Rural racial disparities and barriers in mammography utilization among Medicare beneficiaries in Texas: A longitudinal study

Abstract

This study examined rural racial/ethnic disparities in long-term mammography screening practices among Medicare beneficiaries. A retrospective longitudinal study was conducted using 100% Texas Medicare data for women aged 65–74 who enrolled in Medicare between 2010–2013. Of the 114,939 eligible women, 21.2% of Hispanics, 33.3% of non-Hispanic Blacks (NHB), and 38.4% non-Hispanic Whites (NHW) in rural areas were regular users of mammography, compared to 33.5%, 44.9%, and 45.3% of their counterparts in urban areas, respectively. Stratification analyses showed rural Hispanics and NHB were 33% (95% CI, 25% - 40%) and 22% (95% CI, 6% - 36%) less likely to be regular users of mammography compared to their urban counterparts. Major barriers to routine mammography screening included the lack of a primary care provider, frequent hospitalization, and comorbidity. The findings of this study highlight the importance of addressing rural racial disparities in mammography utilization among older women to ensure equitable screening practices for all populations.

Introduction

Breast cancer screening and early detection have demonstrated considerable success and hold great potential as key components of breast cancer control in the future.¹ Several randomized trials and epidemiological studies have shown that mammography is the only screening test that has been proven to lower mortality rates associated with breast cancer.^2,3 Accordingly, the U.S. Preventive Services Task Force (USPSTF) recommends that average-risk women aged 50 to 74 years undergo mammography screening every two years.⁴ Breast cancer statistics from 2022 indicate that approximately half of breast cancer deaths occur in women 70 years or older.⁵

Despite the proven benefits of mammography screening in terms of improved survival rates, these advantages have not been equally distributed among diverse racial groups and geographically disadvantaged populations.^6–8 For instance, non-Hispanic black (NHB) women have a higher likelihood of late-stage breast cancer diagnoses and cancer-related mortality compared to non-Hispanic white (NHW) women due in part to disparities in mammogram access.^8,9 Similar disparities have been observed among Hispanic women, who are more likely to present with larger tumors of higher histologic grade and unfavorable molecular subtypes of breast cancer compared to NHW.^9–11 A recent study conducted using Texas Behavioral Risk Factor Surveillance System (BRFSS) survey data demonstrated significant and persistent rural disparities in mammography screening.¹² While many studies have investigated racial and urban-rural disparities in mammography screening, limited research has examined the interaction of race/ethnicity and urban-rural areas regarding mammography utilization. Additionally, the majority of studies on mammography screening behavior have utilized a cross-sectional design, which is limited in measuring adherence and patterns of long-term mammography screening.^13,14

According to Andersen’s Behavioral Model of Health Services Utilization, the use of screening mammography services is influenced by predisposing factors (e.g., sociodemographic), enabling factors (e.g., healthcare access), and perceived need for care (e.g., breast cancer risk and comorbidity).^15,16 Given that low socioeconomic status, social injustice, and cultural factors contribute to breast cancer disparities,⁹ we hypothesized that rural minority women, particularly Hispanic and NHB, would experience greater disparities in long-term mammography screening compared to NHW women. Additionally, socioeconomic status, healthcare access, and health status could serve as significant barriers for geriatric women in complying with long-term mammography screening, based on previous studies.^3,17–19

The purpose of this study was to examine longitudinal patterns in mammography screening practices and investigate rural racial/ethnic disparity and barriers to these practices among female Medicare beneficiaries. Using 100% Texas Medicare data, we examined individuals’ repeat mammography screening behavior every two years from the age of 65 and up to 10 years. Additionally, we explored factors at the individual and county levels associated with long-term adherence to mammography screening practices.

Methods

Data source

We used 100% Texas enrollment and Medicare claims data from 2010–2019 including Medicare Beneficiary Summary files, Inpatient Claims files, Skilled Nursing Facilities claims files, Outpatient Claims files, and Medicare Carrier files. All data were obtained through a data use agreement with the Center for Medicare and Medicaid Services. The study protocol underwent a thorough review and received approval from the Institutional Review Board (IRB:22–0031) at the University of Texas Medical Branch.

