Geospatial analysis as a tool for identification of potential targetable regions for lung cancer screening interventions in Massachusetts

Introduction

Background

Non-small cell lung cancer (NSCLC) is the second most common cancer in both men and women in the United States. The American Cancer Association estimates that approximately 227,000 new cases will be diagnosed in 2025, with ~125,000 (55.1%) resulting in death (1). Current guidelines by the United States Preventive Services Task Force (USPSTF) recommend lung cancer screening (LCS) with low dose computed tomography (CT) scan for patients aged 50–80 years old with a 20-pack year smoking history, who currently smoke, or have quit within the past 15 years (2). Nevertheless, there is ongoing concern that certain high-risk patients within Black populations that do not fit the screening guidelines may be missed (3-5), especially given their lack of baseline representation in the National Lung Screening Trial (NLST) (6).

Although NSCLC diagnoses among patients under young patients is uncommon, Black patients <50 yeas old are overrepresented in this cohort (6). Using a large community-based cancer registry encompassing 20 years (1973–1992), Ramalingam et al. demonstrated that the percentage of young Black patients was significantly greater than that of older patients (28.7% vs. 21.9%; P<0.001). Additionally, over the 20-year review period, there was a greater increase in the younger group (27.8% to 33.0%) than in the older group (20.1% to 21.7%) (7). Black race was also found to be an independent negative prognostic factor for NSCLC.

Geographic information science (GIScience) applies Geographic Information Systems (GIS) and other spatial concepts and methods in research to capture, store, review, and display data related to positions on Earth’s surface (8,9). Geospatial analysis, the problem-solving subset of GIScience that examines and interprets data having a geographic component is frequently utilized to identify patterns, trends and relationships in cancer incidence and cancer care distribution both domestically (10,11), and abroad (12,13). It can also shed light on the residual effects of previous geographic barriers including redlining, that may affect cancer screening and care delivery (14). When specifically analyzing lung cancer patterns, though the use of geospatial tools is well documented for screening (15-17), there are few examples of its use to detect incidence and staging trends, specifically within Black neighborhoods.

Rationale and knowledge gap

There are currently no studies examining the social factors affecting the risk and outcomes related to lung cancer in young patients, especially as it pertains to race. This patient population does not currently fall within screening guidelines, thus there is no established system for investigation of their outcomes. As a result, they are not identified until it is often too late for them to reap the full benefits of early diagnosis. Understanding the potential effects of the specific social determinant of geography/place of residence on lung cancer development in patients under 50 years old would allow for the identification of targetable risk factors that can be used to identify high risk patients and appropriately direct them to early screening.

Objective

The aims of this review were to first, use geospatial analysis to understand associations between zip code and stage at diagnosis of lung cancer in Massachusetts. Second, demonstrate an association between zip code and age at diagnosis in Massachusetts [younger (<50 years old) as compared to older (>50 years old) patients]. Each aim would additionally be investigated to determine if there were associations based on race (Black vs. White), socioeconomic status and lack of high school diploma. We hypothesized geospatial analysis will help evaluate the association between a patient’s zip code and later stage at diagnosis as well as understand the geographic distribution of increased rates of lung cancer related deaths in patients under the age of 50 years old as compared to those over the age of 50 years old. These findings together would help identify target locations for efforts at early lung cancer diagnosis among high-risk groups. We present this article in accordance with the STROBE reporting checklist (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-1369/rc).

Methods

Cohort selection and definitions

The Massachusetts Cancer Registry (MCR) was queried for lung cancer cases in non-Hispanic White and Black patients ≥18 years old from 2004 to 2020. Cases were identified using the “International Classification of Diseases for Oncology—Third Edition [2008]” Surveillance, Epidemiology and End Results (SEER) recode, Lung and Bronchus (C34.0–C34.9). Lung cancer cases were histologically defined by “International Classification of Diseases for Oncology—Third Edition” histology codes: small cell carcinoma (8040–8045), adenocarcinoma (8140, 8141, 8143, 8147, 8250–8255, 8260, 8310, 8430, 8480, 8481, 8490, 8570, 8571, 8572, 8573, 8574), squamous cell carcinoma (8052, 8070–8078, 8083, 8084), adenosquamous cell carcinoma (8560), large cell carcinoma (8012, 8013), carcinoma not otherwise specified (NOS; 8010, 8046, 8050, 8051, 8575), pulmonary neuroendocrine carcinoma (8240–8249) and undifferentiated carcinoma (8020). SEER summary stages 0 or 1 (local spread) were considered early-stage, while SEER summary stages 2–7 (regional or distant spread) were considered late-stage (Figure 1). Prior to analysis, cases missing staging data or diagnostic methodology were excluded. Zip code-level socioeconomic data was obtained from the US Census Bureau. Data on median income and educational attainment was taken from 5-year American Community Survey data collected from 2016 to 2020. Population data was taken from decennial census data collected in 2010. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. This study was approved by the Mass General Brigham institutional review board on March 10, 2023 (No. 2023P000515) and by the Massachusetts Department of Public Health on June 29, 2023 (No. 2021994).

