Breast cancer diagnoses and poor health outcomes disproportionately affect women by race/ethnicity [1]. For example, non-Hispanic White (NHW) women have the highest breast cancer incidence rate, but non-Hispanic Black (NHB) women have the highest mortality rate [1]. From 1998 to 2018, breast cancer incidence in Hispanic and NHW women decreased [2], the incidence for NHB, non-Hispanic American Indian/Alaska Native (NH AI/AN), and non-Hispanic Asian or Pacific Islander (NH API) women increased [2]. For example, during the last two decades, the absolute change in breast cancer incidence rate per 100,000 for NH API women increased by 21.1, while the incidence for NHW women decreased by 11.5 [2].

Breast cancer screening can detect cancer at earlier stages when it is easier to treat [3]. The US Preventive Services Task Force (USPSTF) currently recommends that all women aged 40–74 years receive biennial screening mammography [4]. Although the Affordable Care Act (ACA) requires most insurers to cover mammograms at no cost for women starting at age 40 years, women who are uninsured or underinsured may not have access to affordable screenings and diagnostic testing, which may lead to more advanced-stage breast cancer diagnoses than among those who receive screenings as recommended by the USPSTF [3–5]. The National Breast and Cervical Cancer Early Detection Program (NBCCEDP) (hereafter called Program) aims to reduce breast and cervical cancer morbidity and mortality by providing screening and diagnostic services to women between the ages 40–64 years (breast) and ages 21–64 years (cervical) who have low-income and are uninsured, or underinsured [6]. Participants diagnosed by the Program with cancer or pre-cancer are referred for treatment under the Medicaid Program waiver [6]. Women 65 years or older are eligible for insurance by Medicare but may still receive services through the Program if they don’t have Medicare [7].

A prior modeling evaluation of the Program (1991–2006) found that breast cancer screenings through the NBCCEDP resulted in an additional increase in life expectancy that ranged from 0.031 to 0.081 life-years (LYs) saved per woman screened [8]. Similarly, the same study reported that among 1.8 million women who participated in the Program, 100,800 LYs were saved compared with women who did not participate [8]. This was the first model-based study to quantify the Program’s impact compared with women who did not participate [8]. Since these findings were published, there have been many improvements in technology related to breast cancer screening and changes in recommendations for screening, diagnosis, and treatment [9, 10]. For example, X-ray machines were originally used for mammograms until the Food and Drug Administration (FDA) approved full-field digital mammography in 2000 [9]. The improvements in breast screening technologies have helped to provide a clearer image of the breast for more accurate diagnoses of cancer [10]. The advancement in breast cancer treatment with precision medicine, starting with FDA approval of anti-human epidermal growth factor receptor 2 (HER2) antibodies in 1998 and tyrosine kinase inhibitors in 2007, has significantly increased breast cancer survival [11, 12].

In addition to the changes in breast cancer screening, diagnosis, and treatment technologies, in the past decade, the healthcare system has also changed with the implementation of the ACA [13, 14]. Similarly, the USPSTF recommendations for mammograms have also changed [4, 15]. Based on these changes, the previously estimated impact of the NBCCEDP, measured by LYs saved, may be outdated. Further, a previous model-based study of the Program did not consider differences by race/ethnicity [8].

The objectives of this study are to estimate the following health outcomes: (a) the number of screenings received per woman, (b) LYs saved per woman screened, and (c) the number of screenings per LY saved for a woman who participated in the Program. All the outcomes were compared with the counterfactual scenario if the Program was absent. These estimates were stratified by race/ethnicity—NHW, NHB, NH AI/AN and API, and Hispanic. Because of the small sample size, NH AI/AN and API were combined for the stability of the estimates and henceforth donated as “NH AIAN/API.”

This study provides estimates of the health outcomes of NBCCEDP on breast cancer screening in low-income, uninsured women in the past 10 years. Overall women who participated in the Program received more screenings and gained more LYs than women in no Program scenario. Compared with no Program, women screened through NBCCEDP by all races/ethnicities gained an average of 0.027 LYs (or the equivalent of 10 days) per woman. Among racial/ethnic groups, NHB women were estimated to have the highest benefit, with 0.075 LYs saved (or 27.44 days). They also have the least number of screenings required to gain 1 additional LY by participating in the Program. The estimated LYs saved presented in this paper are based on approximately 15% of the eligible women currently reached by the Program [13]. If the proportion of eligible women screened by the NBCCEDP were to increase, the estimated preventive health benefits for these women would significantly improve [28].

