1 Introduction
Like other chronic diseases, cancer incidence and mortality rates reflect life course social determinants of health. A large body of biomarker research on the “biological embedding of experience” has established the close and persistent connections between gene expression, epigenetics and social forces (McDade & Harris, 2018). The complex interaction between childhood adversity, chronic stress, and low control over life circumstances characteristic of lower social class position explains the ubiquitous social gradient in chronic disease prevalence and premature mortality (Jones et al., 2019; McCartney et al., 2019). In the United States the social class gradient in life expectancy has become very pronounced in recent decades with growing income inequality and the continuing pervasive effects of structural racism on the health of racial and ethnic minority populations (Bailey et al., 2021; Hittner & Adam, 2020; Kawachi et al., 2005; Harris, Majmundar & Becker, eds., 2021; Zimmerman & Anderson, 2019). Cancer data, in a standardized tumor registry format that combines patient characteristics at diagnosis with rigorous long-term follow-up, provides many opportunities to model the social patterning of cancer incidence, treatment and survival (Krieger et al., 2019).
This chapter provides findings from two recent investigations that illustrate the potential of modeling cancer health disparities. The studies are representative of the work of the National Institutes of Health-funded Chicago Cancer Health Disparities Collaborative (CHEC), a research consortium between Northwestern University, the University of Illinois at Chicago and Northeastern Illinois University. CHEC scholars combine community-engaged research with epidemiologic analyses focused on the social patterning of cancer incidence and outcomes. The two studies presented here model socioeconomic and racial disparities in breast and lung cancer, respectively, using publicly available, de-identified data.
Lung cancer, closely linked to smoking, has long been known to have a strong socioeconomic gradient in incidence related to the fact that lower income Americans are much more likely to be smokers (Barbeau et al., 2004). Conversely, breast cancer is more common among women from higher socioeconomic status communities, related to risk factors like higher alcohol consumption, fewer children, having children at a later age, and greater use of birth control pills and postmenopausal hormones (Robert et al., 2004). This has remained true even after a mid-2000s reduction in white women’s cancer incidence related to reductions in hormone replacement therapy (National Cancer Institute, 2014; Krieger et al., 2010). Just as lung cancer incidence reflects the historic toll of smoking for different birth cohorts, breast cancer incidence and staging have also evolved in tandem with changing social conditions (Krieger et al., 2010, 2011). It was therefore of interest to analyze recent data on social disparities in outcomes within these two disparate patient populations. Both studies use publicly available, de-identified cancer data to shed light on the specific social gradient of each type of cancer. Each study provides evidence of the socioeconomic, racial, and ethnic patterning of cancer outcomes in the United States. These findings, which highlight the extent of unfair and avoidable differences in population health, have important implications for cancer prevention and control going forward.
2 Socioeconomic Status and Breast Cancer Outcomes
Racial differences in breast cancer mortality between White and Black women in the United States have been attributed to the fact that minority women were consistently diagnosed with higher stage cancer and often received less than optimal treatment (Clegg et al., 2009; DeSantis et al., 2010; Gumpertz et al., 2006; Krieger et al., 2012). However, much less is known about socioeconomic status (SES) disparities in outcomes for women diagnosed with breast cancer. Both breast cancer specific and all-cause mortality have been shown to vary by socioeconomic status in earlier studies (Albano et al., 2007; Byers et al., 2008). However, more recent population-based breast cancer mortality rates, measured across income quintiles, appear to have largely converged across SES categories (Albano et al., 2007).
Our breast cancer outcomes study was undertaken to estimate the effect of socioeconomic status (SES) on all cause mortality among women diagnosed with breast cancer. We were interested in modeling the independent effects of socioeconomic status at the time of diagnosis after controlling for health insurance status, race and ethnicity, breast cancer stage at diagnosis, and surgical and adjuvant treatment received (Feinglass et al., 2015). Our survival estimates were based on vital status follow-up through 2011 of over 582,000 female patients using records from the National Cancer Data Base (NCDB). We presented estimates of SES associations with all-cause mortality during a period of important changes in breast cancer diagnosis and treatment, and coinciding with a significant reduction in average person-years of life lost due to breast cancer in the US (Soneji et al., 2014).
3 Breast Cancer Study Methods: Data Source and Patient Sample
The NCDB is a joint project of the American Cancer Society and the Commission of Cancer of the American College of Surgeons (http://ncdbpuf.facs.org). NCDB hospital-based cancer registries include patient demographics, American Joint Committee on Cancer staging and surgical and adjuvant treatments. Our sample included all female patients diagnosed with breast cancer at 1630 NCDB reporting hospitals with up to 176 month follow-up through 2011, for female patients diagnosed in 1998–2006. NCDB de-identified data were ruled exempt by the Northwestern University Institutional Review Board.
We categorized patients’ age group and race and ethnicity (non-Hispanic White non-Hispanic Black, Hispanic, Asian, and other/unknown). Pathological staging was used whenever available; if missing, clinical staging was used. Treatment variables included primary surgery type (lumpectomy, mastectomy, or no or unknown primary surgery), radiation therapy, chemotherapy, or hormone therapy. One sensitivity study included the Charlson/Deyo Comorbidity Score, which is based on codes for chronic diseases, was trichotomized as 0, 1, or 2 or greater (Deyo et al., 1992) for patients diagnosed in 2003–2006 (32.8% of the sample). We also tested the sensitivity of our final model with analyses restricted to the 82.1% of patients diagnosed with Stage I–IV breast cancer, excluding patients diagnosed with DCIS.