Figure 1 Staging definitions based on SEER staging system and (right) associated groupings based on current AJCC staging systems (18). AJCC, American Joint Committee on Cancer; SEER, Surveillance, Epidemiology and End Results.

Inclusion criteria were as follows: over 18 years old, malignant cancer of the lung or bronchus, Black or White race, and non-Hispanic identity. Dates of diagnosis were between January 1, 2004 and December 31, 2020. Only Black and White patients were included based on data that suggests ongoing racial disparities in lung cancer and quality of life outcomes between these groups (19-22). Additionally, the numbers of enrolled patients in other racial minority groups (e.g., Hispanic) were anticipated to be insufficient for analysis and would not provide data that is representative of the state’s population.

The primary exposures of interest were patient’s race (Black or White) and age (18–50 and 51–80 years old). Both exposures were stratified by zip code to identify any trends or patterns in stage at diagnosis based on place of residence. The primary outcome of interest was stage of diagnosis. Staging data was collected by the reporting hospital at the time of tissue diagnosis. Utilizing the staging system reported by the SEER program, patients were categorized into one of the following seven categories: 0 (in situ), 1 (localized only), 2 (regional by direct extension only), 3 (regional by lymph node involvement only), 4 (regional both by lymph node involvement and direct extension), 5 (regional, NOS), 7 [distant site(s)/node(s)]. To correlate the SEER staging system with what is currently used clinically, stages 0/1 were defined as local or early disease; stages 2–5 were defined as regional disease and stage 7 was defined as distant disease. Stages 2–7 were also grouped as late stage (Figure 1). The secondary outcome of interest was vital status (mortality) based on the reported status at the time of discharge. This was analyzed and reported as a dichotomous variable (dead vs. alive) by age category (≤50 and >50 years old) at the time the data was collected. Mortality was reported as proportions.

Statistical analysis

Patients in the dataset were stratified into groups based on race (Black vs. White) and age (≤50 vs. >50 years old). Clinical and demographic features (age, sex, marital status, stage at diagnosis, diagnostic method, tobacco use history) were compared across these groups using Wilcoxon rank-sum or chi-squared tests, as appropriate.

The prevalence ratio of late-to-early-stage disease was defined as the ratio of the number of late-stage lung cancer diagnoses to the number of early-stage lung cancer diagnoses within a given zip code using ArcGIS technology (23,24). Linear regression analysis was used to model associations between late-to-early-stage prevalence ratio and zip-code level socioeconomic indices such as median income and percent of residents without a high school diploma. Logistic regression was used to determine the odds of patients in a given zip code presenting with regional or distant versus localized lung cancer based on race and age strata. Logistic regression was also used to determine odds of death, stratified by age, race and zip code.

Results

Demographics

The mean age [interquartile range (IQR)] at diagnosis amongst all patients and amongst White patients was 70 (IQR, 62–77) years old (Table 1). This was statistically significantly older than the average age among Black patients [66 (IQR, 58–74) years old, P<0.001]. There was a higher percentage of women amongst the White patient cohort (53.4%) and a higher percentage of men amongst the Black cohort (51.7%). These differences were also noted to be statistically significant between races (P<0.001). Patients across all groups presented most commonly with distant disease (47.6%) as compared to regional or local (25% vs. 27.4%, respectively). Black patients as compared to White patients more commonly had distant disease (51.3% vs. 47.5%).

Age-based analysis highlighted a statistically significantly higher percentage of women than men in the White cohort as compared to the Black cohort under 50 years old (White females 58.1%, Black females 51%, P=0.04). Amongst patients 50 years old and older, the majority of patients in the Black cohort were male (52%) as compared to a majority being female in the White cohort over 50 years old (53.2%).

