Abstract
Objective: A significant proportion of breast cancer patients experience psychiatric morbidity. The present study compared the psychopathological profile (depression, anxiety and general psychopathology) of Greek women with breast cancer with a group of healthy controls. Materials and Methods: Patients (n=109) were recruited from a specialized oncology breast cancer department and healthy controls (n=71) from a breast outpatient clinic. General psychopathology was assessed by the SCL-90-R. The Montgomery-Asberg Depression Rating Scale (MADRS) and the Spielberger State-Trait Anxiety Inventory (STAI) were used for assessing depression and anxiety. Demographics and clinical characteristics were also recorded. Data were modeled using multiple regression analysis. Results: The mean age was 54.7±18.1 years for the control group and 51.2±9.5 years for the patient group (p=0.288). Mean scores on SCL-90-R, MADRS and STAI were significantly higher in the cancer group compared to controls (p<0.05). Multiple regression analysis revealed that breast cancer was independently and positively associated with all psychological measures (p<0.05). Regression coefficients ranged from 0.19 (SCL-90-R, psychotism) to 0.33 (MADRS). Lower anger/aggressiveness and anxiety were found in highly educated women; divorced/widowed women scored higher on obsessionality and MADRS compared to married women. Psychiatric treatment was associated with higher scores on somatization, depression, phobic anxiety and general psychopathology. Conclusion: Anxiety, depression, and overall psychopathology are more frequent in breast cancer patients compared to controls. Disease makes a larger independent contribution to all psychopathological measures than any other investigated variable. Therefore, breast cancer patients should be closely followed up in order to identify and timely treat any mental health problems that may arise.
There is extensive literature on the psychological impact of cancer in breast cancer patients. Reviews of existing research estimate that a third to half of women with breast cancer experience psychological distress and that multiple antecedent and concomitant factors influence psychological functioning of patients over time (1, 2). Psychological distress experienced by cancer patients has been thoroughly studied throughout the various stages of the disease. Interestingly, although several studies have shown that cancer patients in general, breast cancer patients included, were more depressed than the general population, a meta-analytical review showed that no significant differences were found in terms of anxiety and overall psychological distress (3). In patients with a prognosis of six months or less, prevalence of depression according to the ICD-10 diagnostic criteria was 22% (4), while studies that have focused on psychological distress during the course of the disease have shown that psychological distress increases significantly in the last two to three months before death and especially in the last month of life (5-7).
Regarding breast cancer in particular, up to one third of patients may suffer from psychological morbidity during the year following an initial operation (4, 8, 9); moreover, it is acknowledged that adverse effects of adjuvant cancer treatments may also play a part in psychological adjustment (10, 11), with problems persisting in a significant proportion of women up to several years following treatment (12, 13). Furthermore, several psychosocial and demographic variables have been identified as affecting psychological morbidity, such as age at diagnosis (14, 15), marital status (16, 17), health complaints after surgery unrelated to cancer (18), and premorbid personality characteristics (16, 19). However, in the majority of studies assessing psychopathology in women suffering from breast cancer, either within group comparisons are made at different time points, i.e. before and after the onset of the disease or during its progress through different stages, or between the study group and a group of patients suffering from another disease. Few studies have included a control group of healthy women and usually their principal objective was not to measure psychopathology per se (20-22).
The objective of the present study was to compare a group of women suffering from breast cancer and a control group of healthy women or women suffering from benign breast diseases in terms of depression, anxiety and general psychopathology. A further aim was to identify potential risk factors for the development of psychological morbidity, as its early detection and preventive therapeutic interventions have been found in longitudinal studies to improve outcome (5).
Materials and Methods
Participants and study design. Participants were 180 women: a group of women suffering from breast cancer (n=109) and a control group of healthy women or women suffering from benign breast diseases (n=71). The breast cancer patients were attending a Specialized Oncology Breast Cancer Clinic of a General Hospital (Helena's Women's General Hospital) and were studied at different stages of the disease, starting one month after mastectomy or tumor dissection and/or the beginning of adjuvant chemotherapy, including the advanced or terminal stages. The control group comprised healthy women randomly selected who were attending the Breast Outpatient Clinic of another facility specialized in breast diseases for routine check-ups; none of them was receiving treatment for any breast conditions. Initially, 150 women with breast cancer and 100 healthy controls were recruited from these two specialized breast clinics. The population was selected using proportional random sampling based on the expected age distribution and sociodemographic matching of the cases. Dropouts were due either to women declining participation or handing back incomplete questionnaires.
