Elsevier

The Lancet

Volume 354, Issue 9187, 16 October 1999, Pages 1331-1336
The Lancet

Articles
Influence of psychological response on survival in breast cancer: a population-based cohort study

https://doi.org/10.1016/S0140-6736(98)11392-2Get rights and content

Summary

Background

The psychological response to breast cancer, such as a fighting spirit or an attitude of helplessness and hopelessness toward the disease, has been suggested as a prognostic factor with an influence on survival. We have investigated the effect of psychological response on disease outcome in a large cohort of women with early-stage breast cancer.

Methods

578 women with early-stage breast cancer were enrolled in a prospective survival study. Psychological response was measured by the mental adjustment to cancer (MAC) scale, the Courtauld emotional control (CEC) scale, and the hospital anxiety and depression (HAD) scale 4–12 weeks and 12 months after diagnosis. The women were followed up for at least 5 years. Cox's proportional-hazards regression was used to obtain the hazard ratios for the measures of psychological response, with adjustment for known clinical factors associated with survival.

Finding

At 5 years, 395 women were alive and without relapse, 50 were alive with relapse, and 133 had died. There was a significantly increased risk of death from all causes by 5 years in women with a high score on the HAD scale category of depression (hazard ratio 3·59 [95% CI 1·39–9·24]). There was a significantly increased risk of relapse or death at 5 years in women with high scores on the helplessness and hopelessness category of the MAC scale compared with those with a low score in this category (1·55 [1·07–2·25]). There were no significant results found for the category of “fighting spirit”.

Interpretation

For 5-year event-free survival a high helplessness/hopelessness score has a moderate but detrimental effect. A high score for depression is linked to a significantly reduced chance of survival; however, this result is based on a small number of patients and should be interpreted with caution.

Introduction

There have been many suggestions of a role for psychological response in the outcome of cancer, but little scientific evidence. A few studies examined the influence of psychiatric symptoms on disease outcome in patients with breast cancer. The presence of psychiatric symptoms, including emotional distress, was significantly related to poorer prognosis in two studies,1, 2 whereas three found no relation between psychiatric symptoms and disease outcome.3, 4, 5 These studies were limited by their small sample size (most had fewer than 200 patients) or short period of follow-up, so reliable assessment of the effect of psychological response in women with early-stage breast cancer is difficult.

A few studies suggested that women who develop breast cancer may be more likely than those who do not to suppress negative emotions6, 7, 8–the type C personality. Subsequent research suggested a cancer-prone personality type characterised by abnormal inhibition of emotions and inability to express anger.9 Despite flimsy evidence for the cancer-prone personality, this research has been enthusiastically received by the lay press and by practitioners of alternative therapies. This psychological response has been implicated as a factor that may influence the development of cancer, but whether it is also associated with survival is not known. In one of the earliest prospective studies of psychological response and survival in breast cancer, Greer and colleagues found a significant link between prognosis and psychological response.10 Patients whose psychological responses were categorised as fighting spirit or denial were significantly more likely to be alive and relapse-free 5 years after diagnosis of an early-stage breast cancer than those whose response was of helplessness or stoic acceptance. These results were still significant at the 10-year follow-up.11 Nevertheless, this early study has several shortcomings: there were only 57 patients and there was no adjustment for the important prognostic variable of lymph-node status, because few data on node status were available at the beginning of the study. Furthermore, the original assessment of psychological response was restricted to a brief open-ended question rather than a more detailed and psychometrically reliable measure.

Our study aimed to replicate these findings in a large cohort, with a prevalidated assessment of psychological response, and control for all known prognostic factors. In addition, we investigated whether there were any associations between the individual dimensions of the mental adjustment to cancer (MAC) scale and overall or event-free survival, and similarly for the Courtauld emotional control (CEC) scale. Women were followed up for a minimum of 5 years.

Section snippets

Participants

Women with early-stage breast cancer (stages I and II) who presented consecutively at the Royal Marsden Hospital, London and Sutton, UK, were invited to participate in the study. Eligibility criteria for the cohort were: age between 18 and 75 years; ability to complete a questionnaire; awareness of their cancer diagnosis; no apparent serious intellectual impairment; and diagnosis 4–12 weeks before the inclusion date. All patients meeting the eligibility criteria were identified from hospital

Procedure

Patients were assessed by self-administered questionnaires selected for validity, reliability, and previous use in this population. Participants completed the study measures at 4–12 weeks after diagnosis and 1 year later. The patients were recruited into the study in the context of screening for eligibility for a psychological intervention study, and were unaware of the hypotheses being investigated in this follow-up study. This approach reduced any bias in the reporting of psychological

Statistical methods

The MAC scale was analysed in two different ways. First, to replicate the analysis of Greer and colleagues,10 each patient was classified as having one predominant response from the categories of fighting spirit, helplessness or hopelessness, anxious preoccupation, and fatalism. To classify a patient's predominant response, each of the MAC subscale scores was converted to a z-score according to the mean and SD. Each patient was then assigned to a group representing her predominant response

Results

578 breast-cancer patients were enrolled into the study (table 1). The mean age of the women was 55 years (SD 10·6). Of the 415 women for whom social class could be defined, 282 (68%) were in classes I to III-non manual. 64% were married.

The overall median follow-up was 4·93 years (IQR 4·58–5·18). 160 (27·7%) women relapsed and in these women the median time to relapse was 2·12 years (1·08–3·25). 133 (23·0%) had died by 5 years of follow-up, 122 (21·1%) of breast cancer. In the remaining 11

Discussion

In this large population-based cohort study, the psychological data are largely free from measurement bias because the women were approached in the context of screening for an intervention trial21 rather than an observational study of the association between their psychological response and survival, and so they were unaware of the study hypothesis. Only one patient was lost to follow-up after 5 years. Several questionnaires were used in the data collection, and the psychological assessments

References (21)

There are more references available in the full text version of this article.

Cited by (0)

View full text