When looking at tobacco use, the highest percentage of patients overall (51.6%) and amongst White patients (52.2%) were former smokers. This differed from Black patients amongst whom the majority of patients were current smokers (43.4%). Notably, Black patients had almost twice as many never smokers as White patients (12.1% vs. 7.1%, P<0.001).

Socioeconomic factors

Linear regression comparing the ratio of late-to-early-stage diagnosis with median income demonstrated a negative correlation (β =−0.013, P=6.49e−08), suggesting that as patients median income increased, the prevalence ratio of late-to-early-stage disease decreased (Figure 2). Additionally, when comparing late-to-early-stage diagnosis with percentage of the population without a high school diploma, there was a positive correlation (β =0.023, P=0.04), suggesting that as the ratio of late-to-early-stage disease increased, there was also an increase in the percentage of patients without a high school diploma.

Figure 2 Linear regression analysis comparing the prevalence ratio of late-to-early disease and median income in 2020 (top) or percent of population without an HSD (bottom). Associated trends (left) and correlation coefficients/statistical details (right) are presented. HSD, high school diploma.

Geospatial distribution

Analysis of geospatial occurrence by stage demonstrated higher rates of late-stage disease in Western Massachusetts and in the Greater Boston metro area. The distribution of late-stage disease correlated to areas with lower educational attainment and lower median income at the zip code level (Figures 3,4). Likewise, geospatial mapping demonstrated higher prevalence of late-stage disease in Greater Boston communities with a higher proportion of Black residents which also corresponds with zip codes with lower median income and lower educational attainment.

Figure 3 Trends in median income (dollars, $) (A) and education (B) by zip code as compared to late-stage disease (C). Findings suggest higher median income in Eastern Massachusetts as compared to higher percentage of the population without a high school diploma (education) in Western Massachusetts and more late-stage diagnosis in Western Massachusetts. HS, high school.

Figure 4 Geospatial analysis comparing percentage of population without an HSD (top left), with median income (dollars, $) (bottom left) and reported rates of cancer among Black patients (right). HSD, high school diploma.

Stage at diagnosis and vital status

When analyzing stage at diagnosis, there were no statistically significant differences between races. Black patients under 50 years old presented with late-stage disease 80.1% of the time as compared to 79.3% of the time with White patients under 50 years old (P=0.85). In patients over 50 years old, 73.7% of Black patients and 72.2% of White patients presented with late-stage disease (P=0.09). Odds of death in Black versus White patients over the age of 50 years old demonstrated an odds ratio of 3.5 [95% confidence interval (CI): 0.43–33.7] for Dorchester, 2.3 (95% CI: 0.09–176.2) for Stoughton and 1.5 (95% CI: 0.07–104.6) for Waltham. Comparatively, the odds of death in Black versus White patients in patients 50 years old and younger were highest in Springfield (1.9; 95% CI: 0.10–36.4), Brockton (1.2; 95% CI: 0.09–68.8) and Jamaica Plain (1.1; 95% CI: 0.07–22.0).

Logistic regression demonstrated that in patients over the age of 50 years old, the top 3 cities with the highest odds ratio of distant versus local disease in Black versus White patients at the time of diagnosis were Springfield (8.5; 95% CI: 0.99–407.0), Brighton (4.7; 95% CI: 0.62–208.9) and Taunton (4.5; 95% CI: 1.05–41.7). Amongst patients ≤50 years old, the top 3 cities with the highest odds of distant versus local disease in Black versus White patients at the time of diagnosis were Springfield (1.9; 95% CI: 0.1–36.4), Brockton (1.2; 95% CI: 0.09–68.8) and Jamaica Plain (1.1; 95% CI: 0.07–22.0).

Discussion

Lung cancer incidence in minority communities is still under-researched, especially in high-risk neighborhoods that do not meet screening criteria. In this cross-sectional retrospective review of lung cancer patients in the MCR, we aimed to use geospatial analysis to understand associations between a patient’s zip code, stage at diagnosis, age, race and other social determinants to help identify potential targets for early detection interventions. Overall patterns of geographic distribution of lung cancer in Massachusetts patients suggested correlations between sociodemographic factors (income/level of education/race) and late-stage lung cancer diagnosis. We observed higher rates of late-stage disease in Western Massachusetts, corresponding to areas with lower educational attainment and lower median income at zip code level. Likewise, we appreciated a higher prevalence of late-stage disease in Greater Boston communities with a higher proportion of Black residents, lower income and lower education. These findings suggest that geospatial analysis can shed light on important community factors that may increase potential lung cancer risk, much like it has with other cancers.