Study population

The study population consisted of female Medicare beneficiaries aged 65–74 from Texas who initially enrolled in Medicare between 2010 and 2013. To maintain consistency, we defined eligible enrollees as those who had enrolled in Medicare at any time during the previous year for four cohorts from 2010 to 2013. All beneficiaries included in the study were 65 years old as of the first day of each cohort year. The study’s inclusion criteria required continuous enrollment in both Medicare Part A and Part B, without any Health Maintenance Organization (HMO) enrollment during the study period. Beneficiaries who were deceased before the age of 67 years, diagnosed with breast cancer, or had undergone mastectomy before the age of 67 years were excluded from the study to ensure a minimal 2-year study period for assessing mammography utilization. Additionally, beneficiaries with missing values for education and income were also excluded. Please refer to Appendix Table 1 for details on selecting study subjects. The final sample size consisted of 114,939 female Medicare beneficiaries between the ages of 65 and 74.

Measures

The outcome of interest was mammography screening following the guidelines set by the USPSTF for breast cancer screening. Mammography utilizations were identified using Current Procedural Terminology (CPT) codes 77052, 77057, 77063, 77065, 77066, 77067 or Healthcare Common Procedure Coding System (HCPCS) codes G0202, G0204, G0206, or G0279. The mammography utilization status of all subjects was assessed every 2 years starting from the beginning of the four cohort years (2010–2013) and continuing until their endpoint follow-up. The endpoint follow-up of beneficiaries was censored if they experienced death, were diagnosed with breast cancer, underwent changes in residence between urban and rural areas, or reached the end of the study period (12/31/2019), whichever occurred first. Any follow-up period of less than 2 years was excluded from the assessment of mammography utilization. Subjects were categorized into three groups based on their mammography utilization: regular users, indicating subjects who complied with breast cancer screening guidelines and had at least one mammogram every 2 years; less frequent users, indicating subjects who had undergone a mammogram but not consistently every 2 years (applied to subjects with at least 4 years of follow-up); and nonusers, indicating subjects who had never undergone a mammogram.

Covariates were selected based on Andersen’s Behavioral Model of Health Services Utilization and previous studies, including sociodemographic factors, healthcare access, and comorbidities.^15–19 We extracted demographic data of Medicare beneficiaries, including age, sex, race and ethnicity, vital status, and zip codes, from the Medicare administrative files. Healthcare utilization, including mammography, hospitalization, and primary care provider (PCP) visits, was extracted from Medicare claims files. To enhance the designation of race/ethnicity, we used the Research Triangle Institute race code and further categorized it into four groups: Non-Hispanic White (NHW), Non-Hispanic Black (NHB), Hispanic, and Other.²⁰ The USDA 2013 Rural-Urban Continuum Codes (UUCC) were used to differentiate metropolitan counties based on the population size of their metro area and classify nonmetropolitan counties according to the degree of urbanization and adjacency to a metro area. They were further grouped into two categories: urban (UUCC codes 1–3) and rural (UUCC codes 4–9).²¹ Taking residence changes into consideration, we assessed the subjects’ residence status every 2 years in conjunction with their mammography utilization. The percentage of high school graduates and median household income were obtained from U.S. Census Bureau American Community Survey (ACS) 5-year estimates from 2014, based on beneficiaries’ ZIP code.²² To calculate the Charlson’s Comorbidity Index (CCI) scores, we used claims data from the first year of enrollment, which were then grouped into four categories (0, 1, 2, 3+). In order to evaluate health status and healthcare utilization, we examined the number of hospitalizations and whether the beneficiaries had a Primary Care Provider (PCP) during the first year after Medicare enrollment. The definition of PCP has been described elsewhere.²³ If a beneficiary visited a PCP on two or more occasions in an outpatient setting, they were considered to have an established PCP.

Statistical analysis

The mammography utilization was first assessed annually (Appendix Fig. 1) and then further examined longitudinally within each 2-year study period for all beneficiaries until reaching their maximum follow-up. The mammography utilization rates were further stratified per race/ethnicity, geographic areas, and the integration of race/ethnicity and geographic areas. For regression analyses, the beneficiaries’ mammography screening status at the endpoint of follow-up was considered to assess long-term compliance in mammography screening practices. Multinomial logistic regression models were employed to estimate odds ratio and 95% confidence interval for mammography utilization comparing regular users and less frequent users to nonusers, while adjusting for covariates. Furthermore, we explored the interaction of geographic areas (urban and rural) and race/ethnicity. All statistical analyses were performed using SAS/STAT (SAS/STAT 9.4, SAS Institute Inc., Cary, NC, USA).

