Introduction

Breast cancer is a significant cause of morbidity and mortality among women; it is the second leading cause of death due to cancer in women, exceeded only by lung cancer (American Cancer Society 2007) [1]. The fact that patients with apparently similar tumours at the time of presentation differ significantly in the time of relapse and overall survival implies that the determinants of survival could be broader than initially assumed in the purely medical framework. The question of whether psychosocial parameters could have an effect on the clinical outcome of cancer in general has yielded a large body of research devoted to this issue and has divided the medical community into believers and non-believers. We reviewed articles on the psychosocial correlates of cancer relapse and survival as well as on the impact of psychological intervention on survival from 1979 to 2006.

Materials and methods

Literature search

We performed a systematic search of the literature to obtain original studies that investigated possible associations between psychosocial/psychological factors and survival among patients with breast cancer. The relevant studies were identified by the use of the PubMed database [2]. In our primary search, conducted on 16 to 17 May 2006, we used two different strategies with the following key words: first, breast cancer AND (prognosis OR outcome) AND (psychological OR psychosocial OR cognitive) factors; and second, breast cancer AND (prognosis OR outcome) AND personality. Afterwards, we attempted additional searches using various combinations that included the following key words as well: positive outcome, optimism, breast cancer diagnosis, social support, decreased mortality, confounding variables, patient autonomy, patient support, disease recovery, social networks, quality of life, beneficial effect, survivors. Finally, we extended our search by looking into the PsychINFO database to find relevant studies [3].

Study selection

Our systematic review included original studies written in the English language; thus, reviews, editorials, letters, and case reports were excluded from our study. Studies included in our review contained quantitative data regarding the association of psychological/psychosocial factors with breast cancer disease outcomes (survival and recurrence). Moreover, we included studies that examined the impact of psychological intervention programs on patient survival. We did not use any limitations regarding study sample size, study design, and specific measures of outcomes used in the various studies. We excluded studies that focused on the impact of breast surgery and adjuvant therapy on patients' quality of life, psychological and psychosocial well-being, and survival. Studies referring solely to the influence of biological and organic factors on survival were not included in our review. We also left out studies that concentrated on the impact of psychological, psychosocial, or cognitive factors on the onset of the disease. Finally, we did not include studies that focused on the evaluation of intervention programs to enhance self-management and coping with illness.

Data extraction

From the studies that were included in our review we extracted data regarding the date of publication, the type of the study, the setting of the study, the study population, the aim of the study, the follow-up period, the method used to assess and measure psychological and psychosocial factors, the disease outcome (that is, survival/disease free survival and recurrence), the specific psychological/psychosocial factors examined as potentially influencing breast cancer survival or prognosis, and the main findings regarding statistically significant associations between those factors and breast cancer prognosis/outcome.

Results

In total we identified 37 studies examining psychosocial factors and breast cancer survival; 31 descriptive analytical studies and 6 studies with psychological intervention. In Tables 1 and 2 we summarize the 31 studies regarding psychosocial/psychological factors and survival or recurrence among patients with breast cancer [4–34]. In Tables 3 and 4 we summarize the six studies regarding the impact of psychosocial intervention on survival of breast cancer patients [35–40]. In Table 5 we summarize the association of psychosocial factors with survival/recurrence. A total number of 61,611 female patients with breast cancer were examined regarding psychosocial factors and survival (52,857 of them in retrospective studies). Of the total 31 identified descriptive analytical studies, 15 were conducted in the US [5–8, 10, 11, 15, 16, 22, 24, 26, 27, 31, 32, 34], 2 in Canada [9, 29], 1 in Israel [18], 2 in Australia [23, 30], 1 in Japan [19] and the rest of them in Europe (6 in the UK, [4, 12–14, 21, 25] 1 in Switzerland and Germany [17], 1 in Denmark [28], 1 in Belgium [20] and 1 in Finland [33]). Two studies examined the same cohort of women to assess the impact of psychosocial factors on survival [26] and the effect of cancer specific beliefs on survival [27]. Out of the 31 studies that did not employ psychological intervention, 12 examined patients with all stages of breast cancer [6, 7, 9, 12, 15, 19, 22, 26–28, 32, 34]. Five and two studies examined patients with metastatic/recurrent cancer [5, 8, 11, 23, 24] and invasive stage I or II to III cancer [20, 31], respectively, whereas twelve studies examined patients with early stage cancer [4, 10, 13, 14, 16–18, 21, 25, 29, 30, 33] characterized either as operable, stage I and II in the TNM classification (that is, the staging system based on tumour size, lymph node involvement, and the presence of metastasis), or localized regional and non-metastatic. Fourteen studies focused on patients' reactions to having cancer and their feelings whereas nine focused on social factors (that is, marital status, social ties, religion) that can offer significant support to breast cancer patients, and seven studies focused on both. One study examines the effect of beliefs about cancer curability.