Multivariable survival analyses were controlled for regional location of the treating hospital (large urban region, medium urban region, small urban region, rural region, or unknown), and whether a hospital had an academic/research designation or was a community institution. We created three time periods (1998–2000, 2001–2003, 2004–2006) to control for trends in diagnosis and treatment over the study period. We excluded records for patients with missing zip codes (n = 28,410, 4.65%) or stage at diagnosis (n = 22,239, 3.68%).
3.1 Creating a Socioeconomic Status Measure
In the United States, population-based direct measures of social class or social position are scarce or non-existent. Usually social class is inferred from (usually self-reported) household income or from an individual’s educational attainment level. For hospital data, researchers have to attribute individual patient education or income to the patient’s postal zip code average. Postal zip code, which has been mapped to census tract data as Zip Code Tabulation Areas (ZCTAs) is the smallest publicly available census data that can be matched to patient’s residential zip code. The NCDB includes patient ZCTA quartile of education and quartile of income as two separate variables.
Jointly including both ZCTA education and income quartiles, even if interaction terms are estimated, fails to fully measure the synergistic effect of these measures of socioeconomic status (SES). To illustrate this, our study constructed a six-level measure from combined zip code quartiles of census-based median income and educational attainment at the time of diagnosis. To validate this monotonic SES scale, we first used Cox Proportional Hazards models to rank hazard ratios for all 16 combinations of income and education zip code quartiles. Based on those results, we then aggregated patients into five SES categories with almost completely non-overlapping hazard ratio 95% confidence intervals. The reference for analyses was patients living in the highest income and highest education quartile (about one-third of the sample). Finally, we included a variable for patients who were uninsured or had Medicaid coverage at the time of their diagnosis as an additional indicator of SES that has been directly associated with higher breast cancer death rates (DeSantis et al., 2010).
4 Survival Modelling
Cox proportional hazards regression was used to calculate initial hazard ratios for our SES measure after confirming proportional hazards assumptions graphically. The Kaplan Meier estimator and log rank test were used to test the significance of bivariate survival probabilities. Chi square tests of proportions were used to test the significance of baseline SES differences. Hierarchical Cox proportional hazards models were then used to test the significance of SES controlled for other patient and hospital covariates, with standard errors adjusted for intra-group correlation (clustering) within hospitals using STATA Version 12 (College Station, Texas) software. Differences across SES category hazard ratios were examined sequentially before and after adding insurance status, race and ethnicity, stage at diagnosis, and finally, treatment modalities.
5 Results of Breast Cancer Survival Modelling
Overall five and ten year survival probabilities for all 582,396 female breast cancer patients were 84.6% and 69.2%; respectively. Survival probabilities were 93.5% and 82.2% for the 104,055 patients (17.9%) diagnosed with DCIS, 11.3% and 13% higher survival than invasive cancer patients. While only 7.3% of breast cancer patients were from the lowest quartile education and income ZTCAs, 32.5% were from highest quartile education and income ZCTAs. As expected, Black and Hispanic patients composed much larger proportions of lower SES categories, with Blacks composing 27.3% of the lowest SES category. Conversely, 89.8% of the highest SES patients were non-Hispanic Whites (p < 0.001). Five year survival for the highest SES group was 87.8% as compared to 79.5% for the lowest SES group; at 10 years the difference was 10% (71.5% to 61.5%, p < 0.001). Highest SES patients had an 11.5% greater use of lumpectomy versus mastectomy and 1.8% lower proportion of no or unknown surgery, an 8.1% greater use of radiation therapy, a 2.4% greater use of chemotherapy and 4.8% greater use of hormone therapy as compared to lowest SES patients. Table 1.1 shows the original 16 possible categories collapsed across six categories with virtually non-overlapping hazard ratios. There was a clear SES gradient in survival with a 69% greater hazard ratio for the lowest as compared to the highest SES category (all comparisons p < 0.001).
Mortality hazard ratios for patients diagnosed with breast cancer across six zip code income and education categories
Table 1.2 presents the final Cox proportional hazards model results. The highest to lowest SES hazard ratio was 1.27 (95% CI 1.23–1.30), with decreasing hazard ratios across SES categories to about 12% higher hazard ratios for the fourth and fifth highest SES categories. As expected, older age was strongly associated with lower survival. Patients covered by Medicaid or who were uninsured had a 47% higher hazard ratio than those with other insurance. Black patients had a 24% higher hazard ratio than non-Hispanic Whites, while Hispanic, Asian and other or unknown ethnicity had significantly lower hazard ratios than Whites. As compared to patients diagnosed with DCIS, hazard ratios sharply increased for higher stage patients reaching 15.9 for stage IV patients. Time period indicators were not significant, while being diagnosed at an academic hospital or a hospital in a rural area were both protective. Patients with no or unknown surgery had an 82% higher hazard ratio than patients undergoing mastectomy, but there were no significant survival differences between patients undergoing lumpectomy versus mastectomy. Patients who received chemotherapy, radiation or hormone treatment all had better survival than those who received no adjuvant treatment.
