br The modest gains in
The modest gains in BC screening among low-income women in this study were comparable to find-ings from Massachusetts and a previous study of BRFSS data,14,31 though greater improvements were observed in Oregon.30 The lack of significant DD could be due to more widespread and historical sup-port for mammography in low-income populations through initiatives like CDC’s National Breast and
Cervical Cancer Early Detection Program, programs offered by non-profits, and mobile mammography clinics.34,35 Baseline BC screening prevalence was also higher among low-income women compared with CRC screening, leaving less room for improvement. Furthermore, there may be more financial and logistic barriers for colonoscopy as it Cycloheximide is more expensive and requires more preparation than mammography.36,37
Despite improvements in CRC screening among low-income adults, inequalities in screening remain substan-tial, even in states that expanded Medicaid. Overall, in 2016, about half of low-income adults were UTD with CRC screening compared with about two thirds among those with higher incomes. BC screening disparities were less pronounced, but considerable. Provider recom-mendation is a strong facilitator of cancer screening,38,39 and there may be missed opportunities within healthcare encounters to recommend screening in this popula-tion.40−42 Further, a recent study found that physicians ordered fewer preventive services in Medicaid than
privately insured patients.43 It is also possible that some patients may have received a recommendation to be screened, but did not follow through. Other barriers include lack of awareness, logistic hurdles, and cultural beliefs about screening.44−47
The focus of the current study was to examine the association between CRC and BC screening and the tim-ing of Medicaid expansion in low-income adults. Other ACA provisions could influence results as the ACA pro-vided subsidies to low- and middle-income adults with-out alternatives to purchase insurance beginning in 2014. Use of marketplace insurance has grown faster in non-expansion than expansion states, but <4% of adults nationwide have purchased insurance through this mechanism.48,49 Further, the ACA removed cost sharing for preventive services for privately insured adults in
Several limitations influence the interpretation of the cur-rent findings. First, screening data were self-reported and subject to recall bias, though the sensitivity for mammog-raphy and colonoscopy, the most commonly used CRC screening test, is relatively high at 90% and 79%, respec-tively.52 The authors also assessed recent colonoscopy as an outcome and found similar results. BRFSS response rates were approximately 50%, but analyses were weighted to account for non-response, mitigating cyclin bias. Further, there may be unmeasured confounders such as local pro-grams aimed at improving screening rates not captured in adjusted models. Major changes in BRFSS sampling pre-cluded direct comparison before and after 2011; thus, data were not available for pre-expansion VE states. Further-more, the authors were not able to examine changes in insurance type (e.g., Medicaid versus private plans), as these data were not collected throughout the study period. A person could have been insured previously and received a colonoscopy, but this was not captured in the current study. Additionally, household income was used as a proxy for FPL, though an income <$25,000 had a sensitiv-ity and specificity of ≥90% as a surrogate of <138% FPL.
Following the ACA, CRC and BC screening among low-income adults aged 50−64 years increased in Medicaid expansion states. However, the increases were largest and significantly greater for recent CRC screening in VE states compared with non-expansion states, suggesting that large-scale improvements in cancer screening fol-lowing expansion of insurance coverage may take several years. In addition, by 2016, socioeconomic disparities in
CRC and BC screening narrowed more rapidly in Med-icaid expansion states than in non-expansion states, though inequalities persist in all states. In addition to expanding insurance, other strategies, including physi-cian recommendation and outreach, are needed to improve cancer screening in underserved populations. Further research on the potential public health impact of expansions is needed.
Stacey Fedewa, Robin Yabroff, Ann Goding Sauer, Xuesong Han, and Ahmedin Jemal are employed by the American Cancer Soci-ety as part of the Surveillance and Health Services Research Program, which received a grant from Merck, Inc. for intramural research. However, their salaries are funded solely through American Cancer Society funds.