Impact of Access to Breast Care For West Texas Program on Early Detection and Regional Breast Cancer Mortality

Abstract

Introduction

This study aimed to assess whether the Access to Breast Care for West Texas (ABC4WT) program impacted regional breast cancer detection and mortality in the Texas Council of Governments (COG)1 region.

Methods

Interrupted time series analyses were utilized to evaluate the impact of the intervention. Spearman’s rank correlation and cross-orrelation analyses were performed to assess the relationship between the total number of screenings and (i) the total number of breast cancer detected and (ii) the proportion of early-stage cancer detected and the (pre-whitened) residuals. A three-way interaction model compared pre-and post-intervention mortality in COG 1 with the COG 9 region (control).

Results

Increased screening rate was associated with increased breast and early-stage cancer incidences (P = .001 and P = .002, respectively). There were significant positive cross-correlations between the total number of screenings and the total number of breast cancer detected (r = .996) and the proportion of early-stage cancer detected (r = .709) without a lag even after pre-whitening. Univariate analysis showed that regional mortality decreased with time (P < .001) and after intervention (P = .001). Multivariate analysis did not show any significant difference in time (P = .594), intervention (P = .453), and time and intervention interaction (P = .273). The three-way interaction model showed no difference in the baseline mortality and pre-intervention trend difference in COG 1 and COG 9 regions. However, there was a significant pre-post intervention trend difference in mortality COG 1 compared to the COG 9 region (P = .041).

Conclusion

Implementing the ABC4WT program was associated with the early detection of breast cancer and reducing regional mortality in the COG 1 region.

Keywords

breast cancer cancer screening outreach mortality access to breast care for West Texas program prevention

Introduction

Breast cancer is 1 of the commonest cancers, predominantly affecting the female population in the United States (US) and Texas.¹ The age-adjusted mortality rate in Texas in 2018 was around 20.1 per 100,000, and the incidence was 115.5 per 100,000.² Even though breast cancer-related mortality has declined over the years and the incidence of breast cancer has remained stable, there are notable differences between rural and urban settings.¹ In rural settings, more people are uninsured, have high unemployment rates, and have low levels of education.³ Furthermore, they have difficulty in reaching out to healthcare facilities to obtain routine care and screening.⁴ Therefore, many patients present when symptomatic or are lost for follow-up after an abnormal mammogram screening. For instance, from 2014 to 2018, the incidence of breast cancer in rural areas was 104.5, compared to 115.4 per 100,000 in urban areas.⁵ This noteworthy difference in incidence could be due to variability in healthcare access, including screening and early detection. Furthermore, a smaller proportion of cases were diagnosed at a localized stage in rural counties (58%) than urban areas (61%), indicating room for improvement by local outreach programs.⁵

While screening mammograms are recommended for all women older than 50 according to US Preventive Services Task Force⁶ or for those older than 40 according to the American Cancer Society,⁷ many women in underserved communities do not undergo routine mammography screening. Leung et al⁸ reported that women residing in rural areas were less likely than those in urban regions to have ever had a mammogram (odds ratio (OR) [95% confidence interval] = .74, [.62-.89]) or to have an up-to-date mammogram (OR = .59, [.49-.70]).⁸ One particular reason is that many rural people lack early access to health care due to a lack of insurance coverage. Evidence suggests that disenrollment from insurance plans like Medicare or having no insurance contributes to late detection of cancers and treatment delay, which heavily burden the cancer outcome, especially in the case of breast cancer in rural communities.⁹ Several outreach programs have been functioning nationwide to increase access to breast cancer screening services and reduce cancer-related health care burden.^10,11 However, the national screening program serves only 10.6% of the underserved population. Furthermore, the impact of the federal screening program on the mortality rate has been questioned.¹²

Similarly, Texas Medicaid Breast and Cervical Cancer Services (BCCS) is a statewide preventive program. Unfortunately, BCCS serves only a limited number of counties in Texas, which does not provide adequate coverage, especially in rural areas. For instance, in 2010, only one-third of the counties in Texas had a BCCS provider.¹³ Therefore, the importance of local outreach programs has been emphasized.^13,14

The Cancer Prevention & Research Institute of Texas (CPRIT) is the most significant statewide cancer research and prevention investment in the history of the US. Several cancer-related outreach programs have been implemented via CPRIT funding, and it has been serving Texas for 20 years. Access to Breast Care for West Texas (ABC4WT) program continues to serve in the Texas panhandle area - Council of Governments (COG) 1 since 2010. In this study, we evaluated the outcome of our outreach program. We compared regional breast cancer mortality data and our quantitative assessments of the program to assess whether it impacts the regional mortality rate, the total number of breast cancer detected, and the proportion of early-stage cancer detected. Furthermore, we have compared mortality rates in the COG 1 region with a similar area in West Texas, which has not been served by the ABC4WT program (COG 9) (Figure 1).