In the United States, access to cancer screening is dependent upon having health insurance [29]. As a result, preventive healthcare measures such as cancer screening or immunizations may be underused [30]. For more than 30 years, the NBCCEDP, through its vast network of public and private health care providers, has played an indispensable role in delivering breast and cervical cancer screening and diagnostic services at no cost to populations with low-income and who are uninsured [8, 13, 31–33]. These efforts have substantial impacts on these populations who have historically experienced high rates of poverty [34] and low rates of health insurance coverage [35]. As reported in this study, NHB and Hispanic women had the highest number of LYs saved, which is an indicator of improved life expectancy for these populations. Further, our findings indicate that when resources are allocated effectively and efficiently to subpopulations that have historically worse health outcomes, the returns on such investments are substantial. Moreover, the results suggest that ongoing investment in outreach and support services for these groups will lead to significant health improvements in the future. Additionally, screening and diagnosing breast cancer at earlier stages not only improves prognosis but also reduces out-of-pocket expenses and financial hardship, often referred to as ‘financial toxicity,’ associated with treatment and its lasting effects [36, 37]. Through screening and early detection provided by the NBCCEDP, eligible women could be spared from the disproportionate financial burdens experienced by many low-income, uninsured, or underinsured women.

At first glance, the estimated LYs saved per woman who participated in the Program might appear small compared to other national preventive health programs, such as immunization programs [38, 39]. However, comparison with other national prevention programs is difficult for many reasons, including differences in study characteristics, data used, populations, and study methodology. These factors notwithstanding, we compared our results with those from previously published studies to contextualize and evaluate the credibility of our findings. This is presented in Table 4, which displays a comparison of the findings in LYs saved per person alongside outcomes from selected population-based preventive health services [8, 31, 38–43]. These data provide some validation for the estimated LYs saved for breast cancer screening in the Program, which compares favorably to other preventive health services, such as colorectal and lung cancer screening. In colorectal screening, the population aged ≥ 50 years screened with colonoscopy every 10 years or every three or five years with adenoma, was reported to save 0.0220 LYs (or 8.04 days) [42]. Similarly, lung cancer screening with low-dose computed tomography was reported to save 0.0316 LYs (or 11.54 days) [41]. On the other hand, breast cancer screening through NBCCEDP gains fewer LYs than other preventive health services. For example, all childhood immunizations were reported to save 0.1233–0.1273 LYs (or 45.04–46.50 days) [38, 39]. Furthermore, current smokers aged 55 years old attending smoking cessation counseling in a lung cancer screening setting were reported to save 0.1188 LYs (or 43.40 days) [40].

Several factors could have contributed to our estimate of 0.0274 LYs saved (or 10.02 days gained), which is 48.9% lower than the previously reported estimate of 0.056 LYs saved (or 20.45 days) [8] for women of all racial/ethnic groups participating in the Program. Among the factors are advancements in breast cancer screening, diagnostic, and treatment technologies that have improved early detection and reduction in morbidity and mortality [9, 44–49]. For example, digital mammography provides clearer images than traditional film mammography and allows for better visualization of abnormal results. Furthermore, targeted treatments such as human epidermal growth factor receptor 2 drugs (e.g., trastuzumab) and endocrine therapies (e.g., tamoxifen and aromatase inhibitors) have also significantly contributed to improved health outcomes in the past years [50–53]. These advancements in medical imaging technologies imply that more women may be screened per life-year saved. For instance, in this study, we found that, on average, 17 screenings would be required to save 1 LY in the Program. The impact of these imaging technologies in reducing cancer-related mortality has been remarkable, however, in the past decade, improvements in access to healthcare services through insurance coverage, adherence to recommended screenings and treatments [4], and individual awareness of overall health and well-being may have also contributed to improved breast cancer health outcomes.

Another factor could be that we used 10 years of data in the current study, while the previous study used 16 years. Other factors contributing to the differences in the reported LYs saved could be differences in study methodology. The previous study adopted a microsimulation model developed by the Cancer Intervention and Surveillance Modeling Network (CISNET), which contains breast cancer natural history, screening history, breast cancer mortality, early detection and treatment, and other-cause mortality modules [46, 54–56]. The previous study did not use USCS data for survival because survival data were not consistently available for all central cancer registries at the time of that analysis. So, the breast cancer survival estimates used in the previous study were based on the parameters obtained in the CISNET models, which reflected the survival of women with breast cancer in the general population, not women who met the eligibility for Program services. In our current study, we used a time-to-event model, which contains screening history as well as early detection and treatment modules. We did not include a natural history component because we wanted to quantify the impact of the Program’s efforts to provide breast cancer screening and diagnostic services at no cost, regardless of the natural history related to breast cancer. We also wanted to estimate the benefits of Program participation by race/ethnicity, which was not included in prior research.