Results

There were 114,939 eligible female Medicare beneficiaries in our study cohort. The characteristics of the study sample and the mammography uptake status at the endpoint of follow-up for all beneficiaries were displayed in Table 1. Overall, the majority of the sample were NHW (78.07%), lived in urban (79.55%), had a CCI score of zero, were not hospitalized within the first year of Medicare enrollment (90.84%), and had an established PCP (52.31%). Regarding mammography utilization, Hispanic women had the lowest percentage of regular users (31.06%) among all racial groups. NHB (43.57%) and NHW (43.76%) had a similar percentage of regular users. Among the beneficiaries, 36.9% of rural individuals were regular users of mammography compared to 43.48% of urban beneficiaries.

Table 1.

Characteristics of study sample by mammography utilization at the endpoint of follow-up

	Total	Mammography Utilization (%)
	N (%)	Nonusers	Regular Users	Less Frequent users
Race/Ethnicity
White	89,728 (78.07%)	22.68	43.76	33.56
Hispanic	14,874 (12.94%)	30.81	31.06	38.13
Black	6,411 (5.58%)	24.8	43.57	39.79
Other	3,926 (3.42%)	26.64	39.79	33.57
Location
Urban	91,432 (79.55%)	22.88	43.48	33.64
Rural	23,507 (20.45%)	28.3	36.09	35.61
Follow-up periods
Median (IQR)	114,939 (100.00%)	23.99	41.97	34.04
Cohort year
2010	25,687 (22.35%)	22.36	38.13	39.51
2011	25,058 (21.80%)	24.18	41.5	34.32
2012	30,552 (26.58%)	23.35	41.75	34.9
2013	33,642 (29.27%)	25.66	45.46	28.89
Median Income
1st Quartile	28,819 (25.07%)	28.51	35.55	35.93
2nd Quartile	28,695 (24.97%)	25.53	39.47	35
3rd Quartile	28,709 (24.98%)	23.14	43.68	33.17
4th Quartile	28,716 (24.98%)	18.74	49.2	32.06
Education
1st Quartile	29,257 (25.45%)	29.2	34.63	36.17
2nd Quartile	28,903 (25.15%)	25.46	39.45	35.09
3rd Quartile	28,235 (24.57%)	22.78	43.49	33.73
4th Quartile	28,544 (24.83%)	18.35	50.54	31.11
CCI
0	80,807 (70.30%)	24.44	42.63	32.93
1	22,376 (19.47%)	19.61	43.12	37.26
2	6,629 (5.77%)	23.55	40.04	36.42
≥3	5,127 (4.46%)	36.47	29.12	34.41
Hospitalization
0	104,412 (90.84%)	23.64	42.45	33.91
1	8,146 (7.09%)	24.12	39.81	36.07
≥2	2,381 (2.07%)	38.64	28.18	33.18
PCP
No	54,819 (47.69%)	33.15	33.86	33
Yes	60,120 (52.31%)	15.63	49.37	35

Abbreviations: CCI, Charlson Comorbidity Index; IQR, interquartile range; PCP, primary care provider.

The longitudinal analyses using 2-year interval revealed a gradual decline in mammography utilization trends among regular users, decreasing from 60.59% to 38.07% over a follow-up period of 2 to 10 years, with a decline rate of 5.63% every two years (Fig. 1a). At the 10-year follow-up mark, 44.09% of these beneficiaries were categorized as less frequent users of mammography, while 17.84% had never undergone a mammogram between the ages of 65–74 (Fig. 1b). Medicare beneficiaries residing in rural areas consistently exhibited lower screening rates compared to their urban counterparts (Fig. 1b). Among all racial/ethnic groups, Hispanic women had the lowest rates of screening rates among all racial/ethnic groups (Fig. 1c). Specifically, among regular users of mammography, rural Hispanic women consistently had the lowest long-term mammography screening rates from 2 to 10 years, followed by rural NHB, in comparison to their counterparts in urban areas, as well as NHW women in both urban and rural areas (Fig. 1d).

Fig. 1.

Percentage of women underwent mammography screening using 2-year interval (a) among all study sample categorized as regular users, less frequent users, and nonusers; (b-d) among regular users and stratified by: (b) geographic areas (urban vs rural); (c) Race/Ethnicity; and (d) the integration of race/ethnicity (NHW: Non-Hispanic White, NHB: Non-Hispanic Black, and Hispanic) and geographic areas (Rural-Urban).