Figure 1.

Regional differences in breast cancer mortality in the 2000-2009 period vs 2010-2019 period in council of governments 1- panhandle regional planning commission (COG 1) and council of governments 9- permian basin regional planning commission region 9 (COG 9) regions.

Methods

Intervention

ABC4WT is involved in outreach, awareness campaign, education, and service programs. The program has particularly focused on rural West Texas communities, in addition to metropolitan areas. The main objectives of the ABC4WT program were: (i) to build a West Texas breast cancer community network, (ii) to increase breast cancer screening rates through education and outreach, (iii) to enhance comprehensive counseling for high-risk women and offer chemoprevention, and (iv) to prevent advanced breast cancer in targeted underserved West Texas populations.

Data Sources and Study Population

As the study involved secondary analysis of the information retrieved from publicly available anonymized datasets, ethical review and approval were waived. Details regarding services provided by the ABC4WT program, including the number of screenings, were obtained by ABC4WT prospective data from outreach.¹⁵ The number of cancer patients and their stages at diagnosis in the Texas panhandle area were obtained from Cancer Registry.¹⁶ Stages 0, 1, and 2 were considered early-stage breast cancer, and stages 3 and 4 were considered late. COG 9 (Permian Basin Regional Planning Commission Region 9) was selected as the control region. It is geographically located in the West Texas region and demographically similar to COG 1 region in terms of the proportion of rural countries, population at risk, and area^17,18 (Figure 1). Age-adjusted breast cancer mortality rates for COG 1 and COG 9 regions and the number of breast cancer detected were obtained by Texas Cancer Registry.¹⁹

Statistical Methods

First, interrupted time series with linear regression models^20,21 were used to assess the impact of the intervention (ABC4WT program) on the age-adjusted mortality rate in the COG 1 region. Models were constructed after checking for autocorrelation. The Durbin-Watson test was used to check autocorrelation in fitted models. As the breast cancer incidence and mortality time series were stationary, the analyses were conducted using interaction models by checking time and intervention and time and intervention interaction. We used a multivariate linear regression that modeled the interaction between time and intervention to predict mortality.

Second, multivariate regression models with interaction were constructed to examine the association between (i) breast cancer incidence and (ii) early-stage breast cancer incidence and the screening rate. We used a multivariate linear regression that modeled the interaction between time and screening rate to predict the incidences. Spearman’s rank correlation coefficient was calculated to assess the relationship between the total number of patients screened and (i) the total number of breast cancer detected, and (ii) the proportion of early-stage cancer detected. The total number of screening time series and (i) the total number of breast cancer detected over the years, and (ii) the proportion of early-stage cancers detected over the years were assessed for cross-correlation. Each pair of time series were compared and explored the temporal relationship through the cross-correlation function.²²

Finally, a three-way interaction model was constructed to assess the changes in mortality in the COG 1 region compared to the COG 9 region, which is considered a control region where the intervention (ie, the ABC4WT program) was not implemented. In this model, the pre-intervention trend difference (ie, the change in mortality over time in COG 1 before executing the intervention compared to the change in mortality over time in COG 9 region) was compared to the post-intervention trend difference (ie, the change in mortality over time in COG 1 before implementing the intervention comparing to change in mortality over time in COG 9 region). This linear regression model independent variables were time, region, and pre-or post-implementation of the intervention. Statistical analyses were performed using R statistical software (version 4.0.2).

Results

COG 1 region comprises 26 counties, and 80% were considered rural, and only 5 counties were considered urban (Figure 1, Table 1), whereas COG 9 region has 17 counties, out of which 82% (14) were deemed to be rural. Roughly 3,240,317 indirect contacts were made through social and traditional media. Over the 10 years, around 39 621 direct contacts were made through health fairs and outreach events. A total of 108,993 were screened for breast cancer, which was 5.1% of the population at risk in the COG 1 region. Screening rates and incidence for each stage of breast cancer in the COG 1 region during 2000-2019 are shown in Figure 2.

Table 1.

Regional Comparison.