Candidates for the study were approached through a letter asking for participation, followed if necessary by a second letter, up to five telephone calls at different times of the day and week, and as a last resort a home visit. A face-to-face meeting was arranged and a structured interview covering demographics, risk factors for breast cancer, other co-existing medical conditions and medication was administered. A battery of psychometric instruments was also given: the self-rated SCL-90-R, the State-Trait Anxiety Inventory (STAI), and the Montgomery-Asberg Depression Rating Scale (MADRS), which was scored by the interviewer. A single assessment of the participants was performed for the purpose of the present study. Data collection was carried out between October 2003 and October 2004 by the same psychiatrist-researcher (M. F.) throughout the study. Both groups were followed up for one year. Inclusion criteria for the breast cancer group were: age 30-70 years and absence of cognitive impairment. For the control group, apart from the above, absence of any active benign breast disease and/or treatment during the past five years was required.
The study protocol was approved by the hospital Ethical Committee and the study was conducted in accordance with the existing hospital regulations and the “Ethical Principles for Medical Research Involving Human Subjects” (World Medical Association Declaration of Helsinki - 52nd WMA General Assembly, Edinburgh, Scotland, October 2000). Written informed consent was obtained from each participant prior to enrollment in the study.
Measures of psychopathology. For the assessment of anxiety, depression, and general psychopathology, the following psychometric instruments were used: a) Spielberger's State-Trait Anxiety Inventory (STAI): This is a 40-statement self-report inventory designed for the assessment of state (S-Anxiety) and trait (T-Anxiety) anxiety in adults, which has been extensively used in research and clinical practice. It evaluates how the respondent feels “right now, at this moment”, as well as how he feels “generally”. Each STAI item is given a weighted score of 1 to 4 (23, 24); b) Montgomery-Asberg Depression Rating Scale (MADRS): This clinician-rated 10-item scale has been designed to measure the severity of depressive symptoms. Because there is little emphasis on somatic symptoms, it can also be used for the assessment of depression in people with physical illness. It measures the severity (on a scale from 0 to 6) of a number of symptoms, including mood and sadness, tension, sleep, appetite, energy, concentration, suicidal ideation and restlessness (25, 26); c) Symptom Checklist-90-R (SCL-90-R): The SCL-90-R is a self-rated 90-item checklist objectively evaluating a broad range of symptoms of psychopathology. It measures nine symptom dimensions and is designed to provide an overview of a patient's psychological symptoms and their intensity at a given time point. The symptom scales measured are: somatization, interpersonal sensitivity, obsessive-compulsive symptoms, depression, anxiety, hostility, phobic anxiety, paranoid ideation, and psychotism. The SCL-90-R has a high reliability and validity, and has been widely used (27, 28).
Statistical analysis. Quantitative variables are expressed as means (standard deviation), while qualitative variables are expressed as absolute and relative frequencies. For the comparison of means between two groups, Student's t-tests were used. Pearson correlation coefficients were used to explore the association of continuous variables. For the comparison of proportions, chi-square tests were used.
The association between breast cancer and SCL-90-R, S-Anxiety, T-Anxiety, and MADRS scales was modeled using multiple linear regression analyses. All models were adjusted for age, education level, employment status, family status, number of children, smoking, alcohol consumption, age at menarche, psychiatric treatment and the presence of thyroid, cardiac or gynecological disease. Standardized regression coefficients (β), which indicate the relative magnitude of prediction of each independent variable, and standard errors were computed from the results of the linear regression analyses. Hypothesized interactions of variables in the models were not significant. All reported p-values are two-tailed. Statistical significance was set at p<0.05 and analyses were conducted using SPSS statistical software (version 13.0).
