The self-reported likelihood of patient delay in breast cancer: New thoughts for early detection

Background. Delayed presentation of self-discovered breast symptoms influences stage of cancer at diagnosis and decreases breast cancer survival. Methods. A total of 699 asymptomatic women (black, white, and Latino), recruited in community settings and stratified by age, income, and educational level, were surveyed for their likelihood to delay (J-Delay scale) in the event of a breast symptom discovery. Models of likelihood were tested with logistic regreission analyses. Results. A total of 166 women (23.7%) reported likelihood to delay. Lower income, lower educational level, self identification as Latino or black, experienced prejudice in care delivery, perceived lack of access to health care, fatalism about breast cancer, poor health care utilization habits, self-care behavior, spouse/partner and employer perceived constraints, problem-solving style, and a lack of knowledge of breast cancer's presenting symptoms were associated with likelihood to delay. A combined sample multiple logistic regression model correctly predicted 40.6% of women reporting a likelihood to delay, 94.9% of those not likely to delay, and 82.4% (551 of 669) of cases overall. Conclusions. Self-reported likelihood of patient delay is measurable in advance of symptom occurrence, and this measure is consistent with behavioral and knowledge variables previously linked with advanced breast cancer at diagnosis. (C) 2002 American Health Foundation and Elsevier Science (USA).