This study has highlighted patterns of stage distribution as it compares to geography in addition to race and socioeconomic factors of income and education. We observed late-stage cancer diagnosis occurring at higher rates in areas where there was a higher concentration of Black patients. This finding is consistent with prior work by Krieger et al. who investigated stage at diagnosis based on historical redlining patterns in the state of Massachusetts (25). With the knowledge that neighborhoods whose residents were Black, disproportionately low income or foreign-born residents tended to live in “less desirable” or more racially segregated areas, they investigated whether there was a correlation between these areas and stage at diagnosis. They found that net of age, sex/gender, and race/ethnicity, residing in a previously Home Owners’ Loan Corporation-redlined area imposed an elevated risk for late stage at diagnosis, even for residents of census tracts with present-day economic and racial privilege. We additionally found a higher rate of late-stage diagnosis in Western Massachusetts which was similarly distributed with the level of educational attainment and income in that region of the state. Some notable zip codes that our analysis identified for patients 50 years old and younger were Springfield, Brockton and Jamaica Plain. Closer analysis of patient populations within these cities will allow for targeted efforts to increase awareness around the need for screening and target communities for intervention.

A third notable finding of this study is that Black patients in Massachusetts present at a younger average age than White patients. The mean age was 70 (IQR, 62–77) years old in the overall cohort and amongst White patients; which was statistically significantly older than the average age among Black patients (66 years old, IQR, 58–74 years old; P<0.001). These findings are consistent with prior literature demonstrating that Black patients are diagnosed with lung cancer at a younger age than White patients. Richmond et al. reported an average age at diagnosis of 66.9 years old for Black patients and 68.2 years old in White patients (26-28). Similarly, when looking at lung cancer in patients living in the Southeastern United States, the Southern Community Cohort Study found an average age at diagnosis of 64.4±8.75 years old for White patients and 61.7±8.75 years old for Black patients (26). Importantly, when comparing the rate of diagnosis in patients under 50 years old based on race, we see that the percentage of Black patients diagnosed with lung cancer under the age of 50 years old is almost twice as high as that for White patients (8.2% vs. 4.4%, P<0.001). Considering that our study population was 95.9% White and 4.1% Black, it is notable that a greater proportion of Black patients were 50 years old and younger.

While our investigation demonstrated a higher percentage of Black patients under 50 years old being diagnosed with lung cancer, this did not translate to higher mortality in the patient cohort under 50 years old. This is also consistent with prior literature which demonstrates that although young patients present with more advanced disease, they often have comparable or better survival than older patients (7,29,30). This would suggest that the impact of earlier diagnosis in young patients would present as a benefit in minimizing the comorbidities patients suffer as a result of their disease, as opposed to prolonging life. Additionally, the absence of comorbidities or pre-existing conditions in those with earlier presentation, may also contribute to better outcomes (31,32).

Conclusions

Through this retrospective review, we have identified geographically based differences in the rate and stage of lung cancer diagnosis in Black neighborhoods across Massachusetts. For our cohort aged 50 years old and older, we believe that this data could be utilized to improve patient/provider education around LCS (Figure 5) both within and outside of the hospital/center locations. Next steps include refining geospatial analysis of lung cancer distribution to aid in prioritization of locations for LCS.

Figure 5 Moran’s I testing of rates of late-stage lung cancer demonstrate clustering of the hot-spots in specific areas (in- and around the Greater Boston area), regions that can be targeted for lung cancer screening.

Implementing access to screening in out-of-hospital/community settings using mobile technology has been shown to improve uptake in underserved populations, and qualitative analysis has demonstrated that patients would potentially embrace this mode of early detection (33). Mobile Health Clinics (MHC), which provide “one-stop screening” capabilities in community settings, serve as a link between clinical and the areas they serve by addressing both medical and social determinants of health and tackling health issues on a community-wide level (34). Furthermore, evidence suggest that MHCs represent a cost-effective care delivery model that improves health outcomes in underserved groups (35,36), and may also help the patients aged 50 years old and older.

More work is needed to explore screening options for patients under 50 years old of age who would not qualify for lung cancer and live in a geographic hot-spot, and current research and advocacy efforts are underway to potentially lower eligibility to include these patients. Ultimately, increasing screening participation within high-risk communities will likely need a multi-faceted approach that involves in- and out-of-hospital collaboration to meet the needs of both our eligible (age 50 years old and above) and not-yet eligible cohorts to increase uptake in early detection and save lives.