Table 1 Characteristics of studies examining psychosocial factors associated with breast cancer survival
Full size table
Table 2 Methods and outcomes of studies examining the association of psychosocial factors and breast cancer survival
Full size table
Table 3 Characteristics of studies examining the effect of psychological intervention on breast cancer survival
Full size table
Table 4 Methods and outcomes of studies examining the effect of psychological intervention on breast cancer survival
Full size table
Table 5 Association of the examined psychosocial factors with survival/recurrence
Full size table

Out of the 31 studies, 25 (80.6%) revealed that various psychological factors are significantly associated with survival/recurrence. Increased survival is associated with role functioning (one study [29]), fighting spirit (one study [30]), joy (one study [11]), depression (one study [5]), perceived social support (two studies [15, 26]), social support (four studies [5, 15, 31, 34]), minimization (two studies [23, 33]), adjustment (one study [18]), denial (one study [29]), anger (one study [5]), guilt (one study [5]), extroversion (one study [9]), expressive activities (one study [9]), participation in religious/nonreligious groups (one study [6]), psychiatric symptoms (one study [5]), hobbies (one study [19]), female child (one study [19]), and marriage (three studies [6, 19, 32]). Decreased survival is associated with stressful events (one study [29]), anxiety/stress (one study [7]), hopelessness (one study [7]), higher perceived emotional support (one study [33]), depression (five studies [11, 21, 24, 28, 33]), repressive defensiveness (three studies [7, 24, 33]), positive constructive daydreaming (one study [7]), denial/avoidance (one study [33]). Cognitive functioning (one study [29]), stressful events (one study [12]), anxiety (one study [20]), hopelessness (three studies [4, 10, 21]), fatalism (one study [4]) and anger/hostility (one study [9]) were associated with higher recurrence rates. Fighting spirit (one study [4]), stressful events (two studies [13, 25]), denial (one study [4]) and expressive activities (one study [9]) significantly reduced recurrence rates.

As seen in Table 5, for most of the separate psychosocial factors examined with regard to breast cancer survival and/or recurrence there was only one study supporting the association (either a positive or a negative impact on survival/recurrence). For several psychosocial factors (specifically, depression, social support, marriage, perceived support, stressful events, denial/avoidance, minimizing, repressive defensiveness, anger/hostility and hopelessness/helplessness) more than one study reporting a statistically significant association with breast cancer survival. For five of these eight psychosocial factors (stressful events, depression, perceived social support, anger/hostility, denial/avoidance) there are contradictory results (showing both negative and positive effects) on survival/recurrence. For example, while five studies show that depression is associated with decreased survival, there is one study where depression seems to increase survival. Perceived social support, social support, minimization and marriage were associated with prolonged survival in more than one study. Similarly, repressive defensiveness and depression are associated with decreased survival in more than one study. As far as recurrence is concerned, only stressful events appeared to reduce recurrence rates while hopelessness appeared to increase recurrence rates in more than one study.

In the 6 studies that used psychological intervention, a total of 389 female patients with breast cancer were examined (34 of them in a retrospective study). Psychological intervention varied from study to study, including group therapy, behavioral/cognitive therapy, peer support/family therapy, counseling, mental imagery, meditation, and psychology classes. Three of these studies were conducted in the US [35, 36, 39], two in Canada [37, 40] and one in Australia [38]. Four studies examined patients with metastatic breast cancer, one patients with early breast cancer and one patients with breast cancer of any stage. Two studies showed that there is a survival benefit for those patients that received psychological intevention [35, 39], while in four studies there was no influence on survival [36–38, 40].

Discussion

From the results of our review of the literature there are conflicting data regarding the possible association of psychosocial factors and survival. Various methodological issues of the examined studies have to be taken into account to interpret their results. Hence, while most of the examined studies in our review are prospective, the majority of the patients studied were in the retrospective studies. This means that the sample size of most of the prospective studies is relatively small. The follow-up period of the reviewed studies also differs significantly. One must be cautious when drawing conclusions from a variety of studies that used different assessment tools to evaluate the influence of psychosocial factors on the survival of patients with breast cancer. For example, in the study by Goodwin and colleagues [29] a psychosocial parameter (avoidance) was found to be important when assessed with a particular questionnaire, but this finding was not substantiated when using another questionnaire. Moreover, in the same study, the number of statistically significant associations of psychosocial factors with breast cancer survival actually found was less than that expected by chance.

