Development of a cancer specific body-image instrument

Cancer and its treatments have been shown to have a negative psychological effect on many patients. One effect of cancer that is often described anecdotally is body image disturbance. A moderate amount of research has been conducted on the relationship between body image and cancer. However, the results of these studies are limited because until recently a psychometrically validated measure designed to assess body image disturbance in cancer patients did not exist. Instead, when assessing body image disturbance, many researchers used measures intended for other populations or they have created a few questions for their study to assess body image which typically lack standardization and have not been validated. This paper discusses relevant theories of body image and scale construction. Through the use of the literature an empirically-driven body image instrument was developed for use in those who currently have cancer or have a history of cancer. In addition, tests of validity and reliability were conducted to determine the usefulness and factor structure of the measure. Initially, a panel of experts and cancer patients were asked to generate relevant items for a body image measure, resulting in 64 test items. Those items were reviewed by a separate panel of experts and rated on necessity, relevance, and clarity. The total number of items was reduced from 64 to 39 during this step. Convergent and discriminant validity was tested by comparing the new measure to other measures of body image and depression and anxiety. As hypothesized, the new measure demonstrated strong correlations with the other measures of body image and modest correlations with the measures of affect. Stability of the measure was examined through the use of test-retest methodology. Breast cancer patients were asked to complete the measure twice over a two week timeframe. The new measure demonstrated excellent stability (alpha=.96).Finally, a factor analysis was conducted, yielding a seven factor measure. The results of this step are considered preliminary because the study population isn’t entirely representative of those the measure is designed to assess. Furthermore, a larger study population is typically used for a factor analysis.Ph.D., Clinical Psychology -- Drexel University, 200