchers found that their intervention boosted the immune system, but that boost was not correlated with the increased survival rates they saw later.

"The leap across the biobehavioral boundary should be made with great caution and immense humility," says the University of Rochester's Morrow.

Others are bolder. Spiegel, for one, believes there's plenty of evidence that the immune system is involved in fighting cancer--several new cancer drugs even work by way of the immune system.

At the same time, he notes that psychoneuroimmune effects on cancer have not been proven and he doesn't rule out other mechanisms by which psychosocial interventions may impact health. For one, the interventions may influence the endocrine system, which is involved in stress responses and could be linked somehow to cancer progression.

And it may be that there's no biological mechanism at all, says Andersen. In fact, it's most plausible that psychosocial interventions reduce stress and enhance positive health behaviors, such as diet, exercise and compliance with cancer therapy.

"I would not underestimate the impact of behavioral variables in keeping people on target with their therapies," she says. "If behavioral interventions can affect that, it's a pretty cheap investment."

There may also be a link between immune function and treatment compliance, says Andersen. It could be that therapy boosts the immune system, not enough to fight the cancer, but just enough so that patients can tolerate a full course of cancer treatment. There's no evidence for this hypothesis yet, she admits, but there is a link between survival and treatment compliance. And one of the main reasons patients can't comply with a full treatment regimen is that their immune system becomes too compromised, says Andersen.

At the moment, there are more questions than answers about whether psychosocial interventions can help fight cancer, says Michael Stefanek, PhD, director of the Basic Biobehavioral Research Branch at the National Cancer Institute, who started a study of his own in this area when he was at Johns Hopkins University.

When Spiegel's study came out, "some dismissed it out of hand--looking for weaknesses in the methodology," says Stefanek. "Others embraced it enthusiastically without looking closely at the data. Now we're looking more carefully and asking, is there a difference [in survival]? If there is, what's the mechanism? And is the effect related to a certain diagnosis?"

APA Home Page . Search . Site Map