In this survey study conducted in Oregon through random sampling, our major finding was that lack of college degree education was associated with no mammogram screening in the previous two years, supporting existing findings. This is a clinically relevant modifiable risk factor, as screening attendance also predicted outcomes of breast cancer itself.
Education is a commonly used proxy variable for socioeconomic status (SES) (15, 16) and is less likely than other variables to be influenced by diseases in adulthood (17). During 2015–2019, the percentage of the population 25 years and older with at least a college graduate decree increased to 32.1% compared to 27.5% from 2005–2019 (18). Other SES factors in our study, including occupation status and health insurance, were also significantly correlated with education and suggest additional significant disparities in timely use of mammogram screening between SES groups in Oregon.
Our finding is consistent with a previous meta-analysis that women with highest education were more likely to adhere to breast cancer screening guidelines (19). In contrast, lower education was associated with other breast cancer risk factors such as smoking (20), decreased physical activity (21), and obesity (22). Therefore, improving mammogram use through targeting people with lower education may necessitate a multi-faceted approach to address lifestyle habits, pre-existing health conditions, and environmental and psychological factors to eliminate disparities due to education.
Women may refrain from engaging in breast cancer presentational mammogram screening due to lack of knowledge regarding the risks of breast cancer. In a study examining women’s views on breast cancer primary prevention and sources of health care information, 75% of women aged 50–74 years would prefer to receive health information from within, as opposed to outside, health care settings (23). Therefore, further incorporating breast cancer risk assessments into the primary health care settings may improve mammogram usage.
While there is existing research that has shown evidence of higher education being linked with higher risk of breast cancer (24–26), in particular, in situ breast cancer (26), these findings must be considered within context. Beyond the reproductive factors such as nulliparity status and age at first birth, women with more education tend to undergo more frequent mammogram screenings, and consequently have more opportunities to detect cancer in screening and receive cancer diagnoses at earlier stages.
Our finding that women with lower education participated less in mammography screening supports calls for increased outreach efforts to these women. Research has suggested reluctance towards preventative care, from cancer screenings to annual wellness visits, could stem from perceptual factors such as anticipated shaming by medical professionals and perceptions related to SES status. In a qualitative study on how SES status affects patient perceptions of health care, most subjects believed that their SES influenced the health care they received (27). Additionally, physicians perceive and treat patients of low SES differently than those of high SES (28), but few acknowledge or notice their susceptibility to this implicit bias (29). Given that attitudes and biases of healthcare provider’s may affect clinical decision-making including appropriate mammogram screening recommendations, the lower rates of mammogram screening among lower SES women reflect the impact of not only patient attitudes and experiences, but also those of the providers making medical recommendations and orders.
We also identified lack of insurance as another potentially significant risk factor in the unadjusted model. Although the significance disappeared after adjusting for education, perhaps due to our small sample size, the magnitude of the association was consistent with current literature associating lack of insurance with lower screening rates (30). Further, in insured women, lower family income has been associated with lower screening rates; a Medicare study reported women with a family income under the federal poverty level screened at a depressed rate of 51% (8, 9). Additionally, those with fee-for-service care are less likely to screen than peers in health maintenance organizations, and those with public insurance are less likely than peers with private insurance (9). Further investigation into this association with a larger population can further support this connection.
Notably, we found no difference between our urban and rural populations. This may reflect successful efforts to encourage screening in rural Oregon, given previous literature has suggested a disparity and encouraged intervention (31, 32). Also, neither mammogram screening knowledge nor perception of cancer beliefs and cancer attitudes were associated with insufficient screening, suggesting that external factors such as access to health care or lack of health care provider recommendation or guidance may play more critical roles.
The results from this study are helpful in identifying intervention points to improve timely mammogram screening among the general public. These interventions may target both the general public at the eligible screening age as well as healthcare providers. While educating the women who need mammogram screening and lack adequate education on the risks posed by breast cancer may encourage some to change their screening behaviors, others may not or cannot change their behaviors. Health behaviors theories, including the Theory of Care Seeking Behavior, were developed to explain the reasons why people do or do not participate in health promotional programs such as mammogram screening (33). These theories are valuable in understanding and promoting participation in mammogram screening from its broader construct by incorporating habitual and external factors (34). From the perspective of healthcare providers, strategies for supporting and engaging women with lower educational attainment should be emphasized and implemented broadly in the healthcare setting to reduce harmful yet unintended stigmatization by healthcare providers that leads to less preventative care participation (35). These strategies include but are not limited to using non-judgmental vocabulary, and avoiding shame, blame or guilt.
Our study has many strengths, including the use of multiple survey administration and recruitment methodologies and analyses with a weighting strategy. Our limitations include low response rate and the lack of minority population participation. The latter may indicate inaccurate representation of the catchment area’s general population, and our study had a slightly higher proportion of respondents with health care coverage than the Center for Disease Control’s (CDC’s) Behavioral Risk Factor Surveillance System (BRFSS) Oregon’s estimates as well. Further investigation should also account for gender and race discordance between patient and provider to inform methods to increase both rates of screening recommendation and participation.