Cox regression results for all cause mortality for patients diagnosed with breast cancer
6 Breast Cancer and Socioeconomic Status
In this study, patients’ insurance, race, stage at diagnosis, and treatment modalities accounted for about two-thirds of the initially observed SES gradient. Even after controlling for these factors, patients from the lowest income and education ZCTAs still had a 27% higher hazard ratio than the highest SES patients. In secondary analyses, we found that comorbidity, while itself highly predictive of mortality, explains only a very small proportion of the remaining SES mortality gap. Despite better survival among patients with DCIS, NCDB patients with invasive cancer were also found to have a very similar SES ‘gradient’ in mortality. SES disparities in treatment quality had a much weaker impact on survival than social factors.
Historical studies of breast cancer incidence and mortality reveal multiple and complex ‘natural histories’ of breast cancer. Biomarkers at presentation, cancer reoccurrence, and cancer mortality rates have exhibited contingent time trends (Krieger, 2013). Albano et al. analyzed population-based 2001 breast cancer death rates for women ages 25–64 by educational attainment within black and white race, and found the educational disparity was over twice as great among White versus Black women (Albano et al., 2007). Analyzing more recent county level, population-based breast cancer mortality rates, Krieger et al. (2012) found a pattern of increasing (1960–1990) and then decreasing (1990–2006) disparities in standardized breast cancer mortality rates across quintiles of county median household income for both Blacks and Whites (Krieger et al., 2012). In this study, the association of SES with all-cause mortality, above and beyond the effects of race and ethnicity, health insurance, stage at diagnosis and treatment disparities was enduring throughout the study period across both the DCIS and invasive cancer patient cohorts.
7 Lung Cancer Disparities in Illinois
Our study of lung cancer disparities in the state of Illinois was conducted as part of a larger CHEC community based participatory research initiative which includes the NIH funded Supporting High Risk African American Men in Research, Engagement & Decision Making (SHARED) project, a lung cancer control study based on African American men as citizen scientist study partners. Illinois is a large, diverse state with over 12.5 million residents, over 14% Black and 17% Hispanic. To understand the epidemiologic background to racial health disparities in our state we assembled recent publicly available Illinois smoking, cancer registry and lung cancer hospital care information by Illinois resident’s race and ethnicity (Golecha et al., 2021). Our findings provide the health equity modelling features needed to further analyze disparities in diagnosis, treatment, and screening rates.
8 Epidemiology of Lung Cancer
Related to declining smoking rates in the United States and improved care for those with lung cancer, there has been a 5% decline in lung cancer mortality in men and a 4% decline in women since 2013. However, around one quarter of all cancer related deaths in the United States and Illinois are still attributable to lung cancer. Data from 2017 indicate that there have been more lung cancer related deaths than deaths from breast, prostate, colorectal and brain cancer combined (Howlader et al., 2019; Siegel, Miller & Jemal, 2020). When detected early, lung cancer has the potential to be effectively treated (Li et al., 2016). The five year survival rate is 57% when diagnosed at local stage, unfortunately, approximately 57% of diagnoses are made in distant stage, where the five-year survival rate is approximately 5% (Siegel, Miller & Jemal, 2020). Recent data from 10 states from the Behavioral Risk Factor Surveillance System (BRFSS) survey found that only one in eight current or former smokers who met United States Preventive Services Task Force criteria for screening reported lung cancer screening in the last year (Richards, 2020).
Racial disparities in lung cancer incidence, mortality, surgical treatment and screening have been reported since the late 1990s. Non-Hispanic Blacks are both at higher risk for lung cancer than Whites in the United States, present with more advanced disease and have a worse probability of survival once diagnosed (Mulligan et al., 2006; Underwood et al., 2012). It is important to note here that like lung cancer, other disease-specific racial disparities reflect structural racism, the social conditions in which the Black population has had to exist. For example, like many other health conditions, racial disparities in lung cancer are known to be exacerbated by residential segregation (Hayanga et al., 2013). Black residents in the most racially segregated neighborhoods in the United States had a 10% higher lung cancer mortality rate compared Blacks living in the least racially segregated neighborhoods (O’Keefe et al., 2015). Hispanic Americans, particularly those of Mexican origin, have lower smoking rates than white Americans and a younger population. Hispanics have a little more than half the lung cancer incidence and one-third the lung cancer mortality of NH Whites (Miller et al., 2018).
Recent national Surveillance, Epidemiology, and End Results (SEER) findings demonstrated a substantial decrease in age-adjusted lung cancer incidence and mortality between Blacks and Whites between 2000 and 2016. These national data suggest disparities related to lung cancer incidence and mortality are narrowing at the national level. We undertook this study to determine the extent to which lung cancer disparities in Illinois have followed or diverged from these national trends. We used recent, publicly available Illinois smoking, cancer registry and hospital care data.
9 Lung Cancer Study Methods
We obtained data for incidence, mortality, and stage of diagnosis from the Illinois Department of Public Health’s Illinois State Cancer Registry (ISCR) online database. Patients diagnosed with cancer are identified by the ISCR from hospital tumor registries, free standing clinics, radiation treatment facilities, laboratories, and physician offices. The incidence rate was calculated as the average annual age-adjusted (to the 2000 US standard population) rate per 100,000 Illinois residents for the years 2012 to 2016, the most recently available data. Lung cancer mortality data was available for 2016, including extent (stage) of disease at the time of diagnosis categorized as local (if a malignancy limited to origin organ), regional (if tumor extends beyond origin organ’s limits), distant (if tumor that has spread to distant sites, remote from primary tumor), or unknown stage. Stage at diagnosis is provided by race and ethnicity, with cases where patient ethnicity could not be determined reported as “other” or “unknown” included in the “all races” category.