Despite these limitations, some useful data regarding the role of psychosocial factors in breast cancer survival become evident. There is an overlap between studies that showed statistically significant associations and those that did not. This means that while a study may have reported a statistically significant association for a psychosocial parameter, it did not show significant correlations for other parameters. The small patient numbers in these studies may have precluded statistical significance. In the majority of the studies that showed a statistically significant association between a psychosocial factor and survival/recurrence, this association was either positive (increased survival/recurrence) or negative. For only four of the studied psychosocial parameters (depression, perceived social support, anger/hostility, denial/avoidance) were both positive and negative results evident. Thus, with regarding to depression, four studies showed increased patient mortality, five studies showed no effect at all and one study showed a contradictory effect, that is, better survival.

Parameters associated with better breast cancer prognosis are social support, marriage, minimizing and denial. Data from the literature regarding the effect of these factors on the survival of other types of cancer vary. Thus, in a recent systematic review no statistically significant influence of psychological coping on cancer survival (various types of cancer included) was found [41]. On the contrary, data from patients with melanoma show that married patients and patients adjusting with minimization also survived longer [42]. In our review, depression and constraint of emotions are associated with decreased breast cancer survival; the role of these factors has not been verified in all studies. In a prospective study of patients with various types of cancer, depression was found to be associated with cancer prognosis after 30 months of follow-up [43].

Other studies included in our review have shown limited and/or contradictory data for parameters such as fighting spirit, having a female child, showing adjustment, having hobbies, coping, role functioning, stressful events, stoic acceptance, fatalism, locus of control and anger.

Understanding the definitions of these psychosocial parameters is crucial. Closely related terms are not synonymous. Thus, one has to differentiate denial and minimizing (associated with better survival) from constraint of emotions (associated with shorter survival). Denial refers to the belief that one has no disease, while minimization refers to minimizing the seriousness of the disease. The constraint of emotions is not synonymous with denial/minimization. One may constrain his/her emotions while not denying the disease.

One cannot emphasize the need for repeated evaluations of the psychosocial factor over a protracted time period. This holds true as these factors may change over time, that is, they are dynamic in nature. Chronic versus acute changes in the psychosocial factors were not a particular focus in the reviewed studies, apart from one. Neither was any specific mention of the histological type of breast cancer made in these studies. In contrast, the majority of the studies reported on breast cancer stage; it seems that no association between psychosocial factors and the stage of the neoplasm (metastatic versus non-metastatic) was evident.

When examining the effects of various psychosocial factors on the outcome of patients with breast cancer, other possibly interacting factors (either psychosocial or biological) may be deemed to be confounders. It should be kept in mind that when excluding these factors that are potentially in interplay with the examined parameter from the analysis, the effect of the psychosocial factor might not become evident. For example, in breast cancer patients with a hormone (estrogen and/or progesterone) receptor positive status (biological factor), life events were related to recurrence of breast cancer, while such a relationship did not occur in women with hormone receptor negative breast cancer [12]. In an analogy, the effect of one psychosocial parameter may only become evident when examined in the context of other interacting psychosocial parameters. An example is shown in the study by Weihs and colleagues [24]; the effect of chronic anxiety varies when various levels of constraints of emotion are examined. Other studies do not confirm the interplay of psychosocial and biological factors [16].

The effect of various psychological interventions on the survival of breast cancer was examined in six studies and two of them showed a positive result. While psychological interventions improve the quality of life of patients with breast cancer, [35] their effect on actual survival is less obvious. Thus, it seems that psychological interventions have minimal effect on the prognosis of breast cancer patients. One has to acknowledge that the numbers of patients in the studies using psychological intervention are too small to reach definite conclusions. Data from the literature regarding the effect of psychosocial interventions on cancer survival are scarce; even data on the effects of these interventions on quality of life are contradictory [44–46].

Conclusion

The majority of studies show a significant relationship between psychosocial factors and survival, but the actual psychosocial variables related to survival are not consistently measured across studies and the findings for many of the psychosocial variables with survival/recurrence are not consistent across studies. In particular, more research is probably warranted regarding the role of social support, marriage, minimizing, denial, depression and constraint of emotions on breast cancer survival. Adequately powered multicentre studies, the use of a few valid assessment tools and meta-analytical approaches may be necessary to show the potential roles of various psychosocial factors in breast cancer outcome.