	Council of governments 1- panhandle regional planning commission (COG 1)		Council of governments 9- permian basin regional planning commission region 9(COG 9)
Area	25,752 sq. miles		23,443 sq. miles
Number of counties	26 counties: 21 (80.8%) rural 5 urban (19.2%)		17 counties: 14 (82%) rural 3 urban (18%)
Population (2020 est.)	434, 358		487, 845
Year	2000-2009	2010-2019	2000-2009	2010-2019
Population at risk	2,043,052	2,138,810	1,923,087	2,241,277
Total number of deaths	551	494	470	455
Crude mortality rate	27	23.1	24.4	20.3
Age-adjusted rate [95% CI]	23.7 [21.8, 25.8]	19.6 [17.4, 21.2]	23.6 [21.5, 25.8]	20.1 [18.1, 22.1]

Figure 2.

Screening rates and incidence for each stage of breast cancer in COG 1 region during 2000-2019.

A multivariate linear regression model with an interaction between time and the rate of screening was constructed to examine associations between the breast cancer incidence over the years in the COG 1 region with the rate of screening; no significant differences were seen between screening and time (P = .204 and P = .737, respectively). However, significant time and screening intervention indicated that an increased rate of screening by the ABC4WT program was associated with increased detection of breast cancer (Figure 2 and Table 2). Likewise, Spearman’s rank correlation analysis showed a positive correlation between the total number of screening time series and the total number of breast cancer detected (r = .996, P < .001; Figure S1 in Supplement). There was a positive lagging between time series when the total number of screening time series and the total number of breast cancer detected over the years were assessed for cross-correlation. However, pre-whitening had eliminated the lagging. A significant positive cross-correlation remained between the total number of breast cancer detected in the COG 1 region and the total number of screenings at 0 lag.

Table 2.

Multivariate Regression Models Were Constructed to Predict the Breast Cancer Incidence in COG 1 and the Incidence of Early-Stage Breast Cancer.

Variable	Estimate	SE	t-value	P-value
Model 1-association between the breast cancer incidence in COG 1 and the total number of patients screened
Intercept	129.300	4.854	26.637	<.001
Year	−.260	.774	−.336	.737
Number of screenings per 100 000	−.004	.003	−1.271	.204
Year X number of screenings per 100 000	.001	.000	3.430	.001
Model 2-association between the early-stage breast cancer incidence in COG 1 and the total number of patients screened
Intercept	99.980	4.942	20.231	<.001
Year	−1.004	.788	−1.274	.203
Number of screenings per 100 000	<.001	.003	−.142	.887
Year X number of screenings per 100 000	.001	<.001	3.085	.002

SE-Standard error; Durbin-Watson test: Model 1- P = .272, Model 2- P = .806.

When multivariate linear regression was constructed to examine the early-stage breast cancer incidence over the years and the rate of screenings, a significant positive effect was seen in the time and rate of screening interaction, indicating that screening increased the detection of early-stage breast cancer (P = .002) (Table 2). Furthermore, the proportion of early-stage cancer detected over the years was positively correlated with the total number of screening time series without lagging between the 2 series. Spearman’s rank correlation showed a positive correlation (r = .709, P = .028, Figure S1 in Supplement).

Univariate models showed a significant difference between the intervention vs mortality rates in COG 1 region and time vs morality rates in COG 1 region. The mortality rate decreased with each unit increased in time (P < .001). Similarly, the mortality rate in 2010-2019 reduced significantly compared to the 2000-2009 period (P = .001), indicating the mortality rate decreased after implementing the ABC4WT program. However, the multivariate analysis did not show any significant difference between time, intervention, and time and intervention interaction (P > .05) (Table 3, Figure 3).

Table 3.

Interaction Model Comparing Pre-and Post-Intervention Age-Adjusted Death Rate in COG 1 Region.

	Estimate	SE	t-value	P-value
Univariate model 1
(Intercept)	25.506	1.054	24.206	<.001
Year	−.445	.095	−4.698	<.001
Univariate model 2
(Intercept)	23.630	.845	27.961	<.001
Intervention	−4.712	1.195	−3.942	.001
Multivariate interaction model
(Intercept)	24.447	1.694	14.433	<.001
Year	−.149	.273	−.544	.594
Intervention	3.561	4.625	.770	.453
Year X intervention	−.438	.386	−1.134	.273

Intervention- ABC4WTprogram; SE-Standard error; Durbin-Watson test: Univariate model 1- P = .592, Univariate model 1- P = .864, Multivariate Interaction model-P = .632.