Illinois Hospital Association Comparative Health Care and Hospital Data Reporting Services (COMPdata) administrative discharge data from 199 non-federal Illinois hospitals were obtained for all patients with codes for malignant neoplasm of the bronchus or lung coded admitted from 2016–2018. We also identified patients undergoing lung resection surgery, which was only performed at 87 Illinois hospitals. Finally, we identified outpatient low dose computerized tomography screening (LDCT) screening. Only 114 Illinois hospitals performed LDCT in by 2018.
Because smoking history is integrally related to lung cancer incidence, we also present survey data on current or past smoking among Illinois residents age 35 or older. These data were from the 2017 Illinois Behavioral Risk Factor Surveillance System (BRFSS) survey. The BRFSS sampling methodology has been adjusted to increase the representativeness of low income and minority populations. Data were collected from 1,856 telephone interviews representative of 2,864,367 Illinois residents age 35 and older. Ever smoking was defined as having smoked at least 100 cigarettes (approximately five packs).
To compute hospital admission, surgical admission and screening rates, we obtained population denominator estimates for Illinois residents age 35 and older for all Illinois residents and for non-Hispanic White, non-Hispanic Black, and Hispanic residents using 2017 five-year averaged American Community Survey census data. We used the hospital use numerator data to construct medical admission, surgical admission, and screening rates per 10,000. We then compared rate ratios for racial and ethnic groups for each lung cancer outcome and for prevalence of ever smoking. The significance of differences in rate ratios was determined using chi square tests. All analyses were done with Stata Version 15 (College Station, TX). All data were publicly available and de-identified and thus IRB exempt.
10 Results of the Lung Cancer Study
As shown in Table 1.3, the age adjusted annual incidence of lung cancer in Illinois between 2012–2016 was 64.7 per 100,000 Illinois residents, but it was 75.7 per 100,000 for Blacks and only 27.2 per 100,000 for Hispanics.
There were 6,242 total Illinois lung cancer deaths in 2016. The age-adjusted mortality rate was 16.8% higher for Black versus White Illinois residents. Black and especially Hispanic residents had higher proportions of patients diagnosed at distant stage (all comparisons p < 0.001).
Illinois state cancer registry data on lung and bronchus cancer incidence, mortality and stage at diagnosis by race and ethnicity
Average annual rates per 10,000 for hospital admissions, lung resection procedures and low dose CT screening for Illinois residents coded as having lung cancer (2016–2018)a
Average annual rates per 10,000 for hospital admissions, lung resection procedures and low dose CT screening for Illinois residents coded as having lung cancer (2016–2018)a
The overall medical admission rate for NH Black patients (36.2 per 10,000 population) was 35% higher than for NH White patients (26.8 per 10,000 population). Conversely the rate of admission for medical treatment was 70% lower for Hispanic than for NH White patients (8.1 versus 26.8 per 10,000 population). The proportion of patients with a principal diagnosis of lung cancer was virtually identical for NH Blacks and NH Whites (27.8% versus 27.7%). Surgical admissions, which are a surrogate for earlier stage at diagnosis and more effective treatment, were almost 25% lower for Non-Hispanic Black patients and extremely rare among Hispanic patients (0.46 per 10,000). There were 36,515 LDCT screenings at Illinois hospitals from 2016–2018. The rate of LDCT screenings was almost twice as high for non-Hispanic Whites as compared to non-Hispanic Black patients and over seven times the rate for Hispanics (all comparisons p < 0.001).
10.1 Smoking Rates
BRFSS data show that approximately 41% of the Illinois population age 35 or older were self-reported “ever smokers”.
Ever-smoker rates for non-Hispanic Blacks were lower than for NH Whites (45.8% to 39.4%) and much higher than Hispanics (27.2%, p < 0.001). However, non-Hispanic Blacks did have slightly higher ever smoker rates among the age 55–74 population.
Percent of smokers who had ever smoked 100 cigarettes by age Note: N = 1,856 telephone interviews from the 2017 behavioral risk surveillance system survey
Source: Author’s elaboration
10.2 Lung Cancer Race and Ethnicity Rate Ratios
Figure 1.2 displays NH Black and Hispanic to NH White rate ratios for incidence, mortality, distant stage at diagnosis (for those with lung cancer), medical admissions, surgical admissions, and screening.
The incidence, mortality, distant stage at diagnosis, and medical admissions rate ratios for NH Blacks to NH Whites were all greater than 1.0, while the rate ratios for surgical admissions and screening were less than 1.0. For Hispanics to NH Whites, the rate ratio was only higher than 1.0 for diagnosis at distant stage, with incidence, mortality, medical admissions, surgical admissions, and screening the rate ratios were all < 1.0.
10.3 Illinois Lung Cancer Disparities in Context
Our study illustrates the continued presence of racial and ethnic disparities in lung cancer outcomes and care in Illinois. NH Blacks were found to have a higher incidence of lung cancer and had increased risk for mortality, late-stage diagnosis, and medical hospitalization rates while having lower surgical admission and screening rates. Hispanics had the lowest lung cancer incidence and lung cancer related medical admissions rate, which may be related to much lower rates of smoking. However, the higher rate of advanced stage diagnosis may indicate that Hispanic Illinois residents may be less likely to have medical care encounters resulting in routine imaging for other indications, which might lead to the identification of clinically asymptomatic lung cancers. This is consistent with Hispanics’ well-known differential access to primary care and health insurance (Velasco-Mondragon et al., 2016).