Our study has some limitations. First, our analysis focused only on LYs saved and did not account for health-related quality of life (HRQoL) of the Program participants. The characteristics of HRQoL may include social functioning, physical mobility, capacity for self-care, mental health, pain from mammography screening, and potential harms arising from false-positive diagnoses or overdiagnosis (such as the identification of tumors that would never have become a clinical problem in the patient’s lifetime). These HRQoL measures are usually used as an adjustment factor to the estimated LYs saved— hence called quality-adjusted life-years (QALYs) saved. Second, our study includes only the Program’s direct effects on breast cancer survival and does not consider how the Program may have helped uninsured women gain access to the healthcare system, leading to better care for conditions besides breast cancer. For example, the Well-Integrated Screening and Evaluation for Women Across the Nation (WISEWOMAN) program provides heart disease and stroke prevention screening to Program-eligible women [57]. Under the Breast and Cervical Cancer Prevention and Treatment Act of 2000 (Public Law 106–354), Program participants also have the ability to gain access to breast cancer treatment through Medicaid, which covers many other conditions. Third, we also note that in this study, we did not directly include the benefits of general preventive health services such as reducing obesity, improving physical activity, and reducing excessive alcohol consumption [58]. These primary prevention activities are part of the mission of the CDC’s National Comprehensive Cancer Control Program, which provides health prevention education to everyone, including Program-eligible women [59]. Fourth, as stated earlier, the reported population average of LYs saved appears small because (1) breast cancer does not develop in 7 out of 8 women, (2) in general, screening detects about 87% of cancers early [60], (3) some women in whom cancers are screen-detected would have survived in the absence of screening, and (4) some women in whom cancer is screen-detected early will still die. Fifth, as stated earlier, we tested the Cox proportional hazards model assumption, which indicated a violation. However, many studies have reported that this assumption is often overlooked or fails to hold in medical research [61–67]. Nevertheless, the estimated health outcomes LYs saved for the eligible women in the NBCCEDP were not compromised. Sixth, we acknowledge that while time-to-event models effectively capture short- to medium-term outcomes, they may not fully reflect the long-term effects of screening. Another issue with the data used to derive input parameters (e.g., survival gain with the NBCCEDP) is loss to follow-up, which can introduce bias and may lead to an underestimation of the true effects of screening interventions. Additionally, time-to-event models struggle with time-varying effects. Despite these limitations, a recent study demonstrated that the time-to-event approach could produce results similar to those of the Markov model [68]. Our use of a time-to-event model provides robust estimates for evaluating the NBCCEDP’s impact on breast cancer screening outcomes within a relevant timeframe.

Finally, we did not distinguish the effects of different insurance coverage (e.g., state Medicaid expansion for women less than age 65 years or Medicare coverage for women 65 years or older) or different screening technologies—such as mammography, digital mammography, or other breast cancer screening methods—on the Program’s effectiveness in breast cancer survival. In the NBCCEDP, it is difficult to determine whether the eligible women are uninsured or underinsured. Based on previous studies, this may have affected the outcomes reported in this study [69]. Another limitation of this study is that the program lacks the ability to monitor the long-term breast cancer screening behavior of the eligible women.

Similar to past analyses, our estimates suggest that implementing the Program has improved the LYs of women who participated in the Program during the study period. The estimated improvement in LYs is more profound for NHB and Hispanic women, who usually face significant societal and structural barriers to obtaining affordable, preventive health care services like mammography screening [70, 71]. Our results suggest that the Program is beneficial for all eligible women by upholding its purpose to help improve health outcomes for those receiving preventive health services through timely screening and diagnosis, access to quality care, and by focusing on organizational-level factors that influence screening and early detection [6, 72]. The findings also demonstrate the combined and ongoing efforts by the CDC, states, tribal, and territorial programs, healthcare providers, and community partners to improve access to preventive cancer screening services for low-income, uninsured, or underinsured populations.

In this and other studies, it is noteworthy to emphasize that USPSTF recommends breast cancer screening starting at age 40 years [4]. However, over the past two decades, there has been growing evidence that cases of breast cancer diagnosed before age 40 are increasing [73, 74], and the economic burden associated with breast cancer in this age group is substantial [75]. This trend underscores the need for more sensitive and effective screening methodologies that can be applied to younger populations, particularly for those with more aggressive forms of breast cancer, who may be diagnosed at later stages due to the lack of routine screening recommendations for this age group.

It is important to note that in our study, NHB and Hispanic women have the highest number of estimated LYs (or days) saved and the lowest number of screenings required to achieve an additional LY. These findings indicate that when resources are allocated effectively and efficiently to subpopulations that have historically worse health outcomes, the returns on such investments are substantial. Moreover, the results suggest that ongoing investment in outreach and support services for these groups will lead to significant health improvements in the future. They are encouraging and are consistent with the Program’s efforts over the past 30 years to effectively and efficiently reduce or eliminate health differences that contribute to worse health outcomes for specific populations. Such populations may include people living in rural communities and racial/ethnic groups who have not benefited from the progress made against cancer. The findings reported in this study emphasize the importance of providing preventive health care to populations who might not otherwise have access to these services.

Supplementary Information The online version contains supplementary material available at https://doi.org/10.1007/s10552-025-02006-2.