Development and validation of a telephone questionnaire to characterize lymphedema in women treated for breast cancer

Background and Purpose. Accurate and economical characterization of lymphed ema is needed for population-based studies of incidence and risk. The purpo se of this study was to develop and validate a telephone questionnaire for characterizing lymphedema. Subjects. Forty-three women who were treated pre viously for breast cancer and who were recruited from physical therapy prac tices and a cancer support organization were studied. Methods. Questionnair e assessment of the presence and degree of lymphedema was compared with phy sical therapists' diagnoses, based primarily on circumferential measurement s. Twenty-five of the 43 subjects were measured independently by 2 physical therapists to assess interobserver agreement. Results. Interobserver agree ment on clinical assessments of the presence and degree of lymphedema was h igh (20/25, weighted kappa = .80); all of the disagreements were between ju dgments of whether there was no lymphedema or mild lymphedema. For the diag nosis of at least moderate lymphedema (differential in the circumferences o f the upper extremities greater than 2 cm), sensitivity of the questionnair e varied from 0.86 to 0.92 and specificity was 0.90. However, sensitivity ( varying from 0.93 to 0.96) was higher than specificity (varying from 0.69 t o 0.75) for the diagnosis of any lymphedema. Discussion and Conclusion. A f ew straightforward questions exhibited excellent agreement with physical th erapists' assessments for identifying at least moderate lymphedema.