Relationship between numeracy and breast cancer patients' estimates of adjuvant treatment benefit.

586 Background: Based on clinical anecdotes and existing literature, the general population has trouble understanding and making use of statistical information. We explored the accuracy of breast cancer patients' (pts) estimates of treatment outcomes (probabilities of remaining cancer free versus recurring) relative to data provided by 'Adjuvant! Online' decision aid program, and whether accuracy differed by pts numeracy skills (i.e. use of mathematical concepts and operations). METHODS: 43 ER+ women with early stage, node-negative breast cancer were recruited (mean age 56, 72% Caucasian). After signing an informed consent and completing an assessment of numeracy (Lipkus et al., 2001), pts discussed case-specific Adjuvant! print-outs with an oncologist. Pts then estimated their chances of being cancer free after receiving no further treatment, hormonal therapy only, chemotherapy only, or both. Pts also were asked to select the treatment option that afforded their best chance of remaining cancer free. RESULTS: Compared to the estimates provided by Adjuvant! Online, pts underestimated their chance of being cancer free after receiving no further treatment (M=62 vs. 48 out of 100), hormonal therapy only (M=72 vs. 45), chemotherapy only, (M=71 vs.42), and combined therapy (M=78 vs. 54). 63% correctly selected the treatment option that afforded the highest estimate of being cancer free. More numerate patients were more likely to correctly specify which treatment option provided the best chance of being cancer free (OR=0.67, 95% CI: 0.50, 0.89, p<.006); and less likely to give personal estimates of being cancer free inconsistent with estimates provided by Adjuvant! for hormonal therapy (OR=0.17, 95% CI: 0.04, 0.72, p<.02), chemotherapy (OR=0.62, 95% CI: 0.39, 0.99, p<.05) and combined therapy (OR=0.44, 95% CI: 0.23, 0.85, p<.02). CONCLUSIONS: Findings suggest that pt numeracy skills are important in comprehending statistical data about adjuvant treatment outcomes. Numeracy skills may have implications for how statistical information about treatment is presented and discussed, affecting processes of informed decision-making and pt quality of life. No significant financial relationships to disclose.

Duke Scholars

Author Gretchen Genevieve Kimmick Medicine, Medical Oncology