Figure 3.

The mortality rate in COG 1 region. In solid black line shows how the slope of mortality changed after implementing the ABC4WT program. Values predicted by a model in the absence of the intervention (counterfactual) are shown in gray.

Mortality during pre-intervention and post-intervention times in COG 1 was compared with COG 9. A three-way interaction model showed no difference in time, region (COG 1), and time and region (COG 1) interaction (P > .05). Therefore, it was evident that the baseline mortality rates and the slopes (pre-intervention trend difference) in COG 1 and COG 9 regions between 2000 and 2009 were not different. However, after implementing the intervention (between 2010 and 2019), the difference in slopes of mortality rates between COG 1 and COG 9 had a significant negative effect after the initiation of intervention compared to pre-intervention. Hence, there was a statistically significant pre-post intervention trend difference (P = .041) (Table 4, Figure 4.). Therefore, it was evident that the mortality in the COG 1 region decreased during the 2010-2019 time period compared to mortality in the COG 9 region. This change in slope was significantly different from the change in slope from 2000 to 2009. In conclusion, implementing the ABC4WT program in the COG 1 region was associated with reducing the regional mortality rate.

Table 4.

Interaction Model Comparing Mortality Rate Between COG 1 Region vs Control (COG 9) Before and After Implementing the ABC4WT Outreach Program.

	Estimate	SE	t-value	P-value
Intercept	25.780	1.608	16.029	<.001
Year	−.460	.301	−1.527	.137
Region (COG 1)	−1.482	2.275	−.651	.519
Post-intervention	−7.479	2.275	−3.288	.002
Year x region (COG 1)	.312	.426	.731	.470
Year x post-intervention	.844	.426	1.980	.056
Region (COG 1) x post-intervention	4.738	3.217	1.473	.151
Year x region (COG 1) x post-intervention	−1.282	.603	−2.127	.041

Intervention- ABC4WTprogram; SE-Standard error; Durbin-Watson test: P = .956.

Figure 4.

Scatter plot with fitted regression lines presenting the interaction effect of time and regional differences on the age-adjusted mortality rate of breast cancer patients before and after implementing the intervention (ie, ABC4WT program). The intervention group is COG 1 region, and the control group is COG 9 region. Before implementing the ABC4WT program, the mortality rates in COG 1 and COG 9 regions were not different. Following the ABC4WT program, there was a significant pre-post trend difference indicating the age-adjusted mortality rate was decreased compared to control (P = .041).

Discussion

In this study, we have tried to assess the impact of the ABC4WT program that has served the Texas Panhandle, which primarily comprises rural communities, for over a decade. ABC4WT program includes outreach, education, and service programs where we mainly focus on increasing awareness and participation in screening. We found that with increased screening by the ABC4WT program, the detection of breast cancer, specifically the incidence of early-stage cancer, has increased markedly. Furthermore, we found that the age-adjusted mortality rate in COG 1 region is low compared to COG 9 region, where the ABC4WT program has not been implemented.

The majority of Texas counties are rural. There are 172 rural counties, and about 11% of Texans live in these counties. COG 1 comprises mainly rural counties and a few urban counties, rendering COG 1 region vulnerable to rural and urban challenges. ABC4WT project was funded by CPRIT in 2010 with a crucial intent to improve early breast cancer detection in this region. Therefore, we focused on increasing the breast cancer screening rate in the target population group in Texas Panhandle for secondary prevention via early detection. According to our study, we successfully increased breast cancer detection, specifically at an early stage.

Mammography rates remain lower among uninsured, low-income, and low-education groups even though overall mammography rates have increased over 30 years.²³ Therefore, age and racial and social disparities have been identified as factors governing the differences in cancer incidence rates.^2,24,25 It was found that rural counties have a lower per-capita income and a lower percentage of adults below the poverty level, and a higher unemployment rate. These variations between rural vs urban have differing impacts on cancer diagnosis. Many studies revealed differences in adherence to breast cancer screening among women living in rural vs urban communities.^26,27 Leung et al⁸ showed that the stage at which breast cancer is diagnosed differs between women living in urban vs rural areas. Here, women in rural settings present at late stages of breast cancer more than in urban settings. For instance, this very notion is evident in West Texas.²⁸ Several studies have shown that rural communities in West Texas are reluctant to obtain breast cancer screening, leading to later presentation and poor outcomes.^29,30 Interestingly, we found a strong positive correlation between screening rates and the incidence of breast cancer in the COG 1 region, specifically early-stage breast cancer. Hence, it is apparent that the ABC4WT program also successfully reached this difficult-to-reach underserved population. The difference in breast cancer screening participation in rural and urban communities could be linked to a lack of proper screening facilities or unequal distribution.³¹ By implementing a successful outreach program, we hope to have subdued that challenge to a certain extent.