Rate ratio for Illinois lung cancer-related outcomes for Black related to Non-Hispanic Whites and Hispanics related to Non-Hispanic Whites
Source: Author’s elaboration
As compared to our Illinois findings, national on-line SEER data show a much narrower gap in incidence rates, mortality, and stage at diagnosis between NH Blacks and NH Whites. Based on 2016 data from SEER, NH Blacks had a higher age-adjusted incidence rate (56.8 per 100,000) than NH Whites (55.1 per 100,000) (Surveillance, Epidemiology, and End Results (SEER) Program, n.d.) which reflects a significantly lower rate ratio (1.07) than what we found for Illinois (1.17). While NH Blacks also have a higher national lung cancer mortality rate (49.6 per 100,000) in comparison to NH Whites (47.7 per 100,000) (Surveillance, Epidemiology, and End Results (SEER) Program, n.d.), this again reflects a much lower national rate ratio (1.04) than what we found in Illinois (1.24). These findings indicate the urgent need for interventions at the state and local level to address disparities in lung cancer care, where most programs to address these gaps are ultimately enacted.
Our results indicated slightly lower self-reported smoking rates for NH Black Illinois residents. This is consistent with historical research indicating that going back 40–50 years, Black Americans have consistently consumed fewer cigarettes than Whites (Ryan, 2018). However, Black smokers have a longer duration of smoking and are diagnosed with lung cancer at an earlier age, and smoking duration may be more closely associated with lung cancer incidence than pack years (Ryan, 2018). Smoking cessation may not be as successful in the Black population for reasons related to greater social stress, less medical assistance in quitting, and unequal access to healthcare (Bach et al., 2004; Shavers & Brown, 2002). Thus Black smokers do not benefit as much from the roughly 20 year linear decrease in the odds of lung cancer after a smoker quits (Ryan, 2018).
It is also been proposed that Black smokers are more susceptible to the development of smoking-induced lung cancer due to differing nicotine metabolism pathways which lead to differences in the uptake of carcinogens (Haiman et al., 2006). Blacks have higher rates of smoking more than 30 cigarettes per day, at which point metabolic pathways become saturated and toxicity increases (Haiman et al., 2006). This is supported by findings that Black smokers inhale higher amounts of nicotine per cigarette smoked when compared to Whites, a marker for extraction of carcinogens (Trinidad et al., 2010). Finally, the toll of workplace exposure to carcinogens may play a role in so far as Black workers are disproportionally represented in the least safe occupations (Stellman & Stellman, 1996).
Our findings from Illinois that both Hispanics and Non-Hispanic Blacks were more likely to be diagnosed at a later stage in comparison to Whites confirm previous studies which controlled for socioeconomic factors and tumor histology types (Chen et al., 2015). Diagnosis at later stage is likely related to poorer access to primary healthcare and much higher rates of lack of health insurance, with Hispanics having the highest rate of uninsurance. Bach and colleagues described how Black lung cancer patients were highly concentrated among a small subgroup of non-board certified physicians, and were more often treated by physicians who themselves reported challenges in gaining access to high quality services for their patients (Bach et al., 2004).
Illinois disparities in surgical admission rates echo a 1999 study done by Bach et al. on treatment for early stage non-small cell lung cancer (Bach et al., 1999). These findings were replicated in a 2009 study by Farjah et al. which found 14% racial difference among patients who were all recommended to receive surgical therapy (Farjah et al., 2009), and in a 2015 study done by Chen et al. finding that both Hispanics and NH Blacks had lower odds for receiving treatment at earlier stages even after adjusting for socioeconomic factors and tumor histology (Chen et al., 2015). Findings from a study done by Soneji et al., found that Blacks and Hispanics with early stage lung cancer had lower surgical resection rates, and that Black patients who did receive early stage lung cancer treatment experienced worse overall survival than White patients (Soneji et al., 2017). Black patients may be less likely to consent to surgical therapy, reflecting a historic lack of trust in the US healthcare system (Corbie-Smith et al., 1999; Cykert & Phifer, 2003; Gordon et al., 2006; Margolis et al., 2003). Black patients may also have less access to hospitals and surgeons providing the highest quality cancer care (Bach et al., 2004; Shavers & Brown, 2002).
Our results also align with previous findings of racial and ethnic disparities in lung cancer screening. A survey conducted by Japuntich et al. found that of among patients meeting USPSTF criteria, non-Black patients were 2.8 times more likely to report having been screened, despite screening being covered by the Affordable Care Act (Japuntich et al., 2018). One barrier to screening is that former smokers may not believe they are susceptible to lung cancer (Delmerico et al., 2014). Rates of primary care physician referral for screening continue to remain low (Coughlin et al., 2014; O’Keefe et al., 2015).
Illinois has failed to close the lung cancer racial disparities gap and lags behind the rest of the country. Lung cancer disparities, like health status disparities in general, are rooted in the social determinants of health and will likely remain to the extent that politically patterned social, economic, and environmental inequality, based in the concentration of poor, highly segregated communities with concentrated poverty and high rates of smoking, continues to pervade American society.