Table 2 presents the results of multinomial regression assessing factors associated with mammography utilization among Medicare beneficiaries after adjusting for covariates. Overall, Hispanic female Medicare beneficiaries were 46% (95% CI, 43%-50%) less likely to use screening mammography regularly compared to NHW. Beneficiaries with established PCP care were 3.31 (95% CI, 3.2–3.42) times more likely to be regular users of mammography, and 2.27 (95% CI, 2.2–2.35) times more likely to be less frequently users of mammography, compared to Medicare beneficiaries without a PCP. Beneficiaries who were hospitalized more than twice a year or had a CCI score of ≥3 were 45% (95% CI, 38%-51%) and 50% (95% CI, 46%-54%) less likely to use screening mammography regularly, respectively, compared to those without hospitalization or with a CCI score of zero. Of the sociodemographic factors, Medicare beneficiaries in the highest quartile of education level or income were 1.65 (95% CI, 1.55–1.76) and 1.28 (95% CI, 1.2–1.37) times more likely to use screening mammography regularly, respectively, compared to those in the lowest quartile of education level or income. Interestingly, there was no significant rural-urban difference in long-term mammography utilization among regular users after adjusting for covariates.

Table 2.

Logistic regression assessing variables associated with mammography utilization among Medicare beneficiaries, compared to nonusers

	Logistic Regression OR (95% CI)
	Regular Users	Less Frequent Users
Race/Ethnicity
White	Ref	Ref
Hispanic	0.54 (0.50, 0.57)	0.81 (0.77, 0.86)
Black	0.94 (0.85, 1.04)	0.89 (0.80, 0.98)
Other	0.83 (0.70, 0.99)	0.91 (0.77, 1.09)
Location
Urban	Ref	Ref
Rural	0.98 (0.88, 1.09)	1.14 (1.03, 1.27)
Following up periods
Median (IQR)	1.27 (1.25, 1.29)	2.23 (2.18, 2.28)
Cohort year
2010	Ref	Ref
2011	1.20 (1.14, 1.26)	1.54 (1.46, 1.62)
2012	1.24 (1.18, 1.30)	1.60 (1.52, 1.69)
2013	1.47 (1.40, 1.55)	2.39 (2.25, 2.54)
Median Income
1st Quartile	Ref	Ref
2nd Quartile	1.11 (1.06, 1.16)	1.07 (1.02, 1.12)
3rd Quartile	1.09 (1.03, 1.15)	1.03 (0.97, 1.09)
4th Quartile	1.28 (1.20, 1.37)	1.18 (1.10, 1.26)
Education
1st Quartile	Ref	Ref
2nd Quartile	1.11 (1.05, 1.16)	1.04 (1.00, 1.10)
3rd Quartile	1.20 (1.14, 1.27)	1.04 (0.98, 1.10)
4th Quartile	1.65 (1.55, 1.76)	1.15 (1.07, 1.23)
CCI
0	Ref	Ref
1	1.04 (1.00, 1.09)	1.29 (1.24, 1.35)
2	0.85 (0.80, 0.92)	1.17 (1.09, 1.26)
3+	0.50 (0.46, 0.54)	0.98 (0.90, 1.06)
Hospitalization
0	Ref	Ref
1	0.94 (0.88, 1.00)	1.05 (0.98, 1.12)
2+	0.55 (0.49, 0.62)	0.82 (0.73, 0.92)
PCP
No	Ref	Ref
Yes	3.31 (3.20, 3.42)	2.27 (2.20, 2.35)

Abbreviations: CCI, Charlson Comorbidity Index; CI, confidence interval; IQR, interquartile range; PCP, primary care provider.

To further examine the rural racial/ethnic disparity, mammography utilization was stratified by the interaction of race/ethnicity and urban-rural areas. Overall, 21.2% of Hispanics, 33.3% of NHB, and 38.4% NHW individuals in rural areas had regular screening mammography, compared to 33.5% of Hispanic, 44.9% of NHB, and 45.3% of NHW individuals in urban areas (Table 3A). Stratification analysis between urban and rural areas for all racial/ethnic groups among mammography regular users showed that Hispanic and NHB Medicare beneficiaries from rural areas were 33% (95% CI, 25%-40%) and 22% (95% CI, 6%-36%) less likely to have screening mammography regularly than their counterparts from urban areas (Table 3B). For NHW beneficiaries who were mammography regular users, there was no significant difference between those living in urban and rural areas (OR, 1.04, 95% CI, .99 – 1.09) after adjusting for covariates (Table 3B).

Table 3.