Acknowledgments

We would like to acknowledge the support of the Massachusetts Department of Health in providing access to the state’s database. We would also like to acknowledge members of the Jacklitsch lab for their feedback and support during the writing of this manuscript.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-1369/rc

Peer Review File: Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-1369/prf

Funding: None.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-1369/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. This study was approved by the Mass General Brigham institutional review board on March 10, 2023 (No. 2023P000515) and by the Massachusetts Department of Public Health on June 29, 2023 (No. 2021994).

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. American Cancer Society. Lung Cancer Statistics. How Common is Lung Cancer? Accessed February 10, 2024. Available online: https://www.cancer.org/cancer/types/lung-cancer/about/key-statistics.html
2. US Preventive Services Task Force. Screening for Lung Cancer: US Preventive Services Task Force Recommendation Statement. JAMA 2021;325:962-70. [Crossref] [PubMed]
3. Potter AL, Yang CJ, Woolpert KM, et al. Evaluating Eligibility of US Black Women Under USPSTF Lung Cancer Screening Guidelines. JAMA Oncol 2022;8:163-4. [Crossref] [PubMed]
4. Potter AL, Kuhan S, Senthil P, et al. The 2021 US Preventive Services Task Force lung cancer screening eligibility criteria disproportionately exclude younger Black patients with lung cancer. Cancer 2025;131:e35676. [Crossref] [PubMed]
5. National Lung Screening Trial Research Team. Reduced lung-cancer mortality with low-dose computed tomographic screening. N Engl J Med 2011;365:395-409. [Crossref] [PubMed]
6. Schwartz AG, Swanson GM. Lung carcinoma in African Americans and whites. A population-based study in metropolitan Detroit, Michigan. Cancer 1997;79:45-52. [Crossref] [PubMed]
7. Ramalingam S, Pawlish K, Gadgeel S, et al. Lung cancer in young patients: analysis of a Surveillance, Epidemiology, and End Results database. J Clin Oncol 1998;16:651-7. [Crossref] [PubMed]
8. Sahar L, Foster SL, Sherman RL, et al. GIScience and cancer: State of the art and trends for cancer surveillance and epidemiology. Cancer 2019;125:2544-60. [Crossref] [PubMed]
9. National Geographic Society. GIS (Geographic Information System). Accessed February 3, 2025. Available online: https://education.nationalgeographic.org/resource/geographic-information-system-gis
10. Hashtarkhani S, Zhou Y, Kumsa FA, et al. Analyzing Geospatial and Socioeconomic Disparities in Breast Cancer Screening Among Populations in the United States: Machine Learning Approach. JMIR Cancer 2025;11:e59882. [Crossref] [PubMed]
11. Rathod R, Beauchamp A, Shahan K, et al. Neighborhood Factors Related to Cancer Screening in Texas: A Spatioecological Study. Am J Prev Med 2025;68:695-706. [Crossref] [PubMed]
12. Garg P, Krishnamoorthy Y, Halder P, et al. Urban-rural disparities in cervical cancer screening among Indian women between 30-49 years: a geospatial and decomposition analysis using a nationally representative survey. BMC Cancer 2025;25:67. [Crossref] [PubMed]
13. Cramb SM, Cameron J, Dasgupta P, et al. Geospatial patterns by cancer stage across Australia for three common cancers. Cancer Epidemiol 2025;94:102738. [Crossref] [PubMed]
14. Loehrer AP, Weiss JE, Chatoorgoon KK, et al. Residential Redlining, Neighborhood Trajectory, and Equity of Breast and Colorectal Cancer Care. Ann Surg 2024;279:1054-61. [Crossref] [PubMed]
15. Welch AC, Gorden JA, Mooney SJ, et al. Understanding Washington State’s Low Uptake of Lung Cancer Screening in Two Steps: A Geospatial Analysis of Patient Travel Time and Health Care Availability of Imaging Sites. Chest 2024;166:622-31. [Crossref] [PubMed]
16. Sahar L, Douangchai Wills VL, Liu KK, et al. Using Geospatial Analysis to Evaluate Access to Lung Cancer Screening in the United States. Chest 2021;159:833-44. [Crossref] [PubMed]
