Socioeconomic disparities in the decline in invasive breast cancer incidence

Abstract

Breast cancer incidence in the United States has declined dramatically since the year 2002. To improve our understanding of the underlying factors driving breast cancer trends, we explored potential socioeconomic disparities in the recent decline in incidence. We examined the decline in breast cancer incidence according to county-level socioeconomic indicators using data from the Surveillance, Epidemiology and End Results (SEER) program. Since socioeconomic status is associated with mammography screening, we also examined the relation between incidence of ductal carcinoma in situ (DCIS; a strong marker of mammography utilization) and the decline in invasive breast cancer. The reduction in invasive breast cancer incidence between 1998-2001 and 2003-2006 in the SEER 9 registries was greatest among women living in counties with higher median household income (-16% change for > or = $85,000 vs. -4% for <$35,000; P(trend) < 0.01) and a higher percentage of adults aged 25 years or older with a bachelor's degree (-13% change for > or = 40% vs. -8% for <15%; P(trend) < 0.01). Counties with higher DCIS incidence during 1985-2001 had a larger decrease in invasive breast cancer incidence (absolute decrease 1.7 percentage points greater per 5 per 100,000 increase in DCIS incidence; P = 0.01). This association was present for both ER-positive and ER-negative invasive cancers (P < 0.05). In summary, the decline in breast cancer incidence has been largest in areas with high socioeconomic status and high screening utilization rates. These results are consistent with the hypothesis that a saturation of screening mammography utilization contributed to the overall decline in breast cancer incidence.

Figure 1

Invasive breast cancer incidence in the SEER 9 registries during both 1998–2001 (dashed line) and 2003–2006 (solid line) is positively associated with year 2000 county-level median household income (A) and percent of adults aged 25 and older with a bachelor’s degree (B).

Figure 2

The percent decline invasive breast cancer incidence (solid line) was largest among women living in counties with higher median household income (A) and counties with a higher percent of adults aged 25 and older with a bachelor’s degree (B). These same socioeconomic indicators were positively associated with incidence of ductal carcinoma in situ (DCIS; dashed line) during 1985–2001. The error bars represent 95% confidence intervals.

Figure 3

The percent decline in invasive breast cancer incidence tended to be largest in counties with the highest incidence of ductal carcinoma in situ (DCIS) during 1985–2001. Circle sizes are proportional to the year 2000 county population. The Pearson correlation coefficient was −0.18 (P = 0.01).

Figure 4

The association between incidence of ductal carcinoma in situ and the percent decline in invasive breast cancer was similar for both estrogen receptor (ER) positive (A) and ER-negative breast cancers (B). Circle sizes are proportional to the year 2000 county population. The Pearson correlation coefficients were −0.22 (P = 0.002) and −0.15 (P = 0.04) for ER-positive and ER-negative cancers, respectively.