Results
The sample consisted of 180 women of mean age 50.6 years (SD=9.5 years); 109 of them had breast cancer. Sample characteristics are presented in Table I. As shown, all participants were matched for age and sociodemographic level. The group of women with breast cancer had lower proportions of smoking and alcohol consumption than the control group, whereas thyroid and cardiac disease were more frequent in that group. Mean scores on the various dimensions of SCL-90-R, S-Anxiety, T-Anxiety, and MADRS of the two groups are presented in Table II. Women with breast cancer had significantly higher scores on S-Anxiety, T-Anxiety, MADRS and all SCL-90-R scales in bivariate analysis. Results of multiple linear regression analyses with S-Anxiety, T-Anxiety, MADRS and all SCL-90-R scales as dependent variables are shown in Tables III, IVa, and IVb. Controlling for all other variables, breast cancer was the only variable that was significantly associated with all aforementioned parameters. Furthermore, psychiatric treatment was associated with higher scores on somatization, depression, phobic anxiety and the general index of SCL-90-R. Age at menarche was found to be positively associated with depression and psychotism dimensions. Also, divorced/ widowed women had higher scores on obsessionality and the MADRS compared to married women. Adjusting for all independent factors, the score on anger/aggressiveness dimension was lower for women with a middle or high educational level compared to those with a low educational level and decreased with increasing age. In multiple regression analysis, women of middle or high educational level had significantly lower S-Anxiety and T-Anxiety scores compared to those with a low educational level. Finally, the standardized regression coefficients indicated that the presence of breast cancer made larger independent contributions to all scales than any of the other variables.
Discussion
Although numerous studies have investigated psychiatric morbidity in breast cancer patients, in particular depression and anxiety, to our knowledge this is one of the few reports (21, 29) that straightforwardly focuses on the psychological impact of breast cancer by comparing women suffering from the disease with a healthy control sample. Most research uses either within group comparisons at different time points, or comparisons between the breast cancer group and a group of patients suffering from another disease. In the latter case, there is disagreement regarding the appropriate comparison group for breast carcinoma (30, 31). In the present study this considerable shortcoming has been overcome by including as controls women either free from any breast disease or suffering from minor benign breast diseases on a regular follow-up and without any treatment.
Women suffering from breast cancer had significantly higher anxiety, depression, and general symptoms of psychopathology compared to the control group, which corroborates previous findings in this population (32, 33). It is noteworthy that all measures of psychopathology, as assessed by the SCL-90, were increased. This indicates that disease activates complex mechanisms giving rise to a more general psychopathology than mere anxiety and depressive reactions; alternatively, anxiety and depression may elicit, through multiple pathways, psychopathology. This is important because, to our knowledge, this is one of the few existing studies (34, 35) which directly focuses on the evaluation of depression and anxiety in association with general psychopathology in breast cancer patients.
Intense feelings of tension, nervousness and worry, i.e. state anxiety, is an expected and well-known psychological response to a malignant disease and it is related to the emotional turmoil that follows diagnosis, anticipated prognosis, treatment, and its side-effects (36-38). This was also recorded in our study. Moreover, trait anxiety was found to be higher in the breast cancer patients compared to the control group, which probably indicates that personality characteristics may become more prominent when exposed to stress. In a similar vein, symptoms of depression as measured by the MADRS were significantly more intense in the patient than in the control group, which is consistent with the high depression scores measured with different instruments (39-42).
However, time of assessment of psychological reactions may be decisive, because as shown by various investigators, coping responses to stress may change over time following the initial diagnosis (33, 43); this suggests that women gradually may come to terms with their illness and feel less anxious and less passive in accepting it, but at the same time somehow less positive and willing to face their condition (34, 44). Sometimes overoptimism is also seen in such circumstances, which comprises a component of denial through which individuals may attempt to minimize the severity of their condition (44-47). Besides the aforementioned psychological reactions to a life-threatening disease, one should also consider the fact that biological components associated with the disease itself, as well as with its treatment, could be responsible for depressive symptoms often observed in cancer patients. Thus, hormonal changes (e.g. abrupt changes in steroid levels) occurring for various reasons appear to be associated with symptoms of mood disorders in a subgroup of women (48, 49). Of interest is also the observation that several chronic diseases, including different types of cancer, are more or less frequently complicated with depression, with some indications suggesting that its prevalence in cancer patients may depend on tumor localization (50-52). Unfortunately, our study design does not allow any conclusions regarding this complex interaction of factors generating depressive symptoms at different stages of breast cancer.
