ASSESSING THE QUALITY OF BREAST CANCER SURGERY DECISIONS IN A NATIONAL MEDICARE SAMPLE

Purpose: Many studies of decision making focus on younger populations who receive care in academic medical centers. The purpose of this study was to examine decision quality in a national Medicare population.

Methods: A cross sectional survey of breast cancer patients who had mastectomy or lumpectomy were identified through Medicare records. Patients had to be at least 66 yrs old and to have been continuously enrolled in fee for service Medicare for at least a year at the time of their surgery. The mailed survey contained items covering knowledge of treatments and outcomes, involvement in the interaction, and other aspects of the decision.  Here, we present total knowledge scores (on a scale of 0 to 100%) and involvement scores (on a scale of 0 to 100%) with higher scores indicating more knowledge and more involvement. We examined factors associated with knowledge using linear regression.

Results: We received 914 responses (507/667 (76%) lumpectomy and 407/557 (73%) mastectomy). The sample was 91% White, median age 76, and 18% reported a college degree or more. Mastectomy patients were more likely to be non White, older and have lower education compared to lumpectomy patients. The total knowledge score was 53% (SD 20%) and this varied by treatment (49% for mastectomy and 56% for lumpectomy patients (p<0.001)). About half (54%) understood that survival was the same for mastectomy and lumpectomy. Only 29% knew that waiting four weeks to make a decision would have little or no effect on the outcome. On average, patients reported limited involvement in decisions, and this did not vary by treatment (39% for lumpectomy and 43% for mastectomy, p=0.10). For example, while 78% reported a discussion of pros, only 34% reported discussion of cons of the treatments. Only 26% reported that their providers asked them for their treatment preference. Several factors were associated with higher knowledge scores in the regression model, including younger age, higher education, having lumpectomy and higher involvement score.  

Conclusions: Women with breast cancer in Medicare had signficant knowledge gaps regarding treatment options and outcomes. Participants report variable involvement in decisions. Older patients and those with lower education may need additional support in order to ensure they are informed and involved in decisions.