Although studies showed a reduction in later-stage breast cancer,^32-36 some studies pointed out that early detection did not decrease late-stage breast cancer and showed substantial overdiagnosis occurred by screening.^37-39 Reason for this controversy may be that screening benefits are not straightforward and are subject to several factors, including differences in patient population (age, racial/ethnic, and socioeconomic disparities) and study design.³⁵ For instance, Broeders et al⁴⁰ suggest methodological differences, such as the length of the follow-up period, the definition of advanced BC, and the definition of the age groups, which limit the interpretability of results coming from observational studies on this notion. According to our study, we have seen an increase in total breast and early-stage cancer but no reduction in late-stage cancer incidence. As shown in previous studies,⁴¹ lack of adequate follow-up period could be the underlying cause for not seeing a reduction.

Before implementing the ABC4WT outreach program, breast cancer mortality for 1990-1999 and 2000-2010 was 27.3 per 100,000 and 23.1 per 100,000, respectively. From 2010 to 2018 mortality rate was 19.1 per 100,000.¹⁹ So, it is apparent that there has been a decline in the mortality rate over time, probably due to the overall improvement of healthcare facilities and patient awareness. Increased detection and quick diagnosis of early-stage tumors that otherwise would have metastasis may have contributed to improving motility. Early diagnosis can lead to early access to treatment as well. Our study supports the temporal relationship between the decline in mortality and the comprehensive ABC4WT program. We explored if the ABC4WT program has altered the trajectory of the decline, which means a decrease in mortality rate furthered over time once the outreach program was implemented. Despite the difference in the overall mortality rate between 2000-2009 vs 2010-2019, there was no significant impact on the mortality rate reduction in the COG 1 region compared to pre-vs post-intervention. Failure to show a difference in the slope could be due to the inadequacy of time points since we have only had the ABC4WT program for nearly a decade, and to see an actual benefit, we might have to follow up on this for a longer duration. Since breast cancer has a lag time, we cannot expect a reduction in breast cancer mortality as soon as detection is improved. One study showed that it takes at least 2 years from the intervention for any difference to emerge in mortality rates by screening and early detection.⁴² Hence, a longer time is required to get a meaningful difference in mortality rate at a population level. Nonetheless, we saw a significant impact on mortality when we compared the mortality rate in COG 1 with a mortality rate of a similar region that did not have the ABC4WT program, which provides evidence to suggest that the ABC4WT program was effective.

There is a disparity regarding treatment adherence between rural and urban populations. Rural residents face limited access to health services, fewer providers, higher costs, longer distances to travel to reach healthcare services with limited transportation options, and inadequate or under-utilized infrastructure.^43,44 Therefore, due to the issue of accessibility, rural residents are less likely to receive treatment consistently, which leads to poor disease outcomes.⁴ For instance, Thomas et al⁴⁵ revealed that older age, increased poverty, and rural geographical location are barriers to chemotherapy in West Texas. Similarly, Robertson et al⁴⁶ showed that the distance to cancer centers is positively associated with poorer outcomes. Therefore, 1 possible explanation for decreasing mortality following the ABC4WT program is increased outreach, leading to increased treatment adherence.

The importance of local outreach programs has been emphasized because, despite the availability of appropriate services and access mechanisms placed by nationwide or statewide programs, the reach of these services is compromised without awareness and targeted change in local cultures and beliefs toward screening.^12,13 Evidence suggests that effective patient navigation improves cancer screening, reduces breast care delays, helps patients overcome barriers to accessing care, and provides psychosocial support and care coordination.⁴⁷ In the ABC4WT program, we had local community activists sensitive to local cultures trained to approach their communities, effective patient navigation services, and an organizational structure focused on public-private-community partnerships. These mechanisms introduced systems change where timely communication and flow of information were tremendously helpful. For instance, women could get screening at their local facilities, with suspicious screening results being evaluated by university-based experts. These women were then navigated through the system via diagnostic work-up and referral to appropriate treatment including financial counseling and support.