17. Tailor TD, Tong BC, Gao J, et al. A Geospatial Analysis of Factors Affecting Access to CT Facilities: Implications for Lung Cancer Screening. J Am Coll Radiol 2019;16:1663-8. [Crossref] [PubMed]
18. Mirsadraee S, Oswal D, Alizadeh Y, et al. The 7th lung cancer TNM classification and staging system: Review of the changes and implications. World J Radiol 2012;4:128-34. [Crossref] [PubMed]
19. Shaw V, Zhang B, Tang M, et al. Racial and socioeconomic disparities in survival improvement of eight cancers. BJC Rep 2024;2:21. [Crossref] [PubMed]
20. Kapp CM, Green C, Thiboutot J, et al. Understanding the Social Risk Factors That Avert Equitable Lung Cancer Care. Clin Lung Cancer 2024;25:699-704. [Crossref] [PubMed]
21. Chaballout BH, Wu TC, Farrell MJ, et al. Trends in racial and ethnic disparities in health-related quality of life in older adults with lung cancer. J Geriatr Oncol 2024;15:102066. [Crossref] [PubMed]
22. Uprety D, Seaton R, Hadid T, et al. Racial and socioeconomic disparities in survival among patients with metastatic non-small cell lung cancer. J Natl Cancer Inst 2024;116:1697-704. [Crossref] [PubMed]
23. Goldberg DW, Cockburn MG. Improving Geocode Accuracy with Candidate Selection Criteria. Transactions in GIS 2010;14:149-76.
24. McKeon TP, Hwang WT, Ding Z, et al. Environmental exposomics and lung cancer risk assessment in the Philadelphia metropolitan area using ZIP code-level hazard indices. Environ Sci Pollut Res Int 2021;28:31758-69. [Crossref] [PubMed]
25. Krieger N, Wright E, Chen JT, et al. Cancer Stage at Diagnosis, Historical Redlining, and Current Neighborhood Characteristics: Breast, Cervical, Lung, and Colorectal Cancers, Massachusetts, 2001-2015. Am J Epidemiol 2020;189:1065-75. [Crossref] [PubMed]
26. Richmond J, Murray MH, Milder CM, et al. Racial Disparities in Lung Cancer Stage of Diagnosis Among Adults Living in the Southeastern United States. Chest 2023;163:1314-27. [Crossref] [PubMed]
27. Varlotto JM, McKie K, Voland RP, et al. The Role of Race and Economic Characteristics in the Presentation and Survival of Patients With Surgically Resected Non-Small Cell Lung Cancer. Front Oncol 2018;8:146. [Crossref] [PubMed]
28. Varlotto JM, Voland R, McKie K, et al. Population-based differences in the outcome and presentation of lung cancer patients based upon racial, histologic, and economic factors in all lung patients and those with metastatic disease. Cancer Med 2018;7:1211-20. [Crossref] [PubMed]
29. Tominaga K, Mori K, Yokoi K, et al. Lung cancer in patients under 50 years old. Jpn J Cancer Res 1999;90:490-5. [Crossref] [PubMed]
30. Thomas A, Chen Y, Yu T, et al. Trends and Characteristics of Young Non-Small Cell Lung Cancer Patients in the United States. Front Oncol 2015;5:113. [Crossref] [PubMed]
31. Shibazaki T, Odaka M, Noda Y, et al. Effect of comorbidities on long-term outcomes after thoracoscopic surgery for stage I non-small cell lung cancer patients with chronic obstructive pulmonary disease. J Thorac Dis 2018;10:909-19. [Crossref] [PubMed]
32. Iachina M, Jakobsen E, Møller H, et al. The effect of different comorbidities on survival of non-small cells lung cancer patients. Lung 2015;193:291-7. [Crossref] [PubMed]
33. Hasson RM, Fay KA, Phillips JD, et al. Rural barriers to early lung cancer detection: Exploring access to lung cancer screening programs in New Hampshire and Vermont. Am J Surg 2021;221:725-30. [Crossref] [PubMed]
34. Yu SWY, Hill C, Ricks ML, et al. The scope and impact of mobile health clinics in the United States: a literature review. Int J Equity Health 2017;16:178. [Crossref] [PubMed]
35. Vang S, Margolies LR, Jandorf L. Mobile Mammography Participation Among Medically Underserved Women: A Systematic Review. Prev Chronic Dis 2018;15:E140. [Crossref] [PubMed]
36. Greenwald ZR, El-Zein M, Bouten S, et al. Mobile Screening Units for the Early Detection of Cancer: A Systematic Review. Cancer Epidemiol Biomarkers Prev 2017;26:1679-94. [Crossref] [PubMed]