11 Final Conclusion
These two studies on cancer health disparities provide a very introductory framework for more detailed statistical modeling that can drill down to the interactive complexity of cancer epidemiology. This is now being done in biomarker studies that describe the complex physiological effects that embody social experience (McDade & Harris, 2018). While social epidemiology research continues to document how conditions ‘outside the body’ get ‘under the skin’, the need to reduce the social conditions which produce population health disparities remains the foremost public health priority.
References
Albano, J., Ward, E., Jemal, A., Anderson, R., Cokkinides, V., Murray, T., Henley, J., Liff, J., & Thun, M. (2007). Cancer mortality in the United States by education level and race. Journal of the National Cancer Institute, 99(18), 1384–1394. https://doi.org/10.1093/JNCI/DJM127.
Bach, P. B., Pham, H. H., Schrag, D., Tate, R. C., & Hargraves, J. L. (2004). Primary care physicians who treat blacks and whites. New England Journal of Medicine 351(6), 575–584. https://doi.org/10.1056/NEJMSA040609.
Bach, P., Cramer, L., Warren, J., & Begg, C. (1999). Racial differences in the treatment of early-stage lung cancer. The New England Journal of Medicine, 341(16), 1198–1205. https://doi.org/10.1056/NEJM199910143411606.
Bailey, Z., Feldman, J., & Bassett, M. (2021). How structural racism works – racist policies as a root cause of U.S. racial health inequities. The New England Journal of Medicine, 384(8), 768–773. https://doi.org/10.1056/NEJMMS2025396.
Barbeau, E. M., Krieger, N., & Soobader, M.-J. (2004). Working class matters: Socioeconomic disadvantage, race/ethnicity, gender, and smoking in NHIS 2000. American Journal of Public Health, 94(2), 269. https://doi.org/10.2105/AJPH.94.2.269.
Byers, T., Wolf, H., Bauer, K., Bolick-Aldrich, S., Chen, V., Finch, J., Fulton, J., Schymura, M., Shen, T., Van Heest, S., & Yin, X. (2008). The impact of socioeconomic status on survival after cancer in the United States: findings from the National Program of Cancer Registries Patterns of Care Study. Cancer, 113(3), 582–591. https://doi.org/10.1002/CNCR.23567.
Chen, Y., Chang, A., Chen, C., & et al. (2015). 1036 Impact of race and ethnicity on stage at diagnosis of non-small cell lung cancer (NSCLC) and receipt of treatment in contemporary United States population: A SEER-based analysis. European Journal of Cancer, 51.
Clegg, L., Reichman, M., Miller, B., Hankey, B., Singh, G., Lin, Y., Goodman, M., Lynch, C., Schwartz, S., Chen, V., Bernstein, L., Gomez, S., Graff, J., Lin, C., Johnson, N., & Edwards, B. (2009). Impact of socioeconomic status on cancer incidence and stage at diagnosis: Selected findings from the surveillance, epidemiology, and end results: national longitudinal mortality study. Cancer Causes & Control: CCC, 20(4), 417–435. https://doi.org/10.1007/S10552-008-9256-0.
Corbie-Smith, G., Thomas, S. B., Williams, M. V., & Moody-Ayers, S. (1999). Attitudes and beliefs of African Americans toward participation in medical research. Journal of General Internal Medicine, 14(9), 537. https://doi.org/10.1046/J.1525-1497.1999.07048.X.
Coughlin, S., Matthews-Juarez, P., Juarez, P., Melton, C., & King, M. (2014). Opportunities to address lung cancer disparities among African Americans. Cancer Medicine, 3(6), 1467–1476. https://doi.org/10.1002/CAM4.348.
Cykert, S., & Phifer, N. (2003). Surgical decisions for early stage, non-small cell lung cancer: Which racially sensitive perceptions of cancer are likely to explain racial variation in surgery? Medical Decision Making: An International Journal of the Society for Medical Decision Making, 23(2), 167–176. https://doi.org/10.1177/0272989X03251244.
Delmerico, J., Hyland, A., Celestino, P., Reid, M., & Cummings, K. (2014). Patient willingness and barriers to receiving a CT scan for lung cancer screening. Lung Cancer (Amsterdam, Netherlands), 84(3), 307–309. https://doi.org/10.1016/J.LUNGCAN.2014.03.003.
DeSantis, C., Jemal, A., & Ward, E. (2010). Disparities in breast cancer prognostic factors by race, insurance status, and education. Cancer Causes & Control: CCC, 21(9), 1445–1450. https://doi.org/10.1007/S10552-010-9572-Z.
Deyo, R., Cherkin, D., & Ciol, M. (1992). Adapting a clinical comorbidity index for use with ICD-9-CM administrative databases. Journal of Clinical Epidemiology, 45(6), 613–619. https://doi.org/10.1016/0895-4356(92)90133-8.
Farjah, F., Wood, D., Yanez, N., Vaughan, T., Symons, R., Krishnadasan, B., & Flu, D. (2009). Racial disparities among patients with lung cancer who were recommended operative therapy. Archives of Surgery (Chicago, Ill.: 1960), 144(1), 14–18. https://doi.org/10.1001/ARCHSURG.2008.519.
Feinglass, J., Rydzewski, N., & Yang, A. (2015). The socioeconomic gradient in all-cause mortality for women with breast cancer: findings from the 1998 to 2006 National Cancer Data Base with follow-up through 2011. Annals of Epidemiology, 25(8), 549–555. https://doi.org/10.1016/J.ANNEPIDEM.2015.02.006.