Partnerships with community-based organizations, health care systems, and insurers are encouraged to set up a successful outreach program.⁴⁷ In the ABC4WT program, we have tried to educate the community via reaching the public through community-based events (health fairs, faith-based events), multimedia, and social media. We conducted both group education sessions and one-on-one education sessions. We have achieved network building with partners in the Panhandle region. Several barriers were resolved through repeated negotiations, emphasizing community service and networking with neutral agencies in the Panhandle site. Border et al²⁹ showed that the Hispanic community is an underserved population in breast cancer screening in West Texas. However, we have been very successful in reaching the Hispanic community. We experienced some hindrance in some smaller pockets of refugees and other African cultures. To overcome that barrier, we used peer trainers, Women Inspiring, Serving and Educating (WISE) Women who reside in communities to educate women about the importance of breast health made inroads in communities, especially African American and Hispanic populations.

Furthermore, innovative means of reaching potential groups of individuals have been encouraged. For instance, McElfish et al¹⁴ showed that a mobile mammovan recruited the rural underserved population better, which was an innovative intervention. Similarly, we used a different tactic to overcome a reduced follow-up rate after initial contact for scheduling mammogram appointments. Here, we negotiated blocks of time with mammogram providers that allowed us to do “on the spot” scheduling at the first encounter with women either through phone, one-on-one or out in the field at group events.

The study had several strengths and limitations. Regarding strengths, we used an interrupted time series design, a robust study design for studying the impact of health interventions.^20,48,49 Then, a control group was used to compare the effects of our outreach program. Next, we used cancer incidence to explore the outcomes of the ABC4WT program, explicitly focusing on early-stage breast cancer. Regarding limitations, we had only a limited number of data points in our time series. Therefore, we could not avoid sample size limitations, which is necessary to generate sufficient power to estimate the regression coefficients.^50,51 Finally, we can only show an association between mortality rates and the implementation of the ABC4WT program. We cannot prove causation as cancer mortality has various contributory factors.

In conclusion, the ABC4WT program successfully served the West Texas population. We believe that our program significantly impacts the detection of early-stage breast cancer and reduces overall mortality. The intervention, screening strategies, and results would have critical implications for future public health policymaking, and the findings of this study can be generalized to the population with geographic and race-specific similarities.

Supplemental Material

Supplemental Material - Impact of Access to Breast Care For West Texas Program on Early Detection and Regional Breast Cancer Mortality

Supplemental Material for Impact of Access to Breast Care For West Texas Program on Early Detection and Regional Breast Cancer Mortality by Chathurika S Dhanasekara, MD, PhD, Hafiz Khan, PhD, and Rakhshanda L Rahman, MD, FRCS, FACS in Cancer Control

Footnotes

Acknowledgments

The study group would like to thank the Cancer Prevention & Research Institute of Texas (CPRIT) for funding the Access to Breast Care for West Texas (ABC4WT) program.

Author Contributions

Conceptualization, RLR, and CSD; Methodology, CSD; Software, CSD; Validation, CSD, HK, and RLR.; Formal Analysis, CSD; Investigation, CSD; Resources, RLR; Data Curation, CSD; Writing – Original Draft Preparation, CSD; Writing – Review & Editing, CSD, HK, and RLR; Visualization, CSD; Supervision, RLR; Project Administration, RLR All authors have read and agreed to the published version of the manuscript.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Ethical Approval

Ethical review and approval were waived for this study as the study involved secondary analysis of the information retrieved from publicly available anonymized datasets.

ORCID iDs

Chathurika S Dhanasekara

Hafiz Khan

Supplemental Material

Supplemental material for this article is available online.

References

Weir

Thompson

Soman

Møller

Leadbetter

. The past, present, and future of cancer incidence in the United States: 1975 through 2020. Cancer 2015;121(11):1827-1837.

U.S. Cancer Statistics Working Group. U.S. Cancer , Statistics Data Visualizations Tool, Based on 2020 Submission Data (1999-2018). U.S. Department of Health and Human Services, Centers for Disease Control and Prevention and National Cancer Institute. Available online. www.cdc.gov/cancer/dataviz. Updated June 2021. accessed February 11, 2022.

Guy

Tangka

FKL

Hall

Miller

Royalty

. The reach and health impacts of the national breast and cervical cancer early detection program. CCC 2015;26(5):649-650.

Meden

St John-Larkin

Hermes

Sommerschield

. Relationship between travel distance and utilization of breast cancer treatment in rural northern Michigan. JAMA 2002;287(1):111.

Texas Cancer Registry . Cancer epidemiology and surveillance branch. Cancer in Texas 2021. Austin, TX: Texas Department of State Health Services; 2021.

