The relationship between depression and cancer survival

The relationship between depression and cancer survival

Milo, Karen M

Numerous studies have shown that psychosocial variables are associated with medical outcomes in cancer patients and other patient populations. For example, cancer patients who are married or have other sources of social support have been found to have lower rates of morbidity and/or mortality.1,2 Studies with various patient populations have shown a relationship between style of coping and survival, with most studies suggesting that denial is protective.3,4 Patients’ outlook on their disease (e.g., stoic acceptance versus fighting spirit) has been shown to predict medical outcomes in many though not all studies investigating these factors in patients with cancer.5 Psychiatric symptoms and negative mood states (e.g., depression and anxiety) have been associated with less favorable medical outcomes in various patient populations including HIV6 and cardiac7 patients. In research involving cancer patients, the relationship between emotional distress and medical outcomes has been less clear.


Of twenty-one studies investigating the relationship between depression or related psychological variables and survival in cancer patients (Table 1), the majority of those that included measures of depression found that depression is associated with decreased survival.8-15 Three studies failed to find any signifiant relationship between depression and survival.14,15,16 One of the studies that failed to find a significant relationship revealed a trend toward decreased survival14 and another revealed a trend toward increased survival15 in cancer patients who scored higher on measures of depression. Moreover, two studies indicated that higher scores on measures of depression were related to increased survival.3,16

Studies of other psychological variables related to depression and survival in cancer patients also have produced inconsistent results. Three studies linked hopelessness and helplessness to decreased survival,3,4,17 while one found no significant relationship between hopelessness and survival.18 One study of breast cancer patients linked joy to increased survival.19

Some of the inconsistency can be attributed to methodological differences. Some studies predicted only group membership (e.g., survivors versus non-survivors; long-term survivors versus short-term survivors) rather than using more powerful survival analyses. Many had small sample sizes or few subjects who were depressed. The researchers employed various instruments to measure depression, some of which can be criticized for being insensitive, nonspecific (e.g., combining depression and anxiety or depression and coping), or susceptible to response bias from self-reports. The timing of the assessments varied: some subjects were assessed shortly after diagnosis; others were assessed much later in treatment. Emotional distress tends to be highest shortly after diagnosis20 and possibly just before bone marrow transplant.21 Additionally, key information about the patients often was unreported. For instance, the medical variables in some studies may not have been sensitive enough to capture medical prognostic factors. The studies typically did not report what, if any, psychiatric or psychosocial treatment patients identified as depressed received. If patients received treatment and recovered from depression, this could obscure a relationship between prior depression and later survival. Finally, the patient populations were heterogeneous, and the relationship between depression and survival may differ for patients with different types of cancer.


Of the 21 studies that investigated the relationship between emotional variables and survival in cancer patients, five involved patients who were treated with bone marrow or stem cell transplant. These studies of patients with leukemia and other malignancies have yielded more consistent results. In 1991, Colon and his colleagues demonstrated that depressed mood prior to undergoing allogeneic bone marrow transplant was associated with shorter survival times following transplant in 100 patients with acute leukemia.8 In 1998, a study of 100 breast cancer patients found that depression was associated with shorter survival times following autologous stem cell transplant.22 Specifically, patients with clinically significant elevations on a depression measure prior to transplant had a 49% greater chance of dying within the next two years. A subsequent study of 200 cancer patients undergoing bone marrow or stem cell transplant for various malignancies found that depression was associated with a 25% lower survival rate.21 Andrykowski and colleagues24 failed to find such a relationship in 42 patients with acute or chronic leukemia who underwent allogeneic bone marrow transplant. However, these authors found that anxious preoccupation was associated with shorter survival times. In another study, decreased hopefulness was found to predict shorter survival times in 31 patients who underwent allogeneic bone marrow transplant.25 Taken together, these studies suggest that there is a significant relationship between emotional distress, especially depression, and survival following bone marrow transplant after controlling for relevant medical factors.


To date, studies investigating the relationship between emotional distress, particularly depression, and cancer survival have yielded mixed results. The majority supported the conclusion that depression has a negative effect on survival; some studies failed to find a significant relationship; a few found the opposite relationship. The data on patients who underwent bone marrow transplant for leukemia or stem cell transplant for other malignancies, though sparse, consistently showed that emotional distress (either depression or anxiety) was related to decreased survival.

There are several reasons to suspect that a predictive relationship exists between depression and cancer survival. First, studies of patients with other diseases have demonstrated an association between depression and survival. Second, depression has been shown to affect both immune and endocrine functioning, which in turn, can affect cancer progression.28 Third, depression and anxiety have been shown to affect compliance with medical treatment, though not necessarily in a consistent direction.26,27 In the studies reviewed, information regarding the psychosocial treatments provided to patients who were significantly distressed typically was not provided. Thus, it is possible that psychiatric problems that are successfully treated do not affect later survival. Patients who exhibit clinically significant levels of emotional distress after adjusting to the diagnosis cancer should be treated with psychiatric and/or psychological interventions to decrease their symptoms and possibly to extend their survival. Patients with milder symptoms might benefit from support groups or psychotherapy groups designed for cancer patients.29

Copyright Rhode Island Medical Society Aug 2003

Provided by ProQuest Information and Learning Company. All rights Reserved.