Golecha, R., Eskender, M., Odell, D., & Feinglass, J. (2021). Racial and ethnic disparities in Illinois lung cancer incidence, mortality stage at diagnosis, surgical treatment, and screening. Journal of Health Disparities Research and Practice, 14(1). https://digitalscholarship.unlv.edu/jhdrp/vol14/iss1/2.
Gordon, H., Street, R., Sharf, B., Kelly, P., & Souchek, J. (2006). Racial differences in trust and lung cancer patients’ perceptions of physician communication. Journal of Clinical Oncology: Official Journal of the American Society of Clinical Oncology, 24(6), 904–909. https://doi.org/10.1200/JCO.2005.03.1955.
Gumpertz, M., Pickle, L., Miller, B., & Bell, B. (2006). Geographic patterns of advanced breast cancer in Los Angeles: Associations with biological and sociodemographic factors (United States). Cancer Causes & Control: CCC, 17(3), 325–339. https://doi.org/10.1007/S10552-005-0513-1.
Haiman, C., Stram, D., Wilkens, L., Pike, M., Kolonel, L., Henderson, B., & Le Marchand, L. (2006). Ethnic and racial differences in the smoking-related risk of lung cancer. The New England Journal of Medicine, 354(4), 333–342. https://doi.org/10.1056/NEJMOA033250.
Harris, K. M., Majmundar, M. K., & Becker, T. (Eds.) (2021). High and rising mortality rates among working-age adults. Washington DC: The National Academies Press https://doi.org/10.17226/25976.
Hayanga, A. J., Zeliadt, S. B., & Backhus, L. M. (2013). Residential segregation and lung cancer mortality in the United States. JAMA Surgery, 148(1), 37–42. https://doi.org/10.1001/JAMASURGERY.2013.408.
Hittner, E., & Adam, E. (2020). Emotional pathways to the biological embodiment of racial discrimination experiences. Psychosomatic Medicine, 82(4), 420–431. https://doi.org/10.1097/PSY.0000000000000792.
Howlader, N., Noone, A., & Krapcho, M. (2019). Cancer Statistics Review, 1975–2016 – SEER Statistics. https://seer.cancer.gov/archive/csr/1975_2016/.
Japuntich, S., Krieger, N., Salvas, A., & Carey, M. (2018). Racial disparities in lung cancer screening: An exploratory investigation. Journal of the National Medical Association, 110(5), 424–427. https://doi.org/10.1016/J.JNMA.2017.09.003.
Jones, N., Gilman, S., Cheng, T., Drury, S., Hill, C., & Geronimus, A. (2019). Life course approaches to the causes of health disparities. American Journal of Public Health, 109(S1), S48–S55. https://doi.org/10.2105/AJPH.2018.304738.
Kawachi, I., Daniels, N., & Robinson, D. (2005). Health disparities by race and class: Why both matter. Health Affairs (Project Hope), 24(2), 343–352. https://doi.org/10.1377/HLTHAFF.24.2.343.
Krieger, N. (2019). Theoretical frameworks and cancer inequities. In Vaccarella S., Lortet-Tieulent, J., Saracci, R., Conway, D. I., Straif, K., & Wild, C. P. (Eds.), Reducing social inequalities in cancer: evidence and priorities for research (pp. 111–120). Lyon: International Agency for Research on Cancer.
Krieger, N., Chen, J., Kosheleva, A., & Waterman, P. (2012). Shrinking, widening, reversing, and stagnating trends in US socioeconomic inequities in cancer mortality for the total, black, and white populations: 1960–2006. Cancer Causes & Control: CCC, 23(2), 297–319. https://doi.org/10.1007/S10552-011-9879-4.
Krieger, N. (2013). History, Biology, and Health Inequities: Emergent Embodied Phenotypes and the Illustrative Case of the Breast Cancer Estrogen Receptor. American Journal of Public Health, 103(1), 22–27. https://doi.org/10.2105/AJPH.2012.300967.
Krieger, N., Chen, J. T., & Waterman, P. D. (2010). Decline in US breast cancer rates after the Women’s Health Initiative: Socioeconomic and racial/ethnic differentials. American Journal of Public Health, 100(Suppl 1), S132. https://doi.org/10.2105/AJPH.2009.181628.
Krieger, N., Chen, J. T., & Waterman, P. D. (2011). Temporal trends in the black/white breast cancer case ratio for estrogen receptor status: disparities are historically contingent, not innate. Cancer Causes & Control: CCC, 22(3), 511. https://doi.org/10.1007/S10552-010-9710-7.
Li, Z., Luo, L., Hu, Y., Chen, H., Den, Y., Tang, L., Liu, B., Liu, D., & Zhang, X. (2016). Lung cancer screening: a systematic review of clinical practice guidelines. International Journal of Clinical Practice, 70(1), 20–30. https://doi.org/10.1111/IJCP.12744.
Margolis, M., Christie, J., Silvestri, G., Kaiser, L., Santiago, S., & Hansen-Flaschen, J. (2003). Racial differences pertaining to a belief about lung cancer surgery: results of a multicenter survey. Annals of Internal Medicine, 139(7). https://doi.org/10.7326/0003-4819-139-7-200310070-00007.
McCartney, G., Bartley, M., Dundas, R., Katikireddi, S. V., Mitchell, R., Popham, F., Walsh, D., & Wami, W. (2019). Theorising social class and its application to the study of health inequalities. SSM – Population Health, 7, 100315. https://doi.org/10.1016/J.SSMPH.2018.10.015.