Siu

US Preventive Services Task Force . Screening for breast cancer: US preventive services task force recommendation statement. Ann Intern Med 2016;164(4):279-296.

Oeffinger

Fontham

ETH

Etzioni

, et al. Breast cancer screening for women at average risk: 2015 guideline update from the American cancer society. JAMA 2015;314(15):1599-1614.

Leung

McKenzie

Martin

McLaughlin

. Effect of rurality on screening for breast cancer: A systematic review and meta-analysis comparing mammography. Rural Rem Health 2014;14(2):260-272.

Tarazi

Bradley

Bear

Harless

Sabik

. Impact of medicaid disenrollment in Tennessee on breast cancer stage at diagnosis and treatment. Cancer 2017;123(17):3312-3319.

10.

Lee

SJC

Higashi

Inrig

, et al. County-level outcomes of a rural breast cancer screening outreach strategy: A decentralized hub-and-spoke model (BSPAN2). Transl Behav Med 2017;7(2):349-357.

11.

Howard

Tangka

FKL

Royalty

, et al. Breast cancer screening of underserved women in the USA: Results from the National Breast and Cervical Cancer Early Detection Program, 1998–2012. CCC 2015;26(5):657-668.

12.

Howard

Ekwueme

Gardner

Tangka

Miller

. The impact of a national program to provide free mammograms to low-income, uninsured women on breast cancer mortality rates. Cancer 2010;116(19):4456-4462.

13.

Inrig

Tiro

Melhado

Argenbright

Craddock Lee

. Evaluating a de-centralized regional delivery system for breast cancer screening and patient navigation for the rural underserved. Tex Public Health J 2014;66(2):25.

14.

McElfish

Lee

Runnells

Henry-Tillman

Kadlubar

. Mobile mammography screening as an opportunity to increase access of rural women to breast cancer research studies. Breast Cancer. 2019;13:1178223419876296.

15.

Cancer Prevention and Research Institute of Texas . Final progress report. Available online. https://cpritgrants.org/. Accessed August 18, 2021.

16.

National Cancer Institute . Cancer registry. Available online. https://seer.cancer.gov/statistics/. Accessed on Feruary 18, 2022.

17.

Texas Demographic Center . Redistricting Data for Texas Counties, 2000-2020. Available online. https://demographics.texas.gov/InteractiveTools/2021/CBRedistrictingCounty. Accessed on March 22, 2022.

18.

Texas Association of Regional Councils . Texas Association of Regional Councils. Available online. https://www.txregionalcouncil.org. Accessed 22 March 2022.

19.

Texas Cancer Registry . Age-Adjusted cancer mortality rates by council of government in Texas, female breast 2000 - 2018. Available online. http://cancer-rates.info/tx/. Accessed September 23, 2021.

20.

McDowall

McCleary

Bartos

. Interrupted Time Series Analysis. Oxford, UK: Oxford University Press; 2019.

21.

Penfold

Zhang

. Use of interrupted time series analysis in evaluating health care quality improvements. Acad Pediatr 2013;13(6):S38-S44.

22.

Biglan

Ary

Wagenaar

. The value of interrupted time-series experiments for community intervention research. Prev Sci 2000;1(1):31-49.

23.

Polgreen

Yang

Bohnett

Cavanaugh

. A time-series analysis of clostridium difficile and its seasonal association with influenza. Infect Control Hosp Epidemiol. 2010;31(4):382-387.

24.

National Cancer Institute . Cancer trends progress report. Available online. https://progressreport.cancer.gov. Updated July, 2021. Accessed February 16, 2022.

25.

Keegan

THM

Kurian

Gali

, et al. Racial/ethnic and socioeconomic differences in short-term breast cancer survival among women in an integrated health system. Am J Publ Health 2015;105(5):938-946.

26.

DeSantis

Gaudet

, et al. Breast cancer statistics, 2019. CA A Cancer J Clin 2019;69(6):438-451.

27.

Doescher

Jackson

. Trends in cervical and breast cancer screening practices among women in rural and urban areas of the United States. J Publ Health Manag Pract 2009;15(3):200-209.

28.

Shete

Deng

Shannon

, et al. Differences in breast and colorectal cancer screening adherence among women residing in urban and rural communities in the United States. JAMA Netw Open. 2021;4(10):e2128000.

29.

Pruitt

Eberth

Morris

Grinsfelder

Cuate

. Rural-urban differences in late-stage breast cancer: do associations differ by rural-urban classification system? Tex Public Health J 2015;67(2):19.