In the present study, anger/aggressiveness were found to be significantly increased in the breast cancer group compared to the control group. The association between breast cancer and aggression/hostility is unclear and relatively little researched. However, there is evidence that there might be a link between the two and specifically with the suppression of these feelings (53-55). Furthermore, psychological responses to breast cancer such as “helplessness/hopelessness” have been postulated to be factors influencing survival, although the causal mechanisms are unclear (44, 53, 56). Helplessness and repression seem to be two key factors closely associated and interrelated with anger/aggressiveness, as repressed hostility, negative emotions and a feeling of loss of control often described by breast cancer patients are likely to contribute through a vicious circle to unfavorable prognosis, while denial/minimizing have been reported to be associated with a more favorable prognosis (57, 58). In this line of thought, the increased psychotism [a personality trait defined by Eysenck as rigidity and stubbornness (59)] and obsessionality observed in the SCL-90 subscales of the breast cancer group in our study, may reflect the emotional restriction, suppression of negative feelings, and wish of control present in the patient group. This particular association has not been previously reported and its significance needs to be further clarified.
So far, no psychological factor has been convincingly demonstrated to influence cancer development; however, some factors putatively play a role. The so-called “cancer-prone personality”, whose core characteristics are those of emotional suppression and emotional control (i.e. suppressing negative emotions of anger and hostility, abrogating one's needs in favor of the needs of the others and an attitude of helplessness or hopelessness), has been suggested to actually predispose some individuals to developing cancer or to prompt disease progression (47, 60-62). Nevertheless, the findings remain controversial (63-65).
In the present study, several known risk factors were considered, including age at menarche, other coexisting medical conditions, previous psychiatric history and treatment, and various sociodemographic variables. Age at menarche, which is considered to be important for younger women and is known to be a strong and established factor associated with breast cancer (66, 67), was found to positively correlate with depression and psychotism. Women who had undergone psychiatric treatment were found to have high somatization, phobic anxiety, depression, and general psychopathology scores on the SCL-90. This is in agreement with existing studies showing that previous psychological treatment was associated with depression and/or anxiety around diagnosis, with some studies suggesting that the risk factors for these conditions are related to the patient itself rather than to disease or treatment (68, 69).
Regarding the role of various sociodemographic variables, including educational level, employment status, and family status, for the development of psychopathology in the two studied groups, it was observed that divorced/widowed women with breast cancer scored higher on obsessionality and depression, which indicates a higher vulnerability associated with such life events. Anger/aggressiveness scores were found to be lower in women with middle or high educational level, which possibly suggests that these women have a more developed and sophisticated coping system and more access to information compared to women with a lower educational level. Similarly, in the multiple regression analysis, women belonging to the middle-high educational level group were shown to have significantly lower state and trait anxiety scores, which might be explained by a combination of the above alongside a more stoic approach to illness.
In conclusion, the present study testifies that anxiety, depression, and overall psychopathology, as assessed by the SCL-90, are more frequent in breast cancer patients than in a control population and that disease makes a larger independent contribution to all psychopathological measures than any of the other investigated variables. Its main limitations were the cross-sectional design with no long-term follow-up, recruitment of participants from two specialized breast clinics, and that no detailed psychiatric history was available for the participants at the time of recruitment. However, our findings suggest that this highly vulnerable patient group should be followed up after the diagnosis and during the course of the illness by mental health professionals in order to identify and timely treat any mental health problems that may arise. Further research is needed to identify the profile of psychological symptoms and problems and their complex interrelationship that frequently occurs following diagnosis and treatment of breast cancer.
Footnotes
-
Conflict of Interest
No potential conflicts of interest exist regarding this study.
- Received March 7, 2010.
- Revision received June 8, 2010.
- Accepted June 14, 2010.
- Copyright © 2010 International Institute of Anticancer Research (Dr. John G. Delinassios), All rights reserved