An examination of the psychosocial characteristics of women who develop cancer

Abstract

The link between psychological variables, particularly a depressive reaction, and the development of cancer has been noted anecdotally over many centuries. More recently, scientific studies have identified specific psychosocial variables which may be precursors to the development of cancer, not necessarily causes, but factors which appear to contribute to the emergence of cancer, if the necessary predisposing physiological abnormalities or carcinogenic agent are also present. Psychological variables, specifically inability to express emotions, denial and repression, and social variables such as life stress in general, an adverse life history pattern, or the experience of a significant loss,have been documented but the findings of some studies have been contradictory. Another group has examined the individual's reactions to social variables, for example depression and feelings of powerlessness, and the variables which contribute to increase the vulnerability of the individual to the effects of life stress in general, and specific kinds of life stress. This study was an attempt to predict which of a group of women presenting to outpatient clinics at Wellington Hospital with possible cancerous physiological abnormalities would be found on histological examination to have cancer. Feelings of hopelessness provided the basis for this prediction. It was hypothesized that other psychological and social variables would also differentiate the carcinoma from the non-malignant subjects, separately and as a group. Those examined included depression, locus of control or mastery of life, experience of life change and reaction to this, experience of loss, ability to express emotions, typical response to problems and demographic data, age and marital status. The interrelationships of these variables were also investigated. Seventy-eight subjects took part. Fifty presented to the Gynaecological Outpatient Clinic for diagnosis of abnormalities found from cervical smears. Twenty-eight presented to the Cancer (or Combined) Treatment Clinic with a variety of physiological abnormalities including skin lesions and breast lumps. A correct prediction of diagnosis was made, on the basis of hopelessness, for 65.4% of the subjects (p = 0.02). When the gynaecological sample was analyzed separately, this rate was improved to 70% (p = 0.01). The prediction rate was not statistically significant for the CTC sample on its own. Experience of loss, feelings of no control over life, age, and marital status were the only psychosocial variables which differentiated between the diagnostic groups. Using a discriminant function analysis, the set of independent variables separated the subjects into two diagnostic groups - cancer (invasive and carcinoma-in-situ combined) and no cancer, at a statistically significant level. Correct classification was made for 77% of the sample (p = 0.01) using all predictor variables, and for 69% of the sample (p = 0.005) using only three variables, internal/external locus of control, experience of no control, and loss. A large number of significant relationships was found between the independent variables. Some of these would be expected as they were alternative methods of measuring a particular variable. Others, however, provided further more precise understanding of what may be involved in such concepts as hopelessness and depression.