30.

Borders

Warner

Sutkin

. Satisfaction with health care and cancer screening practices among women in a largely rural region of West Texas. Prev Med. 2003;36(6):652-658.

31.

Khan

Rasmussen

Gabbidon

, et al. Disparities in breast cancer survivors in rural West Texas. Cancer Control. 2021;28:10732748211042125.

32.

Elting

Cooksley

Bekele

, et al. Mammography capacity: Impact on screening rates and breast cancer stage at diagnosis. Am J Prev Med 2009;37(2):102-108.

33.

Nyström

Wall

Rutqvist

, et al. Breast cancer screening with mammography: Overview of Swedish randomised trials. Lancet 1993;341(8851):973-978.

34.

Allan

Ball

Alston

. Developing sustainable models of rural health care: A community development approach. Rural Rem Health 2007;7(4):1-13.

35.

Alston

. From local to global: Making social policy more effective for rural community capacity building. Aust Soc Work 2002;55(3):214-226.

36.

Thomas

Class

Gandhi

Bambhroliya

K-A

Brewster

. Demographic, clinical, and geographical factors associated with lack of receipt of physician recommended chemotherapy in women with breast cancer in Texas. CCC 2019;30(4):409-415.

37.

Robertson

Campbell

Smith

, et al. Factors influencing time from presentation to treatment of colorectal and breast cancer in urban and rural areas. Br J Cancer 2004;90(8):1479-1485.

38.

Henderson

Tossas-Milligan

Martinez

, et al. Implementation of an integrated framework for a breast cancer screening and navigation program for women from underresourced communities. Cancer 2020;126:2481-2493.

39.

Cook

Campbell

Shadish

. Experimental and Quasi-Experimental Designs for Generalized Causal Inference. Boston, MA: Houghton Mifflin Boston; 2002.

40.

San Sebastián

Mosquera

Gustafsson

. Do cardiovascular disease prevention programs in northern Sweden impact on population health? An interrupted time series analysis. BMC Publ Health 2019;19(1):1-10.

41.

Hyndman

Kostenko

. Minimum sample size requirements for seasonal forecasting models. Foresight 2007;6(Spring):12-15.

42.

Wagner

Soumerai

Zhang

Ross-Degnan

. Segmented regression analysis of interrupted time series studies in medication use research. J Clin Pharm Therapeut 2002;27(4):299-309.

43.

Helvie

Chang

Hendrick

Banerjee

. Reduction in late‐stage breast cancer incidence in the mammography era: implications for overdiagnosis of invasive cancer. Cancer 2014;120(17):2649-2656.

44.

Anderson

Jatoi

Devesa

. Assessing the impact of screening mammography: breast cancer incidence and mortality rates in Connecticut (1943–2002). Breast Cancer Res Treat 2006;99(3):333-340.

45.

Katalinic

Eisemann

Kraywinkel

Noftz

Hübner

. Breast cancer incidence and mortality before and after implementation of the German mammography screening program. Int J Cancer 2020;147(3):709-718.

46.

Efegoma

. Effect of Mammography Screening on Incidence and Mortality of Breast Cancer in Alberta. [Master’s thesis]. Calgary, Alberta.: Cumming School of Medicine; 2020.

47.

Duffy

Tabár

Yen

AMF

, et al. Mammography screening reduces rates of advanced and fatal breast cancers: Results in 549,091 women. Cancer 2020;126(13):2971-2979.

48.

Bleyer

Welch

. Effect of three decades of screening mammography on breast-cancer incidence. N Engl J Med 2012;367(21):1998-2005.

49.

Choi

Yoon

Song

, et al. Effect of mammography screening on stage at breast cancer diagnosis: results from the Korea National Cancer Screening Program. Sci Rep 2018;8(1):1-8.

50.

Autier

Boniol

Koechlin

Pizot

Boniol

. Effectiveness of and overdiagnosis from mammography screening in the Netherlands: population based study. BMJ 2017;359(j5224).

51.

Broeders

Allgood

Duffy

, et al. The impact of mammography screening programmes on incidence of advanced breast cancer in Europe: a literature review. 2018;18(1):1-11. BMC Cancer 2018;18(1):1-11.

52.

Tabár

Vitak

Chen

TH-H

, et al. Swedish two-county trial: impact of mammographic screening on breast cancer mortality during 3 decades. Radiology 2011;260(3):658.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.63 MB