Mammography utilization status stratified by the interaction of race/ethnicity and geographic areas (A), and stratification analyses between urban (reference) and rural for regular users and less frequent users, compared to nonusers (B).

A
	Total (N)	Regular Users (n, %)	Less Frequent Users (n, %)	Nonusers (n, %)
NHW/Urban	70,139	31,749 (45.3%)	23,239 (33.1%)	15,151 (21.6%)
NHW/Rural	19,589	7,517 (38.4%)	6,873 (35.1%)	5,199 (26.5%)
NHB/Urban	5,652	2,540 (44.9%)	1,768 (31.3%)	1,344 (23.8%)
NHB/Rural	759	253 (33.3%)	260 (34.3%)	246 (32.4%)
Hispanic/Urban	11,962	4,002 (33.5%)	4,523 (37.8%)	3,437 (28.7%)
Hispanic/Rural	2,912	618 (21.2%)	1,148 (39.4%)	1,146 (39.4%)
Other/Urban	3,679	1,466 (39.9%)	1,229 (33.4%)	984 (26.8%)
Other/Rural	247	96 (38.9%)	89 (36.0%)	62 (25.1%)
B
Race/Ethnicity	Regular users OR (95% CI)		Less frequent users OR (95% CI)
NHW	1.04 (0.99, 1.09)		1.10 (1.04, 1.15)
Hispanic	0.67 (0.60, 0.75)		0.95 (0.86, 1.05)
NHB	0.78 (0.64, 0.94)		1.02 (0.84, 1.25)
Other	1.71 (1.21, 2.41)		1.58 (1.11, 2.25)

Abbreviations: NHB, non-Hispanic black; NHW, non-Hispanic white; CI, confidence interval; OR, odds ratio

Discussion

This study reveals a significant rural racial disparity in longitudinal adherence to mammography screening guidelines among female Medicare beneficiaries. Overall, there is a disparity in long-term mammography utilization among rural racial/ethnic minority groups in Texas. Specifically, among Hispanic and NHB female Medicare beneficiaries, those living in rural areas are significantly less likely to have undergone mammography screening regularly compared to their urban counterparts. Rural Hispanic and NHB women have a lower percentage of regular users of mammography compared to NHW women in both rural and urban areas. These findings align with our hypothesis and are consistent with a recent cross-sectional study using Texas BRFSS survey data.¹² Compared to previous studies that have yielded inconsistent results in examining racial/ethnic and urban-rural disparities,^8,14,17,24 our study provides insights into the interaction of geographic areas and race/ethnicity disparities in screening mammography practices. For example, Ahmed et al.’s systematic review and meta-analysis reported similar racial disparity in mammography screening for NHB but not for Hispanic women compared to NHW among women older than 65 years.⁸ Thomson et al. found no racial and urban-rural difference in biennial mammography screening in Virginia.²⁵ These disparities and differences in mammography screening practices could be attributed to the increased diversity in rural populations across the US.²⁶ It is worth noting that Texas has the largest rural population (16.3%) and a significant proportion of Hispanic population (40.2%).²⁶ Therefore, these findings help fill the knowledge gap in examining the interaction between race/ethnicity and geographic areas regarding mammography screening disparities among older women.

An overall decline in mammography screening trends was observed among regular users over a 10-year follow-up period. The decline rate of 5.63% every two years indicates a concerning decrease in the number of regular mammography users. This raises questions about the barriers contributing to reduced screening rates and their potential impact on early detection and breast cancer outcomes, particularly considering mammography’s higher sensitivity, specificity, and cancer detection rates in older women compared to younger women.^18,27

In the present study, we identified several influential factors that affect long-term mammography screening compliance, aligning well with Andersen’s Behavioral Model of Health Services Utilization. Having an established PCP (enabling factor) increased the likelihood of being regular users of mammography, consistent with Flores et al.’s recent findings that patients in the high PCP interaction group had increased long-term adherence to recommended screening mammography.¹⁹ Hospitalization frequency and comorbidity burden (need factors) also affected mammography utilization, potentially due to health complications or reduced accessibility during hospital stays. For example, Zhang et al.’s¹⁸ recent study found that older women with a high burden of functional limitations were less likely to be adherent to screening mammography recommendations. Medicare beneficiaries in counties with higher county levels of education or income were more likely to undergo regular screening, while concerns about healthcare costs had a negative impact on mammography utilization.²⁸ Low socioeconomic status (predisposing factors), such as food-insecure women, had 54% lower odds of reporting breast cancer screening in the past two years compared to food-secure women.²⁹ These findings indicate that socioeconomic and general health status significantly influenced long-term mammography compliance among older women.