McDade, T. W., & Harris, K. M. (2018). The biosocial approach to human development, Behavior, and health across the life course. The Russell Sage Foundation Journal of the Social Sciences: RSF, 4(4), 2. https://doi.org/10.7758/RSF.2018.4.4.01.
Miller, K., Goding Sauer, A., Ortiz, A. P., Fedewa, S. A., Pinheiro, P. S., Tortolero-Luna, G., Martinez-Tyson, D., Jemal, A., & Siegel, R. L. (2018). Cancer Statistics for Hispanics/Latinos, 2018. CA: A Cancer Journal for Clinicians, 68(6), 425–445. https://doi.org/10.3322/CAAC.21494.
Mulligan, C., Meram, A., Proctor, C., Wu, H., Zhu, K., & Marrogi, A. (2006). Unlimited access to care: effect on racial disparity and prognostic factors in lung cancer. Cancer Epidemiology, Biomarkers & Prevention: A Publication of the American Association for Cancer Research, Cosponsored by the American Society of Preventive Oncology, 15(1), 25–31. https://doi.org/10.1158/1055-9965.EPI-05-0537.
National Cancer Institute. (2014). Cancer statistics review, 1975–2011 – previous version – SEER Cancer statistics review. (n.d.). Retrieved August 25, 2021, from https://seer.cancer.gov/archive/csr/1975_2011/.
O’Keefe, E. B., Meltzer, J. P., & Bethea, T. N. (2015). Health disparities and cancer: Racial disparities in cancer mortality in the United States, 2000–2010. Frontiers in Public Health, 3(51): 1–15. https://doi.org/10.3389/FPUBH.2015.00051.
Richards, T. B. (2020). Screening for lung cancer – 10 states, 2017. MMWR. Morbidity and Mortality Weekly Report, 69(8), 201–206. https://doi.org/10.15585/MMWR.MM6908A1.
Robert, S., Strombom, I., Trentham-Dietz, A., Hampton, J., McElroy, J., Newcomb, P., & Remington, P. (2004). Socioeconomic risk factors for breast cancer: Distinguishing individual- and community-level effects. Epidemiology (Cambridge, Mass.), 15(4), 442–450. https://doi.org/10.1097/01.EDE.0000129512.61698.03.
Ryan, B. (2018). Lung cancer health disparities. Carcinogenesis, 39(6), 741–751. https://doi.org/10.1093/CARCIN/BGY047.
Surveillance, Epidemiology, and End Results (SEER) Program. SEER*Stat Database: Incidence-Based Mortality – SEER 18 Regs (Excl Louisiana) Research Data, Nov 2018 Sub (2000–2016) <Katrina/Rita Population Adjustment> – Linked To County Attributes – Total U.S., 1969–2017 Counties. National Cancer Institute, DCCPS, Su. (n.d.). Retrieved August 25, 2021, from https://seer.cancer.gov/data/hurricane.html.
Shavers, V., & Brown, M. (2002). Racial and ethnic disparities in the receipt of cancer treatment. Journal of the National Cancer Institute, 94(5), 334–357. https://doi.org/10.1093/JNCI/94.5.334.
Siegel, R. L., Miller, K. D., Jemal, A. (2020). Cancer statistics, 2020. CA: A Cancer Journal for Clinicians, 70(1), 7–30. https://doi.org/10.3322/CAAC.21590.
Soneji, S., Beltrán-Sánchez, H., & Sox, H. (2014). Assessing progress in reducing the burden of cancer mortality, 1985–2005. Journal of Clinical Oncology: Official Journal of the American Society of Clinical Oncology, 32(5), 444–448. https://doi.org/10.1200/JCO.2013.50.8952.
Soneji, S., Tanner, N., Silvestri, G., Lathan, C., & Black, W. (2017). Racial and ethnic disparities in early-stage lung cancer survival. Chest, 152(3), 587–597. https://doi.org/10.1016/J.CHEST.2017.03.059.
Stellman, J., & Stellman, S. (1996). Cancer and the workplace. CA: A Cancer Journal for Clinicians, 46(2), 1984–1988. https://doi.org/10.3322/CANJCLIN.46.2.70.
Trinidad, D. R., Pérez-Stable, E. J., Messer, K., White, M. M., & Pierce, J. P. (2010). Menthol cigarettes and smoking cessation among racial/ethnic groups in the United States. Addiction (Abingdon, England), 105(0 1), 84. https://doi.org/10.1111/J.1360-0443.2010.03187.X.
Underwood, J., Townsend, J., Tai, E., Davis, S., Stewart, S., White, A., Momin, B., & Fairley, T. (2012). Racial and regional disparities in lung cancer incidence. Cancer, 118(7), 1910–1918. https://doi.org/10.1002/CNCR.26479.
Velasco-Mondragon, E., Jimenez, A., Palladino-Davis, A., Davis, D., & Escamilla-Cejudo, J. (2016). Hispanic health in the USA: A scoping review of the literature. Public Health Reviews, 37(1). https://doi.org/10.1186/S40985-016-0043-2.
Zimmerman, F. J., & Anderson, N. W. (2019). Trends in health equity in the United States by race/ethnicity, sex, and income, 1993–2017. JAMA Network Open, 2(6), e196386–e196386. https://doi.org/10.1001/JAMANETWORKOPEN.2019.6386.