Fear of recurrence of breast cancer. Validation of the dutch version of the concerns about recurrence scale (cars)the concerns about recurrence scale (cars) is an instrument measuring fear of recurrence of cancer in former breast cancer patients. It assesses the general level of fear of recurrence, and fear of recurrence as related to four different domains: health, death, womanhood and role functioning. In the present study we evaluate the psychometric properties of the dutch version of the cars (cars-dlv) in a sample of 136 breast cancer patients who had undergone curative treatment. Explorative factor analysis on the cars-dlv indicated a slightly different factor structure than that previously found: the items related to concerns about health and death collapsed into one subscale, whereas the items related to concerns about role functioning loaded on two different factors. However, based on similarity in content and the small number of items we merged these factors to a single domain of concerns about role functioning. The overall concern subscale and the three domain specific concern subscales proved highly internally consistent and also had good test-retest stability. Construct validity of the cars-dlv appeared adequate, as indicated by the correlations with related questionnaires (general anxiety, pain catastrophizing and optimism). Moderate to high levels of overall fear of recurrence were found in 66% of the breast cancer survivors. Fear of the consequence of recurrence for health was most prominent. Pain was a strong predictor of overall fear and of fear on the three sub domains of the cars. The prevalence of fear decreased significantly with age. Fear of recurrence was negatively correlated with health-related quality of life, and fear of the consequence of recurrence for role functioning appeared to be the strongest predictor.the results indicate that the cars-dlv may be a valuable instrument to measure women’s fears for recurrence of breast cancer. However, the distinctness of the four domains of fear and their specific predictors and/or consequences should be substantiated further. Moreover, the large number of items especially on the health subscale and their high inter-correlations suggest that a shorter version may be recommendable.