Objectives. Women's perception of the likelihood of developing breast cancer is generally inaccurate since it has been consistently found that they overestimate their risk. Prior explanations of this phenomenon are incomplete. By integrating both personal and situational characteristics, this study attempts to develop a theoretical explanation of women's perception of their risk of developing breast cancer.

Methods. Cross-sectional survey data from a sample of 163 women whose sisters were diagnosed with breast cancer at age 50 or younger in the San Francisco Bay Area are used. Simple and multiple regression models were estimated to test the association between perceived risks and personal (cognitive and affective) and situational (familial and socio-economical) characteristics. A stepwise regression analysis was conducted to identify the most explanatory model of the predictors for perceived risk of breast cancer. Finally, a hierarchical regression analysis was used to compare relative effects of the independent variables.

Results. Women with a family history of breast cancer overestimated their risk of developing breast cancer by an average of 25 percentage points. Both personal (use of avoidance coping strategy, intrusive thinking, and worry of developing breast cancer) and situational variables (multiple family histories and frequent visit with affected sister) were statistically associated with overestimation of the risk of developing breast cancer. Having more than one first degree relative with breast cancer was the most powerful predictor (β=16.0, p-value<0.001). More variance was explained by situational factors than by personal factors. The amount of social contact with the affected sister consistently increased the unaffected sister's perceived risk of developing breast cancer.

Conclusion. It can be concluded that (1) overestimation of the risk of developing breast cancer is explained by family history and the amount of contact with the affected sister and (2) both psychological and situational factors need to be considered to help women's perceived risk of developing breast cancer to be closer to their actual risk.