In urban areas of Texas, NHB and NHW had similar mammography screening rates, consistent with national reports.⁵ However, NHB women in rural areas had the second lowest screening rate, after Hispanics. Coupled with NHB women having a 40% higher mortality rate than NHW women,⁵ they are at increased risk of experiencing the worst cancer outcomes. While mobile mammography has helped underserved communities,³⁰ it may not sufficiently improve long-term compliance. For instance, Stanley et al.’s study found that patients using the mobile units were less likely to adhere to screening mammography guidelines compared to patients screened at the cancer center.³¹ Targeted and effectiveness mammography screening programs are needed to ensure fairness and justice in breast health for minority populations in rural areas.³

In our study, we observed a large gap in rural racial/ethnic disparities among female Medicare beneficiaries in Texas. Their mammography screening rates are below the national average. For example, Bronner et al. reported an increase from 63% to 65% in screening rates for Medicare beneficiaries aged 67 to 69 from 2011 to 2018.³² This suggests that rural areas in Texas face additional challenges in accessing mammography services. The accelerated closure of rural hospitals in the state may exacerbate existing disparities for minority populations, as mammography has been reported as one of the most commonly available imaging services at rural hospitals.^33,34

The findings of this study have several clinical implications. Firstly, healthcare professionals must recognize the importance of improving long-term compliance with mammography screening among older women in rural minority populations, a group at the highest risk of breast cancer mortality. Secondly, efforts to reduce rural racial disparities among older women should not only address barriers related to healthcare access and socioeconomic factors but also consider their general health status, comorbidities, and life expectancy. Moreover, there is a growing body of evidence recognizing potential negative consequences of mammography screening, particularly among older women, which may include overdiagnosis, false-positive, anxiety, and radiation exposure.^35,36 Therefore, policies addressing these disparities are crucial for balancing the benefits of screening and the potential risks of overuse. Considering there is typically a few years’ delay from the detection of breast cancer through screening to symptomatic breast cancer, the life expectancy and the quality of life of older women need to be carefully evaluated before making mammography recommendations.^36,37 A major obstacle to routine mammography screening for rural monitories is the lack of an established primary care provider. To address these challenges and promote equitable access, various strategies in promoting healthcare access, including telemedicine, IT-based healthcare communication, and federal-designated rural healthcare facilities such as Critical Access Hospital (CAH), Medicare-certified rural health clinics (RHC), federally qualified health centers (FQHC), and Rural Emergency Hospital (REH),^38,39 need to be tested within rural communities.

Limitations

Limitations of this study should be considered. Firstly, the use of Medicare claims data for mammography utilization is subject to errors or omissions, which may lead to underestimation or overestimation of the significance of the research findings. Secondly, this study focused on female Medicare beneficiaries aged 65–74 with fee-for-service plans, which may limit the generalizability of findings to the general population and those with HMO coverage. The results may not be applicable to other age groups or individuals with different insurance coverage. Thirdly, since this study was conducted exclusively among Medicare beneficiaries in Texas, the rural racial/ethnic disparity observed among Hispanic and NHB women may not be generalizable to Hispanic and NHB women residing in other states. Regional variations and contextual factors could influence the observed disparities. Fourthly, county-level education and income were used as covariates to measure socioeconomic status. However, these measures may not fully capture the individual-level socioeconomic status, and there could be potential variations within counties that were not accounted for. Finally, because this study utilized only pre-pandemic Medicare data, it cannot capture the impact of COVID and associated policies, such as the postponement of non-emergent outpatient radiology services, on mammography screening practices. Future research should consider evaluating the effectiveness of interventions (e.g., telemedicine, designation of rural healthcare services) to increase rural healthcare access in promoting routine mammography screening practices among disadvantaged rural populations.

Conclusions

This study provides insights into the interaction of race/ethnicity and geographic areas on mammography screening disparities. Our findings demonstrate significant racial/ethnic disparities in long-term mammography utilization among rural Hispanic and NHB female Medicare Beneficiaries. Lack of a regular PCP and overall health status indicated by hospitalization and comorbidity, were identified as major barriers to long-term adhering to mammography screening guidelines. These findings highlight the importance of addressing rural racial/ethnic disparities and implementing targeted interventions to enhance screening mammography utilization and to achieve equitable screening practices for all populations.