Increasing Breast Cancer Screening for Underserved Women: Redefining “Hard-to-Reach” By Specific Program Outcomes

Background:  Promoting regular breast cancer screening (mammography) among hard-to-reach women is an important, yet challenging task. In mammography literature, hard-to-reach women have generally been characterized as older women of color with lower income, based on attributes of non-screened women. Despite substantial efforts to reach this group of women, lingering disparities in screening have been reported. We argue that focusing on screening outcome does not sufficiently address an important, but rarely scrutinized question: who are “hard-to-reach” women? Specifically, we examine whether attributes of these women are universal across different program outcomes—intermediate communicative outcomes (inquiry calls), intermediate behavioral outcomes (scheduling), and ultimate behavioral outcomes (rescreening)—or specific to each of them.

Program background:  Minnesota’s Sage Program is part of the National Breast and Cervical Cancer Early Detection Program, which provides low-income, uninsured, and underinsured women access to breast and cervical cancer screening. Sage screens more than 17,000 women each year at over 430 clinics throughout the state. Sage continually evaluates its recruitment strategies’ effectiveness and cost-effectiveness. The present study looks to further improve intervention strategies based on Sage’s accumulated profile of hard-to-reach women.

Evaluation Methods and Results:  Our sample includes 15,304 women previously screened through Sage who received a reminder letter between June 2011 and June 2012. From this sample, 41% of women (n=6,215) received a rescreening mammogram while the remaining women were not rescreened through Sage. In contrast to conventional dichotomous screening program outcomes (i.e., screening/no screening), we further broke down non-screeners into “non-responders” (women who neither called nor were screened; n=7,211); “information seekers,” (women who called and were eligible for Sage but did not schedule an appointment; n=128); and “no-shows” (women who scheduled an appointment but did not get rescreened during the study timeframe; n=1,148). Women who called and were Sage-ineligible (n=602) were excluded from the analyses.  Logistic regression analyses indicated that women who are older, more educated, insured, have less Sage screening experience, and no reported breast cancer symptoms were more likely to be “non-responders.” Older women with less Sage experience were more likely to be “information seekers,” whereas younger and non-White women with less Sage experience were more likely to be “no-shows.” Importantly, while the subgroups of hard-to-reach women were a function of different factors, less screening experience with Sage was a consistent factor across the three subgroups.

Conclusions:  We identified determinants that offer insights on reaching subgroups of women at intermediate program outcomes. Whereas Sage serves hard-to-reach women by definition, we found significant differences in attributes of women who reach various program outcomes, suggesting the definition of “hard-to-reach” should be outcome-specific and may require tailored interventions to increase screening.

Implications for research and/or practice:  Screening-based characterizations of hard-to-reach women emphasize disparities in ethnicity and income. However, categorizing these women using only dichotomous screening outcomes is an incomplete assessment. The use of intermediate outcomes in defining hard-to-reach women provides more sensitivity to understanding the characteristics and screening barriers of women in each sub-group. The present analyses indicate this is a feasible approach to increase the utility of an evaluation scheme for informing future Sage recruitment strategies.