Neuropsychological data collected in 28 breast cancer patients before and after chemotherapy were compared to those of a local healthy control group, a local disease control group, and published norms, respectively, in order to determine whether the nature of the control group influenced outcome. The frequency of decline in the chemotherapy group was 21% and significantly higher than that of the control group whether referenced to the healthy controls, disease controls, or published norms. These results suggest that published norms may be adequate to demonstrate cancer-related cognitive impairment provided that cognitive function is measured adequately, practice effects and base rates are taken into account, and demographic factors that might influence practice effects are considered.